CROJFE - Volume 30, Issue 2 by Studio Bernardic

Orginal scientific papers – Izvorni znanstveni radovi RAFFAELE SPINELLI, CARLA NATI A Low-Investment Fully Mechanized Operation for the Pure Selection Thinning of Pine Plantations ............ 89 Snižavanje ulaganja za potpuno mehanizirano pridobivanje drvnoga iverja u selektivnim proredama borovih plantaža SAŠA BOGDAN, MARIO ŠPOR^I], ANTE SELETKOVI], MLADEN IVANKOVI] Biomass Production of the Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations and Mixed Plantations with Willow Clones (Salix sp.) in Croatia ............................ 99 Proizvodnja biomase crne johe (Alnus glutinosa /L./ Gaertn.) u ~istim kulturama i mješovitim kulturama s klonovima vrbe (Salix sp.) u Hrvatskoj

Issue 2

2009

MATEVŽ MIHELI^, JANEZ KR^ Analysis of the Inclusion of Wood Forwarding Into a Skidding Model ............................... 113 Analiza uklju~ivanja izvoženja drva forvarderom i traktorskom ekipažom u model privla~enja drva ŽELJKO TOMAŠI], MARIJAN ŠUŠNJAR, DUBRAVKO HORVAT, ZDRAVKO PANDUR Forces Affecting Timber Skidding ........................................................ 127 Utjecajne sile pri privla~enju drva RAMIN NAGHDI, MAJID LOTFALIAN, IRAJ BAGHERI, AGHIL MORADMAND JALALI Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration ..................... 141 Štete na tlu i pomlatku pri privla~enju drva skiderima i animalnom vu~om SELCUK GUMUS, BURAK ARICAK, KORHAN ENEZ, H. HULUSI ACAR Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works ....................... 151 Analiza ošte~enja stabala uzrokovanih odronima prilikom izgradnje šumske ceste YURI GERASIMOV, ANTON SOKOLOV Ergonomic Characterization of Harvesting Work in Karelia ...................................... 159 Ergonomsko ozna~ivanje radova na pridobivanju drva u Kareliji IGOR POTO^NIK, TIBOR PENTEK, ANTON POJE Severity Analysis of Accidents in Forest Operations ............................................ 171 Analiza težine nesre~a pri šumskim radovima

Preliminary notes – Prethodno priop}enje JOZEF SUCHOMEL, KATARÍNA BELANOVÁ Influence of Selected Meteorological Phenomena on Work Injury Frequency in Timber Harvesting Process ........................................................... 185 Utjecaj vremenskih prilika na u~estalost ozlje|ivanja u postupcima pridobivanja drva

ISSN 1845-5719

9 771845 571000

Original scientific paper – Izvorni znanstveni rad

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of Pine Plantations Raffaele Spinelli, Carla Nati Abstract – Nacrtak Over the last decades many Mediterranean sites were planted with conifers, which offered fast growth, good fibre quality and the capacity to grow on poor soils. If not thinned, these stands cannot develop strong and healthy trees and they become vulnerable to stress agents. The authors designed a complete operation for the thinning of conifer plantations under favourable terrain conditions by a feller-buncher, a farm tractor equipped with a skidding grapple, mounted on the three-point hitch and a trailer-mounted drum chipper. The whole investment in mechanical equipment was very limited, in the order of 338,000, including the truck and a tractor to move the chipper. Felling and extraction were quite balanced, with an average productivity around 40 trees hour–1. A spreadsheet was developed in order to determine the effect of tract size on biomass delivered cost, a crucial issue in Non Industrial Private Forests (NIPF) where the problems inherent to the generally small tract size are compounded by the limited quantity and value of the harvest. Keywords: feller-buncher, Non-Industrial Private Forestry, biomass, whole-tree harvesting, thinning

1. Introduction – Uvod Starting from the Second World War, an intensive activity of reforestation has been conducted all across Europe, with a number of different goals, including soil protection, the promotion of employment in rural areas and the establishment of a strategic reserve of a precious raw material. Most sites were planted with conifers, which offered fast growth, good fiber quality and the capacity to grow on poor soils. Species differed according to the climatic conditions of the sites: while Norway spruce and Scots Pine were most popular in Central Europe, Austrian Pine (P. nigra Arnorld) and Mediterranean pines (P. pinaster Aiton, P. pinea L., P. halepensis Mill; etc.) were prevalent further south, due to their better adaptation to dry summer conditions. Most efforts were successful, and plantation continued well into the early 1980s. However, the sound development of these stands depended on a timely thinning schedule. If not thinned, these stands cannot develop strong and healthy trees and become vulnerable to stress agents, such as drought and parasites. Thinning repCroat. j. for. eng. 30(2009)2

resents a problematic issue, due to its very low economic sustainability: on one hand, the operation generally produces a harvest of little commercial value, often suitable for energy production only; on the other hand, the low harvest intensity and small tree size contribute to reduce operational efficiency, thus raising production costs (Kärhä et al. 2003). Recently, the growing demand for energy biomass is offering an opportunity to thinning operations, which may accrue the double benefit of improving stand stability and of mobilizing significant amounts of biomass. In turn, this may help preventing the growing competition between new forest energy users and the traditional consumers of wood fiber, which starts to be felt across Europe (Lundmark 2006). Thinning operations lend themselves to the application of whole-tree harvesting (WTH), with significant gains in terms of increased productivity and reduced harvesting cost. Whole-tree harvesting involves the extraction of whole trees and their eventual processing at the landing, offering the advantage of simplified in-forest handling. First documented in the US (Kammenga 1983), the application of

R. Spinelli and C. Nati

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

WTH to thinning operations is often associated to whole-tree chipping, and its basic set-up has proven so effective to remain virtually unchanged and appreciated until our days (Mitchell and Gallagher 2007). However, standard WTH operations resort to heavy and expensive machinery, and can only be deployed under the conditions of a heavy geometric thinning, and of a significant capital commitment. This may pose a problem to small enterprises that lack the investment capacity required for purchasing modern forest machinery. The alternative – practiced by many – is to conduct a motor-manual operation, where trees are felled by chainsaw and dragged to the roadside with small tractors and winches. Much work has been devoted to developing cost-effective motor-manual thinning operation, but the ever increasing labor cost makes it a very hard challenge. Besides, mechanization is certainly preferable in terms of operator comfort and work safety (Bell 2002), which play a major role in the current technology shift. On the other hand, there is no need for heavy machines when handling small trees harvested in a thinning operation, so that dedicated thinning machinery could be downsized and simplified. Ideally, one may use a widely available general purpose prime mover to further reduce investment cost. Today, the market offers some interesting felling implements for application to small skid-steer loaders in the 4-tonnes class (Windell and Bradshaw 2000), and adapted skid-steer loaders have shown some potential in the thinning of North-American forests (Becker et al. 2006). The goal of the study was to develop and test a WTH system with the following specifications: 1) ability to carry out a pure selection thinning, without opening any strip roads; 2) complete mechanization of all work steps, with no personnel on the ground

(for increased operator safety and comfort); 3) moderate capital cost, obtained by widespread use of lightweight general-purpose equipment.

2. Materials – Materijal The authors designed a complete operation for the thinning of conifer plantations under favorable terrain conditions. A feller-buncher was obtained by applying a disc-saw with accumulating arms to the universal implement hitch of a 58 kW, 4-tonne tracked skid-steer loader. A 44 kW farm tractor was equipped with a skidding grapple, mounted on the three-point hitch. Both machines were not wider than 2 m, so that they could sneak into the standard 3 m inter row of most plantations. The third element in the operation was a trailer-mounted drum chipper, powered by a 162 kW independent engine and equipped with its own integral loader (Table 1). The operating principle was very simple: the feller-buncher reversed along every other inter row picking selected trees on both sides and laying 3–5 tree bunches in the middle of the inter row. The skidder collected the bunches and dragged them to the roadside landing, in front of the chipper. When enough trees were available, the chipper was started and whole-tree chips were blown directly into transportation vehicles parked alongside. The whole investment in mechanical equipment was very limited, in the order of 338.000, including the truck and a tractor to move the chipper. The trial was carried out in a pine plantation in the Province of Campobasso (Central Italy). The stand had been established with Pinus halepensis Mill. 25 years earlier and had never been thinned before. The terrain was relatively even, with a moderate slope gradient. The silvicultural prescription consisted in

Table 1 Machinery used for the thinning trial Tablica 1. Kori{teni strojevi pri proredi Process – Postupak Prime mover – Osnovni stroj Brand – Proizvo|a~ Model – Model Engine power – Snaga motora, kW Width – [irina, m Weight – Masa, kg Implement – Radno oru|e Brand – Proizvo|a~ Model – Model Weight – Masa, kg

Felling – Sje~a Skid-steer loader – Mali utovariva~ Bobcat T-250 Hi-Flow 60 2.03 4240 Disc saw – Kru`na pila Davco QC 1400 650

Extraction – Privla~enje Farm tractor – Poljoprivredni traktor Valpadana 6064 44 1.05 1280 Log grappe – Hvatalo Japa SG 235

Chipping – Iveranje Drum clipper – Bubnjasti ivera~ Pezzolato PTH 700 162 2.40 6700 Loader – Dizalica Dalla Bona AS 510 1200

Croat. j. for. eng. 30(2009)2

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

Table 2 Site description Tablica 2. Opis sastojine

3. Methods – Metode

Place name – Mjesto

Liscione

Municipality – Lokalna samouprava

Guardialfiera

Province – Pokrajina

Surface area – Povr{ina, ha

1.03

Altitude – Nadmorska visina, m

310

Species – Vrsta

Pinus halepensis Mill.

Age, years – Dob, godine

Treatment – Uzgojni postupak

Thinning – Proreda

–1

Removal – Sje}a

R. Spinelli and C. Nati

Trees – Stabala, n ha

463 –1

Biomass – Biomasa, odt ha

27.5

DBH – Prsni promjer, cm

14.9

Height – Visina, m

11.8

–1

0.060

Residual, Trees – Preostala stabla, n ha–1

1207

Mass, odt tree – Masa, odt/stablo Slope gradient – Nagib terena, %

Terrain roughness, class Povr{inske prepreke, razred

Moisture content of whole-tree chips*, % Mokrina iverja izra|enoga iz stabla*, %

47.4

odt – oven-dry tonne – masa suhe tvari, t * determined according to the European Standard CEN/TS 14774-2 * utvr|eno prema Europskoj normi CEN/TS 14774-2

a selection thinning, with the purpose of removing 28% of the trees, chosen among weak, dead and malformed trees. Site characteristics are described in Table 2. The operation produced 27.5 oven-dry tonnes (odt) of whole-tree biomass per hectare.

All trees to be removed were marked and identified with a numerical code painted on their bark with fluorescent paint. The diameter at breast height (DBH) of all such trees was measured and noted together with the identification code, so that its value could be associated to processing time without interfering with the work process. The authors carried out a time-motion study, designed to evaluate machine productivity (Bergstrand 1991). The study was conducted separately for all the main work steps: felling, extraction and chipping. Each processing cycle was stop watched individually, using Husky Hunter hand-held field computers running the dedicated Siwork3 time study software (Kofman 1995). A cycle was defined as the time to process a single tree, a single tractor trip and a single chip load, respectively, for felling-bunching, extraction and chipping. Extraction distances were measured with a cotton-thread hip chain, for each tractor trip. Productive time was separated from delay time (Bjorheden et al. 1995). In order to translate DBH records into mass values, the Authors used the double-entry tariff tables developed by Castellani et al. (1982) for P. halepensis, after building a DBH-height curve with the heights of 30 sample trees. Volume figures were converted into dry weight by applying a base density value of 510 kg/m3 (Giordano 1988). Moisture content was determined according to the European Standard CEN/TS 14774-2, on 10 chip samples. Machine costs were calculated with the method described by Miyata (1980), on the assumptions shown in Table 3. The calculated operational cost per Scheduled Machine Hour (SMH) was increased by 20% in order to include administration costs.

Table 3 Machine costing: assumptions and final result Tablica 3. Izra~un tro{kova strojnoga rada – ulazni parametri i rezultati Machine – Stroj Feller-buncher Sje~no vozilo Tractor with grappe Traktor s hvatalom Mobile chipper* Mobilni ivera~* Truck Kamion

Investment, Nabavna vrijednost,

Service life, years Uporabno razdoblje, god.

Utilization, SMH year–1 Iskori{tenost, sati/god.

Crew Broj radnika

Hourly cost, SMH–1 Tro{ak, /sat

58.000

800

40.000

800

140.000

800

100.000

1800

* includes a 65 kW farm tractor for towing the clipper – uklju~uje poljoprivredni traktor snage 65 kW za premje{tanje ivera~a

Croat. j. for. eng. 30(2009)2

R. Spinelli and C. Nati

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

Regression analysis of time-study data allowed developing a set of equations capable of predicting cycle time (and therefore productivity) as a function of statistically significant independent variables.

4. Results of experiments – Rezultati istra`ivanja The Whole-tree harvesting method allows recovering a significant amount of biomass even from the thinning in a young stand, and that is especially important with P. halepensis, generally characterized by poor form and therefore only capable of yielding a relatively small amount of conventional low quality roundwood products. The graph in Fig. 1 shows the incidence of branch material (diameter <3 cm) over the total tree mass, as a function of DBH. The trial demonstrated that the skid-steer feller-buncher can maneuver in the 3 m wide inter rows, cutting removal trees and forming bunches of 4–6 trees, thus making extraction much easier. Occasional problems were caused by the very strong base sweep (pistol-butt) of some trees, which made it difficult to grab and cut the stem at the same time. In fact, the saw and the grab-arms are aligned on the same axis, and any significant deviation of the tree stem from the rectilinear form complicates the operation. In this case the best solution was to cut the stem above the base sweep and then finish the work with a second cut to cut the stump to the ground level. If the skid-steer is not fitted with appropriate forestry guarding around the cab and the hydraulic hoses, it is advisable to prune the trees along the working side of the row, in order to avoid damage. Extraction requires a tractor, able to drag a load of about 500–600 kg. Ideally, this machine should be

Fig. 1 Incidence of the branch portion (diameter <3 cm) on the total tree mass, weighted with portable scales Slika 1. Udjel granjevine (promjera < 3 cm) u ukupnoj masi stabla, mjereno prijenosnom vagom heavy and powerful enough to tow a small trailer for relocating the feller-buncher. This way one would obtain a self-contained operation, capable of quick moving between sites and therefore specifically suited to harvesting the small-size tracts characterizing much of the non-industrial private forestry (NIPF) so abundant across Europe. A special warning must be given about the fire-hazard represented by the operation of high-speed disc saws during summer. The contact of the fast rotating disc with stones can generate sparks, which

Table 4 Recorded productivity data Tablica 4. Ostvarena proizvodnost Process – Postupak Prime mover – Osnovni stroj Brand – Proizvo|a~ Model – Model Tree DBH – Prsni promjer stabla, cm Tree mass – Masa stabla, odt Extraction distance – Udaljenost privla~enja, m Delays – Prekidi, % Productivity, trees SMH–1 – Proizvodnost, stabala/sat Productivity, odt SMH–1 – Proizvodnost, odt/sat Preparation, hours day–1 – Priprema, sat/dan

Felling – Sje~a Skid-steer loader Mali utovariva~ Bobcat T-250 Hi-Flow 14.9 0.060 – 20 45.7 2.7 0.75

Extraction – Privla~enje Farm tractor Poljoprivredni traktor Valpadana 6064 14.9 0.060 300 20 36.4 2.2 0.50

Chipping – Iveranje Drum clipper Bubnjasti ivera~ Pezzolato PTH 700 14.9 0.060 – 35 66.8 4.0 1.00

Croat. j. for. eng. 30(2009)2

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

may start a fire under the appropriate soil, fuel and weather conditions. Since we are proposing this system for Mediterranean stands, it is important that readers realize the risks involved when the system is used during the dry summers so common in this region. For safeguard, disc saw operations should be avoided during the days when the fire risk is highest, and fire-extinguishers should always be available on the machine and at the landing. This is also needed when operating a chipper during the hot summer days, since chippers tend to develop a lot of flammable dust and engine heat, and are particularly prone to catch fire. Trial results are reported in Table 4, which shows the gross productivity recorded for each process stage: these figures are calculated on the overall time consumption, including delays but excluding preparation and relocation time. It must be noted that felling and extraction are quite balanced, with an average productivity around 40 trees hour–1. On the other hand, chipping proceeds at a much higher rate, almost twice as high. Therefore, one can choose three different operating modes: 1) stop and go, where the chipper sits idle at the landing and is turned on only when the other two machines have accumulated enough wood to fill a truck load; in the meantime the chipper operator can perform other jobs, such as removing big branches in the alternate inter rows trafficked by the feller-buncher. This operating mode is not the most effective, but is the simplest and most expedient; 2) dispatching two feller-bunchers and two tractors to serve one chipper, which however requires good organizational capacity and a reasonably large woodlot size; 3) organizing chipping as a separate operation, which is moved to a landing only after the trees have been felled, extracted and

R. Spinelli and C. Nati

piled, possibly allowing for partial drying of the biomass before chipping. In this case it is necessary to equip the felling and extraction operation with a loader for stacking the extracted trees. This could also be done by fitting the farm tractor with a front-end loading fork and/or by carrying a quick-couple fork for the skid-steer, so that at regular intervals the machine can drop the disc saw, attach the fork and stack the trees being skidded.

5. Simulation with a spreadsheet model Simulacija modela tabli~nim kalkulatorom The different options mentioned above have not been all tested directly, but they have been simulated after building a simple spreadsheet model with the productivity data recorded during the study. These data were statistically analyzed to estimate any relation existing between time consumption and working conditions (SAS 1999), and the resulting equations were organized in a spreadsheet model capable of returning a specific harvesting cost figure for the specific operational and economic assumptions made by each user. The model includes an estimate of relocation time and cost (Table 5), which is spread over the total amount of product obtained from each tract. The model can simulate options 1 (stop and go chipping) and 3 (chipping as a separate operation), on the assumption that option 2 is not compatible with both the firm organization and the tract size most common in NIPF. Then a comparative simulation was run under the following assumptions: tract size of 4 ha, DBH of removal tree of 15 cm, extraction and transportation

Table 5 Relocation time and cost Tablica 5. Vrijeme premje{tanja i tro{kovi

Tractor and grapple with feller-buncher Traktor s hvatalom i sje~no vozilo

Chipper towed by farm tractor Ivera~ vu~en poljoprivrednim traktorom

Croat. j. for. eng. 30(2009)2

Travel speed – Brzina kretanja, km h–1 Relocation distance – Udaljenost premje{tanja, km Load and unload, hour – Utovar i istovar, sati Travel time, hour – Vrijeme kretanja, sati Working cost, hour–1 – Tro{ak rada, /sat Relocation cost – Tro{ak premje{tanja, Travel speed – Brzina kretanja, km h–1 Relocation distance – Udaljenost premje{tanja, km Load and unload, hour – Utovar i istovar, sati Travel time, hour – Vrijeme kretanja, sati Working cost, hour–1 – Tro{ak rada, /sat Relocation cost – Tro{ak premje{tanja,

35 30 1.5 0.9 95 223 35 30 1.5 0.9 133 312

R. Spinelli and C. Nati

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

distance of 300 m and 24 km, respectively. Delivered price of the biomass was assumed to reach 85 odt–1. Simulation results are shown in Table 6, showing a thinning cost of approximately 400 ha–1. The different options appear similar to each other, with the »stop and go« method costing only 3% more than the other two. This explains why so many enterprises adopt this method, as already pointed out in other studies (Spinelli and Hartsough 2001). If the chipper operator can be employed in other activities, a discontinuous chipping pattern will only affect the depreciation cost of the equipment, with a limited impact on the whole production cost, provided that the initial investment paid for the machine is not too high, as in the case of the medium-size trailer-mounted chipper used in the study. A further advantage of this work pattern system can be the relative independence of chipping and transport. Of course, this simulation suffers from the typical limit of all deterministic models dealing with interactive work, and may not fully reflect the inefficiencies of a system where all links are interdependent. Nevertheless, it offers a first attempt to estimating system performance, and its findings seem to be corroborated by the observation of common practice. Although common practice is not necessarily the benchmark for the best possible use of resources, it still provides a reasonably good guarantee for an acceptably efficient work method. Table 5 allows determining the effect of tract size on biomass delivered cost. This is a crucial issue in NIPF where the problems inherent to the generally small tract size are compounded by the limited quan-

Fig. 2 Delivered cost as a function of woodlot size Slika 2. Tro{kovi proizvodnje ovisno o veli~ini {umoposjeda tity and value of the harvest (Kittredge et al. 1996). The operation designed by the authors is very agile, but even its minimal relocation cost has a considerable impact on total harvesting cost, due to the very low removal levels. The graph in Fig. 2 shows a strong effect of tract size on overall delivered cost up for lot surface areas below 4 ha: on larger tracts, harvesting cost is much less dependent on harvest surface, indicating a better amortization of relocation cost.

Table 6 Comparison of two different operating modes Tablica 6. Usporedba sustava rada Input – Ulaz Relocation distance – Udaljenost premje{tanja, km Tract size – Veli~ina posjeda, ha Removal, trees ha–1 – Sje~a, stabala/ha Avg. DBH Removal – Pros. prsni promjer, cm Avg. Mass Removal – Pros. masa stabla, odt Extraction distance – Udaljenost privla~enja, m Distance, forest road – Udaljenost, {umska cesta, km Distance, mountain road – Udaljenost, lokalna cesta, km Distance, state road – Udaljenost, dr`avna cesta, km Total harvest – Ukupno posje~eno, odt Total harvest, trees – Ukupno posje~eno, stabala Chip price, delivered, odt–1 Cijena isporu~enoga iverja, odt–1

30 4 400 15 0.061 300 0 0 24 97 1600

Results – Rezultati Felling Sje~a

hours – sati

Extraction Privla~enje

hours – sati

Chipping Iveranje

hours – sati

Transport Prijevoz

hours – sati

Relocation – Premje{tanje Total cost – Ukupni tro{ak

–1

odt

–1

Profit – Dobit

odt ha

–1

Stand-by chipper Iveranje u sustavu

Independent chipping Neovisno iveranje

39 2122 47 1957 28 3373 38 2054 536 104 –19 –451

49 2713 47 1957 28 2527 38 2054 536 101 –16 –387

Croat. j. for. eng. 30(2009)2

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

R. Spinelli and C. Nati

6. Discussion and Conclusion – Rasprava i zaklju~ci

Fig. 3 Financial result as a function of removal tree DBH and the price of dry chips (85 or 100 per oven- dry ton) Slika 3. Poslovni u~inak kao funkcija prsnoga promjera oborenih stabala prema cijeni drvnoga iverja od 85 /odt i i 100 /odt On a similar note, one can explore the effect of tree size on harvesting cost. A simulation was run under the same assumptions shown in Table 6, except for removal tree DBH, which ranged between 10 and 22 cm. Two different levels were set for chip price: the current 85 odt–1 and the projected 100 odt–1. The graph in Fig. 3 shows that under current chip prices, the thinning operations start to accrue some profits only when DBH is bigger than 19 cm. However, the removal of such large trees may not be suited to a first thinning, but rather to a second thinning. If chip prices will actually reach the projected 100 odt–1, then thinning breaks even for a DBH of 15 cm, which is much more compatible with the needs of a first selection thinning. Transport distance also affects the total harvesting and delivery cost, which increases by 20% when the distance increases from 20 to 60 km. Using greater trucks-and-trailer units with a larger payload, instead of simple trucks, is crucial to the reduction of transportation cost. Organizing a short-distance supply chain is always a good way to contain delivered cost, but the main cost component still remains harvesting and processing and therefore the primary focus should be on targeting those stands that offer favorable harvesting conditions (i.e. flat terrain, large tract and stem size), even if farther away. Croat. j. for. eng. 30(2009)2

The operation designed and tested by the authors offers a low investment option for the fully mechanized, pure selection thinning of conifer plantations in flat terrain. Such operation is also characterized by high mobility, which represents a strategic requirement imposed by the fragmentation of NIPFs. Whole-tree harvesting allows increasing the amount of merchantable product obtained from the thinning, while decreasing production cost. It also results in the almost complete removal of slash, which is a cost-effective way to reduce fire risk (Hartsough et al. 2008), which is particularly high in conifer plantations and in the Mediterranean region (Lovreglio et al. 1999). Since the industrial chipper is almost twice as productive as the other units, and the typically small tract size discourages doubling up the felling and extraction teams, chipping must either be separated from the other operations and occur at a later stage, or the chipper must work in a stand-by mode. Both options were simulated through a simple spreadsheet model, showing that operation in a stand-by mode is not much more expensive than the apparently more rational two-stage operation. In any case, thinning cost is relatively low, and the operation can break even if the projected 100 odt–1 price target for forest chips is eventually reached. Such a result would be remarkable, considering the extremely challenging specifications of low investment cost, full mechanization of the operation and capacity to perform pure selection thinning. It should be stressed at this stage that this harvesting method is only suitable to flat or gently sloping terrain, and it is recommended for conifer plantations – especially Mediterranean pines planted along the coast or on low hills. The full mechanization achieved with this operation can be found attractive by local firms that operate in rural areas, since it can offer a relatively comfortable work place and a qualified employment to young loggers, while requiring reasonably low capital commitment.

7. References – Literatura Becker, P., Jensen, J., Meinert, D., 2006: Conventional and Mechanized Logging Compared for Ozark Hardwood Forest Thinning: Productivity, Economics, and Environmental Impact. Northern Journal of Applied Forestry 23(4): 264–272.

R. Spinelli and C. Nati

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

Bell, J., 2002: Changes in logging injury rates associated with the use of feller-bunchers in West Virginia. Journal of Safety Research 33(4): 463–471. Bergstrand, K. G., 1991: Planning and analysis of forestry operation studies. Skogsarbeten Bulletin 17: 1–63. Björheden, R., Apel, K., Shiba, M., Thompson, M. A., 1995: IUFRO Forest work study nomenclature. Swedish University of Agricultural Science, Dept. of Operational Efficiency, Garpenberg, 16 p. Castellani, C., Ghidini, G., Tosi, V., 1982: Tavole dendrometriche ed alsometrica del pino d’Aleppo (Pinus halepensis Mill.) valevoli in Italia (Tree growth rate tables of Pinus halepensis valid for Italy). Annali dell’Istituto Sperimentale per l’Assestamento Forestale e per l’Alpicoltura, Vol. VIII: 5–44. Giordano, G., 1988: Tecnologia del Legno (Wood technology). Vol. III: 896–897. UTET, Torino.

Kittredge, D., Mauri, M., McGuire, E., 1996: Decreasing woodlot size and the future of timber sales in Massachusetts: when is an operation too small? Northern Journal of Applied Forestry 13(2): 96–101. Kofman, P., 1995: Siwork 3: User Guide. Danish Forest and Landscape Research Institute, Vejle, Denmark, 37 pp. Lovreglio, R., Fidanza, F., Leone, V., 1999: Un modello per la stima della sopravvivenza post-incendio in Pinus halepensis Mill (A model for estimating survival of Pinus halepensis after fire). L’Italia Forestale e Montana 4: 178–190. Lundmark, R., 2006: Cost structure and competition for forest-based biomass. Scandinavian Journal of Forest Research 21(3): 271–280. Miyata, E. S., 1980: Determining fixed and operating costs of logging equipment. General Technical Report NC-55. Forest Service North Central Forest Experiment Station, St. Paul, MN, 14 pp.

Hartsough, B., Abrams, S., Barbour, J., Drews, E., McIver, J., Moghaddas, J., Schwilk, D., Stephens, S., 2008: The economics of alternative fuel reduction treatments in western United States dry forests: financial and policy implications from the National Fire and Fire Surrogate Study. Forest policy and Economics 10(6): 344–354.

Mitchell, D., Gallagher, T., 2007: Chipping Whole Trees for Fuel Chips: A Production Study. Southern Journal of Applied Forestry 31(4): 176–180.

Kammenga, J., 1983: Whole-tree utilization system for thinning young Douglas-fir. Journal of Forestry 81(4): 220–224.

Spinelli, R., Hartsough, B., 2001: Indagine sulla cippatura in Italia (A chipping survey in Italy). CNR–IRL Contributi Scientifico-Pratici XLI, Firenze, 112 pp.

Kärhä, K., Jouhiaho, A., Mutikainen, A., Mattila, S., 2003: Mechanized energy wood harvesting from early thinning. International Journal of Forest Engineering 16(1): 23–36.

Windell, K., Bradshaw, S., 2000: Under story biomass reduction methods and equipment catalogue. USDA Forest Service. TR-0051-2826 MTD C. Missoula, MT. 156 p.

SAS Institute Inc. 1999. StatView Reference. SAS Publishing, Cary, NC. ISBN-1-58025-162-5, pp. 84–93.

Sa`etak

Sniàvanje ulaganja za potpuno mehanizirano pridobivanje drvnoga iverja u selektivnim proredama borovih plantaà Pridobivanje drva iz proreda s jedne strane ima vrlo nisku ekonomsku odrìvost zbog male komercijalne vrijednosti, naj~e{}e energijskoga drva ili iverja, dok s druge strane niska sje~na gusto}a i prosje~na veli~ina stabala pridonose smanjenju radne u~inkovitosti i pove}anju tro{kova proizvodnje. Zbog pove}ane potra`nje za energijskim drvom u proredama se sve vi{e koristi stablovna metoda sje~e uz pomo} koje se smanjuju tro{kovi rada i pove}ava se proizvodnost. Zbog visokih tro{kova strojnoga rada kod stablovne se metode ~esto poku{avalo uvesti i ru~no-strojni na~in rada motornom pilom, no zbog njegove sve ve}e cijene ni takav na~in nije u potpunosti isplativ. Cilj je ovoga istraìvanja razviti i ispitati primjenjivost stablovne metode sje~e u pridobivanju energijskoga drva iz neindustrijskih privatnih {uma u Italiji pri ~emu su trebali biti zadovoljeni ovi uvjeti: Þmogu}nost izvo|enja selektivnih proreda bez dodatnoga otvaranja {uma, Þpotpuno mehanizirani rad bez radnika neposredno na tlu (pove}anje sigurnosti i za{tite na radu), Þkori{tenje lak{ih specijaliziranih {umskih vozila radi smanjenja tro{kova. Pri stablovnoj metodi pridobivanja energijskoga drva kori{tena su sljede}a specijalizirana {umska vozila na opisane na~ine: 1) sje~na glava postavljena na mali utovariva~ (Bobcat) za sje~u te uhrpavanje oborenih stabala uzdu` prorednih redova, 2) prilago|eni poljoprivredni traktor opremljen hvatalom za privla~enje drva do pomo}noga stovari{ta, 3) pokretni ivera~ opremljen hidrauli~nom dizalicom. Svojstva su vozila prikazana u tablici 1. Istraìvanje je provedeno u plantaì alepskoga bora Pinus halepensis Mill. u pokrajini Campobasso u sredi{njoj Italiji. Prije toga u plantaì nije bilo proreda te je metodom negativne selekcije odabrano i posje~eno 28 % ukupne

Croat. j. for. eng. 30(2009)2

A Low-Investment Fully Mechanized Operation for Pure Selection Thinning of ... (89–97)

R. Spinelli and C. Nati

drvne zalihe, {to je 27,5 t suhe drvne tvari/ha (tablica 2). Pomo}u studije rada i vremena odre|ena je i produktivnost vozila po 3 faze rada: 1) ru{enje stabala, 2) privla~enje drva, 3) iveranje drva. Obujam je stabala dobiven pomo}u dvoulaznih tablica autora Castellani i dr. (1982) za alepski bor. Sadràj vlage u drvu je odre|en na 10 uzoraka drvnoga iverja prema europskoj normi CEN/TS 14774-2, uz gusto}u drva od 510 kg/m3. Tro{kovi strojnoga rada izra~unati su Miyatinom metodom (1980) te su uve}ani za dodatnih 20 % zbog tro{kova administracije (tablica 3). Pomo}u stablovne metode u proredama mladih sastojina alepskoga bora mogu}e je pridobiti zna~ajnu koli~inu biomase, {to je i vidljivo na slici 1 koja prikazuje postotni udio mase grana debljine do 3 cm u ukupnoj masi stabala, ovisno o njihovu prsnom promjeru. U tablici 4 prikazana je proizvodnost strojeva po radnim sastavnicama. Sje~om i privla~enjem drva postiè se pribli`no jednaka proizvodnost rada (oko 40 stabala/h), dok se iveranjem drva postiè gotovo dvostruko ve}a proizvodnost. Odre|ene su 3 ina~ice sustava rada: Þ1. Ivera~ je postavljen na pomo}nom stovari{tu, ali se stavlja u pogon tek kada je sru{eno i privu~eno dovoljno stabala za puni tovar kamiona. Þ2. Skupni rad dvaju malih utovariva~a sa sje~nim glavama i dvama adaptiranim poljoprivrednim traktorima s hvatalima te jednim ivera~em za koji je potrebna visoka organiziranost. Þ3. Iveranje je drva zasebna radna operacija te se ivera~ dovozi na pomo}no stovari{te i stavlja u pogon tek nakon zavr{enoga ru{enja i privla~enja stabala te djelomi~noga prosu{ivanja drva. Sve tri ina~ice rada nisu neposredno ispitane, ve} je uz pomo} algoritama i statisti~ke analize podataka procijenjena povezanost utro{ka vremena rada i radnih uvjeta. U tablici 5 prikazani su tro{kovi po ina~icama rada br. 1 i 3 uz procijenjeno vrijeme i tro{ak premje{tanja vozila, dok ina~ica br. 2 nije uskla|ena s organizacijom rada te prosje~nom veli~inom {umskih posjeda u neindustrijskim privatnim {umama u Italiji (NIPF – Non Industrial Private Forests). U tablici 6 prikazani su podaci ispitivanja radnih ina~ica uz ove ulazne podatke: veli~ina posjeda od 4 ha, prosje~ni prsni promjer stabla od 15 cm, prosje~na udaljenost privla~enja od 300 m i prosje~na udaljenost daljinskoga transporta od 24 km. Pokazalo se da 1. ina~ica rada ima tro{kove tek 3 % ve}e od ostalih dviju ina~ica te se dolazi do zaklju~ka da ako se radnik na ivera~u drva zaposli i na drugim poslovima na radili{tu dok ivera~ nije u pogonu, ukupno }e pove}anje tro{kova ove ina~ice rada biti neznatno. Tro{kovi premje{tanja vozila imaju velik utjecaj na ukupne tro{kove pridobivanja energijskoga drva. Na slici 2 prikazana je ovisnost veli~ine posjeda na ukupne tro{kove po jedinici suhe drvne tvari energijskoga drva te je vidljivo da su tro{kovi pridobivanja niì na posjedima ve}im od 4 ha. Na slici 3 vidljivo je da }e pri trenuta~noj cijeni drvnoga iverja od 85 /t suhe drvne tvari dobit u poslovanju biti tek kod stabala s prsnim promjerom ve}im od 19 cm, dok se pri projiciranoj cijeni od 100 /t suhe drvne tvari postiè dobit kod stabala s prsnim promjerom od 15 cm. Tro{kovi daljinskoga transporta uvelike utje~u na ukupne tro{kove pridobivanja drvnoga iverja, i to pove}anjem od 20 % pri pove}anju udaljenosti transporta od 20 do 60 km. Istraìvanje je pokazalo da su najmanji tro{kovi i najve}a proizvodnost u pridobivanju energijskoga drva iz selektivnih proreda plantaà alepskoga bora pri kori{tenju stablovne metode, {to ujedno i umanjuje rizik od poàra jer se sva granjevina rabi za pridobivanje drvnoga iverja, pri odvajanju rada ivera~a od ostalih radnih operacija te pri veli~ini {umskih posjeda iznad 4 ha. Klju~ne rije~i: sje~na glava, neindustrijske privatne {ume, biomasa, stablovna metoda, prorede

Authors' address – Adresa autorâ:

Received (Primljeno): July 31, 2009 Accepted (Prihva}eno): November 15, 2009 Croat. j. for. eng. 30(2009)2

Raffaele Spinelli, PhD. e-mail: spinelli@ivalsa.cnr.it Carla Nati, PhD. e-mail: nati@ivalsa.cnr.it CNR – Ivalsa via Madonna del Piano Pal. F I-50019 Sesto Fiorentino ITALY

Original scientific paper – Izvorni znanstveni rad

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations and Mixed Plantations with Willow Clones (Salix sp.) in Croatia Sa{a Bogdan, Mario [por~i}, Ante Seletkovi}, Mladen Ivankovi} Abstract – Nacrtak In the last two decades, Croatian energy policy was directed towards the increase of renewable energy sources in the total balance. It resulted in an increased interest for breeding fast-growing hardwoods in short rotations. Although common alder is not so productive in short rotations as some willows and poplars, insufficient development of the biomass market and increased awareness for conservation of domestic forest tree species, makes it favorable for raising plantations as it is the indigenous forest tree species that can be grown in longer rotations, if needed. The estimate of above-ground dry biomass per tree (trunk and branches up to 7 cm in diameter) as well as production of biomass per hectare was made in 4 experimental plantations. The research included two mixed plantations of common alder open-pollinated families with willow clones (Salix sp.) and two pure common alder plantations. Experimental plantations, aged 14 and 16 years, are located in two different sites. The estimated family mean values for dry biomass varied between 12.5 and 70.9 kg per tree. The results showed considerable differences between families as well as between different silvicultural treatments. The estimated production of the common alder dry biomass varied between 27.4 and 87.5 t/ha with mean annual increments (MAI) between 1.9 and 6.3 t ha –1 year –1. It was shown that willow clones have a negative influence on the alder biomass in mixed plantations, at the studied plantation age. Willow clones have shown greater biomass production, compared with common alder, in spite the fact that its planting density was far lower. Although the planting density of common alder trees was lower compared to other studies, the average biomass production is within the framework of other results, which indicates significant potential of biomass production in local conditions. The additive variance was not statistically significant for the biomass trait, which was probably caused by the dominant influence of microenvironmental factors (weed, game) during the growth, but also by a small number of studied families. Statistically significant interactions of investigated families with silvicultural treatments and sites have been shown, which directs to the genetically caused differences in the adaptability of families on studied site and growth differences. Key words: open-pollinated families, genetic test, variation, genotype ´ environment interactions

1. Introduction – Uvod The common alder (Alnus glutinosa /L./ Gaertn.) is spread all over Europe, from Ireland to the western Siberia, as far as the northern Africa and up to Croat. j. for. eng. 30(2009)2

60° in the northern Europe (Krstini} 1994). In Croatia, common alder grows in lowland forests along the rivers in pure or mixed stands with narrow-leaved ash (Fraxinus angustifolia Vahl.), European white elm (Ulmus laevis Pall.) and pedunculate oak

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

(Quercus robur L.). In some parts of northern Croatia along the river Drava, on humus-gley soil, common alder is the primary element of the association described as Carici elongatae Alnetum glutinosae W. Koch 1926. Localities of the phytocenosis belong to the most productive stands of common alder, with the tree heights above 30 m and volumes greater than 500 m3/ha (Vukeli} and Rau{ 1998). The Croatian energy policy is today directed towards increased efficiency, security of supply and diversification, market deregulation, and the use of renewals and environmental protection. The government launched a number of National Energy Programs in order to reach the goals of the energy policy, one of which (named BIOEN) is directly aimed at biomass and waste utilization (Domac et al. 1998, 2001). In regard to this, research work on the breeding of fast-growing hardwoods for the biomass production in short rotations has started. Due to its biological characteristics common alder has a potential as a species for biomass production in short rotations. Earlier investigations in Croatia (Komlenovi} et al. 1996, Kajba et al. 1998, Bogdan 2002, Kajba et al. 2004, 2006), as well as in other countries (Saarsalmi 1995, Hytönen 1996) showed that common alder does not have such a high production capacity as some willow and poplar clones. However, considering the trend of conservation of natural forest species and undeveloped biomass market in Croatia, it is prospective for raising plantations as it is indigenous species and, depending on the market oscillations, can be grown in longer rotations. Research on the production of alder biomass done so far include results from UK (Ovington 1956, Mitchell et al. 1981, Proe et al. 1999), Netherlands (Meeuwissen and Rottier 1984), Nordic and Baltic countries (Björklund and Ferm 1982, Rytter 1996, Saarsalmi and Mälkönen 1989, Saarsalmi et al. 1992, Korsmo 1995, Tullus et al. 1998, Uri and Tullus 1999, Uri et al. 2002) while the reports from South-Eastern Europe is scarce. The aim of this investigation was to estimate the production of above ground dry biomass in present experimental plantations with open-pollinated families of common alder in Croatia, and to determine the variability of biomass trait within and between the investigated families in various growth and site conditions.

2. Materials and methods – Materijal i metode 2.1 Investigated plantations – Istra`ivane kulture The research was done in four experimental plantations with open-pollinated families of common al-

100

der established at two sites in Croatia. Plants were grown from the seeds gathered in the existing clonal seed orchard. The mixed plantation of common alder with willow clones (CJ1 plantation) is situated at the locality of Crni Jarci (46.0100 N; 17.1705 E). It is 16 years old and includes 18 open-pollinated families and 11 willow clones (Table 1). The trees of the common alder families were planted on a spacing of 4.0 ´ 2.0 m (1250 trees/ha) with ten trees per plot while willow clones were planted on a spacing of 4.0 ´ 4.0 m (625 trees/ha) between rows of alder. The trial was established according to the design of the randomized complete block system with four replications. At the same time, right next to the above mentioned, a pure plantation of common alder (CJ2 plantation) was planted with a spacing of 2.5 ´ 2.0 m (2000 trees/ha) with 15 trees per plot according to the design of the randomized complete block system with four replications. There are 14 alder open-pollinated families in this plantation that are in common with CJ1 plantation. The soil of the location of Crni Jarci belongs to the peat-gley type of hydromorphic soil (histic gleysol, according to the WRB soil classification), of somewhat acid reaction and well supplied with nutritions (Krstini} and Komlenovi} 1986). At the locality of Lisicine (45.6615 N; 17.5072 E) a 14 year old mixed plantation was established (L1 plantation) with common alder open-pollinated families and single white willow clone 'V093'. Com-

Table 1 Investigated willow clones in mixed culture with common alder open–pollinated families – CJ1 (Clone 'V 093' is also planted in mixed plantation L1 – see Material and methods section) Tablica 1. Istra`ivani klonovi vrbe uzgajani u mje{ovitoj kulturi s familijama crne johe – CJ1 (Klon 'V 093' je tako|er posa|en u mje{ovitoj kulturi L1 – v. Materijal i metode) No. Br. 1 2

Clone label Oznaka klona 'BR 1BB' 'MB 15'

Botanical name Botani~ki naziv Salix alba S. alba ´ S. a. var. vitellina

'V 04'

S. alba var. calva ´ S. alba

'V 052'

S. alba var. calva ´ S. alba

5 6 7 8 9

'V 093' 'V 158' 'V 160' 'V 161' 'V 217'

(S. alba ´ S. alba. var. vit.) ´ S. alba Salix alba Salix alba Salix alba

'V 218'

(S. alba ´ S. sitchensis) ´ S. alba

'V 219'

(S. alba ´ S. sitchensis) ´ S. alba

Croat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

mon alder trees were planted with a spacing of 4.0 ´ 1.5 m (1667 trees/ha) with 18 trees per plot while the willow clone was planted with a spacing of 4.0 ´ 4.0 m between the rows of alder. There are 18 alder families in the plantation, but only 13 are in common with CJ1 plantation. Nearby, a pure plantation of common alder was established (L2 plantation) according to the randomized complete block design. It consists of the same families and equal planting pattern as the plantation L1 and it is at the same age. It has nine open-pollinated families in common with the pure plantation CJ2. The soil of this locality is characterized by pseudogley plateau (dystric planosol according to the WRB soil classification). The soil reaction is acidic, and the level of nutrition is lower than the level of adequate supplies for alder (Krstini} and Kajba 1991). In the vicinity of the researched plantations at the Lisicine site there is an artificial lake which secures higher air humidity during the growth period. In all the mentioned plantations, no additional silvicultural (e. g. weed control, tending and fertilization) or protective measures have been implemented.

2.2 Growth measurements – Izmjere rasta Diameter at breast height (DBH) and the height of each single tree were measured in all four plantations. Based on diameters and heights the volumes of trees were calculated using the Schumacher-Hall equations for the volume of common alder (Cestar and Kova~i} 1982) and white willow (Cestar and Kova~i} 1979). By the stated measurements the survival of studied families and clones was also determined.

2.3 Biomass estimate – Procjena biomase The samples, based on which the biomass of the above-ground part of trees was estimated, were wood discs at breast height of the cut model trees. Model trees were determined by DBH variability of each single family and clone. In that way, four model trees were cut, respective of each studied family, clone and different trial. After taking the disks from the model trees, their fresh masses were measured. Minimum and maximum diameters of the discs were taken (dmin. and dmax.), together with the thickness in four points representing the beginning and end points of the minimum and maximum diameter (h1, h2, h3, h4). The volumes of the discs were determined using the formula for the volume of cylinder: Croat. j. for. eng. 30(2009)2

S. Bogdan et al.

[V = ( d / 2 2 )ph; d = ( dmin + dmax ) / 2; h = ( h1 + h2 + h3 + h4 ) / 4] Discs were dried at the temperature of 105°C until reaching the constant mass. Based on the volume of fresh discs and their dry mass, nominal densities were calculated: j n = ms / V f jn nominal density ms mass of dry ring Vf volume of fresh ring Based on mean values of nominal density for each family and clone (calculated as the mean of the nominal densities of the respective discs), and the calculated tree volumes (fresh volume), the dry biomass of each single tree were estimated. Since the calculated tree volumes regard the volume of trunks and branches up to 7 cm in diameter (Cestar and Kova~i} 1979, 1982), the estimate of dry biomass also regards the dry biomass of trunks and branches up to 7 cm in diameter. This method is based on the somewhat modified method of determining the nominal tree density (ISO 3131, 1975). The estimate of the dry biomass per hectare was done based on the mean values of each family and clone in each locality, with respect to actual planting spacing and survival percentage. The biomass was estimated and analyzed for 13 families that are common in both mixed plantations (CJ1 and L1) and 9 common families in both pure plantations (CJ2 and L2).

2.4 Statistical analysis – Statisti~ke analize The statistical analyses were done by the program package SAS System for Windows, ver. 6.12 (SAS Institute Inc. 1989). The descriptive analysis was conducted for determining the mean values, variability range, standard deviations and coefficients of variability. The analysis of variance was done using the GLM procedure in SAS (SAS Institute Inc. 1989). Expected mean squares, their coefficients and F tests were conducted for the purpose of testing the significance of the variances of single effects. Expected mean squares of error deviation were used as the denominator in the F tests. The analysis of variance for the investigated common alder families was calculated separately for each single trial according to the following linear model: Yijk = m + Fi + Rj + FRij + eijk

(1)

101

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Yijk observed measurement of tree ijk m a fixed overall mean Fi effect of family i (i – 1,2,.....,i; depending on the test plot) Rj effect of replication j (j – 1,2,.....j; depending on the test plot) FRij effect of family i in replication j eijk random error The analysis of variance was done for the common nine open-pollinated families, which were tested regarding two different treatments (mixed or pure plantations) with the aim of establishing the effect of the growth treatment according to the linear model: Yijkl = m + Ti + Ri(j) + Fk + TFik + eijkl

(2)

Yijkl observed measurement of tree ijkl m a fixed overall mean Ti effect of the treatment i (i – 1, 2) Fk effect of family k (k – 1, 2,.....,9) Ri(j) effect of replication j in treatment i TFik effect of family k in treatment i eijkl random error The analysis of variance for the common open-pollinated families tested in two localities was done for the purpose of determining the effect of the site as well as the family ´ site interaction, according to the linear model: Yijkl = m + Li + Ri(j) + Fk + LFik + eijkl

(3)

Yijkl observed measurement of tree ijkl m a fixed overall mean Li effect of the locality i (i – 1, 2) Fk effect of family k Ri(j) effect of replication j in a locality i LFik effect of family k in locality i; eijkl random error The analysis of variance for willow clones grown in association with the common alder open-pollinated families (plantation CJ1) was conducted using the following model: Yij = m + Ci + eij

(4)

Yij observed measurement of tree ij m a overall mean Ci effect of the clone i eij random error All analyzed effects, except m, were considered random and normally distributed.

102

2.5 Genetic parameters estimation – Procjene genetskih parametara The REML method of VARCOMP procedure was used to estimate the variance components of all effects. Additive, genetic and phenotypic variances were estimated with the usual formulae (Wright 1976, Falconer and Mackay 1996): sF = 1 / 4 sa

(5)

sPH = sF + sFB + sE

(6)

s C = sG

(7)

Where: sF variance due to family effect sA additive genetic variance sG genetic variance sC variance due to clone effect sPH phenotypic variance sFB variance due to family ´ block interaction sF variance due to random error

3. Results – Rezultati 3.1 Above ground dry biomass – Suha biomasa nadzemnoga dijela stabala A rather big variability can be noticed among the investigated common alder families in all trials (Tables 2, 3), but also a significant variability within the families, which can be seen from the high coeff. of variation. Generally, families 'B18' i 'B21' have shown the largest mean values, which is the indicator of the good general combining ability (GCA) of their mother trees. Comparing the two plantations (CJ1 i CJ2), one can see the slightly higher mean values of the biomass of common alder families grown together with willow clones (Tables 2 and 3). Family 'B18' showed high biomass values in both trials, while the family 'B21' in the mixed plantation showed average value. In the pure plantation Lisicine (L2) the common alder families had the highest mean dry biomass values in regard to all the other studied plantations (Table 3). Comparing the average values of the families in plantations L1 and L2, significant difference can be seen in the production, at the advantage of the pure plantation. The differences are so great that the family with the lowest mean value is better than the best family in the mixed plantation. Negative influence of willow on the total growth, as well as on the common alder biomass production is obvious. Croat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Table 2 Descriptive statistical parameters for the above-ground dry biomass per tree (kg) of the common alder open-pollinated families grown in mixed plantations with willow clones Tablica 2. Deskriptivni statisti~ki parametri familija crne johe uzgajanih u mje{ovitim kulturama s klonovima vrbe za svojstvo suha biomasa po stablu (kg) O. – P. Familija

B1 B14 B15 B16 B17 B18 B21 B23 B29 B32 B4 B5 B7 Mean Arit. sredina B1 B14 B15 B16 B17 B18 B21 B23 B29 B32 B4 B5 B7 Mean Arit. sredina

Range Opseg (xmin. – xmax.)

Mean Stand. Dev. Arit. sredina Stand. dev. 2 (s ) (X)

Mixed culture – Mje{ovita kultura CJ1 1.44 – 65.20 23.84 19.43 5.06 – 88.46 33.18 17.44 5.72 – 80.65 37.79 20.50 12.84 – 81.87 37.18 18.56 1.29 – 91.83 26.63 23.55 8.94 – 98.55 40.13 19.61 4.30 – 80.29 31.47 23.49 5.09 – 65.88 30.31 17.40 7.92 – 88.08 34.75 21.28 2.67 – 91.65 28.19 17.77 6.22 – 74.16 25.68 15.67 3.03 – 44.06 20.75 11.22 4.23 – 133.75 32.38 29.94 30.81

C.V. Koficijent varijacije

20.44

66.34

Mixed culture – Mje{ovita kultura L1 1.73 – 57.76 19.72 11.44 0.05 – 3 8.82 16.84 10.40 1.08 – 48.22 18.13 11.38 0.96 – 82.65 20.00 17.07 0.64 – 62.87 21.74 16.32 3.04 – 72.39 33.12 20.97 5.63 – 68.57 28.21 15.61 2.85 – 41.00 17.93 10.79 0.17 – 71.13 22.60 19.83 2.74 – 61.58 23.56 14.79 2.63 – 57.66 23.55 12.93 0.97 – 29.65 12.55 7.94 0.93 – 74.51 21.89 19.03

58.00 61.76 62.74 85.35 75.08 63.31 55.33 60.20 87.73 62.79 54.89 63.28 86.91

21.71

15.48

71.28

Table 4 shows the results of the studied willow clones grown in the mixed plantation CJ1. It can be noticed that on average willow clones showed superior biomass production compared to common alder families. The clone 'V 219' had the lowest mean biomass, even lower than the majority of the alder families. Croat. j. for. eng. 30(2009)2

Table 3 Descriptive statistical parameters for the above-ground dry biomass per tree (kg) of the common alder open-pollinated families grown in pure plantations Tablica 3. Deskriptivni statisti~ki parametri familija crne johe uzgajanih u ~istim kulturama za svojstvo suha biomasa po stablu (kg) O. – P. Familija

(%)

81.51 52.57 54.24 49.92 88.43 48.87 74.67 57.40 61.24 63.04 61.01 54.06 92.47

S. Bogdan et al.

B1 B14 B15 B17 B18 B21 B32 B5 B7 Mean Arit. sredina B1 B14 B15 B17 B18 B21 B32 B5 B7 Mean Arit. sredina

Range Opseg (xmin. – xmax.)

Mean Stand. Dev. Arit. sredina Stand. dev. (s2) (X)

Pure culture – ^ista kultura CJ2 1.07 – 72.93 23.15 14.96 1.24 – 67.04 24.71 13.97 1.09 – 79.97 29.15 18.43 0.99 – 180.78 34.29 28.52 4.44 – 80.54 34.24 21.09 4.27 – 103.66 37.57 23.42 1.80 – 66.62 26.76 16.81 0.72 – 71.37 28.71 17.32 1.61 – 100.07 30.72 22.55 29.59

20.39

Pure culture – ^ista kultura L2 15.81 – 73.46 49.41 16.63 8.07 – 126.69 63.00 26.85 10.44 – 104.69 55.69 27.49 16.45 – 90.01 55.05 22.18 10.47 – 142.43 70.94 35.91 21.28 – 111.35 63.96 27.71 8.55 – 126.22 53.61 27.67 3.35 – 87.14 37.77 18.77 31.25 – 117.55 67.58 22.77 57.65

27.64

C.V. Koficijent varijacije (%) 64.64 56.54 63.25 83.18 61.60 62.35 62.79 60.34 73.41 68.91 33.65 42.63 49.36 40.28 50.61 43.32 51.61

49.69 33.70 47.94

3.2 Biomass production per hectare Proizvodnja biomase po hektaru The estimated above ground dry biomass production (trunks and branches up to 7cm in diameter) per hectare had the highest mean value in plantation L2 (87.5 t/ha), and lowest in CJ2 (27,4 t/ha) – see Figures 1 and 2. Mean annual increment (MAI) varied between 1.9 and 6.3 t ha–1 year–1. At the locality of Crni Jarci (Fig. 1) studied families showed slightly greater mean biomass production in the plantation where they were grown in association with willow clones, but also with wider planting spacings (4.0 ´ 2.0 m). Some families, in spite of that, have shown better production in treatment without willows ('B17', 'B21', 'B5', 'B7').

103

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Table 4 Descriptive statistical parameters for the above-ground dry biomass per tree (kg) of the investigated willow clones grown in mixed plantations CJ1 Tablica 4. Deskriptivni statisti~ki parametri klonova vrbe uzgajanih u mje{ovitoj kulturi s crnom johom CJ1, za svojstvo suhe biomase (kg) O. – P. Familija 'V 052' 'V 093' 'V 158' 'V 217' 'BR1BB' 'MB 15' 'V 160' 'V 218' 'V 04' 'V 161' 'V 219' Mean Arit. sredina

Range Opseg (xmin. – xmax.) 11.61–376.38 10.50–315.71 16.92–202.78 15.49–183.12 10.47–175.41 20.28–204.22 15.72–173.66 9.74–147.04 6.60–142.59 18.52–69.30 5.36–75.00

Mean Stand. Dev. Arit. sredina Stand. dev (s2) (X)

C.V. Koficijent varijacije

123.43 95.35 83.41 78.57 65.69 65.02 64.99 46.84 40.65 38.02 28.47

109.72 62.89 54.01 48.40 48.95 47.11 37.83 47.63 29.78 17.79 18.62

(%) 88.89 65.95 64.75 61.60 74.51 72.46 58.21 101.70 73.25 46.79 65.40

78.25

63.78

81.51

Total average value of production of dry wood per hectare in the pure plantation L2 was 87.5 t/ha. Comparison of the average values of biomass production in the investigated common alder families in plantations L1 i L2 is presented in Fig 2. In opposite to the Crni Jarci locality, here, both treatments had the same planting spacings. Considerably better production of common alder biomass can be seen in pure plantation, where the families were grown without willows. The white willow clone 'V 093' had an 89% survival rate and the biomass production of 65.2 t/ha with considerably lower tree density (625 trees per hectare) in comparison with the common alder (2000 trees per ha). Average biomass production of studied willow clones per hectare in the mixed plantation CJ1 was 32.8 t/ha (Fig. 3). However, clones 'V 052' and 'V 093' showed markedly higher biomass production that exceeded 50 t/ha. Average survival of common alder open-pollinated families in mixed plantation L1 was somewhat better than in CJ1 (from 83% to 100%). The lowest production of biomass was by family 'B5' (19.3 t/ha), and the highest by 'B21' (47.0 t/ha) – Fig. 4. It can be seen that families showed different pattern of bio-

Fig. 1 Average above ground dry biomass production1 of investigated common alder open-pollinated families grown in two different silvicultural treatments – locality of Crni Jarci Slika 1. Prosje~na proizvodnja suhe biomase nadzemnoga dijela istra`ivanih familija crne johe uzgajanih u dvjema planta`ama na lokalitetu Crni jarci mass production at two sites. For example, while family 'B 21' had the highest biomass production at L1 plantation it showed under average production at CJ1 plantation. Fig. 5 presents the comparison between the average values of dry biomass production of common alder families grown without willow at two localities (CJ2 i L2). Higher productivity of all families at the locality of Lisicine, in spite of younger age, is rather surprising. However it could be explained by better site conditions at the Lisiscine site, but also by wider spacing between trees. Some families have shown considerably greater potential of biomass production (e. g. 'B7' – Fig. 5), their MAI being 7.6 tons ha–1 year–1.

3.3 Genetic variation – Genetska varijabilnost On the basis of the quantitative genetics theory, it is assumed that differences in quantitative traits between trees that belong to various open-pollinated families and grow within the same environmental conditions can be regarded to the genetic differences

above ground biomass refer to estimated biomass of tree trunks and branches up to 7 cm in diameter

104

Croat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Fig. 2 Average above ground dry biomass production of investigated common alder open-pollinated families grown in two different silvicultural treatments – locality Lisicine Slika 2. Prosje~na proizvodnja suhe biomase nadzemnoga dijela istra`ivanih familija crne johe uzgajanih u dvjema planta`ama na lokalitetu Lisi~ine

S. Bogdan et al.

Fig. 4 Average above ground dry biomass production of common alder open-pollinated families. Comparison of the families grown in mixed plantations with willow clones in two different localities Slika 4. Prosje~na proizvodnja suhe biomase nadzemnoga dijela familija crne johe. Usporedba familija uzgajanih u mje{ovitim kulturama s klonovima vrbe na dvama lokalitetima. between their mother trees (Falconer and Mackay 1996). As biomass of the tree is quantitative trait, variance components caused by family effect are considered as a major part of the genetically caused variance of the trait. Table 5 shows that variance components caused by the family effect were not statistically significant and took relatively small part of the total variance. The largest percentage of the family effect (but still not significant) was observed in data from the L2 plantation (Table 5). The major percentages of the total variation were observed for the error variance component but it can be seen that family by replication interactions also had significant effects.

3.4 Family by plantation establishment interaction – Interakcija familija s na~inom osnivanja planta`a

Fig. 3 Average above ground dry biomass production of investigated willow clones grown in mixed plantation with common alder – locality of Crni Jarci (CJ1; 625 trees per ha; 16 yrs old) Slika 3. Prosje~na proizvodnja biomase nadzemnoga dijela stabala istra`ivanih klonova vrbe uzgajanih u mje{ovitoj kulturi s crnom johom – lokalitet Crni jarci (CJ1; 625 stabala po ha; dob 16 godina) Croat. j. for. eng. 30(2009)2

The results of the combined analyses of variance for both plantations at the locality of Crni Jarci (combined CJ1 and CJ2) and Lisicine locality (combined L1 and L2) are shown in Table 6. The aim of those analyses was to determine a significance of the influence of the willow clones on the common alder biomass production as well as of the influence of different spacing of common alder trees on its biomass production (the so called »silvicultural treatment«

105

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

at both sites, which means that families had different responses to the studied silvicultural treatments.

3.5 Family ´ site interaction – Interakcija familija sa stani{nim uvjetima

Fig. 5 Average above ground dry biomass production of common alder open-pollinated families. Comparison of the families grown in pure plantations in two different localities Slika 5. Prosje~na proizvodnja suhe biomase nadzemnoga dijela familija crne johe. Usporedba familija uzgajanih u ~istim kulturama na dvama lokalitetima. influence). It can be seen that treatment has caused significant variation at the locality of Lisicine, but not at the locality of Crni Jarci. Also, it can be noticed that the effects of family by treatment interactions were statistically significant

The results of the combined analyses of variance for both sites but regarding the same silvicultural treatment (combined analyses of CJ1 and L1, as well as combined analyses of CJ2 and L2) are shown in Table 7. The aim of those analyses was to determine the significance of the influence of different site conditions on the common alder biomass production. The mean annual increment (MAI) was analyzed because of the different age of trees The results showed statistically significant differences between the sites (for both types of plantations i.e. mixed and pure plantations), as well as statistically significant family by site interactions (with significance level of 0.05 and 0.01, respectively). The studied common alder families, grown in two pure plantations (CJ 2 and L2), have shown particularly high influence of environment (50% variance component caused by sites).

4. Discussion – Rasprava 4.1 Above ground dry biomass – Suha biomasa nadzemnoga dijela stabala Taking into consideration a more uniform genetic structure of willow in comparison to common alder, great variation of biomass within clones is rather

Table 5 Variance components for the above-ground dry biomass trait of the common alder open-pollinated families in investigated plantations Tablica 5. Komponente varijance za svojstvo suhe biomase nadzemnog dijela stabala familija crne johe u istra`ivanim kulturama

sF sb2 sfxb2 se2 sA2 sPH2 * ** ns 2

106

sf2

sb2

16.04ns

3.8

0.00ns

0.0

6.52 ns

1.6

0.00ns

0.0

1.113 ns

0.5

5.010ns

2.1

55.464ns

7.2

0.000ns

0.0

(%) sfxb2 se2 Mixed culture – Mje{ovita kultura CJ1 25.03* 6.0 378.55 Pure culture – ^ista kultura CJ1 22.18* 5.3 378.98 Mixed culture – Mje{ovita kultura L1 32.723** 13.5 203.624 Pure culture – ^ista kultura L2 73.785** 9.5 644.562

sA2 (%)

sPH2 (%)

90.2

64.16 (15.3%)

419.62 (100%)

93.1

26.08 (6.4%)

407.68 (100%)

84.0

4.45 (1.9%)

237.46 (100%)

83.3

221.86 (28.7%)

773.81 (100%)

variance caused by family effect variance caused by replication variance caused by family ´ replication effect random error variance additive variance phenotypic variance significance level p < 0.05 significance level p < 0.01 not significant

Croat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

S. Bogdan et al.

Table 6 Variance components for the above – ground dry biomass of the studied common alder open – pollinated families. Family by plantation establishment interactions (mixed or pure cultures and different planting spacings – so called silvicultural treatment) Tablica 6. Komponente varijance za suhu biomasu nadzemnoga dijela stabala istra`ivanih familija crne johe. Interakcija familija s na~inom osnivanja kulture (mje{ovita ili ~ista kultura te razli~iti razmak sadnje). sf2

sb2

9.97ns (2.3%)

1.45ns (0.3%)

45.17* (4.1%)

0.00ns (0.0%)

sF2 sb2 sT2 sfxT2 se2 sA2 sPH2 * ** ns

sT2

sfxT2 se2 Locality – Lokalitet (Crni jarci) ns 0.00 (0.0%) 8.95* (2.1%) 412.41 (95.3%) Locality – Lokalitet (Lisi~ine) 635.69** (58.4%) 12.70* (1.17%) 395.12 (36.3%)

sA2 (%)

sPH2(%)

39.88 (9.2%)

431.33 (100%)

180.68 (39.9%)

452.99 (100%)

variance caused by family effect variance caused by replication variance caused by silvicultural treatments (mixed or pure culture) variance caused by family ´ treatment effect random error variance additive variance phenotypic variance significance level p < 0.05 significance level p < 0.01 not significant

surprising. It is most likely caused by the influence of microhabitat factors (e. g. weed, game) and the competition between alder and willow. The estimated mean values of the above ground dry biomass are in accordance with dry weights of the total biomass production above stump level per tree presented by Johansson (2000) for common alder. Though alder biomass production is obviously much lower than willow, it is still significant. Moreover, if one takes into account larger market prices of alder wood, its ecological values (as well as public perception on indigenous forest species) and unde-

veloped biomass market in Croatia, it seems that alder plantations presently have greater potential than willows.

4.2 Biomass production per hectare Proizvodnja biomase po hektaru Greater production of willow in regard to the common alder families, in spite of smaller number of trees per plot (planting spacing 4.0 ´ 4.0 m), is the consequence of their biological differences, i.e. more rapid growth of willows in the first years of life,

Table 7 Variance components for the above-ground dry biomass of the studied common alder open-pollinated families. Family by site interactions (the same silvicultural treatments at different sites) Tablica 7. Komponente varijance za suhu biomasu nadzemnoga dijela stabala istra`ivanih familija crne johe. Interakcija familija sa stani{tem (jednako osnovane kulture na razli~itom stani{tu). sf2

sb2

0.053ns (3.6%)

0.015* (1.0%)

0.084 ns (1.7%)

0.001 ns (0.0%)

sF2 sb2 sL2 sfxL2 se2 sA2 sPH2 * ** ns

sL2

sfxL2 se2 Mixed cultures – Mje{ovite kulture 0.052*(3.6%) 0.037* (2.5%) 1.308 (89.2%) Pure cultures – ^iste kulture 2.497** (50.0%) 0.152** (3.1%) 2.257 (45.2%)

sA2 (%)

sPH2 (%)

0.212 (15.2%)

1.398 (100%)

0.336 (13.5%)

2.493 (100%)

variance caused by family effect variance caused by replication variance caused by sites (Crni Jarci and Lisicine) variance causd by family ´ site interaction random error variance additive variance phenotypic variance significance level p < 0.05 significance level p < 0.01 not significant

Croat. j. for. eng. 30(2009)2

107

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

which was even more likely as the investigated clones are of superior growth even in regard to Salix species. Different authors reported above ground biomass production for common and grey alder stands in Europe (Ovington 1956, Björklund and Ferm 1982, Rytter 1996, Saarsalmi and Mälkönen 1989, Saarsalmi et al. 1992, Korsmo 1995, Tullus et al. 1998, Uri et al. 2002) and USA (Wittwer and Stringer 1985). The reported mean annual increments varied between 1.8 (Meeuwissen and Rottier 1984) and 14.8 tons ha–1 year–1 (Rytter 1996,Tullus et al. 1998). Mitchell et al. (1981) bring results on the above ground biomass production for 12 and 18 year old common alder stands (approximately the same age as plantations used in this study), which were 29.6 t/ha with 2.5 MAI and 87.2 t/ha with 4.8 MAI. The production of common alder biomass in plantations presented here is in the framework of other reports. However, the planting densities in our studied plantations were considerably lower. It shows that there is significant potential in increase of biomass production by planting plantations with greater plant density, under local conditions, especially by combining selection with favorable silvicultural treatments. Additionally, there is a great potential for effective harvesting method that could improve the economical basis for establishment of such plantations (Krpan and Por{insky 2004).

4.3 Genetic variation – Genetska varijabilnost Family effect as an indicator of genetic control was not a statistically significant factor, which shows that the environment (weed, game etc.) had a great impact on researched trait. Experimental error, which represents the variability within families and within replications, had the greatest share in the total variability (Table 5). Small values of the additive variance could have also been influenced by lowered genetic variability because the trials include descendants of a small number of mother trees from the clonal seed orchard, in which the trees were selected by the superior growth traits, and are probably more genetically uniform. However, if one takes into consideration all studied plantations, it can be seen that greater differences in biomass production might be expected in more uniform environments (e.g. see Fig. 2 – production of the families at the plantation L2). Thus, the assumption that environmental impact caused small additive variance seems more reasonable.

108

4.4 Family by plantation establishment interaction – Interakcija familija s na~inom osnivanja planta`a Analyzed families are common in both types of tests (mixed and pure plantations), and the goal of the analyses was to determine the interaction of families and silvicultural treatments, i.e. growing common alder with or without the willow clones. For a more reliable testing of this interaction, other environmental factors should be eliminated. That was not done because of different spacing of alder trees (4.0 ´ 2.0 m and 2.5 ´ 2.0 m). For this reason the evaluation of the interaction is tampered with the influence of different spacings, and for the purpose of the exact evaluation of the influence of willow on the investigated alder families, the research should be repeated. Statistically significant interaction can be explained by differences in the influence of willow together with the influence of spacing on families, i.e. with specific adaptability of families to two different silvicultural treatments. In the plantations L1 and L2, trees were planted in the same spacing, so the established interaction is a better indicator of willow’s influence on the researched families. It was shown that the variance caused by treatments is statistically significant, with significance level of 0.01, which indicates the great influence of willow on all common alder families, which can be seen through decreased growth. However, a statistically significant family by treatment interaction was determined, with the significance level of 0.05, which indicates that the investigated families vary in their ability to compete with the willow. Those results could be useful for selection of families, which showed better production in mixed plantations with willow clones. According to the earlier research done in Croatia (Kajba et al. 1999) it was shown that the mixed plantations have an increased height growth of common alder in regard to the pure plantations, but only until the age of 8 years, after which white willow overgrew and suppressed the alder. In compliance with these reports these authors recommended the establishment of mixed alder and willow plantations in order to increase the total biomass production and enhance biological stability, but with the need to cut the willow by the time of eight years old. Mixed plantations in this study were way over that age, which had a negative influence on the common alder biomass production.

4.5 Family ´ site interaction – Interakcija familija sa stani{nim uvjetima Krstini} and Kajba (1991) investigated the mean annual increment for the height in the same plantaCroat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

tions with the trees aged 4 and 6, and have also determined significant family by site interaction. The authors explained the interaction with the contrast of sites considering the soil type, level of nutritions and humidity, that is, for the different adaptability of the studied families to the mentioned differences. Although the site on the locality of Crni Jarci is of higher quality (taking into consideration the soil type and nutrition) investigated families grew better in the locality of Lisicine, which Krstini} and Kajba (1991) explained by higher humidity during growth period (due to water accumulation nearby). With the given explanation and considering the age of trees, another possible reason for better growth of common alder families at Lisicine locality is wider tree spacing, which allowed better light interception.

5. References – Literatura Björklund, T., Ferm, A., 1982: Pienikokoisen koivun ja harmaalepän biomassa ja tekniset ominaisuudet. Summary: Biomass and technical properties of small–sized birch and gray alder. Folia Forestalia 500: 1–37. Bogdan, S., 2002: Procjena genetskih parametara i produkcije biomase mekih lista~a u pokusnim kulturama kratkih ophodnji. Master thesis. Faculty of Forestry Zagreb, 178 pp. Cestar, D., Kova~i}, \., 1979: Tablice drvnih masa bijele vrbe (Salix alba L.). Radovi 38: 81. [umarski institut, Jastrebarsko. Cestar, D., Kova~i}, \., 1982: Tablice drvnih masa crne johe i bagrema. Radovi 49: 149. [umarski institut, Jastrebarsko. Cotterill, P. P., Zed, P. G., 1980: Estimates of genetic parameters for growth and form traits in four Pinus radiata D. Don progeny tests in South Australia. Aust. For. Res. 10: 155–167. Domac, J., Beronja, M., Dobri~evi}, N., \iki}, M., Grbe{a, D., Jelavi}, V., Juri}, @., Kri~ka, T., Mati}, S., Or{ani}, M., Pavi~i}, N., Pliesti}, S., Salopek, D., Stani~i}, L., Tomi}, F., Tom{i}, @., Vu~i}, V., 1998: Bioen – Program kori{tenja biomase i otpada: Prethodni rezultati i budu}e aktivnosti. Energetski institut »Hrvoje Po`ar«, Zagreb, p. 180. Domac, J., Beronja, M., Fijan, S., Jelavi}, B., Jelavi}, V., Krajnc, N., Kajba, D., Kri~ka, T., Krstulovi}, V., Petri}, H., Raguzin, I., Risovi}, S., Stani~i}, L., [unji}, H., 2001: Bioen – Program kori{tenja energije biomase i otpada, Nove spoznaje i provedba. Energetski institut »Hrvoje Po`ar«, Zagreb, p. 144.

S. Bogdan et al.

ISO 3131, 1975: International Standard 3131. Wood – Determination of density for physical and mechanical tests. International Organization for Standardization. Johansson, T., 2000: Biomass equations for determining fractions of common and grey alders growing on abandoned farmland and some practical implications. Biomass and Bioenergy 18(2): 147–159. Kajba, D., Bogdan, S., Vratari}, P., 1999: Growth of white willow (Salix alba L.) clones in mixed plantation with black alder (Alnus glutinosa /L./ Gaertn.) on the peat-gley type of soil in Podravina (Croatia). [umarski list 123 (11–12): 523–531. Kajba, D., Krstini}, A., Komlenovi}, N., 1998: Arborescent willow biomass production in short rotation. [umarski list 122 (3–4):1 39–145. Kajba, D., Bogdan, S., Kati~i}–Trup~evi}, I., 2004: White willow biomass production in a short rotation clonal test in Croatia. [umarski list 128 (9–10): 509–515. Kajba, D., Bogdan, S., Kati~i}, I., 2006: Biomass production in willow clonal tests on marginal sites in Croatia. Glasnik za {umske pokuse, Posebno izdanje 5: 261–275. Komlenovi}, N., Krstini}, A., Kajba, D., 1996: Posibilities of biomass production in short rotation with arborescent willows in Croatia. In: Mayer, B. (Ed.), Unapre|enje proizvodnje biomase {umskih ekosustava. Vol 1. Hrvatsko {umarsko dru{tvo. Zagreb, p. 9–23. Korsmo, H., 1995: Weight equations for determining biomass fractions of young hardwoods from natural regenerated stands. Scand. J. For. Res. 10 (1–4): 333–346. Krpan, A. P. B., Por{insky, T., 2004: Efficiency of mechanical felling and processing in soft and hardwood broadleaved stands – part 2: Efficiency of harvesters in the culture of soft broadleaf trees. [umarski list 128 (5–6): 233–244. Krstini}, A., 1994: Genetics of Black Alder (Alnus glutinosa /L./ Gaertn.). Ann. Forest 19 (2): 33–72. Krstini}, A., Kajba, D., 1991: Possibilities of genetic gain for vigorous growth of black alder (Alnus glutinosa /L./ Gaertn.) by clonal seed orchards. [umarski list 114 (6–9): 261–272. Krstini}, A., Komlenovi}, N., 1986: The effect of Black Alder (Alnus glutinosa /L./ Gaertn.) on the growth of White Willow (Salix alba L.) clones. Proc. 18th IUFRO World Congress Division 2. Ljubljana. Vol. II: 436–445. Meeuwissen, T. W., Rottier, H., 1984: Development of alder (Alnus glutinosa) coppice. Netherlands J. of Agric. Sci. 32: 240–242.

Falconer, D. S., Mackay, T. F. C., 1996: Introduction to Quantitative Genetics, Longman Group Ltd., p. 464.

Mitchell, C. P., Matthews, J. D., Proe, M.F., Macbrayne, C. G., 1981: ETSU – B1081a. Biofuels Programme. Dep. of For. Aberdeen Univ. ST. Machar Drive, Aberdeen. Tech. Rep, p. 113.

Hytönen, J., 1996: Biomass production and nutrition of short rotation plantations. Academic dissertation. University of Helsinki, Helsinki, Finland.

Ovington, J. D., 1956: The form weights and productivity of tree species grown in close stands. New Phytologist 55: 289–304.

Croat. j. for. eng. 30(2009)2

109

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Proe, M. F., Craig. J., Griffiths, J., Wilson, A., Reid, E., 1999: Comparison of biomass production in coppice and single stem woodland management systems on an imperfectly drained gley soil in central Scotland, Biomass and Bioenergy 17 (2): 141–151.

SAS Institute Inc., 1989: SAS User’s guide: Statistics 1989 Edition. Cary, NC, USA.

Rytter, L., 1996: The potential of grey alder plantation forestry. In: Perrtu, K., Koppel, A. (Eds.), Short Rotation Willow Coppice for Renewable Energy and Improved Environment. Uppsala, pp. 89–94.

Uri, V., Tullus, H., Lõhmus, K., 2002: Biomass production and nutrient accumulation in short-rotation grey alder (Alnus incana /L./ Moench) plantation on abandoned agricultural land. For. Ecol. and Manage. 161 (1–3): 169–179.

Saarsalmi, A., Mälkönen, E., 1989: Harmaalepikon biomassan tuotos ja ravinteiden käyttö. Summary: Biomass production and nutrient consumption in Alnus incana stands. Folia Forestalia 728: 1–16.

Uri, V., Tullus, H., 1999: Grey alder and hybrid alder as short rotation forestry species. In: Overend, R. P., Chornet, E. (Eds.), Proceedings of the 4th Biomass Conference of Americas, Vol. 1. Oakland, CA, pp. 167–173.

Saarsalmi, A., Palmgren, K., Levula, T., 1992: Haarmaalepän ja raduskoivun biomassan tuotos ja ravinteiden käyttö energiapuuviljelmällä. Summary: Biomass production and nutrient consumption of Alnus incana and Betula pendula in energy forestry. Folia Forestalia 797: 1–29.

Vukeli}, J., Rau{, \., 1998: [umarska fitocenologija i {umske zajednice u Hrvatskoj. [umarski fakultet Zagreb, 195–199.

Saarsalmi, A., 1995: Nutrition of deciduous tree species grown in short rotation stands. Academic dissertation, University of Joensuu, Finland.

Tullus, H., Uri, V., Lõhmus, K., Mander, Ü., Keedus, K., 1998: Halli lepa majandamine ja ökoologia. Tartu. Estonian Agricultural University.

Wittwer, R. F., Stringer, J. W., 1985: Biomass production and nutrient accumulation in seedling and coppice hardwood plantations. For. Ecol. and Manage. 13 (3–4): 223–233. Wright, J. W., 1976: Introduction to Forest Genetics. Academic Press, Inc. New York, p. 463.

Sa`etak

Proizvodnja biomase crne johe (Alnus glutinosa /L./ Gaertn.) u ~istim kulturama i mje{ovitim kulturama s klonovima vrbe (Salix sp.) u Hrvatskoj Energetska politika Hrvatske u posljednja je dva desetlje}a usmjerena prema pove}anju udjela kori{tenja obnovljivih izvora energije u ukupnoj energetskoj bilanci. Zbog toga je pove}an interes za oplemenjivanje brzorastu}ih lista~a podizanjem tzv. energetskih kultura radi proizvodnje biomase u kratkim ophodnjama. Crna joha u tom smislu nije toliko produktivna kao neke vrste vrba ili topola, ali uzimaju}i u obzir nedovoljnu razvijenost trì{ta biomasom u Hrvatskoj i u posljednje vrijeme poja~anu svijesti o va`nosti o~uvanja autohtonoga {umskoga drve}a, ta se vrsta name}e pogodnom za podizanje energetskih kultura. U radu su prikazani rezultati istraìvanja proizvodnje biomase u ~etirima pokusnim kulturama crne johe osnovanim s familijama dobivenim slobodnim opra{ivanjem plus stabala iz klonske sjemenske plantaè iz [umarije \ur|evac. Dvije su pokusne kulture osnovane na lokalitetu Crni jarci ([umarija \ur|evac), od kojih je prva (CJ1) mje{ovita kultura crne johe i razli~itih klonova vrbe, dok je druga ~ista kultura s familijama crne johe (CJ2). Druge su dvije kulture osnovane na lokalitetu Lisi~ine ([umarija Vo}in), od kojih je jedna mje{ovita kultura (L1) crne johe s jednim klonom bijele vrbe (oznake 'V 093'), a druga ~ista kultura crne johe (L2). Sve ~etiri pokusne kulture osnovane su sukladno dizajnu potpunoga blok-sustava sa slu~ajnim rasporedom u ~etiri ponavljanja (RCB dizajn). Prve su dvije kulture (CJ1 i CJ2) bile u dobi od 16 godina, dok su druge dvije (L1 i L2) bile u dobi od 14 godina. Razmak sadnje stabala crne johe u kulturi CJ1 bio je 4,0 ´ 2,0 m (1250 stabala po ha), u kulturi CJ2 2,5 ´ 2,0 m (2000 stabala po ha), a u kulturama L1 i L2 4,0 ´ 1,5 m (1667 stabala po ha). Vrba je u obje mje{ovite kulture (CJ1 i L1) posa|ena izme|u redova crne johe u razmaku sadnje 4,0 ´ 4,0 m (627 stabala po ha). Izmjere visine stabala i njihova promjera na prsnoj visini obavljene su na svim preìvjelim stablima u svim ~etirima pokusnim kulturama, pri ~emu je utvr|eno i preìvljavanje familija crne johe odnosno klonova vrbe. Na temelju izmjerenih visina i prsnih promjera izra~unati su volumeni stabala sukladno Schumacher-Hall-ovim jednad`bama prilago|enih koeficijenata za crnu johu i bijelu vrbu. Biomasa je procijenjena modificiranom metodom odre|ivanja nominalne gusto}e stabala, a na temelju prosu{enih kolutova odrezanih s prsne visine modelnih stabala. Prema utvr|enoj varijabilnosti prsnih promjera izabrana su po ~etiri modelna stabla za svaku pojedinu familiju crne johe i za svaki klon vrbe. Nominalna gusto}a kolutova izra~unata je kao omjer njihove suhe mase i volumena u svjeèm stanju. Na temelju tako izra~unatih nominalnih gusto}a izra~unata je suha biomasa

110

Croat. j. for. eng. 30(2009)2

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99â&#x20AC;&#x201C;112)

S. Bogdan et al.

svakoga pojedinoga stabla kao umnoàk procijenjene nominalne gusto}e i volumena stabala. Budu}i da se volumeni stabala odnose na deblo i grane do 7 cm debljine, tako se i procijenjena biomasa odnosi na suhu biomasu debla i grana do 7 cm debljine. Proizvodnja suhe biomase po hektaru izra~unata je na temelju srednjih vrijednosti biomase stabla, razmaka sadnje i preìvljavanja, zasebno za svaku familiju crne johe odnosno klon vrbe. Statisti~ke analize obuhvatile su deskriptivnu analizu i analize varijance za svojstvo suhe biomase. Analize varijance provedene su s ciljevima ispitivanja signifikantnosti razlika me|u istraìvanim familijama crne johe (zna~ajnost varijance uzrokovane efektom familija) i me|u klonovima vrbe; utjecaja okoli{a na svojstvo (ispitivanjem zna~ajnosti varijanci uzrokovanih efektima blokova i varijance ostatka); ispitivanja zna~ajnosti interakcije istraìvanih familija crne johe s razli~itim tretmanima osnivanja kultura (misli se na tretmane s obzirom na ~istu kulturu ili mje{ovitu kulturu crne johe s klonovima vrbe i na razli~ite razmake sadnje crne johe); ispitivanja zna~ajnosti interakcije istraìvanih familija crne johe s razli~itim stani{nim uvjetima (lokalitetima pokusnih nasada). Statisti~ke su analize omogu}ile izra~unavanje tzv. kvantitativnih genetskih parametara pomo}u kojih je kvantificirana genetski uvjetovana komponenta ukupne varijabilnosti svojstva suhe biomase. Rezultati su pokazali izrazito visoku varijabilnost biomase me|u istraìvanim familijama crne johe, ali i visoku varijabilnost unutar svake familije. Prosje~na vrijednost suhe biomase stabla po pokusnim kulturama kretala se od 21,7 kg u kulturi L1 do 57,6 kg u kulturi L2. Gledano po familijama, prosje~na vrijednost suhe biomase stabla kretala se izme|u 20,7 i 40,1 kg u kulturi CJ1, izme|u 12,5 i 33,1 kg u kulturi L1, 23,1 i 37,6 kg u kulturi CJ2 i na kraju izme|u 37,8 i 70,9 kg u kulturi L2. Promatraju}i izra~unate koeficjente varijabilnosti moè se uo~iti visoka unutarfamilijarna varijabilnost toga svojstva (tablice 2 i 3). Istraìvani klonovi vrbe pokazali su zna~ajno ve}e vrijednosti suhe biomase stabla u odnosu na crnu johu (tablica 4). Procijenjena proizvodnja biomase po hektaru bila je najvi{a u kulturi L2 (87,5 t/ha), a najnià u kulturi CJ2 (27,4 t/ha). Proizvodnja po istraìvanim familijama prikazana je na grafikonima 1 i 2. Proizvodnja suhe biomase istraìvanih klonova vrbe u mje{ovitoj plantaì CJ1 kretala se od 6,0 do 55,5 t/ha (grafikon 3). U rezultatima je usporedno prikazana prosje~na proizvodnja biomase istraìvanih familija crne johe koje su rasle u mje{ovitim kulturama CJ1 i L1 i onih koje su bile zajedni~ke u ~istim kulturama CJ2 i L2 (grafikoni 4 i 5). Prema teorijskim postavkama kvantitativne genetike smatra se da su razlike u nekom kvantitativnom svojstvu me|u razli~itim familijama dobivenim slobodnim opra{ivanjem, a koje rastu u istim okoli{nim uvjetima, uvjetovane genetskim razlikama izme|u maj~inskih stabala. Budu}i da je biomasa tipi~no kvantitativno svojstvo, onda se varijance uzrokovane efektom familija smatraju genetski uvjetovanom komponentom varijabilnosti toga svojstva u istraìvanim kulturama. Komponente varijance uzrokovane efektom familija nisu bile statisti~ki zna~ajne ni u jednoj istraìvanoj kulturi (tablica 5). Najvi{e su vrijednosti imale varijance uzrokovane efektima interakcije familija s blokovima i ostatak. Rezultati kombinirane analize varijance za iste familije crne johe koje su rasle u razli~itom tretmanu (~iste ili mje{ovite kulture) prikazani su u tablici 6. Moè se uo~iti da je tretman bio statisti~ki zna~ajan efekt na lokalitetu Lisi~ine. Interakcija istraìvanih familija crne johe s tretmanom (na~inom osnivanja kultura) bila je statisti~ki zna~ajan efekt na obama lokalitetima. Rezultati kombinirane analize varijance za familije koje su u istom tretmanu uzgajane na razli~itim stani{tima (dva lokaliteta) prikazani su u tablici 7. Stani{ta su bila statisti~ki zna~ajan izvor varijabilnosti, a utvr|ena je i statisti~ki zna~ajna interakcija istraìvanih familija i stani{ta. Na temelju dobivenih rezultata raspravljena je proizvodnja biomase crne johe i vrbe na razini stabla i na jedinici povr{ine. Uzimaju}i u obzir genetsku uniformnost vrbe u pokusnim kulturama, iznena|uje visoka varijabilnost suhe biomase stabala unutar klonova. Vjerojatni je uzrok takve pojave kombinirani utjecaj mikrostani{nih prilika, kao {to su kompeticije s korovnom vegetacijom, {teta od glodavaca i divlja~i i kompeticije sa stablima crne johe. Ve}a proizvodnja biomase klonova vrbe u usporedbi s crnom johom, usprkos manjemu broju stabala po jedinici povr{ine, posljedica je njihovih biolo{kih razlika (brì rast u prvim godinama). Iako je crna joha pokazala zna~ajno manji potencijal u odnosu na vrbu, utvr|ena proizvodnja suhe biomase u okvirima je dosada{njih istraìvanja u Europi i SAD-u. Me|utim, treba naglasiti da je gusto}a stabala po jedinici povr{ine u ovom istraìvanju bila mnogo manja u odnosu na ostale radove. Ta ~injenica govori o znatnoj mogu}nosti pove}anja proizvodnje biomase s ve}om gusto}om sadnje i dodatnom primjenom korisnih uzgojno-za{titinih mjera. Mali udio varijance uzrokovane efektom familija (kao pokazatelj genetske komponente varijabilnosti) vjerojatno je posljedica negativnoga utjecaja mikrostani{ta (u obliku korovne vegetacije, glodavaca, divlja~i i kompeticije izme|u stabala). Iako genetski uvjetovana varijanca nije bila statisti~ki zna~ajna, to ne zna~i da biomasa nije pod genetskom kontrolom, ve} to zna~i da je okoli{ zamaskirao genetske razlike me|u istraìvanim familijama crne johe. U prilog tomu ide i ve}i udio varijance uzrokovane efektom familija utvr|en u kulturi L2 u kojoj su okoli{ni uvjeti bili jednolikiji od ostalih kultura.

Croat. j. for. eng. 30(2009)2

111

S. Bogdan et al.

Biomass Production of Common Alder (Alnus glutinosa /L./ Gaertn.) in Pure Plantations ... (99–112)

Utjecaj na~ina osnivanja kulture na istraìvane familije crne johe (pri ~emu se misli na sadnju u ~istim ili mje{ovitim kulturama odnosno u razli~itim razmacima) potvr|eno je zna~ajan. Ta ~injenica upu}uje na razli~itu sposobnost prilagodbe familija na testirane uvjete. Op}enito je utvr|eno da klonovi vrbe negativno utje~u na rast i biomasu crne johe, ali su neke familije pokazale ve}u proizvodnju upravo u mje{ovitoj kulturi. Ta je pojava zabiljeèna na lokalitetu Crni jarci na kojem je joha sa|ena u razli~itim razmacima pa se stoga bolja proizvodnja nekih familija u mje{ovitoj kulturi moè pripisati i ve}emu razmaku me|u stablima. Utjecaj vrbe na proizvodnju biomase crne johe moè se jasnije vidjeti promatraju}i kulture na lokalitetu Lisi~ine jer je tamo joha posa|ena u jednakim razmacima u objema kulturama. O~it je negativan utjecaj klona vrbe na sve istraìvane familije crne johe. Iako se ta informacija ~ini logi~nom, treba napomenuti da je vrijednost mje{ovitih kultura u dodatnoj visokoj proizvodnji biomase vrbe koja moè podi}i ekonomsku vrijednost, ali i biolo{ku stabilnost takvih kultura (u smislu ve}e biolo{ke raznolikosti). Me|utim, ve} su prija{nja istraìvanja pokazala da vrba agresivno nadvladava crnu johu nakon osme godine starosti kulture. Stoga bi radi ve}e ekonomi~nosti i proizvodnosti iz mje{ovitih kultura trebalo odstranjivati vrbu do osme godine starosti. Rezultati su pokazali zna~ajne interakcije istraìvanih familija crne johe s testiranim lokalitetima, {to upu}uje na razli~itu sposobnost prilagodbe familija na stani{ne uvjete. Iako je stani{te u Lisi~inama bolje kakvo}e, neke su familije pokazale ve}u proizvodnju biomase na lokalitetu Crni jarci. Taj podatak ukazuje na mogu}nost selekcije familija koje su bolje prilago|ene na stani{ta slabije kakvo}e. Klju~ne rije~i: familije, geneti~ki test, raznolikost, interakcija genotip ´ okoli{

Authors' address – Adresa autorâ: Asst. Prof. Sa{a Bogdan, PhD. e-mail: sasa.bogdan@zg.htnet.hr Department of Forest Genetics, Dendrology and Botany Asst. Prof. Mario [por~i}, PhD. e-mail: sporcic@sumfak.hr Department of Forest Engineering Asst. Prof. Ante Seletkovi}, PhD. e-mail: aseletkovic@sumfak.hr Department of Forest Management and Remote Sensing Forestry Faculty of Zagreb University Sveto{imunska 25 HR-10000 Zagreb

Received (Primljeno): April 8, 2009 Accepted (Prihva}eno): November 15, 2009

112

Mladen Ivankovi}, PhD. e-mail: mladeni@sumins.hr Croatian Forestry Institute Cvjetno naselje 41 HR-10450 Jastrebarsko CROATIA Croat. j. for. eng. 30(2009)2

Original scientific paper â&#x20AC;&#x201C; Izvorni znanstveni rad

Analysis of Inclusion of Wood Forwarding into a Skidding Model Matev` Miheli~, Janez Kr~ Abstract â&#x20AC;&#x201C; Nacrtak Mechanized felling was introduced in Slovenia at the beginning of this decade, and before that Slovenian forestry relied on motor manual felling and tractor skidding or cable crane yarding. In this research we are dealing with wood forwarding. Model skidding maps were built, using terrain classification and decision support systems for forwarders and tractor trailers. The purpose of this research is to establish how to plan in advance new individual skidding systems, which involves changing the skidding map. In the first phase, the criteria for the selection of skidding means were determined using multicriterial methodology, and then this model was applied to a study area. In the third stage we conducted a comparison between the reference map provided by the Slovenian Forest Service and the model map resulting from steps one and two. Afterwards it was determined which types of terrain were selected for forwarding and which forms of skidding would face the most serious competition from forwarding. It was determined that the terrain allows for more forwarding than suggested by the reference skidding map. In the research area most of the forest is privately-owned, making it very unlikely that large forwarders would be used, and so tractor trailers were included in the model. Including tractor trailers we have established, that tractor skidding increases marginally at the expense of forwarding, in particular in terrain with higher terrain gradient and more difficult working conditions. Keywords: skidding, forwarding, extraction systems, GIS, skidding map, modelling

1. Introduction â&#x20AC;&#x201C; Uvod Up until the beginning of this decade, when mechanized felling was introduced (Ko{ir 2004a, Kr~ and Ko{ir 2003, Ko{ir 2004b), Slovenian forestry relied on motor manual felling and tractor skidding or cable crane skidding. Before 2000 there were only individual demonstrations of contemporary technologies. Shortwood technologies were used only in salvage cuttings of fire and storm damaged sites in the Ljubljana, Slovenj Gradec and Primorsko areas (Maru{i~ 1998, Ko{ir and Robek 2000, Magajna 2002). After 2000 the number of mechanized felling machines started rising, as did the number of forwarders. Naturally, wood forwarding is about more than just forwarders, the highly-efficient machines whose price matches their features and which are used alCroat. j. for. eng. 30(2009)2

most exclusively by larger forest management companies. Increasingly important elements of wood forwarding are also forestry trailers, which are used primarily by small forest management companies. In Slovenia it has been established (@logar 2007) that forwarders can use the existing skid trails for movement within the stand, while in Croatia tractor assemblies and their ecological suitability was assessed ([u{njar et al. 2008). Ecological suitability in lowland forests was also studied on forwarder Timberjack 1710B (Horvat et al. 2004). The properties of physical working environment and technological systems are connected with different types of terrain classifications. There are two kinds of classification systems, functional and descriptive. Functional systems classify forests in terms of potentials and limitations of technical equipment, but their use is dwindling due to

113

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

the development of computers. Descriptive systems, on the other hand, are becoming increasingly useful: they classify the forest with respect to terrain characteristics that affect the use of harvesting technology. A descriptive terrain classification system must allow for a classification with varying degrees of precision and, consequently, different degrees of generalisation (Segebaden et al. 1967, Samset 1971). There have been numerous advancements in the field of terrain classification in the recent years. There has been research into using GIS for opening forests (Tu~ek 1994) and terrain classification has been used for large scale harvesting operations in northern Velebit (Pentek et al. 2008). McDonagh et al. (2004) report simulating efficiency of harvesting systems using models that include terrain as one of the constraints. Digital elevation models (DEM) are being used for trafficability analysis (Kokkila 2002) and improved terrain classification parameters have been applied to harvesting on sensitive sites (Owende et al. 2002). A very important part of ecologically sound and economically feasible wood skidding in general is also the road network that has to be well planned and maintained (Poto~nik 2005). Application of DSS to terrain classification is not new. A similar method was used to plan large scale timber harvesting (Davis and Reisinger 1990). Models were developed that plan harvesting in respect to erosion and landslides (Adams et al. 2003), evaluate the feasibility of logging systems (Cavali 2006) and also predict harvesting efficiency with respect to changing stand and terrain parameters (McDoangh et al. 2004). Kr~ (1996) used DSS approach to predict skidding means in Slovenia, whereas in Italy a model for selecting skidding means in relation to stand and terrain parameters was built (Lubello 2008).

3. Material and methods – Materijal i metode 3.1 Study area – Mjesto istra`ivanja The study area comprised the Bistra-Borovnica Forest Management Unit, which is located on the western edge of the Forest Management Area. The total area of the study area is 6,170 ha, of which 4,641 ha is forest. In nearly half of the study area, 48.9%, the terrain gradient is lower than 20%, 30.1% of the area falls into the terrain gradient bracket 20–40%, 13.8% has a terrain gradient of 40–60% and the rest has a terrain gradient higher than 60%. On the majority of the unit area (55.8%) stoniness is below 20%, it is at between 20% and 40% on 34.4% of the area, and higher than 40% on 9.8% of the area. Stoniness has been assesed as percent of the section’s surface covered by visible stones. Rockiness is lower than 20% on 47.5% of the unit and 43% of the unit has rockiness of between 20% and 40%. On 9.4% of the area rockiness is above 40%. Rockiness has been evaluated similary to stoniness, and was assesed, as a percent of the section covered by rocks, or boulders, that cannot be moved. Average skidding distance depends primarily on the road infrastructure in forests, which is expressed as road density. In the studied forestry management unit the road density is 18.6 m/ha ([u{ter{i~ et al. 2007). In the model, however, skidding distance was calculated as the average shortest skidding distance

2. Aims – Ciljevi Since almost no wood forwarding has been used in Slovenian forestry in the last 30 years, our foresters are having problems determining which areas are optimal for forwarding. This can also be seen in public forest service map, where only some terrains were selected for wood forwarding. The purpose of this research is to establish how to plan in advance new individual skidding systems, which involves changing the skidding map and defining the types of terrain where forwarding is feasible. We also wanted to determine on which types of terrain tractor skidding will face the most serious competition from wood forwarding.

114

Fig. 1 Terrain gradient in the study area Slika 1. Nagib terena istra`ivanoga podru~ja Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113â&#x20AC;&#x201C;125)

M. Miheli~ and J. Kr~

of the forest section to the road multiplied with the factor of skidding distance, which results in model skidding distance. Skidding distance factor represents the ratio between the length of the actual skidding distance and the average shortest skidding distance (Dobre 1980). Road infrastructure in the study area is shown in Fig. 2.

3.2 Data and model â&#x20AC;&#x201C; Podaci i model In the first phase, the criteria for the selection of skidding means were determined (Kr~ 1996) using multicriterial methodology (Saaty 1990), and then this model was applied to a study area. In the third stage, we conducted a comparison between a reference map provided by the Slovenian Forest Service and a model map resulting from steps one and two. The reference skidding map was compared to the model map. The aim of this comparison was to establish whether there were possibilities to use forwarder in parts of the study area, where the reference map does not predict forwarding. The model predicts suitability in respect to terrain classification

Fig. 2 Road infrastructure in the study area Slika 2. Cestovna mre`a istra`ivanoga podru~ja

Fig. 3 Diagram of the research model Slika 3. Dijagram toka istra`ivanoga modela Croat. j. for. eng. 30(2009)2

115

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

for each type of skidding. A diagram of the model used is presented in Fig. 3. Data on the terrain gradient were obtained from the digital elevation model with a resolution of 25 ´ 25 metres owned by the Surveying and Mapping Authority of the Republic of Slovenia; rockiness, stoniness, bedrock hardness and soil depth are data provided by the Slovenian Forest Service, and they are part of the national forest inventory; distance to forest road and skidding direction are derived from the road network and the digital elevation model (Kr~ and Ko{ir 2008). Since the accuracy and reliability of the model depend primarily on the quality of input data the databases were checked for possible errors. Logical values and data completeness were also checked. In the first phase, the skidding map was produced based on the model for predicting skidding systems. A tried-and-tested skidding model designed in the IDRISI environment was used (Kr~ 1996). This model uses the method of multicriteria evaluation, which is a part of decision support systems (DSS) and determines the optimal skidding technology for each cell in the raster GIS environment. In the model, we have relied on several empirical thresholds that determine which skidding system is more suitable on terrains with certain features. In his work, Kr~ (1996) specifies the thresholds for manual, tractor and cable crane skidding. We have to explain, that the term »tractor skidding« refers to all types of skidders used in Slovenia, which include adapted farm tractors with winch, as well as cable skidders.

Fig. 4 Distribution of standardized value for criteria distance from forest road by determination of suitability of wood forwarding Slika 4. Distribucija standardiziranih vrijednosti udaljenosti od {umske prometnice s obzirom na pogodnost izvo`enja drva The thresholds were summarized, and new ones for forwarding and forestry trailer (Fig. 3) were added using Delphi method (Okoli and Pawlowski 2004). Using this method we have determined that the maximum distance from forest road that still makes forwarding feasible was 1,000 metres, and 800 m for forestry trailer.

Fig. 5 Distribution of standardized value for factors of rockiness and stoniness by determination of suitability of wood forwarding Slika 5. Distribucija standardiziranih vrijednosti faktora stjenovitosti i kamenitosti s obzirom na pogodnost izvo`enja drva 116

Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

M. Miheli~ and J. Kr~

Table 1 Assignment of standardized values to terrain gradient for the establishment of positive correlation with individual forwarding systems Tablica 1. Pridruìvanje standardiziranih vrijednosti nagibu terena radi uspostave povezanosti s pojedinim sustavima izvoènja drva Extraction system Sustav privla~enja drva Forwarder downhill Forvarder niz nagib Forwarder uphill Forvarder uz nagib Forestry trailer downhill Traktorska ekipaà niz nagib Forestry trailer uphill Traktorska ekipaà uz nagib

Terrain gradient – Nagib terena 20–30% 30–40% 40–45% 45–50%

0–5%

5–15%

15–20%

Rockiness at 75% and stoniness at 65% were set as upper bound values still permitting forwarding, while these values were set at 40% and 45% for forestry trailer. For soil depth it was posited that shallow soil is the best for manual skidding, tractor skidding and forwarding, and medium shallow soil less so. Deep soil is inappropriate, but it still allows for cable crane to be used. Table 1 displays thresholds for the selection of forwarding system in relation to terrain gradient, while Table 2 provides the standardized value of bedrock hardness as per individual forwarding system. In the second step, we have applied the basic collected data, and also the data obtained in the course of research using multicriterial evaluation in our study area. Basic data used for multicriterial evaluation were rockiness, stoniness, bedrock hardness and soil depth. The processed data used as input were terrain gradient in percent, distance to forest road and skidding direction (up, down). Following the processing, the data on individual weighing factors (Fig. 3) were merged with multi-

50–60%

60–80%

>80%

criterial evaluation into a single index that was used for comparison between skidding systems. The skidding system with the highest index value was selected. In the third step, we have evaluated the model. This was achieved by comparing the model functional classification of the terrain with the reference map obtained from the Slovenian Forest Service. This required an alignment of the classes on the model map with those on the reference map. The model map has the following classes: manual, tractor downhill, tractor uphill, forwarder downhill, forwarder uphill, cable crane downhill and cable crane uphill. To be able to compare the maps, it was necessary to merge the classes tractor straight and tractor downhill in the Forest Service reference map, and the classes forwarder uphill and downhill on the model map, as the reference map only has the class forwarder. In analyzing the functional classification, it should be noted that in the multicriterial evaluation weight of factors terrain gradient and skidding distance together form almost 70% of weight of all factors incorporated in the model. The analysis of the results of

Table 2 Assignment of standardized values of bedrock hardness for the establishment of positive correlation with individual skidding systems Tablica 2. Pridruìvanje standardiziranih vrijednosti nosivosti podloge radi uspostave povezanosti s pojedinim sustavima izvoènja drva Extraction system Sustav privla~enja drva Forwarder downhill Forvarder niz nagib Forwarder uphill Forvarder uz nagib Forestry trailer downhill Traktorska ekipaà niz nagib Forestry trailer uphill Traktorska ekipaà uz nagib

Croat. j. for. eng. 30(2009)2

Hard – Tvrdo

Bedrock hardness – ^vrsto}a podloge Hard/soft – Osrednje

Soft – Meko

117

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

functional classification will therefore focus on the terrain gradient with a weight of 41% and skidding distance, which had a weight of 27%. The differences between the model and the reference map were compared using CROSSTAB. CROSSTAB performs a cross-tabulation analysis that compares images containing categorical variables of two types. Since the results of this module are output in the number of cells, it was also necessary to calculate cell size (0.0625 ha).

4. Results and discussions – Rezultati i rasprava The comparison of the functional classifications shows that in the model classification, manual skidding is limited to extreme terrain gradients, as opposed to the reference classification, where the average terrain gradient is 48.22%. In both cases the average skidding distance is similar – 239.8 metres in the reference classification and slightly more, 257.7 metres, in the model classification. In the model classification manual skidding is assigned only to truly remote areas with a high terrain gradient. Individual classes of the reference functional classification have the following average weighing factor values. This was followed by an analysis of the model functional classification in terms of value of weighing factors. The individual classes of the model functional classification have the following average weighing factor values.

Terrain gradients are quite high in cable crane skidding as well, averaging 47.67% for downhill skidding in the reference classification and substantially more, 68.4%, in the model classification. The distance for downhill skidding is 186.8 m in the reference classification and 380.6 m in the model classification. In cable crane skidding uphill the average terrain gradients are similar in both classifications, 51.4% in the reference and 52.4% in the model classification, whereas the skidding distance is somewhat higher in the model classification, at 180 m. The comparison of tractor skidding shows that the model classification assumes higher average terrain gradients (43.2%) than the reference classification (28.1%). The reasons for that might lie in the fact that model classification designated a greater area for the forwarder and left only areas with the highest terrain gradient for tractor skidding. The average skidding distance is slightly longer in the model classification, at 264 m, compared to 226.3 m in the reference functional classification. In tractor skidding uphill the terrain gradient is similar (22.1% in the reference classification and 27.7% in the model classification). The model classification also determines shorter skidding distances (136.4 m) than the reference classification (176.6 m). There is more forwarding in the model classification and at 19.1% the average terrain gradient is much lower than in the reference classification (24.9%). The model classification attributes to forwarding signifi-

Table 3 Classes of reference functional classification Tablica 3. Razredi referentne namjenske klasifikacije terena Extraction system Sustav privla~enja drva

Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forwarder Forvarder Cable yarder downhill @i~ara niz nagib Cable yarder uphill @i~ara uz nagib

118

Terrain gradient Nagib terena Mean SE Arit. sred. St. pogr. %

Distance to forest road Udaljenost do {um. ceste Mean SE Arit. sred. St. pogr. m

Stoniness Sjenovitost Mean SE Arit. sred. St. pogr. %

Rockiness Kamenitost Mean SE Arit. sred. St. pogr. %

Soil depth Dubina tla Mean SE Arit. sred. St. pogr. class – razred

48.2

17.87

239.8

182.32

20.4

6.04

21.3

6.56

1.4

0.49

28.1

18.06

226.3

179.27

25.4

18.23

24.4

16.51

2.1

0.74

22.1

15.25

176.6

161.13

29.6

21.61

28.2

19.55

2.3

0.72

24.9

14.27

262.8

193.89

19.7

13.21

19.8

12.69

3.0

0.53

47.7

14.89

186.8

147.55

14.9

8.47

9.1

8.13

1.9

0.33

51.4

14.21

154.1

98.45

17.8

3.25

17.7

6.30

1.4

0.58

Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113â&#x20AC;&#x201C;125)

M. Miheli~ and J. Kr~

Table 4 Classes of model functional classification Tablica 4. Razredi funkcionalne klasifikacije terena modela Extraction system Sustav privla~enja drva Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forwarder Forvarder Cable yarder downhill @i~ara niz nagib Cable yarder uphill @i~ara uz nagib

Terrain gradient Nagib terena Mean SE Arit. sred. St. pogr. %

Distance to forest road Udaljenost do {um. ceste Mean SE Arit. sred. St. pogr. m

Stoniness Sjenovitost Mean SE Arit. sred. St. pogr. %

Rockiness Kamenitost Mean SE Arit. sred. St. pogr. %

Soil depth Dubina tla Mean SE Arit. sred. St. pogr. class â&#x20AC;&#x201C; razred

115.2

114.74

257.7

204.77

21.5

17.34

7.6

14.17

2.0

0.00

43.2

12.90

264.0

193.23

19.2

10.89

18.7

10.11

1.9

0.88

27.7

9.62

136.4

138.22

19.4

12.31

18.9

11.76

2.3

0.75

19.1

10.16

206.8

162.61

28.4

20.75

27.4

18.83

2.4

0.65

68.4

11.51

380.6

235.45

22.8

10.21

22.2

10.30

1.2

0.63

52.4

12.45

180.0

153.22

18.6

7.40

18.3

9.28

10.6

0.79

cantly shorter average skidding distances (206.8 m) than the reference classification (262.8 m). Fig. 6 shows the reference skidding map made by the Forest Service (left) and the model of skidding systems for the study area (right). The image shows a skidding model complemented with forwarding.

We have compared the two maps using Idrisi module Crosstab. The results of the comparison are presented in Tables 5 and 6. The model was designed to establish which skidding system would be best for the given terrain. It was established that forwarding appears in areas

Fig. 6 Reference map of skidding systems (Forest Service) and model for skidding with forwarder Slika 6. Referentna karta sustava privla~enja drva (Javna {umarska slu`ba) i karta modela za izvo`enje drva forvarderom Croat. j. for. eng. 30(2009)2

119

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

Reference map – Referentna karta

Table 5 Overlap between the model map and the reference map for the use of forwarder* Tablica 5. Preklapanje izme|u karte modela i referentne karte za uporabu forvardera*

Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forwarder Forvarder Yarder downhill @i~ara niz nagib Yarder uphill @i~ara uz nagib Total Ukupno

Model map – Karta modela Skidder uphill Forwarder Yarder downhill Skider uz nagib Forvarder @i~ara niz nagib

Manual Ru~no

Skidder downhill Skider niz nagib

Yarder uphill @i~ara uz nagib

Total Ukupno

0.0

0.9

0.1

0.0

1.13

27.3

551.1

51.5

81.3

51.0

9.4

771.56

3.1

138.4

71.1

28.0

5.6

7.1

253.25

12.1

1682.9

554.4

313.3

21.1

12.7

2596.50

9.2

90.6

4.6

5.4

14.8

4.9

129.50

8.9

137.0

61.3

10.7

32.4

95.6

345.81

60.56

2600.88

742.94

438.81

124.88

129.69

4097.75

* the quantities are in hectares – vrijednosti izra`ene u hektarima

ture, growing stock, proximity of off-limits areas (military installations), existing infrastructure and microrelief features. These factors have not been included in the model because the relations between the factors and their suitability for different skidding systems have not been determined yet. Including these factors into the model is to be the subject of future research. Differences between the models do af-

with a lower average terrain gradient than tractor skidding, which is shown in Fig. 7. Average skidding distances are shorter than those assigned to forwarding in the reference classification. The reference skidding map shows the current use of skidding systems, which depend on a number of factors. Most of the factors had not been included in the model, for example forest ownership struc-

Reference map – Referentna karta

Table 6 Overlap between the model map and the reference map for the use of forwarder* Tablica 6. Preklapanje izme|u karte modela i referentne karte za uporabu forvardera*

Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forwarder Forvarder Yarder downhill @i~ara niz nagib Yarder uphill @i~ara uz nagib

Model map – Karta modela Skidder uphill Forwarder Skider uz nagib Forvarder

Manual Ru~no

Skidder downhill Skider niz nagib

Yarder downhill @i~ara niz nagib

Yarder uphill @i~ara uz nagib

0.0

81.8

9.1

0.0

3.5

71.4

6.7

10.5

6.6

1.2

54.6

28.1

11.1

2.2

2.8

0.5

64.8

21.4

12.1

0.8

0.5

7.1

70.0

3.6

4.2

11.4

3.8

2.6

39.6

17.7

3.1

9.4

27.6

* the quantities are in percents – vrijednosti izra`ene u postotcima

120

Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113â&#x20AC;&#x201C;125)

M. Miheli~ and J. Kr~

Fig. 7 Terrain gradient for individual skidding systems in reference and model classifications Slika 7. Nagibi terena za primjenu pojedinih sustava privla~enja drva u referentnoj karti i karti modela fect the results of the comparison analysis. The purpose of the constructed model, however, was to consider the possibility of wood forwarding in regard to terrain conditions, which is why the comparison analysis can be conducted. The results of the skidding model with respect to the terrain have shown that the model designates substantially larger areas to forwarding than the reference skidding map. However, it is also true that the reference skidding map considers skidding in combination with mechanized felling and, consequently, with specific stand, ownership, environmental and technical-organisational circumstances. The conclusion for manual skidding is that the model and the reference map do not overlap. The fact is that the most important weighing factor in the model is the terrain gradient; there are few extreme slopes in the Bistra-Borovnica Forest Management Unit, which is why the model determined much fewer areas suitable for manual skidding than the reference map. On the reference map meanwhile, manual skidding is conditional on several factors, primarily microrelief features, which cannot be determined with a digital elevation model with a resolution of 25 Â´ 25 metres. The analysis of tractor skidding uphill and downhill is particularly interesting, as tractor skidding is present on the majority of areas of the reference map, whereas the model shows that much more of the area is suitable for forwarding: in the model forwarding took up 64.7% of the area previously deemed suitable for tractor skidding downhill, and Croat. j. for. eng. 30(2009)2

21.4% of the area previously assigned to tractor skidding uphill. The terrain which was classified as tractor skidding in both classifications was additionally analyzed. In tractor skidding downhill it was established that the terrain was relatively flat, with 80% of

Fig. 8 Skidding by system in the reference and model classifications depending on average terrain gradient Slika 8. Sustavi privla~enja drva pri referentnoj i oblikovanoj razredbi ovisno o prosje~nom nagibu terena 121

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

Reference map – Referentna karta

Table 7 Overlap between the model map and the reference map for the use of tractor trailer* Tablica 7. Preklapanje izme|u karte modela i referentne karte za uporabu traktorskih ekipa`a*

Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forestry trailer Traktorska ekipa`a Yarder downhill @i~ara niz nagib Yarder uphill @i~ara uz nagib

Model map– Karta modela Skidder uphill Forwarder Skider uz nagib Forvarder

Manual Ru~no

Skidder downhill Skider niz nagib

Yarder downhill @i~ara niz nagib

Yarder uphill @i~ara uz nagib

0.0

0.3

0.2

0.0

17.4

379.5

45.3

62.7

41.7

7.0

5.1

450.8

318.6

90.9

8.3

10.0

8.6

1354.4

295.7

258.2

18.9

7.6

18.2

251.9

15.4

13.0

22.9

7.3

9.4

153.1

65.5

11.7

32.9

97.3

* the quantities are in hectares – vrijednosti izra`ene u hektarima

the area falling into the terrain gradient bracket 15–30%. The 15–20% and the 25–30% brackets took up 20% of the area each and the 20–25% bracket 40% of the area. Distance to forest road is below 300 m on 70% of the area; extending the distance to 500 m captures as much as 84% of the area. Stoniness is relatively low, as 73% of the area has a stoniness of between 0% and 20% and another 15% of the area falls into the 20–30% bracket. Rockiness is similar: the 0–20% bracket covers 72% of the area and the 20–30%

bracket another 20%. Most downhill skidding (84.5%) is carried out on hard bedrock and 88% on very shallow or shallow soil. The terrain that produced the selection of tractor skidding uphill in both classifications has a terrain gradient of 20–25% on 42.8% of the area and a more moderate 15–20% on 36.3% of the area. 15% of the skidding is done on terrain gradients lower than 15%. Skidding distances are shorter than in downhill skidding, remaining below 100 metres on as much as

Reference map – Referentna karta

Table 8 Overlap between the model map and the reference map for the use of tractor trailer* Tablica 8. Preklapanje izme|u karte modela i referentne karte za uporabu traktorskih ekipa`a*

Manual Ru~no Skidder downhill Skider niz nagib Skidder uphill Skider uz nagib Forestry trailer Traktorska ekipa`a Yarder downhill @i~ara niz nagib Yarder uphill @i~ara uz nagib

Model map – Karta modela Skidder uphill Forwarder Skider uz nagib Forvarder

Manual Ru~no

Skidder downhill Skider niz nagib

Yarder downhill @i~ara niz nagib

Yarder uphill @i~ara uz nagib

0.0

70.0

30.0

0.0

3.1

68.5

8.2

10.2

7.5

1.3

0.6

51.0

36.0

10.3

0.9

1.1

0.5

69.7

15.2

13.3

1.0

0.4

5.5

76.6

4.7

4.0

7.0

2.3

2.5

41.4

17.7

3.2

8.9

26.3

* the quantities are in percents – vrijednosti izra`ene u postotcima

122

Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

61% of the area; only 15.0% of the skidding is done on distances longer than 200 m. The entire area has hard ground and most of the skidding is done on shallow (37.6%) and medium soil (43.5%). On over 70% of the area rockiness and stoniness are below 20%. In cable crane skidding the overlap between the model and the reference map is relatively small, 73.7% for uphill cable crane skidding and only 11.9% for downhill cable crane skidding. In the model classification a large part of both types of cable crane skidding is taken over by tractor skidding: 40.8% of downhill cable crane skidding and 7.3% of uphill use of the cable crane. There are several reasons, including the relative openness of the unit, small share of deep soil on which the cable crane is the only option, and the relative flatness of the unit, which further reduces the area appropriate for the cable crane. Fig. 8 merges uphill and downhill skidding, showing only skidding by system in order to make more clear the terrain gradient at which forwarding takes over in the reference model, and the fact that in the descriptive classification tractor skidding is present only in terrain that has higher terrain gradients than terrain suitable for forwarding. It was also interesting how the model map changed if forestry trailer was used instead of forwarder. The parameters were changed, and the model – this time for the forestry trailer – compared to the reference map. The results, in Tables 7 and 8 show a small increase of surface assigned to tractor skidding and a minor decrease of surfaces assigned to forestry trailer, while values for cable crane skidding remain at the same level.

5. Conclusion – Zaklju~ci It was determined that the terrain allows for more forwarding than suggested by the reference skidding map. The biggest change was observed in tractor skidding: in the model forwarding it took up to 64.7% of the area previously designated for tractor skidding downhill and 21.4% of the area previously designated for tractor skidding uphill. A more detailed analysis has led to the conclusion that the optimal terrain for tractor skidding downhill is on terrain gradient of below 30% and over a skidding distance of up to 300 m (a distance of up to 500 m is still acceptable). Stoniness and rockiness should be low, at less than 20%, the ground should be hard and the soil shallow. Tractor skidding uphill would be best at terrain gradients of up to 25% and at shorter skidding distances, below 200 m. The ground should be hard, with rockiness and stoniness not exceeding 20%. Croat. j. for. eng. 30(2009)2

M. Miheli~ and J. Kr~

A comparison of cable crane skidding and forwarding shows that the model classification assumes higher average terrain gradients for cable crane skidding (43.19%) than the reference classification, where the average terrain gradient is 28.14%. The reasons may lie in the fact that the model classification designated more areas to skidding with forwarder, leaving only areas with higher terrain gradient for cable crane skidding. The average skidding distance is higher in the model classification (264.03 m) than in the functional classification (226.26 m). The terrain gradient for cable crane skidding uphill is comparable, averaging 22.09% in the reference classification and 27.67% in the model classification. At 136.55 m, the model classification determines shorter skidding distances than the reference classification (176.71 m). In our study area most of the forest is privately-owned, making it very unlikely for large forwarders to be used. The fact is that cooperation between forest owners is very poorly developed in Slovenia, which means that the felling and skidding will be done by small contractors for whom a forestry trailer is a more accessible and economical option than a forwarder. This is the reason why forestry trailer was included in the model. The results show that there is a marginal increase of tractor skidding and a decrease of surface assigned to forestry trailer. Terrains reassigned from forestry trailer to tractor skidding have higher terrain gradients and more difficult working conditions. The model described is very dependent on input data and the suitability criteria, which define individual skidding systems. With future improvements of digital elevation models, the use of LIDAR and inclusion of suitability criteria for other factors into the model we expect higher accuracy of the model results. It would be interesting to compare the results of this model with the model using DEM with a higher resolution. Inclusion of ecological constraints, for instance Natura 2000, and functions of the forests remain challenges for the future.

6. References – Literatura Adams, J. D., Visser, R. J. M., Prisley, S. P., 2003: Modeling Steep Terrain Harvesting Risks Using GIS. Precision forestry, Proceedings of the second international precision forestry symposium, Seattle, Washington, June 15–17, 2003: 99–108. Davis, C. J., Reisinger, T. W., 1990: Evaluating Terrain for Harvesting Equipment Selection. International journal of forest engineering 2(1): 9–16. Cavalli, R., Grigolato, S., Lubello, D., 2006: Planning logging systems through site analysis. IUFRO Precision For-

123

M. Miheli~ and J. Kr~

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

estry Conference 2006, Precision Forestry in plantations, semi-natural and natural forests, March 5–10, 2006, Stellenbosch University: 319–330. Dobre, A., 1980: Odprtost gozdov v Sloveniji. In{titut za gozdno in lesno gospodarstvo, Ljubljana, 142 p. Horvat, D., Por{insky, T., Krpan, A., Pentek, T., [u{njar, M., 2004: Suitability evaluation of forwarders based on morphological analysis. Strojarstvo 46 (4–6): 149–160. Kokkila, M., 2002: Digital elevation models (DEM) in trafficability analysis. ECOWOOD studies. University of Helsinki, 17 p. Ko{ir, B., Robek, R., 2000: Characteristics of the stand and soil damage in cut-to-length thinning on the @ekanc working site (SW Slovenia). Zb. gozd. lesar. 62: 87–115. Ko{ir, B., 2004a: Work efficiency in mechanized cutting. Gozd. vestn. 62(1): 19–24. Ko{ir, B., 2004b: Operational planning in mechanized cutting. Gozd. vestn. 62 (1): 25–31. Kr~, J., Ko{ir, B., 2003: The suitability evaluation of cut-to-length in Slovenia in view of terrain and stand. Zb. gozd. lesar. 71: 5–18. Kr~, J., 1996: A model of timber skidding predicting. Proceedings »Planning and implementing forest operations to achieve sustainable forests« 19th Annual Meeting of COFE& IUFRO SGS3.04–00, July 29–August 1, 1996, Marquette, Michigan USA: 277–282. Kr~, J., Ko{ir, B., 2009: Predicting Wood Skidding Direction on Steep Terrain by DEM and Forest Road Network Extension. Croat. j. for. eng. 29 (1): 1–12. Lubello, D., 2008: A rule-based SDSS for integrated forest harvesting planning. University of Padova, Dissertation: 213 p. Magajna, 2002: Nekatere predhodne gozdnogojitvene usmeritve pri uvajanju strojne se~nje v Sloveniji. Strojna se~nja v Sloveniji, Ljubljana, Gospodarska zbornica Slovenije, Zdru`enje za gozdarstvo: 33–47.

Dynamic Model. Southern Journal of Applied Forestry 28(2): 91–99. Okoli, C., Pawlowski, S. D., 2004: The Delphi method as a research tool: an example, design considerations and applications. Information & Management 42: 15–29. Owende, P. M. O., Lyons, J., Haarla, A. R., Peltola, A., Spinelli, R., Molano, J., Ward, S. M., 2002: Operations protocol for Eco-efficient Wood Harvesting on Sensitive Sites. Project ECOWOOD, Funded under the EU 5th Framework Project: 1–74. Pentek, T., Por{insky, T., [u{njar, M., Stanki}, I., Neve~erel, H., [por~i}, M., 2008: Environmentally Sound Harvesting Technologies in Commercial Forests in the Area of Northern Velebit – Functional Terrain Classification. Period. biol. 110(2): 127–135. Poto~nik, I., Yoshioka, T., Miyamoto, Y., Igarashi, H., Sakai H., 2005: Maintenance of forest road network by natural forest management in Tokyo University Forest in Hokkaido. Croat. j. for. eng. 26(2): 71–78. Saaty, T. L., 1990: How to make a decision: The Analytic Hierarchy Process. European Journal of Operational Research 48: 9–26. Samset, I., 1971: Classification of terrain and operational systems. Joint FAO/ECE/ILO Committee on Forest Working Techniques and Training of Forest Workers, Symposium on forest operations in mountainous regions, 30 aug. – 11 sept. 1971, Krasnodar: 22 p. Segebaden, G. von, Stromnes, R., Winer, H. I., 1967: Proposal for an international system of terrain classification. Proceedings of the XIV IUFRO world congress, III (31–32). [u{njar, M., Horvat, D., Kristi}, A., Pandur, Z., 2008: Morphological analysis of forest tractor assemblies. Croat. j. for. eng. 29(1): 41–51. [u{ter{i~, B., Jeklar, A., Matja{i~, D., Jurjev~i~, B., 2007: Forest management plan of the Bistra – Borovnica management unit 2006–2015. Zavod za gozdove Slovenije, Obmo~na enota Ljubljana, Ljubljana: 175 p.

Maru{i~, J., 1998: Cutting time study of Timberjack harvester FMG 1270 in Forest Management Unit Ravnik. Graduation thesis, Ljubljana: 38 p.

Tu~ek, J., 1994: Using the GIS environment for opening-up forests. Proceedings of the international IUFRO, FAO, FEI seminar on forest operations under mountainous conditions, Harbin, China: 58–65.

McDonagh, K. D., Meller, R. D., Visser, R. J. M., McDonald, T. P., 2004: Harvesting System Simulation Using a Systems

@logar, J., 2007: Suitability of tractor skid roads for forwarding. Graduation thesis, Ljubljana: 57p.

Sa`etak

Analiza uklju~ivanja izvoènja drva forvarderom i traktorskom ekipaòm u model privla~enja drva Strojna se sje~a u Sloveniji po~ela primjenjivati po~etkom ovoga desetlje}a. Prije toga slovensko se {umarstvo oslanjalo na ru~no-strojnu sje~u i privla~enje traktorima ili {umskim ì~arama. Ovo se istraìvanje odnosi na

124

Croat. j. for. eng. 30(2009)2

Analysis of Inclusion of Wood Forwarding into a Skidding Model ... (113–125)

M. Miheli~ and J. Kr~

prou~avanje izvoènja drva forvarderima. Oblikovane su karte na~ina privla~enja uz primjenu razredbe terena i sustava podr{ke odlu~ivanju za forvardere i traktorske ekipaè. Namjena je ovoga istraìvanja uspostava i dono{enje planova za pojedine sustave pridobivanja drva, {to uklju~uje izmjene postoje}ih karata na~ina primarnoga transporta drva uz determinaciju onih terena koji su pogodni za rad forvardera. Istraìvanje je obavljeno u gospodarskoj jedinici Bistra-Borovnica koja je smje{tena u zapadnom dijelu slovenskoga {umskogospodarskoga podru~ja. Povr{ina istraìvanoga podru~ja iznosi 6170 ha. Izra|en je digitalni model reljefa razlu~ivosti 25 ´ 25 metara. Podaci o kamenitosti, stjenovitosti, nosivosti podloge i dubini tla pribavljeni su kao dio nacionalne inventure {uma od Javne {umarske slu`be Slovenije. Udaljenost od {umske ceste i smjer privla~enja dobiveni su iz mreè {umskih prometnica i digitalnoga modela reljefa. Gotovo polovica povr{ine istraìvanoga podru~ja (48,9 %) ima nagib manji od 20 %, 30,1 % pripada u drugu kategoriju nagiba (20–40 %), dok su ve}i nagibi zastupljeni na manjim povr{inama. Kamenitost je utvr|ivana u postotku povr{ine prekrivene kamenom. Ve}ina povr{ine gospodarske jedinice ima kamenitost ispod 20 %, u razredu izme|u 20 % i 40 % kamenitosti nalazi se 34,4 % povr{ine, a u razredu preko 40 % kamenitosti 9,8 % ukupne povr{ine gospodarske jedinice. Stjenovitost je razmatrana sli~no kao i kamenitost, kao postotni odnos ukupne povr{ine prekrivene stijenama koje se ne mogu pomicati. Manja je od 20 % na 47,5 % povr{ine, drugi razred (20–40 %) obuhva}a 43 %, a tre}i (<40 %) obuhva}a 9,4 % istraìvane povr{ine. Ulazni kriterij za odabir na~ina privla~enja odre|en je na osnovi vi{ekriterijske metodologije te primijenjen na istraìvanom podru~ju. Potom su uspore|ene referentna karta Javne {umarske slu`be Slovenije i karta modela. Naposljetku se pristupilo odabiru terenâ pogodnih za izvoènje drva i odre|ivanju koji se ostali na~ini primarnoga transporta drva najozbiljnije suo~avaju s tim oblikom transporta. Ustanovljeno je da svojstva terena dopu{taju izvoènje drva na ve}oj povr{ini nego {to je to prikazano referentnom kartom. Najve}e su promjene pri privla~enju drva traktorima, gdje je 64,7 % povr{ine prvotno namijenjene privla~enju nizbrdo i 21,4 % povr{ine namijenjene privla~enju uzbrdo oblikovanjem pripalo izvoènju drva. Uspore|uju}i podru~ja koja su namijenjena izno{enju ({umske ì~are) s ostalim na~inima primarnoga transporta, utvr|eno je da oblikovani model pretpostavlja ve}e nagibe terena za rad {umskih ì~ara od one prikazane referentnom kartom. Privla~enje drva traktorom uz nagib najprihvatljivije je kod nagiba terena do 25 % i na kra}im udaljenostima privla~enja. Srednje udaljenosti privla~enja ve}e su kod funkcionalne klasifikacije terena (264 m) nego kod one dobivene iz referentne karte gdje iznose 226 metara. Oblikovanjem je utvr|eno da se na istraìvanom podru~ju izvoènje moè primijeniti na ve}oj povr{ini nego {to je to prikazano referentnom kartom Javne {umarske slu`be Slovenije. [ume su istraìvanoga podru~ja ve}inom u privatnom vlasni{tvu, a to ~ini izvoènje drva te{kim forvarderima malo vjerojatnim za primjenu. Treba o~ekivati da }e se sje~a i izradba odvijati ru~no-strojno primjenom motorne pile lan~anice, a izvoènje drva traktorskim ekipaàma ~iji su tro{kovi rada manji nego kod forvardera. Naveden je i osnovni razlog zbog ~ega su traktorske ekipaè uvedene u model. Klju~ne rije~i: vu~a drva, izvoènje drva, sustavi privla~enja drva, GIS, karta privla~enja drva, modeliranje

Authors' address – Adresa autorâ:

Received (Primljeno): February 3, 2009 Accepted (Prihva}eno): November 15, 2009 Croat. j. for. eng. 30(2009)2

Matev` Miheli~, MSc. e-mail: matevz.mihelic@bf.uni-lj.si Assist. Prof. Janez Kr~, PhD. e-mail: janez.krc@bf.uni-lj.si University of Ljubljana, Biotechnical Faculty Department of Forestry and Forest Resources Ve~na pot 83 1000 Ljubljana SLOVENIA

125

Original scientific paper – Izvorni znanstveni rad

Forces Affecting Timber Skidding @eljko Toma{i}, Marijan [u{njar, Dubravko Horvat, Zdravko Pandur Abstract – Nacrtak The research was carried out on the strip road where longitudinal slopes were determined by the leveling method as well as individual distances of the strip road of uniform slope. Skidders were used for carrying out traction tests in downhill and uphill skidding of 9 different loads. Measurement of wheel loads, wheel torques and components of rope force was performed by tensiometric method and remote transferring of measurement signal. According to the results of research, by the increase of longitudinal slope, more tractive force is used for overcoming the terrain slope than for overcoming the traction resistance. In downhill skidding, the horizontal component of the skidder weight (G sina) acts in the skidder travel direction and hence its value is higher than the traction resistance. Torque distribution depends on the skidder wheel vertical load. In uphill skidding, torques increase proportionally to the vertical component of rope forces and adhesive weight of the skidder. In downhill skidding, the skidder wheel torques are negative, because they are not used for achieving wheel tractive force and instead, the transmission of torque through the transmission system causes skidder’s braking performance. The skidder’s need for braking arises under influence of the horizontal component of the skidder weight (G sina) that acts in the direction of the skidder travel and under its effect the traction resistances are overcome. It can be concluded that in case of downhill skidding we cannot speak of achieving real tractive force because the skidder pulls the load only by its weight, and the transmittion from engine to wheels is not used for achieving the tractive force. Key words: skidder, wheel load, wheel torque, pulling force, adhesive weight, slope

1. Introduction – Uvod Skidders, as forest vehicles for timber skidding, are exclusively designed for achieving tractive force through wheel circumference. By the skidder transmission system, torque is transmitted and changed from the drive engine to the wheels. Under the effect of torque, thrust force is generated on the wheel. Horizontal component of the thrust force is partly used for overcoming the rolling resistance of the vehicle (Ff), and the remaining force (Fv) is used for pulling the load, for overcoming the slope and surface obstacles or for vehicle acceleration (Wong 2001, Stoilov 2007). When using skidders equipped with forest winch, timber is extracted with one end of the load lifted off the ground and through the winch rope leaned on the rear part of the vehicle, while the other part of the load is dragged on the soil. The force generated in the rope is used for pulling the part of the load lifted off the ground (vertical component – V) and for Croat. j. for. eng. 30(2009)2

overcoming the traction resistance of the part of load weight dragged on the soil (horizontal component – H). Skidder’s traction characteristic affects the ratio between forces applied on the wheel and forces resisting their action, where the adhesive weight of the vehicle ([u{njar and Horvat 2006) plays an extremely important role. Adhesive weight (Ga) is the sum of vertical loads on skidder driving wheels under conditions of timber harvesting (Toma{i} et al. 2007). Adhesive weight depends on the skidder mass, terrain slope and value of vertical component of rope force, which is primarily affected by the value and orientation of pieces of timber in the load: Ga = G · cosa + V Consequently, adhesive weight differs from the weight of the unloaded skidder (G) because the rear axle is additionally loaded by full vertical component of rope force, which is distributed on rear wheels

127

@. Toma{i} et al.

Fig. 1 Distribution of forces during uphill timber skidding Slika 1. Raspodjela sila pri privla~enju skiderom uz nagib

Forces Affecting Timber Skidding (127–139)

to overcome the traction resistance of the part of load dragged on the soil (H). Based on the above considerations, the highest load is expected at the skidder’s rear axle during uphill skidding. Due to different wheel loads, transmission must enable the distribution of the torque with respect to the wheel load. Consequently with the mechanical transmission system, the distribution of the torque to the wheels should be in accordance with the wheel load distribution. The research of torque on skidder’s wheel shafts (Horvat 1987, [u{njar and Horvat 2006, Maren~e 2005) confirmed the hypothesis that higher torque is applied on the wheel under higher load, where torques at the wheels of the same shaft are balanced. The aim of this paper is to determine the dependence of components of rope force, wheel load distribution and torque on load weight and terrain slope, which represents a significant correlation for the assessment of tractive (exploitation) characteristics of skidders in timber skidding.

through horizontal winch rollers (Hassan and Gustafson 1983). The value of tractive force depends on the adhesive weight of the skidder and hence for obtaining higher tractive force higher adhesive weight is necessary (Sever 1984). Therefore, in timber skidding, the heavier end of timber assortments in the load should be lifted off the ground so as to increase the adhesive weight by higher value of vertical component of rope force. The distribution of loads by skidder axles changes with respect to the load volume and weight, orientation of assortments in the load, travel direction and terrain slope. During travel on longitudinal slope, the rear axle is additionally loaded as the load of the skidder weight is transferred from the front axle due to the effect of the parallel weight component of the skidder (G sina). During uphill timber skidding, the tractive force must overcome the traction resistance of the part of load dragged on the soil (H), as well as the resistance of the horizontal weight component of the skidder (G sina) that pulls the vehicle downhill (Fig. 1). During downhill travel, under the effect of the horizontal weight component of the skidder (G sina), the load of the skidder weight is transferred to the front axle (Fig. 2). In the same way, the horizontal weight component of the skidder acts in the skidder’s travel direction and hence it is only necessary

In this research, the skidder ECOTRAC 55 V equipped with forest winch Hittner 2 ´ 35 kN (Fig. 3) was used. The skidder mass is 3483 kg (62% at the front axle and 38% at the rear axle). The skidder’s driving engine is a 3-cylinder DEUTZ diesel engine, air cooled, displacement 3236 cm3, compression ratio 20:1, power output 40 kW at 2300 min–1 and highest torque of 207 Nm at 1600 min–1. The transfer of power is carried out by mechanical transmission: drive engine ® clutch ® gear box ® drive distribution ® front and rear differentials with individual blockade ® final (planetary) drive in tractor wheels. For measuring the dynamic load of the skidder, measuring parameters were designed or applied for simultaneous determination of components of rope force, load and torque on all driving wheels.

Fig. 2 Distribution of forces during downhill timber skidding Slika 2. Raspodjela sila pri privla~enju skiderom niz nagib

Fig. 3 Mass distribution of skidder Ecotrac 55 V Slika 3. Raspored mase skidera Ecotrac 55 V

128

2. Materials and methods – Materijali i metode

Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

Tensometry is the basic method for measuring mechanical values. Heidl and Husnjak (1992) describe tensometry as a mechanical method for determining length deformation of a structure or model in order to determine the strain on the structure surface. In doing so, measurement transducers are used based on changeable electrical resistance caused by the change of its legth (the so-called »strain gauge«). This method provides the possibility to measure electrically non-electric values. The application of tensometric method enables the measurement without affecting adversely the vehicle structure, but it requires the conversion of vehicle elements into measurement parameters. Tensometric method was used by Sever (1987), Marklund (1987), Horvat (1998), Maren~e (2005) and [u{njar and Horvat (2006), Toma{i} et al. (2008) for measuring wheel torques and wheel load of skidders and forwarders. Horizontal and vertical component of rope force in timber skidding was determined with two transducers HBM 50 kN and HBM 20 kN. Transducers were connected under the angle of 90 degrees and installed on the support structure articulately mounted on vertical rollers of the winch (Fig. 4). Transducers were placed horizontally and vertically with respect to the basis at fully lifted skidder’s rear protective and anchoring blade, corresponding to its position in timber skidding. The measurement of torques was carried out by strain gauges placed opposite the housing of the final planetary drives. For measuring dynamic wheel load, the strain gauges were placed right behind the wheels on the upper part of the shaft housing leading from the differential to the wheel. Strain gauges were connected by conductors to the juncture. Due to wheel rotation it was necessary to install a sliding

Fig. 4 Carrier with traction and tension transducers Slika 4. Nosa~ s vla~no-tla~nim dinamometrima Croat. j. for. eng. 30(2009)2

@. Toma{i} et al.

Fig. 5 Preparation of final (planetary) drive Slika 5. Prepariranje planetarnoga reduktora transducer on each wheel. The carriers of the sliding transducers were installed on the housing of the final drive (Fig. 5). The signal of the strain gauge resistance change is transmitted through the sliding transducer by cables to the amplifier. All measuring transducers and strain gauges are connected to measuring amplifiers HBM Spider 8 installed on the skidder. The use of the radio modem ELPRO 805 U (ELPRO Technologies Pty Ltd.) enabled remote transmission of data. The radio modem was installed on the skidder and connected to the measuring amplifier HBM Spider 8. It received analogically amplified measurement signals and transmitted them through the antenna installed on the roof of the skidder cabin. Another radio modem received the measurement signals and transmitted them into the field computer. The software programme Catman 4.0 (Hottinger Baldwin Messtechnik GmbH) was used for making records of the measurement data at the frequency of 50 Hz. Measurement tranducers were recorded into the software programme by channels and measuring constants obtained by calibration of measurement transducers were then entered. Further data processing of the measuring results was carried out by the software programme Microsoft Excell. During traction tests, changes were measured of wheel load of unloaded skidder on plane ground. In case of reduction of wheel load, the result of measurement will be deducted from the wheel load of the unloaded skidder, i.e. it will be added in case of increase of wheel load. Due to the described way of measurement, it was necessary to establish the mass and wheel load of the unloaded skidder. Although these data are provided by the manufacturer, by installing the measuring equipment, carriers and aux-

129

@. Toma{i} et al.

iliary structures and devices, the basic skidder mass and wheel load increased. The mass and wheel load of an unloaded skidder were determined by four scales by a Sweden manufactured TELUB connected to a measuring amplifier HBM Spider 8, which is directly connected to a laptop so that the measuring results were read by a computer programme Catman 4.0. The measurement showed that the mass of the equipped skidder was 3647 kg, i.e. that the mass was increased by 164 kg compared to the basic mass. The front wheels were additionally loaded, 36 kg by each wheel. Rear wheels have a slightly higher load (45 kg and 47 kg), which is caused by installation of traction and tension transducers for measuring the vertical and horizontal rope force at the rear end. Traction and tension transducers are installed on the carrier connected to the carrier connected to roller bases of the winch on the skidder rear protective and anchoring blade, and since higher loads are possible, the structure of the transducer carrier is massive. Load distribution to wheels remained almost unchanged compared to the distribution of values measured by manufacturers. The skidder front axle, equipped with the measuring devices, is loaded with 61% of the total skidder weight, and the rear axle with 39%, while the load ratio of the front/rear axle of the skidder as delivered by the manufacturer is 62%:38%. The load distribution to wheels with respect to the left and right side of the skidder is 50%:50%.

3. Research results â&#x20AC;&#x201C; Rezultati istra`ivanja The research was carried out on the strip road where longitudinal slopes were determined by the leveling method as well as individual distances of the strip road of uniform slope. Longitudinal slopes increase with the distance from the beginning of the strip road. The lowest route slope was recorded at the beginning of the strip road (2.3%), and then followed the road sections with higher longitudinal slope: 15%, 18.3%, 27.0% and 35.5%. The highest slope of the strip road was selected based on the investigation of limit slopes for skidder movement. MacDonald (1999) recorded the highest limit slope for skidders of 35% downhill, and Inoue and Tsuji (2003) the slope of 45% downhill and 30% uphill. 8 beech logs with the mean diameters ranging between 27 cm and 39 cm were used for the research. Based on the measured dimensions, log volumes were calculated. Log volumes ranged between 0.32 m3 and 0.50 m3. The mass of individual logs was weighed

130

Forces Affecting Timber Skidding (127â&#x20AC;&#x201C;139)

by two scales TELUB connected to the measuring amplifier HBM Spider 8 and laptop. Based on the known volume and mass of logs, the characteristics were determined of the load to be used in traction tests. 9 types of load were selected, with 1 to 4 pieces in a load, of the size between 0.27 and 1.8 m3 and weight ranging between 2.49 kN and 17.38 kN. Skidders were used for carrying out traction tests in downhill and uphill skidding of different loads. The traction test consisted of skidding of a certain load from the starting point at the beginning of the strip road through parts of the strip road with increasing longitudinal slopes until the end of the strip road. If not all terrain slopes could be overcome due to too large load, the traction test would be interrupted at the slope level that could be overcome. Downhill skidding started from the highest slope. Out of 9 uphill traction tests, i.e. skidding of 9 types of load, the skidder reached the top of the slope in six tests. In skidding the three largest loads, successful traction tests were made on the first two uphill slopes. The same occurred with downhill tractive tests: the skidding of three largest loads was only recorded for the two lowest slopes. At the final tractive tests, the wheel load was measured in uphill and downhill unloaded skidder travel. All measuring results were expressed as mean values per traction test with individual load and the determined longitudinal slope of strip road. Table 1 shows the measuring results of the vertical and horizontal components of rope force in timber skidding by the skidder Ecotrac 55V according to strip road slope, skidding direction and load size. In uphill skidding, the values of the vertical component of rope forces are higher than the horizontal component on all longitudinal slopes up to 18.3% for all skidded loads. On longitudinal slopes ranging between 27.0% and 35.5% the horizontal component is higher than the vertical for load weights of 7.83 kN and 9.55 kN. With smaller loads, the vertical component of rope forces remained higher than the horizontal one. These results show that the increase of slope results in decrease of part of load pulled by the rope, i.e. the part of load dragged on the soil is increased. Therefore, higher horizontal force component is required for overcoming the traction resistance of the part of load dragged on the soil. In downhill skidding, the vertical component of rope forces in each traction test is higher than the horizontal component. The increase of downhill slope results in the decrease of values of both components of rope forces, and the described phenomenon is more conspicuous with the horizontal component. Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

@. Toma{i} et al.

Table 1 Components of rope force and tractive forces Tablica 1. Sastavnice sile u u`etu i vu~ne sile

Slope Nagib

2.3%

15.0%

18.3%

27.0%

35.5%

Horizontal component of skidder weight Load weight Te`ina tovara Usporedna sastavnica te`ine skidera Q kN 2.49 3.11 3.69 6.72 7.83 9.55 12.03 14.91 17.38 2.49 3.11 3.69 6.72 7.83 9.55 12.03 14.91 17.38 2.49 3.11 3.69 6.72 7.83 9.55 2.49 3.11 3.69 6.72 7.83 9.55 2.49 3.11 3.69 6.72 7.83 9.55

Gsin kN

0.84

5.39

6.59

9.54

12.24

Croat. j. for. eng. 30(2009)2

Uphill skiding – Privla~enje uz nagib Component of rope force Tractive force Sastavnica sile u u`etu Vu~na sila Vertical Horizontal Fv Usporedna Okomita V 1.494 1.646 2.125 3.813 4.055 4.471 6.621 7.018 9.472 1.496 1.674 2.264 3.858 4.520 4.686 7.027 7.714 10.444 1.502 1.651 2.235 3.618 4.070 4.240 1.485 1.583 2.202 3.664 3.917 4.285 1.503 1.618 2.229 3.734 3.978 4.545

H kN 0.613 1.020 0.732 1.796 2.024 3.140 3.407 4.133 5.593 0.769 1.278 1.302 2.082 2.814 3.397 4.142 5.434 6.461 0.942 1.285 1.319 2.432 3.775 3.960 1.125 1.420 1.503 3.083 4.190 4.816 1.384 1.732 1.730 3.857 4.647 5.926

Downhill skiding – Privla~enje niz nagib Component of rope force Tractive force Sastavnica sile u u`etu Vu~na sila Vertical Horizontal Fv Usporedna Okomita

H + Gsin

1.45 1.86 1.57 2.64 2.87 3.98 4.25 4.97 6.43 6.16 6.67 6.69 7.47 8.20 8.79 9.53 10.82 11.85 7.53 7.87 7.90 9.02 10.36 10.55 10.66 10.96 11.04 12.62 13.73 14.35 13.62 13.97 13.97 16.10 16.89 18.16

1.468 1.576 2.010 3.689 3.777 4.126 6.265 7.239 7.815 1.450 1.564 1.984 3.502 3.584 3.824 6.234 7.055 7.647 1.496 1.681 1.971 3.369 3.881 3.326 1.426 1.487 1.828 3.350 3.523 3.498 1.151 1.234 1.424 3.134 3.480 3.164

H kN 0.357 0.589 0.703 0.973 1.706 1.944 2.418 3.345 5.656 0.147 0.303 0.499 0.450 1.295 1.091 1.643 2.329 4.217 0.215 0.421 0.473 0.343 1.084 0.640 0.029 0.259 0.182 0.118 0.803 0.454 0.030 0.187 0.312 0.103 0.258 0.930

H – Gsin –0.48 –0.25 –0.14 0.13 0.86 1.10 1.58 2.50 4.81 –5.24 –5.09 –4.89 –4.94 –4.10 –4.30 –3.75 –3.06 –1.17 –6.37 –6.16 –6.11 –6.24 –5.50 –5.94 –9.51 –9.28 –9.35 –9.42 –8.73 –9.08 –12.21 –12.05 –11.93 –12.14 –11.98 –11.31

131

@. Toma{i} et al.

Forces Affecting Timber Skidding (127â&#x20AC;&#x201C;139)

The lowest value of the horizontal component of rope forces (0.03 kN) was recorded in downhill skidding with the slope of 27% and 35% of the smallest load of 2.49 kN. In downhill skidding on higher longitudinal slopes, the front end of the load gets closer to the rear end of the skidder, which causes the increase of the vertical component of rope forces that holds the load eleveted off the soil. In doing so, lower horizontal component of forces is required because lower weight of the load is dragged on the soil. It was not possible to tie the loads made of three or four timber assortments to the carrier of the rope tied to transducers exclusively with the heavier end eleveted off the ground. With respect to the position of timber assortments in the load, and the position of the load with resprect to the rear end of the skidder, irregular sequence of values of horizontal component of forces was observed in downhill skidding on the two highest longitudinal slopes. In uphill skidding, the tractive force is used for overcoming the traction resistance of the part of load dragged on the soil (H) and the opposite effect of resistance of horizontal component of skidder weight (G sina) that pulls the skidder downhill, due to the effect of gravitation, in the opposite direction of traction. Table 1 shows the values of this calculation of tractive force. It can be seen from the presented data that the values of horizontal component of skidder weight increase with the increase of the slope of the test skid trail. On lower slopes, the effect of horizontal component of rope forces, used for overcoming the traction resistance, increases, because the hori-

zontal component of the skidder weight has lower values due to small angles. From the slope of 18.3%, the effect of horizontal component of skidder weight on the value of the tractive force is considerably higher than the traction resistance, regardless of the size of the skidded load. In downhill slope, the horizontal component of skidder weight (G sina) acts in the direction of skidder travel, by which traction resistances are overcome more easily. Only on the lowest slope, positive values were recorded of the tractive force, due to low inclination of the skid trail (the value G sina that is deducted from the the horizontal component of the traction resistance is only 0.84 kN). On other skid trail slopes the calculated values of tractive forces were mostly negative. It can be clearly seen from the above data that due to the increase of the load weight (Q), and hence also of the horizontal component of rope forces (H), the negative values of tractive forces are usually decreased, and however compared to the effect of the skidder weight in the traction direction, this effect on the value of the tractive force is incomparably lower. The dependence of the tractive force on the load weight, skidding direction and strip road slope is shown in Fig. 6 and 7. In uphill skidding, the tractive force increases with the increase of the longitudinal slope and load weight. With the increase of the slo-

Fig. 6 Dependance of tractive force on load weight in uphill skidding Slika 6. Ovisnost vu~ne sile o te`ini tereta pri vu~i uz nagib

Fig. 7 Dependance of tractive force on load weight in downhill skidding Slika 7. Ovisnost vu~ne sile o te`ini tereta pri vu~i niz nagib

132

Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

pe, resistances increase caused by the impact of the part of skidder weight, acting contrary to the traction direction, and the increase of load weight increases the values of the horizontal component of rope forces (traction resistance). The skidder’s tractive force must overcome the values of both resistances in uphill skidding. In downhill skidding, based on low inclination of regression lines, it can be established that horizontal component of the skidder weight has a lower impact on the values of tractive forces, than the impact of traction resistance. In this research, the wheel load was determined from the data of dynamic measurement of changes of wheel loads in traction tests from the previoiusly determined values of statistical weight distribution of unloaded skidder. Tables 2 and 3 show the measured skidder’s adhesive weights and axle load distribution in traction tests. The results of measurement of unloaded skidder travel are presented for the purpose of analysing the impact of load weight and strip road slope on axle load distribution. During uphill travel of unloaded skidder, the front axle load decreases with the increase of longitudinal slope. On the highest slope of 35.5%, the rear axle load of the unloaded skidder is higher due to the effect of horizontal component of skidder weight G·sina, which contributes to the transfer of load weight from the front axle to the rear axle. In traction tests, the rear axle load increases with the increase of load weight and slope. The increase of load weight increases the vertical component of rope forces that carries the part of load eleveted off the ground and additionally increases the load on the rear axle. On the lowest slope, the front axle is under higher load with the 4 smallest loads, on the following slope of 15% with the three smallest loads, on slopes of 18.3% and 27% only with the smallest load. On the highest slope, higher load was recorded on the rear axle in all tracton tests. Sever (1984) states that the longitudinal stability of the skidder is questionable when the load ratio of the front and rear axle reaches the value of 1:3.5 or 22%:78%. According to Weise and Nick (2003) at least 10% of the total dynamic load should remain on the front axle so as to enable steering. The highest load of the rear axle was recorded in skidding the largest load of 17.38 kN on the slope of 15% and it was 76% of the total adhesive weight. Under such conditions, 24% of the total adhesive weight remains on the front axle, which is very close to the limit value when the skidder’s longitudinal stability is compromised. This is why the skidder could not pull the three largest loads on the following slope of the strip road of 18.3%. Higher values of the Croat. j. for. eng. 30(2009)2

@. Toma{i} et al.

vertical force component with these loads along with the effect of the horizontal component of the skidder weight G sina on such longitudinal slope would result in the decrease of the front axle load under the limit value of 22% of the total adhesive weight and in the disruption of the skidder’s longitudinal stability. The analysis of the measuring results of downhill traction tests shows higher load of the front axle with the increase of the slope in skidding the same load, but also higher load of the front axle with the decrease of load weight on the same slope. In all traction tests, including the unloaded skidder, on slopes higher than 18.3% in download skidding, higher load of the skidder’s front axle was recorded. It should be noted that in downhill skidding of the largest loads on lower slopes, the load of the skidder’s rear axle is higher than the load of its front axle, i.e. the load of the front axle decreases with respect to the rear axle. It can be concluded that the horizontal component of the skidder weight (G sina) contributses to the load transfer from the rear axle to the front axle, but overcoms the effect of the vertical component of rope force (V) so that higher load is applied on the rear axle. As already explained, the essence of the effect of the skidder transmission system is the transmission of torque from the engine to the wheel. In doing so, the value of torque changes, starting with the gear box, through drive distributio to the final (planetary) drives in skidder wheels. As the transmission system with skidders is mechanical, the torque is tramsmitted mechanically and during this transmission its value changes with respect to wheel load. Along with wheel load distribution Tables 2 and 3 also show the measured mean values of the skidder axle torque distribution. In uphill skidding, uniform increase can be observed of the total values of torques with the increase of load and slope. When analisying the torques of the front and rear axle, it can be clearly seen that the increase of the load results in considerable increase of rear wheel torques, and in the decrease of front wheel torques, which is in accordance with the above considerations on loads. The skidder could not reach the top of the 3 highest slopes with the largest load, but on the last slope of 18.3% that was overcome, the mean torque of 7.0 kNm was achieved at the rear wheels, which is the highest value achieved up to that slope in all uphill traction tests. After that, the highest value of the skidder’s rear axle torque is achieved in traction tests on higher slopes with the heaviest loads with which the said slopes were overcome. In downhill skidding on testing skid trails, a very interesting situation appears in the analysis of the skidder’s wheel torques in timber skidding. On the

133

@. Toma{i} et al.

Forces Affecting Timber Skidding (127–139)

Table 2 Axle load and torque distribution in uphill skidding Tablica 2. Raspodjela optere}enja i momenata po osovinama pri privla~enju uz nagib

Slope Nagib

2.3%

15.0%

18.3%

27.0%

35.5%

134

Load weight Te`ina tovara

Q, kN 0 2.49 3.11 3.69 6.72 7.83 9.55 12.03 14.91 17.38 0 2.49 3.11 3.69 6.72 7.83 9.55 12.03 14.91 17.38 0 2.49 3.11 3.69 6.72 7.83 9.55 0 2.49 3.11 3.69 6.72 7.83 9.55 0 2.49 3.11 3.69 6.72 7.83 9.55

Adhesive weight Adhezijska te`ina Ga, kN 35.479 37.556 37.433 38.032 39.220 42.444 40.835 44.871 43.971 46.439 35.142 37.125 37.754 39.080 40.498 41.974 42.512 43.686 44.714 45.613 34.661 37.398 38.189 39.395 39.643 41.547 40.201 36.695 37.838 37.617 39.188 38.497 40.161 37.736 36.306 37.588 37.088 36.612 39.444 39.863 38.228

Uphill skiding – Privla~enje uz nagib Axle load distribution Raspodjela optere}enja Torgue Moment Front axle Rear axle Prednji most Stra`nji most GaF, % GaR, % M, kNm 60 40 2.401 58 42 3.000 57 43 3.324 57 43 3.141 52 48 3.730 46 54 4.059 47 53 4.419 43 57 5.074 40 60 5.453 34 66 6.298 57 43 5.016 55 45 5.466 53 47 5.958 52 48 6.185 46 54 6.459 42 58 7.143 41 59 7.341 37 63 8.014 31 69 8.559 24 76 9.093 56 44 5.837 53 47 6.246 50 50 6.508 50 50 6.729 47 53 7.439 39 61 8.217 37 63 7.955 53 47 7.025 50 50 8.053 46 54 8.209 48 52 8.422 42 58 9.228 34 66 9.511 33 67 9.846 48 52 8.543 46 54 9.547 43 57 9.956 44 56 9.771 37 63 11.128 30 70 11.250 31 69 12.294

Torgue distribution Raspodjela momenata Front axle Rear axle Prednji most Stra`nji most MF, % M R, % 62 38 60 40 57 43 57 43 52 48 45 55 45 55 40 60 41 59 34 66 56 44 54 46 51 49 50 50 46 54 44 56 42 58 36 64 31 69 23 77 56 44 53 47 50 50 51 49 47 53 38 62 37 63 52 48 49 51 46 54 50 50 42 58 34 66 34 66 46 54 47 53 44 56 45 55 37 63 28 72 31 69

Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

@. Toma{i} et al.

Table 3 Axle load and torque distribution in downhill skidding Tablica 3. Raspodjela optere}anja i momenata po osovinama pri privla~enju niz nagib

Slope Nagib

2.3%

15.0%

18.3%

27.0%

35.5%

Load weight Te`ina tovara

Adhesive weight Adhezijska te`ina

Croat. j. for. eng. 30(2009)2

Ga, kN 38.040 36.369 36.460 36.616 39.311 39.509 39.294 41.394 44.872 45.100 34.593 37.719 37.537 37.188 37.342 39.287 38.699 42.862 42.463 45.291 35.271 37.314 36.743 37.366 37.318 37.606 37.337 34.989 35.729 35.062 36.694 36.659 37.172 38.068 33.742 35.388 36.698 35.846 36.295 37.343 36.187

Downhill skiding – Privla~enje niz nagib Axle load distribution Raspodjela optere}enja Torgue Moment Front axle Rear axle Prednji most Stra`nji most GaF, % GaR, % M, kNm 66 34 1.967 59 41 1.806 58 42 1.973 60 40 2.140 54 46 2.427 49 51 2.900 49 51 2.608 44 56 3.450 42 58 4.058 37 63 4.963 68 32 –0.893 68 32 –0.604 67 33 –0.518 65 35 –0.429 58 42 –0.442 54 46 0.240 57 43 –0.022 50 50 0.504 48 52 0.959 43 57 2.070 67 33 –1.433 69 31 –1.167 66 34 –1.069 63 37 –1.011 59 41 –1.203 57 43 –0.725 59 41 –1.067 69 31 –3.350 73 27 –3.176 71 29 –2.911 64 36 –2.911 66 34 –3.059 59 41 –2.621 65 35 –2.832 73 27 –4.579 76 24 –4.149 71 29 –4.307 67 33 –4.693 69 31 –4.332 66 34 –4.170 67 33 –4.251

Torgue distribution Raspodjela momenata Front axle Rear axle Prednji most Stra`nji most MF, % MR, % 65 35 59 41 55 45 63 37 52 48 47 53 49 51 45 55 41 59 38 62 66 34 68 32 69 31 67 33 59 41 50 50 50 50 49 51 46 54 42 58 65 35 69 31 64 36 63 37 59 41 54 46 56 44 69 31 73 27 72 28 64 36 68 32 59 41 65 35 76 24 77 23 73 27 67 33 70 30 65 35 68 32

135

@. Toma{i} et al.

Forces Affecting Timber Skidding (127–139)

lowest negative slope, the wheel torque is positive. On the following longitudinal slope of 14.9% in skidding smaller loads, negative torque appeared at the skidder shafts, in very low negative amounts that ranged between –0.2 kNm and –0.9 kNm in unloaded skidder travel. In skidding larger loads, the skidder had to achieve the required combination of tractive forces by positive amounts of total wheel torques of 0.2 kNm and up to 2.1 kNm. On the following slope of 18.3%, in skidding of all types and sizes of loads in unloaded skidder travel, all values of torques were recorded in negative amounts. The negative values of total torques are relatively low and they range between –0.7 kNm in skidding the heaviest load, tested on this slope, and –1.4 kNm achieved durting unloaded skidder travel. We have a similar situation with the last two highest slopes, with the difference that the amounts of the achieved negative wheel torques are considerably higher and they range between –2.6 kNm and –4.7 kNm, which implies that the increase of negative slope results in the increase of the achieved negative amounts of torques. This phenomenon of negative skidder wheel torques in downhill skidding was also recorded by Maren~e (2005) and [u{njar and Horvat (2006). In this case, torques are not used for generating wheel

tractive forces, and instead, the transmission of torque through the transmission system causes skidder’s braking performance. The skidder’s need for braking arises under the influence of the horizontal component of the skidder weight (G sina) that pushes the skidder downhill. Torque distribution to skidder axles shows the same ratios as wheel load distribution. Considering the impact of the vertical component of rope forces (V) on the value of the skidder adhesive weight, the analysis was made of torque dependence right on this value (force). Figures 8 and 9 show the data with regression lines. Points on the ordinate axis represent wheel torques during unloaded skidder travel on testing skid trails. At such values, torques enable overcoming of terrain slope and rolling resistance of an unloaded skidder. In uphill skidding (Fig. 8) it can be observed that the increase of the vertical component of rope forces results in the increase of skidder wheel torque. This can be understood because the vertical component of rope forces directly causes the increase of the skidder adhesive weight, and hence also the rolling resistance, which is consequentially related to the increase of torques that must enable the generation of the circumferential force required for overcoming this increased resistance and other traction resis-

Fig. 8 Dependance of wheel torque on vertical component of rope force and uphill slope Slika 8. Ovisnost zakretnoga momenta na kota~ima o vertikalnoj sastavnici sile u u`etu i nagibu u usponu

Fig. 9 Dependance of wheel torque on vertical component of rope force and downhill slope Slika 9. Ovisnost zakretnoga momenta na kota~ima o vertikalnoj sastavnici sile u u`etu i nagibu u padu

136

Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

tances. According to the position and equations of regression lines it can also be read from Fig. 8 that the value of slope and load (load size) undoubtedly affect the value of wheel torque. On lower slopes, the lines are more horizontal (direction coefficients of 0.42 and 0.43), which means that the increase of the vertical component of rope forces result in slower increase of torque. On higher slopes this impact is definitey more significant (0.73). It can be concluded from this analysis that the value of wheel load that represent the pulled load affects considerably the increase of the values of required wheel torque. In downhill skidding, it can be observed in Fig. 9 according to the heights of regression lines that here the values of parallel skidder weight forces have a prevailing impact on the value of the skidder wheel torques. This is confirmed by low values of the correlation coefficient of regression lines. However, the impact of the vertical component of rope forces can be seen in the increase of the rolling resistance, by which the skidder’s downhill travel is slowed, and hence less braking is necessary as well as lower torque. Thus, it can be concluded from the survey of achieved torques in downhill travel that the negative values of wheel torques were decreased in skidding larger loads at the highest slopes.

4. Conclusions – Zaklju~ci The determination of dependence of components of rope forces, skidder load and wheel torques on load weight and terrain slope represents a significant relationship for the possibility of assessing skidder traction characteristics in timber skidding. Torque distribution depends on the skidder wheel vertical load. In uphill skidding, torques increase proportionally to the vertical component of rope forces and adhesive weight of the skidder. According to the results of research, the following conclusions can be made related to the effect of forces with respect to the required tractive force in uphill skidding: by the increase of longitudinal slope, more tractive force is used for overcoming the terrain slope than for overcoming the traction resistance. In downhill skidding, the horizontal component of the skidder weight (G sina) acts in the skidder travel direction and hence its value is higher than the traction resistance. In the same way, with downhill skidding, the skidder wheel torques are negative, because they are not used for achieving wheel tractive force and instead, the transmission of torque through the transmission system causes skidder’s braking performance. The skidder’s need for braking arises under influence of the horizontal component of the skidder Croat. j. for. eng. 30(2009)2

@. Toma{i} et al.

weight (G sina) that acts in the direction of the skidder travel and under its effect the traction resistances are overcome. Further to the above said, it can be concluded that in case of downhill skidding we cannot speak of achieving real tractive force because the skidder pulls the load only by its weight, and the transmittion from engine to wheels is not used for achieving the tractive force.

5. References – Literatura Hassan, A. E., Gustafson, M. L., 1983: Factors Affecting Tree Skidding Forces, Transactions of the ASAE, 26(1): 47–53. Heidl, I., Husnjak, M., 1992: Tenzometrija. Tehni~ka enciklopedija, Leksikografski zavod »Miroslav Krle`a«, Svezak 12: 685–690. Horvat, D., 1987: Skidder Wheel Torque Measuring. Proceedengs of 9th ISTVS International Conference, Barcelona, Vol. II: 531–541. Inoue, M., Tsujii, T., 2003: Management, technology and system design of mechanized forestry in Japan. Textbook of forestry mechanization technology, Forestry Mechanization Society, Tokyo, Japan, Forestry Machine Series No: 92, 1–122. MacDonald, A. J., 1999: Harvesting Systems and Equipment in British Columbia. FERIC, Handbook No. HB-12: 1–197. Maren~e, J., 2005: Spreminjanje tehni~nih parametrov traktorja pri vla~enju lesa – kriterij pri izbiri delovnega sredstva. Disertacija, Biotehni{ka fakulteta Univerze u Ljubljani, Slovenija: 1–271. Marklund, B. O., 1987: Torque distribution on wheeled vehicles affects damage on the forest ground. Proceedings of 9th ISTVS International Conference, Barcelona, Vol. 1: 347–354. Sever, S., 1984: Istra`ivanje nekih eksploatacijskih parametara traktora kod privla~enja drva (Investigations of some tractor's exploitation parameters at wood skidding). Glasnik za {umske pokuse 22: 183–303. Sever, S., 1987: Dynamic loading of skidder axles at wood skidding. Proceedings of the 9th International Conference of the ISTVS, Barcelona, Vol. II: 531–540. Stoilov, S., 2007: Improvement of wheel skidder tractive performance by tire inflation pressure and tire chains. Croatian Journal of Forest Engineering 28 (2): 137–144. [u{njar, M., Horvat, D., 2006: Dinami~ko optere}enje kota~a skidera pri privla~enju drva (Dynamic loading of skidder wheels at timber skidding). Glasnik za {umske pokuse, posebno izdanje 5, 601–615. Toma{i}, @., Horvat, D., [u{njar, M., 2007: Raspodjela optere}enja kota~a skidera pri privla~enju drva (Wheel load distribution of skidders in timber extraction). Nova mehanizacija {umarstva 28 (1): 27–36.

137

@. Toma{i} et al.

Forces Affecting Timber Skidding (127–139)

Weise, G., Nick, L., 2003: Determining the performance and the environmental impact of forest machines – Classification numbers and performance diagrams. Proceedings of Austro2003 – High Tech Forest Operations for Mountainous Terrain, October 5–9, 2003, Schlaegl, Austria, Uni-

versity of Natural Resources and Applied Life Sciences Viena, CD-ROM, 1–10. Wong, J. Y., 2001: Theory of Ground Vehicles. Third Edition. John Wiley & Sons, N.Y., 1–528.

Sa`etak

Utjecajne sile pri privla~enju drva Cilj je rada utvr|ivanje ovisnosti sastavnica sila u uètu, raspodjele optere}enja kota~a i zakretnih momenata o teìni tovara i nagibu terena. Ti su odnosi va`ni za mogu}nost procjene vu~nih (eksploatacijskih) svojstava skidera pri privla~enju drva. Skideri, kao {umska vozila za privla~enje drva, isklju~ivo su namijenjeni postizavanju vu~ne sile koja se ostvaruje preko oboda kota~a. Sustavom transmisije skidera prenosi se i mijenja zakretni moment od pogonskoga motora na kota~e. Zbog djelovanja zakretnoga momenta na kota~u se javlja obodna sila. Horizontalna se sastavnica obodne sile dijelom tro{i za svladavanje otpora kotrljanja vozila (Ff), a ostali dio sile (Fv) sluì za vu~u tereta, svladavanje nagiba i povr{inskih prepreka terena ili ubrzavanje vozila (Wong 2001, Stoilov 2007). U primjeni skidera opremljenih {umskim vitlom drvo se privla~i s jednim krajem tovara odignutim od tla i preko uèta vitla oslonjenim na zadnji kraj vozila, dok se drugi kraj tovara vu~e po tlu. Sila koja se javlja u uètu sluì za no{enje teìne dijela tovara odignuta od tla (vertikalna sastavnica – V) te za svladavanje otpora vu~e dijela teìne tovara oslonjena na tlo (horizontalna sastavnica – H). Pod adhezijskom se teìnom (Ga) razumijeva zbroj okomitih optere}enja na pogonskim kota~ima skidera u uvjetima pridobivanja drva (Toma{i} i dr. 2007). Prema tomu je adhezijska teìna razli~ita od teìne praznoga skidera (G) jer se stra`nji most dodatno optere}uje punim iznosom vertikalne sastavnice sile u uètu, koja se raspore|uje na stra`nje kota~e preko horizontalnih valjaka vitla (Hassan i Gustafson 1983). Raspodjela se optere}enja po mostovima skidera mijenja s obzirom na obujam i teìnu tovara, orijentaciju sortimenata u tovaru, smjer kretanja i veli~inu nagiba terena. Kretanjem po uzdu`nom nagibu dodatno se optere}uje stra`nji most jer dolazi do prijenosa optere}enja teìne skidera s prednjega mosta zbog djelovanja usporedne sastavnice teìne skidera (G sina). Pri privla~enju uz nagib vu~na sila treba svladati vu~ne otpore dijela tovara oslonjenoga na tlo (H), ali i otpor horizontalne sastavnice teìne skidera (G sina) koja povla~i vozilo prema dolje (slika 1). Pri kretanju niz nagib zbog djelovanja horizontalne sastavnice teìne skidera (G sina) optere}enje se teìne skidera prenosi na prednji most (slika 2). Tako|er }e horizonatalna sastavnica teìne skidera djelovati u smjeru kretanja skidera te je potrebno svladati jedino vu~ne otpore dijela tovara oslonjena na tlo (H). Zbog razli~itih optere}enja na kota~ima transmisija mora omogu}iti i raspodjelu zakretnoga momenta s obzirom na optere}enje kota~a. Prema tomu kod mehani~koga sustava transmisije raspodjela momenata po kota~ima treba biti u skladu s raspodjelom optere}enja po kota~ima. U istraìvanju je kori{ten skider ECOTRAC 55 V opremljen {umskim vitlom Hittner 2 ´ 35 kN (slika 3). Za mjerenje dinami~koga optere}enja skidera konstruirana su ili primijenjena mjerila za istodobno odre|ivanje sastavnica sile u uètu, zakretnih momenata i optere}enja na svim pogonskim kota~ima. Istraìvanja su provedena na traktorskom putu gdje su metodom niveliranja utvr|eni uzdu`ni nagibi te pojedine udaljenosti traktorskoga puta jednolika nagiba. Po~etni dio traktorskoga puta ima najmanji nagib trase (2,3 %), a zatim uzastopno slijede dijelovi puta sa sve ve}im uzdu`nim nagibom: 15 %, 18,3 %, 27,0 % i 35,5 %. Odabrano je 9 vrsta tovara, s 1 do 4 komada u tovaru, veli~inom od 0,27 do 1,8 m3 te teìnom od 2,49 kN do 17,38 kN. Skiderom su se izvodili vu~ni pokusi privla~enja razli~itih tovara uzbrdo i nizbrdo. Svi su rezultati mjerenja izraèni u srednjim vrijednostima mjerenja po vu~nom pokusu s pojedinim tovarom i odre|enim uzdu`nim nagibom traktorskoga puta. Rezultati mjerenja okomite i usporedne sastavnice sile u uètu pri privla~enju drva skiderom Ecotrac 55V prikazani su prema uzdu`nomu nagibu traktorskoga puta, smjeru privla~enja i veli~ini tovara u tablici 1. Pri privla~enju uz nagib rezultati pokazuju da se s pove}anjem nagiba smanjuje teìna dijela tovara koji je no{en na uètu, tj. pove}ava se dio teìne tovara koji se oslanja na tlo. Stoga je potrebna ve}a usporedna sastavnica sile za svladavanje otpora vu~e dijela tovara po tlu. Pri privla~enju niz nagib prednji se kraj tereta pribliàva stra`njemu kraju skidera ~ime raste okomita sastavnica sile u uètu koja tovar drì odignutim od tla. Pri tome je potrebna manja usporedna sastavnica sila jer je manja teìna tovara oslonjena na tlo.

138

Croat. j. for. eng. 30(2009)2

Forces Affecting Timber Skidding (127–139)

@. Toma{i} et al.

U tablici 1 prikazani su iznosi izra~una vu~ne sile. Pri privla~enju uz nagib porastom uzdu`noga nagiba ve}i se dio vu~ne sile tro{i za svladavanje nagiba terena nego za svladavanje vu~nih otpora. Pri privla~enju niz nagib usporedna sastavnica teìne skidera (G sina) djeluje u smjeru kretanja skidera, ~ime se olak{ava svladavanje vu~nih otpora. Samo na najmanjem nagibu utvr|ene su pozitivne vrijednosti vu~ne sile. Na ostalim nagibima vlake izra~unate su vrijednosti vu~nih sila, uglavnom negativne. Ovdje se dakle ne radi o ostvarivanju vu~e u pravom smislu jer prevladavaju gravitacijski utjecaji djelovanja sila na nagibu, ve} se vu~a drva ostvaruje djelovanjem teìne skidera u smjeru vu~e. Ovisnost vu~ne sile o teìni tovara, smjeru privla~enja i nagibu traktorskoga puta prikazana je na slikama 6 i 7. Pri privla~enju uz nagib vu~na sila raste s pove}anjem uzdu`noga nagiba i teìne tereta. Pri privla~enju niz nagib moè se utvrditi ve}i utjecaj usporedne sastavnice teìne skidera na vrijednosti vu~nih sila od utjecaja sila vu~nih otpora. U provedenom je istraìvanju optere}enje na kota~ima odre|eno iz podataka dinami~kih mjerenja odstupanja optere}enja na kota~ima u vu~nim pokusima od prije utvr|enih vrijednosti stati~ke raspodjele teìne neoptere}enoga skidera. U tablicama 2 i 3 prikazane su izmjerene adhezijske teìne skidera i raspodjela optere}enja po mostovima u vu~nim pokusima. Radi analize utjecaja teìne tovara i nagiba traktorskoga puta na raspodjelu optere}enja po mostovima prikazani su rezultati mjerenja pri kretanju neoptere}enoga skidera. U vu~nim se pokusima uz nagib optere}enje stra`njega mosta pove}ava s pove}anjem teìne tovara i nagiba. U ra{~lambi mjernih rezultata vu~nih pokusa niz nagib uo~ava se ja~e optere}ivanje prednjega mosta s pove}anjem nagiba pri privla~enju istoga tovara, ali se tako|er pove}ava optere}enje prednjega mosta sa smanjenjem teìne tovara na istom nagibu. U svim vu~nim pokusima niz nagib ve}i od 18,3 % zabiljeèno je ve}e optere}ivanje prednjega mosta skidera. Zanimljivo je primijetiti da pri vu~i najve}ih tovara niz manje nagibe optere}enje stra`njega mosta prema{uje ono na prednjoj osovini skidera, odnosno prednja se osovina rastere}uje vi{e od stra`nje. Zaklju~ak je da horizontalna sastavnica teìne skidera (G sina) pridonosi prijenosu optere}enja sa stra`njega mosta na prednji, ali prevladava djelovanje vertikalne sastavnice sile iz uèta (V) tako da je ve}e optere}enje na stra`njem mostu. U tablicama 2 i 3 usporedno s raspodjelom optere}enja po kota~ima prikazane su mjerene srednje vrijednosti zakretnih momenata raspodjela momenata po mostovima skidera. Raspodjela zakretnih momenata u ovisnosti je o vertikalnom optere}enju na kota~ima skidera. Pri privla~enju uz nagib zakretni momenti proporcionalno rastu s vertikalnom sastavnicom sile u uètu i adhezijskom teìnom skidera. Pri privla~enju niz nagib zakretni su momenti na kota~ima skidera negativni jer ne sluè za ostvarivanje vu~ne sile na kota~ima, ve} se prijenosom zakretnoga momenta kroz sustav transmisije skider ko~i. Potreba za ko~enjem skidera o~ituje se u utjecaju horizontalne sastavnice teìne skidera (G sina) koja djeluje u smjeru kretanja skidera i zbog njezina djelovanja dolazi do svladavanja vu~nih otpora. Iz navedenoga izlazi da se u slu~aju privla~enja niz nagib ne moè govoriti o ostvarivanju prave vu~e jer skider vu~e tovare svojom teìnom, a prijenos snage s pogonskoga motora na kota~e ne koristi se za ostvarivanje vu~ne sile. Klju~ne rije~i: skider, optere}enje kota~a, zakretni momenti, vu~na sila, adhezijska teìna, nagib

Authors' address – Adresa autorâ: @eljko Toma{i}, PhD. e-mail: zeljko.tomasic@hrsume.hr »Hrvatske {ume« d.o.o. Zagreb Headquaters Zagreb Farka{a Vukotinovi}a 2 HR–10 000 Zagreb

Received (Primljeno): June 28. 2009. Accepted (Prihva}eno): November 15. 2009. Croat. j. for. eng. 30(2009)2

Asst. Prof. Marijan [u{njar, PhD. e-mail: susnjar@sumfak.hr Prof. Dubravko Horvat, PhD. e-mail: horvat@sumfak.hr Zdravko Pandur, BSc. e-mail: pandur@sumfak.hr Department of Forest Engineering Forestry Faculty of Zagreb University Sveto{imunska 25 HR–10 000 Zagreb

139

Original scientific paper – Izvorni znanstveni rad

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration Ramin Naghdi, Majid Lotfalian, Iraj Bagheri, Aghil Moradmand Jalali Abstract – Nacrtak Extracting logs from stump to landings causes extensive damages to forest stand and soil. In this research two parcels adjacent to each other were selected in order to assess the effect of traditional and mechanized methods of logging on regeneration and soil compaction. A skid trail and a mule trail with similar longitudinal slope, skidding direction and total volume of extracted wood were chosen in the parcels. The cylindrical sampling method was used to determine wet and dry soil bulk density and the samples were taken at 0–10 cm and 10–20 cm depth from skid and mule trails. The results showed that soil dry bulk density increase in skid and mule trails compared to control at 0–10 and 10–20 cm depth, was significant (p<0.01). This increase in mule trail at 0–10 cm depth was significantly higher than at 10–20 cm depth (p<0.01), but percentage of soil dry bulk density increase compared to control in skid trail at 0–10 cm and 10–20 cm depths was not significantly different. Soil dry bulk density increase compared to control at 0–10 cm depth of mule trail is higher than skid trail, but at 10–20 cm depth the skid trail is higher than mule trail. Systematic random sampling method was used to determine damages to different regeneration groups due to logging operations. The results showed that damages to each regeneration group seedling and small sapling in mule logging method were significantly lower than mechanized logging method. Keywords: animal logging, mechanized logging, natural regeneration damage, soil compaction

1. Introduction – Uvod Transporting logs from stump to landings is one of the most costly and difficult operation phases of forest harvesting. The main impact of logging in forest stand are disturbances of the soil surface, change in soil physical and chemical properties, damage to natural regeneration and residual stand. Skidder traffic causes soil compaction in skid trail and because of this the soil water infiltration and aeration decrease and run off erosion increases (Croke et al. 1999, Grigal 2000, Hamza and Andersson 2005, Buckley et al. 2003, Defossez and Richard 2002, Pinard et al. 2000, Johns et al. 1996). Skid trails, winching areas and landings are generally exposed to soil compaction during logging operations (Bob 2002).The choice of technology and performing operations under suitable terrain and weather condition are most important means of reducing harvesting disturbances on forest stand (Modry and Hubeny 2003). Croat. j. for. eng. 30(2009)2

Ground based skidding system using cable skidders are the only available facilities for mechanized extraction in northern Iran, because it costs less than other mechanized logging systems and is adaptable to all logging methods such as tree length and cut to length methods. On the other hand mule logging system is used to harvest steep terrains without available forest roads (Ghaffariyan 2003). With regards to this, animal logging systems with low undesirable effects are used especially in single selection systems and areas with low road density. Wang (1997) analyzed and compared skidder and animal wood extraction systems in terms of operation cost, disturbances to soil, remaining trees and regeneration in China and reported that animal logging was more effective in the steep terrain harvesting. Rummer et al. (2001) in their study in mountainous forests of Alabama in USA showed that animal logging systems were more economical than mechanized logging systems in harvesting small area, with low stock volume and single selection system.

141

R. Naghdi et al.

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

Williamson and Neilson (2000) measured the amount of compaction due to skidder traffic in skid trails at 10 cm depth of soil. They concluded that after first pass of machinery the amount of soil bulk density increase was 62% compared to control. Shaw and Carter (2002) showed that the change of soil physical properties during logging was a function of number of passes, soil properties and soil moisture content capacity. They used cylinder with 20 cm length and 5 cm diameter for measuring soil compaction. They concluded that soil bulk density at 0–20 cm depth increased compared to control after logging operations. The results of the study carried out by Modry and Hubeny (2003) on the impact of skidder and high-lead system logging on forest soils and advanced regeneration in Czech Republic showed that the skidder had more disturbances effects on regeneration and soil properties than high-lead system. Makineci et al. (2008) in evaluating the effects of wood extraction on soil at 0–5 and 5–10 cm depths showed that soil dry bulk density increases in skid trails compared to control were significantly high. Eric (2006) in evaluating soil compaction at 5, 10 and 20 cm depths also concluded that the amount of soil dry bulk density compared to control was significantly high. Comparing the effect of machines of different weights on soil compaction Kim (2000) showed that soil compaction can occur at 30 cm to 60 cm depth. Hoseini (2002) evaluated ground skidding logging and cable logging systems in Sari northern Iran and showed that the amount of damages to remaining trees and soil were higher in ground skidding logging system than cable logging system. However cable logging systems are not in use anymore, because of high costs of assembling and impossibility to provide adequately trained personnel, and in recent years single selection cutting is used in Iran and cable logging is not feasible. Ghaffariyan (2003) studied the effect of skidding by mule on forest soil of Khairoudkenar Forest, Northern Iran, and concluded that bulk density increase in skid trails after 28 passes was 13.8%. The results of research carried out by Jamshidi (2005) in Sari Forest, Northern Iran, showed that the average bulk density of the soil under skidder traffic was significantly higher than control. While the above average values of bulk density for the skid trails were not significant. Naghdi et al. (2007) analyzed the changes of soil bulk density and relative soil compaction for different number of wheeled skidder passes from stump to landing for two soil types (clay soil with high and low liquid limits). The findings of their research

142

showed that the effect of skidder traffic on increasing soil bulk density at sample locations was significant. In planning forest harvesting in Northern Iran, the only methods used in single selection cutting are the mechanized logging system with the use of cable skidder and traditional logging system using mule for extracting wood. The last case studies in Northern Iran separately evaluated soil ad stand damages by mule and tractor logging in different areas. However this study compares both extraction systems in a unit area to give the logging planners a real evaluation of site damages caused by both logging systems.

2. Materials and methods – Materijali i metode This research was carried out on parcels 923 and 927 of the ninth district of Shafaroud Forest (37°22' to 37°25' N, 48°30' to 48°72' E) in Northern Iran with average altitude of 1250 m above sea level. The slopes of the parcels were 30% to 60%. Parcels 923 and 927 are adjacent to each other with 85 and 94 hectares, respectively. The trees were felled using manual chain saw and log extraction from stump to forest road side landing in traditional logging system (animal power) and ground skidding logging system (mechanized logging) were carried out by mule and Timberjack C450 wheeled skidder, respectively. Traditional and mechanized logging methods were used in parcels 923 and 927, respectively. The forest type was beach forest (Fagus orientalis Lipsky) with the average growing stock of 430 cubic meters per hectare. In this research the amount of damages to different regeneration groups due to mechanized and traditional logging systems were assessed after wood extraction in the form of logs and lumber. The volume of harvested wood in the form of lumber and logs in parcels 923 and 927 were 510 m3 and 590 m3, respectively. Parcel 923 had a mule trail of 1000 m long, where the first 200 m started from road side landing towards the west of the parcel and branched into two 400 m trails towards the south of the parcel. Parcel 927 had seven skid trails, which could be crossed by the skidder and the longest skid trail was 600 m. In this research systematic random sampling method was used to assess the amount of damages to natural regeneration due to logging operations (felling and extraction) in the studied parcels. The sample plots area was 100 m2 and in order to get an appropriate distribution of sample plots a grid cells was set in 100 m by 100 m. Croat. j. for. eng. 30(2009)2

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

R. Naghdi et al.

In order to determine the amount of natural regeneration and assess the amount of damages to it, each plot was evaluated in three stages (before felling, after felling and after extraction). Trees with diameter of less than 7.5 cm are measured as regeneration and three-growth stages are used for measuring them: Þ Seedling, from start of growth to 0.5 m height, Þ Small sapling, growth stage from 0.5 to 2 m height, Þ Large sapling, growth stage more than 2 m height. Also damages to regeneration (seedling, small sapling and large sapling) are qualitatively evaluated in three degrees of bending, wounding and broken, and complete uprooting. Two independent samples were used, t test and Mann–Whitney u-test, to show if there was a significant difference between the amount of damages to regeneration in the traditional and mechanized method of logging. In order to study soil compaction due to log extraction in traditional and mechanized logging methods, in each of the studied parcels one skidding trail was chosen with 200 m long downward skidding, with no lateral slope and with the same average longitudinal slope (24%). These trails were assessed after extracting 100 m3 of wood. An undisturbed area was chosen as control next to these trails. The cylindrical sampling method was used to determine wet and dry soil bulk density; the cylinder soil sampler used was a 5cmdiameter and 10 cm height. The samples were taken at 20 m interval along trails on tire truck and mule foot step at 0–10 cm and 10–20 cm depths. Therefore in each skid and mule trail 20 samples were taken. Also 20 samples were taken along the control of each skid and mule trail at the same distance and depth. Therefore a total of 80 samples were taken for measuring soil bulk density. The formula below was used for measuring wet bulk density: w (1) g w = , (g/cm3) v

where:

Table 1 The number and percentage of regeneration damage after felling Tablica 1. Broj i udio o{te}enja nara{taja nakon sje~e stabala

wet bulk density w sample weight v cylinder volume In order to determine dry bulk density, 100 gram of soil sample was used to measure the soil moisture content and from formula (2) the dry bulk density of the samples is determined. gw , (g/cm3) (2) gd = 1 + w%

Croat. j. for. eng. 30(2009)2

gd gw

dry bulk density wet bulk density w% percentage of moisture content To determine soil physical properties, 80 samples were taken at 20 m interval from sample locations and control. The boreholes were 20 cm depth and 30 cm diameter. The soil texture for each sample was determined, too. Statistical package for social sciences (SPSS) were used for data analysis and Dunken test and least significant difference (LSD) were used for comparing averages.

3. Results and discussion – Rezultati s diskusijom The total plots for assessing regeneration in parels 923 and 927 were 68 and 94 plots for traditional and mechanized logging methods, respectively. Reeneration conditions (trees with diameter less than 7.7 m) were evaluated in the above plots before felling and the total regeneration and different regeneration group dispersion were determined in the studied parcels. The average number of regeneration per hectare in parcels 923 and 927 were 3050 and 3350, respectively. The assessment of the studied parcels after felling and extraction showed that 31 and 42 plots were damaged in parcels 923 (traditional logging) and 927 (mechanized logging), respectively.

3.1 Damages to regeneration due to felling O{te}enja nara{taja zbog sje~e stabala The analysis of the collected data from damaged plots after felling showed that the amount of damages to regeneration in traditional and mechanized logging methods were 15.5 and 12.8%, respectively (Table 1). Comparison of total damaged regeneration after felling in the studied parcels were not significantly different (t58,0.05 = 0.05, p = 0.96).

Parcel – Odjel Total number of regeneration Ukupni broj pomlatka Number of damaged regeneration Broj ošte}enoga pomlataka Percentage Postotak

879

1215

136

156

15.5%

12.8%

143

R. Naghdi et al.

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

3.2 Damages to regeneration due to wood extraction – O{te}enja nara{taja zbog privla~enja drva The results of analyzed data from damaged plots after wood extraction showed that the amount of damages to regeneration (seedling, small sapling and large sapling) in traditional logging method and mechanized logging method were 13.3 and 23.3%, respectively (Table 2). The t test comparison of number of damaged regeneration after wood extraction showed that damages to total regeneration in traditional logging method were significantly lower than mechanized logging method (t71, 0.05 = 4.3, p > 0.001). Assessing the amount of damages to different groups of regeneration after wood extraction showed that damages to each regeneration group seedling and small sapling in traditional logging method were significantly lower than mechanized logging

method (for seedling t71, 0.05 = 2.01, p > 0.05 and for small sapling z = –2.3, p > 0.05). With this respect, the damages to large sapling were not significantly different in the two methods of logging (z=–1.3, p=0.19). Analyzing type of damages to different groups of regeneration due to extraction in mechanized logging showed that the number of broken seedling, small sapling and large sapling were more than bending and complete uprooting (Fig 1).

Table 2 Number and percentage of regeneration damage after wood extraction Tablica 2. Broj i udio o{te}enja nara{taja nakon privla~enja drva Timber extraction Privla~enja drva Total number of regeneration Ukupni broj nara{taja Number of damaged regeneration Broj o{te}enoga nara{taja Percentage Postotak

Mules Mule

Skidders Skideri

745

1210

281

13.3%

23.3%

Fig. 1 Type of regeneration damage due to extraction in mechanized logging method Slika 1. Vrste o{te}enja pomlatka zbog privla~enja drva skiderima

Fig. 2 Soil dry bulk density of skid trail compared to control Slika 2. Usporedba prirodne gusto}e tla na {umskim vlakama i kontrolnim uzorcima 144

Croat. j. for. eng. 30(2009)2

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

3.3 Soil compaction – Zbijanje tla The hydrometer test results showed that the soil texture was silt loam for all samples. In this research parameters such as moisture content, soil dry bulk density and soil dry bulk density increase (%) compared to control were measured and compared in order to determine the amount of soil damages for skid and mule trails from samples. Table 3 shows average figures for each parameter. The results presented in Fig. 2 and Table 3 show that soil dry bulk density changes in skid trail compared to control at 0–10 and 10–20 cm depths were

R. Naghdi et al.

significantly higher (t18,0.01 = –17.5 and t18,0.01 = –10.5, p > 0.01). Fig. 3 and Table 3 show soil dry bulk density changes of mule trail compared to control at 0–10 and 10–20 cm depths. In all samples of mule trail, the soil dry bulk density was significantly higher than control (t18, 0.01 = –10.3 and t18, 0.01 = –7.4, p > 0.01).The results show that soil dry bulk density at 0–10 cm is higher than 10–20 cm depth (t18, 0.01 = 1.8, p > 0.01) (Table 4). This demonstrates that transporting with mule affects the first depth of soil more than the second depth.

Table 3 Soil compaction in the studied trails for mule and tractor logging (with 100 m3 volume of extracted wood) Tablica 3. Zbijanje tla istra`ivanih vlaka pri privla~enju drva mulama i skiderima (100 m3 privu~enoga drva) Average soil dry bulk density Srednja prirodna gusto}a tla Trail Vlaka

g/cm3 Depth 10 cm Dubina 10 cm

Depth 20 cm Dubina 20 cm

0.82

0.97

1.41

1.35

0.97

1.05

1.47

1.62

Mule trail control Kontrolna animalna vlaka Mule trail Animalna vlaka Skid trail control Kontrolna traktorska vlaka Skid trail Traktorska vlaka

Percentage of soil dry bulk density increase compared to control Postotno pove}anje prirodne gusto}e tla u odnosu na kontrolni uzorak Depth 10 cm Depth 20 cm Dubina 10 cm Dubina 20 cm

38.66

29.25

22.12

30.61

Average moisture content Srednja vla`nost tla % Depth 10 cm Dubina 10 cm

Depth 20 cm Dubina 20 cm

24.61

24.85

25.82

25.35

24.84

21.42

26.70

19.92

Fig 3 Soil dry bulk density of mule trail compared to control Slika 3. Usporedba prirodne gusto}e tla na animalnoj vlaci i kontrolnom uzorku Croat. j. for. eng. 30(2009)2

145

R. Naghdi et al.

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

Table 4 Comparison of soil dry bulk density between skid trail and mule trail Tablica 4. Usporedba prirodne gusto}e tla traktorskih i animalnih vlaka Trail Vlaka

t value t vrijednost

Degree of freedom Stupnjevi slobode

F value F vrijednost

Significance Zna~ajnost

Control mule trail – Mule trail (0–10 cm) Kontrolna animalna vlaka – Animalna vlaka (0 – 10 cm)

–10.297

1.054

Significant Zna~ajno

Control skid trail – Skid trail (0–10cm) Kontrolna traktorska vlaka – Traktorska vlaka (0 – 10 cm)

–17.470

2.284

Significant Zna~ajno

Mule trail first depth – Mule trail second depth (0–10 cm) Animalna vlaka, prvi prolaz – Animalna vlaka, drugi prolaz (0 – 10 cm)

1.814

1.692

Significant Zna~ajno

Skid trail first depth – Skid trail second depth Traktorska vlaka, prvi prolaz – Traktorska vlaka, drugi prolaz

–3.227

11.214

Non significant Bez zna~ajnosti

Control mule trail – Mule trail (10–20cm) Kontrolna animalna vlaka – Animalna vlaka (10 – 20 cm)

–7.386

4.168

Significant Zna~ajno

Control skid trail – Skid trail (10–20cm) Kontrolna traktorska vlaka – Traktorska vlaka (10 – 20 cm)

–10.458

13.243

Significant Zna~ajno

The results of Dunken test confirmed that there were significant differences in percentage of dry bulk density increase compared to control of mule and skid trails at two depths with 100 m3 of extracted wood for both methods (Table 5). Percentage of soil dry bulk density increase compared to control at 0–10 cm depth of mule trail is higher than skid trail, but percentage of soil dry bulk density increase compared to control at 10–20 cm depth of skid trail is higher than mule trail (t18, 0.05 = 2.9 and t18, 0.05 = –2.2, p > 0.05) (Table 6). Soil compaction was significantly different in mule trail at two depths (t18, 0.01 = 4.05, p > 0.01) and the amount of compaction at 0–10 cm depth is much higher than 10–20 cm depth. Therefore mule logging has the highest affect on surface soil depth (0–10 cm), but in skid trail the two depths were not significantly different (Table 6).

3.4 Discussion – Rasprava In this research, measuring damages to different regeneration groups after wood extraction showed that in both traditional and mechanized methods of logging, the damages to seedlings were less than small and large saplings, Hoseini (1994) and Ahmadi (1996) have showed the same results. Also damages to all groups of regeneration in mule logging were significantly lower than in tractor logging. Wang (1997) and Rummer et al. (2001) in their research in mountainous forest and selective silviculture method came to the same results. In our case study the reason that the damage to regeneration is higher with mechanized method than with traditional method is because the skid trails are planned and constructed after the felling phase. Therefore, directional felling towards trails is not possible and damages to regenera-

Table 5 Results of analysis of variance between percentages of soil dry bulk density increase compared to control in traditional and mechanized methods at two depths Tablica 5. Analiza varijance postotnoga pove}anja prirodne gusto}e tla vlaka pri privla~enju drva mulama i skiderima te usporedba s kontolnim uzorcima na dvjema dubinama Source of variable Izvor varijablnosti

Soil dry bulk density increase compared to control, % Pove}anje prirodne gusto}e tla u odnosu na kontrolni uzorak, %

Degree of freedom Stupnjevi slobode

Means square Sredina kvadrata

F value F vrijednost

Among groups Izme|u grupa

1293.310

431.103

Significant – Zna~ajno 6.415

Inside groups Unutar grupa

2419.313

67.203

Total Ukupno

3712.623

146

Croat. j. for. eng. 30(2009)2

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

R. Naghdi et al.

Table 6 Comparison of percentage of soil dry bulk density increase between mule and skid trails (t test) Tablica 6. Usporedba postotnoga pove}anja prirodne gusto}e tla vlaka pri privla~enju drva mulama i skiderima (t-test) Trail Vlaka

t value t vrijednost

Degree of freedom Stupnjevi slobode

F value F vrijednost

Significance Zna~ajnost

Mule trail – skid trail (0–10 cm) Animalna vlaka – traktorska vlaka (0 – 10 cm)

2.878

3.433

Respectively significant Vrlo zna~ajno

Mule trail – skid trail (0–20 cm) Animalna vlaka – traktorska vlaka (0 – 20 cm)

–2.182

0.224

Significant Zna~ajno

Mule trail (First depth with second) Animalna vlaka (Prvi prolaz s drugim prolazom)

4.048

0.118

Respectively significant Vrlo zna~ajno

Skid trail (First depth with second) Traktorska vlaka (Prvi prolaz s drugim prolazom)

–0.451

5.479

Non significant Bez zna~ajnosti

tion increases during winching logs. The results of this study showed that the type of damage to regeneration groups is mostly broken and complete uprooting and Modry and Hubeny (2003) showed similar results. The results of this research showed that the amount of soil dry bulk density increase in skidder and mule trails compared to control at 0–10 and 10–20 cm depths, was significant. This increase in mule trail at 0–10 cm depth was significantly higher than at 10–20 cm depth. Soil dry bulk density increase in skid trails at 0–10 and 10–20 cm depths was not significantly different. The comparison of percentage of soil dry bulk density increase in skid and mule trails for 100 m3 of wood extraction at 0–10 cm and 10–20 cm depths showed that this increase was significantly different. This increase at 0–10 cm depth for mule trail was higher than for skid trail. The results of study carried out by Toms (1996) showed that animal skidding causes more compaction of soil surface. McGonagil (1979) concluded that horse skidding causes more compaction of soil surface than subsoil. With the same amount of wood extraction volume (100 m3), mule logging has more passes (traffic) than the skidder, and therefore causes more compaction of soil surface. Comparison of soil dry bulk density increase at 10–20 cm depth for mule and skid trails showed that this increase in skid trail was higher than in mule trail. Although the number of skidder passes was lower, the weight of the loaded skidder caused depth soil compaction. The soil moisture content at 10–20 cm depth was nearer to optimum soil moisture content and maximum compaction occured at this moisture content. Croat. j. for. eng. 30(2009)2

4. Conclusions – Zaklju~ci This study showed that damages to natural regeneration due to logging in traditional method was lower than in mechanized method, but the soil compaction of surface soil (0–10 cm depth) by animal logging was higher than by tractor logging. Therefore regeneration establishment would be difficult. In this study the skid trails were planned after felling, which caused more damages to soil and forest stand. Planning skid trail before felling phase could be applied with directional felling to avoid high stand damages. This study compared both extraction systems in a unit area to give the logging planners a real evaluation of site damages caused by both logging systems. The results of this research can be useful as environmental criteria for future researches to evaluate current logging systems in hilly terrains and to choose the best alternative and develop a decision support system for logging planning in this area.

5. References – Literatura Ahmadi, H., 1996: Study of logging damages on forest stand. MSc thesis, Faculty of Natural Resources, University of Tehran, p. 148. Bob, R., 2002: Forest Operations Technology. Southern Research Station, used forest service, Southern Forest Resource Assessment Draft Report, pp. 341–353. Buckley, D. S., Crow, T. R., Nauertz, E. A., Schulz, K. E., 2003: Influence of skid trails and haul roads on understory plant richness and composition in managed forest landscapes in Upper Michigan, USA. Forest Ecology and Management 175(1–3): 509–520. Croke, J., Hairsine, P., Fogarty, P., 1999: Runoff generation and redistribution in logged eucalyptus forest, south–eastern Australia. Journal of Hydrology 216(1–2): 56–77.

147

R. Naghdi et al.

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

Defossez, P., Richard, G., 2002: Models of soil compaction due to traffic and their evaluation. Soil and Tillage Research 67(1): 41–64.

Herbaceous Cover, Forest Floor and Topsoil on Skid road in an Oak (Quercus petrea L.) Forest. Journal of Terramechanics (in press).

Eric, R. L., 2006: Assessing Soil Disturbances Caused by Forest Machinery. Forest Engineering, UNB, ppt: 1–25 pp.

McGonagil, K., 1979: Production Study Horse and Mule Logging Alabama. US Forest service.

Ghaffariyan, M. R., 2003: Evaluation of production and damages to soil and regeneration due to skidding by mule. MSc thesis, Faculty of Natural Resources, University of Tehran, p. 109.

Modry, M., Hubeny, D., 2003: Impact of Skidder and HighLead System Logging on Forest Soils and Advanced Regeneration. Journal of Forest Science 49(6): 273–280.

Grigal, D. F., 2000: Effects of extensive forest management on soil productivity. Forest Ecology and Management 138(1–3): 167–185.

Naghdi, R., Bagheri, I., Akef, M., Mahdavi, A., 2007: Soil Compaction Caused by 450C Timber Jack Wheeled Skidder (Shefarood forest northern Iran). Journal of forest science 53(7): 314–319.

Hamza, M. A., Anderson, W. K., 2005: Soil compaction in cropping systems. A review of the nature, causes and possible solutions. Soil and Tillage Research 82(2): 121–145. Hoseini, S. M., 1994: Study of logging damages on forest stand in Darabkola forest, northern Iran. MSc thesis, Faculty of Natural Resources, University of Tarbiyet moderras, p. 129. Hoseini, S. M., 2002: Evaluation of cable logging and ground skidding systems in Sari forest. PhD thesis, Faculty of Natural Resources, University of Tarbiyet moderras, p. 110. Jamshidi, A. R., 2005: Effects of ground skidding system on soil physical properties of skid trails and production. MSc thesis, Faculty of Natural Resources. University of Tarbiyet moderras, p. 75. Johns, J. S., Barreto, P., Uhl, C., 1996: Logging damage during planned and unplanned logging operations in the eastern Amazon. Forest Ecology and Management 89(1–3): 59–77. Kim, D. C., 2000: Soil Compaction Impacts on Tree Roots. University of Georgia WARNELL School of Forest Resources Extension Publication For 00–8. Makineci, E., Demir, M., Aydýn, C., Yilmaz, E., 2008: Effects of Timber Skidding on Chemical Characteristics of

Pinard, M. A., Barker, M. G., Tay, J., 2000: Soil disturbance and post–logging forest recovery on bulldozer paths in Sabah, Malaysia. Forest Ecology and Management 130(1–3): 213–225. Rummer, B., Dubois, M., Bliss, J., Toms, C., 2001: A survey of animal powered logging in Alabama. The international mountain logging and 11th Pacific Northwest Symposium, 12 pp. Shaw, J. N., Carter, E. A., 2002: Timber harvesting effects on spatial variability of southeastern U.S. Piedmont soil properties, Soil Science 167(4): 288–302. Toms, C. W., Wilhoit, J. H., Rummer, R. B., 1996: Animal Logging in the Southern United States, ASAE Pap, No, 96–5005, ASAE, St, Joseph, MI, American Society of Agricultural Engineers, 13 pp. Wang, L., 1997: Assessment of Animal Skidding and Machine Skidding, China. Journal of Forest Engineering 8(2): 57–64. Williamson, J., Neilson, W., 2000: The Influence of Forest Site on Rate and Extent of Soil Compaction and Profile Disturbance of Skid Trails during Ground–Based Harvesting. Canadian Journal Forest Research 30: 1196–1205 pp.

Sa`etak

[tete na tlu i pomlatku pri privla~enju drva skiderima i animalnom vu~om Privla~enje drvnih sortimenata od panja do pomo}noga stovari{ta uzrokuje velika o{te}enja u {umskoj sastojini i na tlu, {to se o~ituje u promjenama fizi~kih i kemijskih svojstava tla te o{te}enjima na pomlatku. Istraìvanje je provedeno u dvama susjednim {umskim predjelima u sjevernom Iranu. Cilj je istraìvanja bio ocjena posljedica radova pridobivanja drva na obnovu sastojine i zbijanje tla pri privla~enju drva mulama i skiderima Timberjack C450. Privla~enje drva skiderima s vitlom uglavnom se primjenjuje u sjevernom Iranu zbog manjih tro{kova od drugih sustava privla~enja i zbog toga {to je takav sustav pogodan za privla~enje pri svim metodama sje~e i izradbe. Animalno privla~enje drva uglavnom se koristi u prorednim sastojinama i na podru~jima s malom otvoreno{}u {uma. Na istraìvanim {umskim predjelima odabrane su {umske vlake sa sli~nim uzdu`nim nagibom, smjerom privla~enja i ukupnim obujmom privu~enoga drva. Sjeklo se motornom pilom u sastojini orijentalne bukve (Fagus orientalis Lipsky) s prosje~nom drvnom zalihom od 430 m3/ha. O{te}enost biljaka u razli~itim razvojnim stadijima

148

Croat. j. for. eng. 30(2009)2

Damages of Skidder and Animal Logging to Forest Soils and Natural Regeneration (141–149)

R. Naghdi et al.

utvr|ena je prije i nakon sje~e i izradbe te nakon privla~enja drva. Za procjenu o{te}enja primijenjena je metoda planskoga nasumi~noga uzorka povr{ine 100 m2 u mreì kvadrata 100 ´ 100 m. Radi odre|ivanja zbijenosti tla zbog privla~enja drva mulama i skiderima na svakom istraìvanom {umskom predjelu izabrane su vlake duljine 200 m sa smjerom privla~enja nizbrdo, prosje~nim uzdu`nim nagibom od 24 % i bez popre~noga nagiba na izabranim vlakama. Za odre|ivanje prirodne gusto}e tla kori{tena je metoda uzimanja uzoraka pomo}u valjaka. Uzorci su uzimani na dvjema dubinama tla, od 0 do 10 cm i od 10 do 20 cm na traktorskoj i na animalnoj vlaci nakon 100 m3 privu~enoga drva. Rezultati pokazuju da su o{te}enja na nara{taju zna~ajno nià pri privla~enju drva mulama nego pri privla~enju skiderima (tablice 1 i 2). [to se ti~e zbijenosti tla, rezultati pokazuju da postoji zna~ajna razlika u prirodnoj gusto}i tla, koja je ve}a na svim vlakama u odnosu na kontrolne uzorke koji su uzimani pored vlaka na istim dubinama. Pove}anje je na animalnoj vlaci zna~ajno ve}e na dubini od 0 do 10 cm nego na dubini od 10 do 20 cm, ali postotak pove}anja prirodne gusto}e tla u usporedbi s uzorkom na traktorskoj vlaci na objema dubinama ne razlikuje se zna~ajno. Pove}anje prirodne gusto}e suhoga tla na dubini od 0 do 10 cm s obzirom na kontrolni uzorak ve}e je na animalnoj vlaci, ali na dubini od 10 do 20 cm to je pove}anje ve}e na traktorskoj vlaci. Ovo je istraìvanje pokazalo da su o{te}enja u prirodno obnovljenoj mladoj sastojini manja pri privla~enju drva mulama, ali je zbijenost tla na dubini od 0 do 10 cm ve}a nego pri privla~enju drva skiderima. Tako|er u ovom istraìvanju {umske su vlake bile projektirane nakon sje~e, {to se pokazalo lo{e jer su nastale velike {tete na tlu i u {umskoj sastojini. Da bi se izbjegla velika o{te}enja sastojine, uz projektiranje {umskih vlaka prije sje~e trebalo bi primijeniti i usmjereno obaranje stabala. Rezultati ovoga istraìvanja mogu biti primjenjivi kao okoli{ni kriterij u budu}im istraìvanjima pri vrednovanju trenuta~noga sustava i odabira najboljega na~ina privla~enja drva u brdskim podru~jima. Klju~ne rije~i: animalno privla~enje, privla~enje skiderima, o{te}enje nara{taja, zbijanje tla

Authors' address – Adresa autorâ: Asst. Prof. Ramin Naghdi, PhD. e-mail: rnaghdi@guilan.ac.ir University of Guilan Faculty of Natural Resources Department of Forestry P.O. Box 1144 Somehsara IRAN Asst. Prof. Majid Lotfalian, PhD. e-mail: mlotfalian@yahoo.com Aghil Moradmand Jalali, MSc. e-mail: amjalaly@yahoo.com University of Mazandaran Faculty of Natural Resources Department of Forestry Sari IRAN

Received (Primljeno): July 12, 2008 Accepted (Prihva}eno): November 20, 2009 Croat. j. for. eng. 30(2009)2

Iraj Bagheri, MSc. e-mail: biraj@guilan.ac.ir University of Guialn Faculty of Agriculture Department of Agricultural Mechanization Engineering P.O. Box 41335–3179 Rasht IRAN

149

Original scientific paper – Izvorni znanstveni rad

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works Selcuk Gumus, Burak Aricak, Korhan Enez, H. Hulusi Acar Abstract – Nacrtak Forest roads provide access for people to study, enjoy or contemplate natural ecosystems. Therefore, roads are one of the most important tools needed in forestry. Forest roads are built by excavation of soil and rock. Rockfall occurs during construction works, caused by excavated rock pieces on embankment slopes and by blasting of block rock masses. This study analyzes rockfall damage to trees during forest road construction. Injuries, branches brokenness, trunk wounds and collapse were defined as tree damages because of rockfall during road construction. These damages have been analyzed by site measurements and statistical analysis. In the study area it has been determined that 90.48% of damaged trees are in the first 10 m after the beginning of the embankment slope, and the rest of the trees are positioned between 11 m and 23 m away from the beginning of the embankment slope. The average injury area of the damaged trees was calculated as 1,081 cm2 for the first 10 m from the embankment slope, and an injury area of 1,463 cm2 was calculated for between 11 m and 23 m on the embankment slope. This is a very important result in terms of forest protection. Wooden obstruction or synthetic holders should be used as preventative measures in the first 10 m of the embankment slope of the road to block rockfall or slow it down. Keywords: Forest road, construction, tree damage, rockfall, Turkey

1. Introduction – Uvod The opening-up of forests enables the application of rational forestry, which is more productive and suitable for sustainable use. Therefore, forest roads are one of the most important tools needed. In Turkey, forestry work is carried out in 21 million hectares of forest land scattered in different geographical regions. A good road network is required for performing work in these forest lands with different climates and topographical conditions. Forest villagers also benefit from the road network for their transportation. So, forest roads provide economic, social and cultural benefits for all user groups (Acar and Senturk 2000, Acar and Gumus 2005). According to the Turkish General Directorate of Forestry (GDF), technical and economical management of forests requires 210,000 km of forest roads to be constructed (SPO 2001). Up to now, approximately 150,000 km of forest roads have been constructed. GDF plans to construct 1,000 km of new forest roads per year, so they will be on the agenda for a long time to come. Croat. j. for. eng. 30(2009)2

Forest roads are built by excavation of soil and rock. Rockfall occurs during construction works, caused by excavated rock pieces on embankment slopes and the blasting of block rock masses. In Turkey, the traditional use of bulldozers causes loss of land and damage to trees and forest habitat (Acar and Eker 2007). The ecological balance of forests and trees is adversely affected by rockfall and road construction works (Luce and Wemple 2001, Madej 2001, Tague and Band 2001, Tunay and Melemez 2004,). It has been found that injured trees are more sensitive to insect epidemic (Lemperiere 1994, Fielding and Evans 1997, Ozcan et al. 2006). In the study area particularly, the pest Dendroctonus micans (Great Spruce Bark Beetle) has been a serious problem for a long time. There have been many studies on the formation and mechanics of rockfalls (Larsen and Parks 1997, Marquinez et al. 2003, Heidenreich 2004, Segalini and Giani 2004, Perret et al. 2004, Perret et al. 2006, Dorren et al. 2005, Aydin 2007) and preventative measures have been researched (Berger and Rey 2004, Guzzetti and Reichenbach 2004, Brauner et al. 2005, Stoffel et al. 2006).

151

S. Gumus et al.

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

2. Rockfall and tree damage – Odroni i o{te}enja stabala

Fig. 1 General modes of motion of rocks during their descent on slopes in relation to mean slope gradients Slika 1. Op}eniti na~ini kretanja stijenja pri kotrljanju niz padinu s obzirom na prosje~ni nagib terena There has been no comprehensive study about the effects of rockfall on forest trees as a result of forest road construction. Construction works have a significant adverse effect on forest trees because of the rockfall they create. This effect mostly appears in the form of tree damage. The aim of this study is to analyze tree damage occurring during forest road construction.

After the rock has been detached and starts to move, it descends the slope in different modes of motion. These modes of motion strongly depend on the mean slope gradient (Fig.1). The three most important modes of motion are freefall through the air, bouncing on the slope surface and rolling over the slope surface. Freefall occurs if the slope gradient below the potential falling rocks exceeds 76°, but in different field situations this value varies. Fig. 1 shows that around 70° the motion of the rock gradually transforms from bouncing to falling (Dorren 2003). Forest road constructions produce many rock pieces. If these rock pieces are not carefully placed on the embankment slope, they roll down it. This situation is unavoidable during bulldozer digging or blasting of rock mass. Stone and rockfalls cause tree damage such as torn bark, collapse, broken branches – crown – trunk. Fig. 2 shows samples of such tree damage.

3. Materials and methods – Materijali i metode 3.1. Materials – Materijali This study has been made throughout 3,100 m of forest road construction area in the Gumushane (Kurtun region of Turkey). The road started at 510,050 m, 4,511,145 m coordinates at UTM projection system with ED 50 datum in 37 T square and ended at 510,375 m, 4,510,865 m point location. The

Fig. 2 Samples of tree damage in study area Slika 2. Primjeri o{te}enja na stablima na istra`ivanom podru~ju 152

Croat. j. for. eng. 30(2009)2

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

study area predominantly contains Quercus petraea (Matt.) Liebl. Subsp. iberica (stwen ex M.Beieb.) Krassiln, and Picea orientalis (L.) Link, and other species such as Pinus sylvestris L., Fagus orientalis Lipsky, and Alnus glutinosa (L.) Gaertn. Supsp. barbata (C.A.Mey.) Yalt. There is no forest cover for 0 – 100, 295–475 and 1,390–2,075 meters of the road section under consideration. According to the measurement results, the average diameter and height of trees is as follows: Quercus petraea 15 cm and 6 m, respectively, Picea orientalis 21 cm and 12 m, Fagus orientalis 29 cm and 18 m, Alnus glunitosa 26 cm and 12 m and Pinus sylvestris 36 cm and 20 m. The road route has three different types of forest stand. Quercus forests have Pinus trees as well. Quercus forests are of low density (less than 10% canopy) and approximately 100 years of age. The main parts of the road section are covered by Fagus forest with Picea and Abies species in low proportions. Fagus forest cover has 40–70% of canopy degrees on forest lands. The forest age is approximately 100 year. Both Quercus and Fagus forests have not any shrubs on the floor. The average slope gradient is 75% throughout the constructed road. The study area is classified as 5th class (>70%) very steeply sloping land according to terrain stability class (B.C. Ministry of Forests 1999). This class has been identified as high likelihood of landslides from timber harvesting or road

S. Gumus et al.

construction area by Schwab and Geertsema (Schwab and Geertsema 2009). Excavators and bulldozers were used during road construction. Rocks were blasted with dynamite between 1,015 and 1,182 meters. Tape measure, caliper, height meter and GPS have been used for measurement of the area. Measurement data, standard topographic map with 1/25,000 scale, satellite image, ESRI ArcGIS software and SPSS statistical software have been used to create a database.

3.2. Methods – Metode Tree and site feature measurements have been made to determine and analyze the damage caused by a rockfall. Two hundred and ninety-five damaged trees have been numbered and their type of damage has been determined. Distribution by damage type, relation to construction techniques (excavator – bulldozer, blasting) and the effect of positional values of damaged trees have been investigated by means of the data collected. In the study area, the following features have been measured and recorded: species of damaged trees, diameter and height of trees, whether the tree has fallen or not, impact on bark or otherwise, whether the impact is on the side or front, whether there are broken branches – trunk – crown, the height of the main injury, location of injury, UTM coordinates of damaged trees.

Fig. 3 Damage to trees and its geographical disturbance Slika 3. Zemljopisni prikaz {teta na drve}u Croat. j. for. eng. 30(2009)2

153

S. Gumus et al.

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

Geographical information systems (GIS) and remote sensing databases have been used to determine the positional values of the measured trees. While positional information of the damaged trees has been obtained by GPS, satellite images have been combined with data that have been created on GIS. Consequently, information was obtained about the distance between the road and trees, and values of slopes and width of the construction area (Fig. 3). Spot image (2.5 m resolution, May 2006) has been used to determine the construction area. A digital elevation model (DEM) was prepared in GIS database for calculating the gradient of the area. The relation between damage to trees in stands remaining after construction and the reason for the damage has been examined with the help of this study. For this purpose, the following parameters have been determined: frequency distribution of the damage type, distance between tree and road and construction techniques. The relationship between damaged trees’ distance from road, injury status, fallen, broken branches – crown – trunk situation and the relationship diameter of trees, height of trees and side gradient were tested by independent t test. Consequently, a statistical decision was made according to the accepted hypothesis evaluation of t statistic and significance levels. The relations between rock crash type, side – front – graze, to remaining trees and height-width of main injuries were determined by the correlation analysis. The correlation displayed p<0.05 at significance levels (Ozdamar 2004).

4. Results – Rezultati Two hundred and ninety-five damaged trees have been identified throughout the 3,100 m of forest road construction area. Table 1 shows the type of damage, the number of trees and ratio of distribution. The measurement of injury has not been made for trees, which have fallen or whose bark has been torn. Several trees have broken branches and crown. It has been determined that 90.48% of injured trees are in the first 10 m after the beginning of the embankment slope. Weighted average of the injured surface area on these trees has been calculated as 1,081 cm2. Specific measurements also show as follows: the average height of the injured area is 56 cm, the average diameter of trees is 22 cm, the average tree height is 12 m, and the average slope gradient is 74% (Table 2). It has been determined that tree collapse and fractured branches – trunk – crown are effective for the first 10 m, like the injury area. The rest of the trees are positioned between 11 m and 23 m from the beginning of the embankment

154

Table 1 The distribution of tree damage Tablica 1. Raspodjela o{te}enja na stablima Damage type Vrsta {tete Injuries (torn bark) Ozljeda kore Collapses of tree Izvaljeno stablo Trunk brokenness Slomljeno deblo Branches brokenness Slomljene grane Crown brokenness Slomljena kro{nja

Number of trees Broj stabala

Damage ratio in all trees Udio {tete na svim stablima, %

189

64.07

32.54

3.39

25.76

8.81

slope. Weighted average of the injured surface area on these trees has been calculated as 1,463 cm2. The average height of the injured area is 30 cm, the average diameter of trees is 24 cm, the average tree height is 15 m, and the average slope gradient is 77%. Broken branches were determined for 16 (34%) of 47 measured trees and 6 (12%) trees had crown damage between 1,015 and 1,182 m of the road. Blasting by dynamite for rock excavation was used as a road construction technique at this section of the studied road. There is 1 (2%) tree that has two injuries. Forty-eight (19%) of 248 trees have broken branches and 20 (8%) trees have broken crowns on other parts of the road where excavator and bulldozers were used for rock and soil excavation. Twelve (5%) trees suffered both types of damage. Variables of damage effect were obtained by statistical relationship after determination of damage frequency. When the fallen trees’ diameter, height, distance to road and slope gradient were compared, according to a non-equal variance t Test (t=6.47, sd=291.30, p=0.000ns), the average distance of fallen and remaining trees to roads was different, with p<0.05 significance level. In other words, the distances of trees to roadside are significant in felling of trees. The diameter, height, distance to roadside and slope gradient of broken-trunk trees and remaining trees were evaluated. According to an equal variance t test (t=2.83, sd=294, p=0.005ns), the average tree height of broken-trunk and remaining trees are not the same, with significance level (p<0.05). It is fair to say that, according to the results, small trees are more affected than larger trees. The other factors are not effective on trunk fracture at p<0.05 signifiCroat. j. for. eng. 30(2009)2

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151â&#x20AC;&#x201C;158)

S. Gumus et al.

Distance to road (m) Udaljenost do ceste, m

Number of Injured trees Broj o{te}enih stabala

Average injured surface cm2 Prosje~na povr{ina ozljede (cm2)

Average injured height (m) Prosje~na visina ozljede, m

Number of collapsed trees Broj izvaljenih stabala

Number of broken trunk Broj slomljenih debala

Number of broken crown Broj polomljenih kro{anja

Number of broken branches Broj slomljenih grana

Average tree diameter (cm) Prosje~an promjer stabala, cm

Average tree height (m) Prosje~na visina stabala, cm

Average slope gradient (%) Prosje~an nagib terena, %

Table 2 Frequency distribution and features of damaged trees according to distance to road Tablica 2. Raspodjela u~estalosti i obilje`ja {tete na stablima u ovisnosti o udaljenosti od ceste

<1 1 2 3 4 5 6 7 8 10 11 12 13 14 17 18 19 20 22 23 Total Ukupno

22 46 28 22 24 7 3 12 2 5 4 2

1350 1719 799 660 793 645 707 600 751 995 337 290

1.00 0.88 0.48 0.40 0.86 0.22 0.40 0.46 0.17 0.80 0.10 0.10

2 7 13 11 8 15 15 7 6 3 2 4 1

1 10 3 5 6 1

10 16 20 15 13

1 2 2 1 4 1 1

2400 1525 600 600 2681 4462 1975

1.00 0.50 1.00 0.10 0.50 0.10 0.10

27 23 17 21 20 22 24 23 19 25 15 17 32 32 20 28 30 26 26 23

16.33 14.63 9.58 11.41 10.54 11.52 12.81 12.33 10.14 13.43 7.40 8.50 17.00 16.00 12.00 16.00 18.00 20.75 21.00 21.33

70 71 71 75 74 76 71 80 80 77 78 75 80 75 75 80 75 80 75 80

189

1 1 1

cance level. No correlation was obtained between trunk breaking and tree diameter, and however, it can be assumed that small trees have lower diameter than larger trees. When trees with broken branches were examined according to the variables, the variance of average slope gradient of trees with broken branches and remaining trees was not equal (t=2.017, sd=108.17, p=0.046ns). The average slope gradient affects branch fracture. On the other hand, there is no significant relation between tree diameter, height and distance to roadside and branch fracture at p<0.05 significance level. The variance of average tree height between the broken crown and remaining trees is not equal (t=6.37, sd=52.06, p=0.000ns). The other variables have no effect on broken crown at p<0.05 signifiCroat. j. for. eng. 30(2009)2

4 2

cance level. Small trees have a broken crown more often. This has been especially observed on road segments where dynamite blasting was used for rock excavations. Blasted rock pieces hit the crowns of small trees and damaged them. A comparison was made between height and width of the main injury and impact of rocks and stones with grazing, side and front hit, and build-up of stones and pieces of rock. There is no statistical relation at p>0.05 significance level. The width and height of the main injury do not depend on the impact style of the rocks.

5. Discussion â&#x20AC;&#x201C; Rasprava The objective of this investigation is to define tree damage, analyze the relations between the occur-

155

S. Gumus et al.

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

rence of damage and the site and tree features, and suggest preventive measures against the damage resulting from forest road construction. The damages were defined as injuries, broken branches, broken crown, wounded trunks and collapse of trees. Those were determined in the study area of this investigation. The results show that the main injuries occurred in the first 10 m from the road embankment slope. This is a very important result in terms of forest protection. Preventative measures such as wooden obstructions or synthetic holders to block rockfall or slow it down must be implemented to protect tree trunks especially for up to 1 m from the surface. The slope gradient and distance of trees from the road were the factors affecting the injuries and brokenness. Rolling rock pieces are faster because of slope gradient, and rock pieces become faster and affect trees more seriously at a higher distance than those on steep slope closer to the road. Also, a rise in slope gradient increases the branches brokenness, because rock pieces bounced more at higher slopes and hit the branches. Preventive measures should be taken against rock bouncing where the slope gradient is more than 70%. The technique of forest road construction causes different types of tree damage. Blasting with dynamite for rock excavation resulted in increased branches and crown brokenness. When this technique is used, action should be taken to block the ‘popping-out’ of stones and rocks. Use of dynamite for rock excavation causes approximately double damage to trees than bulldozers do. Excavator is the machine that can provide protection to trees. The excavator should be the choice for forest road construction.

6. Conclusion – Zaklju~ci Forest road construction is one of the major causes of rockfall. The rockfall resulting from forest road construction can cause damages to forest trees. These damages were determined as injuries (torn bark), collapse and broken trunk, branches and crown of trees in the Kurtun study area. The distance of trees to road side, tree heights, slope gradients, rock excavation methods were the affecting factors in damage occurrence. Road construction causes damage to forest trees and this can result in the decrease of growing stock as well as in insect epidemic. Forest road construction must be carried out very carefully taking adequate protective measures.

156

6. References – Literatura Acar, H. H., Eker, M., 2007: Orman yolu yapiminda ekskavatörlerin kullanilmasi ve çevresel açidan yararlari (The use of excavators in forest road construction and their environmental benefits), Journal of Southwest Anatolia Forest Research Institute 5: 97–128. Acar, H. H., Gumus, S., 2005: Forest roads in Turkish forestry, Mehanizacija {umarstva 26 (2): 117–120. Acar, H. H., Senturk, N., 2000: Orman Yol Aglari, Heyelanlar ve Mühendislik Jeolojisi (Forest road network, landslides and geology of engineering), Journal of Faculty of Forestry, Istanbul University, B, 45 (3–4): 71–84. Aydin, A., 2007: Ormanlik Alanlarda Taº ve Kaya Yuvarlanmalari (Rockfalls in forests lands), Journal of faculty of forestry, Istanbul University, 57 (2): 1–16. B. C. Ministry of Forests, 1999: Mapping and Assessing Terrain Stability Guidebook, Second Edittion, Forest Practice Code of British Columbia, 43 p., Victoria. Berger, F., Rey, F., 2004: Mountain Protection Forests against Natural Hazards and Risks: New French Developments by Integrating Forests in Risk Zoning, Natural Hazards 33: 395–404. Brauner, M., Weinmeister, W., Agner, P., Vospemik, S., Hoesle, B., 2005: Forest management decision support for evaluating forest protection effects against rockfall, Forest Ecology and Management 207: 75–85. Dorren, L. K. A., Berger, F., Hir, C., Mermin, E., Tardif, P., 2005: Mechanisms, effects and management implications of rockfall in forests, Forest Ecology and Management 215: 183–195. Dorren, L. K. A., 2003: A review of rockfall mechanics and modeling approaches, Progress in Physical Geography 27 (1): 69–87. Fielding, N. J., Evans, H. F., 1997: Biological control of Dendroctonus micans (Scolytidae) in Great Britain, Biocontrol News and Information 18 (2): 51–60. Guzzetti, F., Reichenbach, P., 2004: Rockfall Hazard and Risk Assessment Along a Transportation Corridor in the Nera Valley, Central Italy, Environmental Management 34 (2): 191–208. Heidenreich, B., 2004: Small and half scale experimental studies of rockfall impacts on sandy slopes. Ph.D.Thesis, No:3059, EPFL, Lausanne. Larsen, M. C., Parks, J. E., 1997: How wide is a road? The association of roads and masswasting in a forested montane environment, Earth Surf. Process. Landforms 22: 835–848. Lemperiere, G., 1994: Ecology of the Great European Spruce Bark Beetle Dendroctonus micans (Kug.), Ecologie 25 (1): 31–38. Luce, C. H., Wemple, B. C., 2001: Introduction to special issue on hydrologic and geomorphic effects of forest roads, Earth Surf. Process. Landforms 26: 111–113. Croat. j. for. eng. 30(2009)2

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

Madej, M. A., 2001: Erosion and sediment delivery following removal of forest roads, Earth Surf. Process. Landforms 26: 175–190. Marquinez, J., Menendez, D. R., Farias, P., Jimenez, S. M., 2003: Predictive GIS – Based Model of Rockfall Activity in Mountain Cliffs, Natural Hazards 30: 341–360. Ozcan, G. E., Eroglu, M., Akinci, H. A., 2006: Ladin ormanlarinda Dendroctonus micans (Kugelann) (Coleoptera: Scolytidae)’in zarar durumu ve Rhizophagus grandis (Gyllenhal) (Coleoptera: Rhizophagidae)’in zararlinin populasyonuna etkisi (Pest status of Dendroctonus micans /Kugelann/ Gyllenhal/Coleoptera: Scolytidae/ and the effect of Rhizophagus grandis /Gyllenhal/ /Coleoptera: Rhizophagidae/ on the population of Dendroctonus micans in the oriental spruce forests), Turkish journal of entomology 30 (1): 11–22. Ozdamar, K., 2004: Paket programlar ile istatistiksel veri analizi 1 (Statistical data analysis by software 1), Kaan Press. 649 p. Eskiºehir 2004. Perret, S., Dolf, F., Kienholz, H., 2004: Rockfall into forests: Analysis and simulation of rockfall trajectories –considerations with respect to mountainous forests in Switzerland, Landslides 1: 123–130. Perret, S., Baumgartner, M., Kienholz, H., 2006: Inventory and analysis of tree injuries in a rockfall – damaged forest stand, Eur J Forest Res 125: 101–110.

S. Gumus et al.

Schwab, J. W., Geertsema, M., 2009: Terrain stability mapping on British Columbia forest lands: an historical perspective, Nat Hazards, DOI 10.1007/s11069–009–9410–3. Segalini, A., Giani, G. P., 2004: Numerical Model for the Analysis of the Evolution Mechanisms of the Grossgufer Rock Slide, Rock Mechanics and Rock Engineering 37 (2): 151–168. SPO, 2001: VIII. Five years period development reports, State Planning Organization – Turkish Republic Prime Ministry Office, SPO P.N: 2531. Forestry Specialist Community Report No: 547, Ankara, p. 158–169. Stoffel, M., Wehrli, A., Kuhne, R., Dorren, L. K. A., Perret, S., Kienholz, H., 2006: Assessing the protective effect of mountain forests against rockfall using a 3D simulation model, Forest Ecology and Management 225: 113–122. Tague, C., Band, L., 2001: Simulating the impact of road construction and forest harvesting on hydrologic response, Earth Surf. Process. Landforms 26: 135–151. Tunay, M., Melemez, K., 2004: Dik egimli arazide orman yol inºaatinin çevresel etkileri (Environmental Effects of Forest Road Construction on Steep Slope), Ekoloji 13 (53): 33–37.

Sa`etak

Analiza o{te}enja stabala uzrokovanih odronima prilikom izgradnje {umske ceste [umske ceste omogu}uju pristup svima koji èle istraìvati ili promatrati prirodne ekosustave ili samo u njima uìvati. Stoga je jedan od najva`nijih objekata u {umarstvu – cesta. [umske se ceste grade iskopavanjem zemlje i kamena. Odroni se javljaju tijekom gra|evinskih radova, pri iskopavanju stjenovitih komada na strmim terenima te ru{enjem ve}ih blokova stijena. Cilj je ovoga istraìvanja definirati o{te}enja stabala, analizirati odnos izme|u pojavljivanja {teta i poloàja te svojstava stabala. Nadalje, predloìt }e se preventivne mjere protiv {tete prilikom izgradnje {umskih cesta. Ovo je istraìvanje napravljeno na gradili{tu {umske ceste u Gumushane (oblast Kurtun) u Turskoj ukupne duljine 3100 metara. Uzrokovane {tete, na istraìvanom podru~ju, definirane su kao ozljede, slomljene grane, debla ili kro{nja te izvaljena stabla. [tete su analizirane prema mjestu mjerenja razli~itim statisti~kim analizama. Na podru~ju istraìvanja utvr|eno je 295 o{te}enih stabala i pritom odre|ene vrste o{te}enja. Raspodjela prema vrsti o{te}enja, odnos prema tehnici izgradnje (bager, miniranje) i u~inak poloàja o{te}enih stabala bili su istraìvani pomo}u prikupljenih podataka. Na istraìvanom je podru~ju utvr|eno kako se 90,48 % o{te}enih stabala nalazi u prvih 10 metara nasipne strane, dok je ostatak stabala smje{ten izme|u 11 i 23 metra od mjesta gradnje. Prosje~na izra~unata povr{ina o{te}enja iznosila je 1081 cm2 na udaljenosti od 10 metara, a 1463 cm2 na udaljenosti od 11 do 23 metra s nasipne strane. Promatrav{i te podatke s podru~ja za{tite {uma izlazi kako bi drvene prepreke ili sinteti~ke drà~e trebalo koristiti na prvih 10 metara nasipne strane kao preventivnu mjeru za spre~avanje ili usporavanje odrona. Nagib padine i udaljenost stabala od mjesta gradnje utjecajni su ~imbenici za broj o{te}enja i sru{enih stabala. Brzina kotrljanja kamenja proporcionalno raste s pove}anjem nagiba i u tom slu~aju kamenje djeluje na udaljena stabla ve}om snagom nego na stabla koja su bliè mjestu gradnje i koja su na manjem nagibu. Pove}anje nagiba Croat. j. for. eng. 30(2009)2

157

S. Gumus et al.

Analysis of Tree Damage Caused by Rockfall at Forest Road Construction Works (151–158)

terena tako|er utje~e na pove}ani broj slomljenih grana jer je i odbijanje od podloge mnogo izra`enije. Preventivne bi se mjere trebale poduzimati na nagibima ve}im od 70 %. Razli~ite tehnike gradnje uzrokuju razliku u opsegu i vrsti {teta na stablima. Miniranje pri probijanju trase ceste najvi{e o{te}uje grane i kro{nje stabala. U takvim je slu~ajevima potrebno poduzeti za{titne mjere radi spre~avanja nekontroliranoga izlijetanja dijelova stijena. Upotrebu eksploziva trebalo bi svesti na najmanju mogu}u mjeru jer su {tete otprilike dvostruko ve}e nego kada se koristi samo bager. Gradnja {umskih cesta o{te}uje dube}a stabla i time se smanjuje drvna zaliha i mogu}nost napada {tetnika na oslabljena stabla, stoga bi svi radovi na izgradnji trebali biti obavljeni {to pa`ljivije, uz upotrebu prikladnih za{titnih mjera. Klju~ne rije~i: {umska cesta, izgradnja, o{te}enje stabala, odron, Turska

Authors' address – Adresa autorâ: Asst. Prof. Selcuk Gumus, PhD. e-mail: sgumus@ktu.edu.tr Prof. H. Hulusi Acar, PhD. e-mail: hlsacar@ktu.edu.tr Karadeniz Technical University Faculty of Forestry Department of Forest Engineering 61080 Trabzon TURKEY

Received (Primljeno): May 14, 2009 Accepted (Prihva}eno): November 16, 2009

158

Ress. Asst. Burak Aricak, PhD. e-mail: baricak@kastamonu.edu.tr Ress. Asst. Korhan Enez, PhD. e-mail: kenez@kastamonu.edu.tr Kastamonu University Faculty of Forestry, Department of Forest Engineering 37150 Kastamonu TURKEY Croat. j. for. eng. 30(2009)2

Original scientific paper – Izvorni znanstveni rad

Ergonomic Characterization of Harvesting Work in Karelia Yuri Gerasimov, Anton Sokolov Abstract – Nacrtak A comparison of the ergonomic performance of 13 harvesting machine models was assessed from an ergonomic viewpoint. The main objective of the study was to compare ergonomic performance to harvesting machine operators’ work and propose viable solutions to improve the work environment. The principal assessed ergonomic requirements were operators’ workspace, operators’ seats, visibility, work postures, whole-body vibration and noise in the cab, all as related to the tasks involved in typical harvesting cycles. Altogether, more than 120 different parameters that impact ergonomics and work conditions were measured directly at workplaces in the actual working conditions. The results were then compared to the effective norms and the degree of compliance with the stipulated values was determined. The obtained estimates for the degree of compliance were integrated. This permits a direct comparison of the work-load on operators with a single integrated indicator (severity). In many respects the ergonomic standard is now good, except for skidders. Visibility and work postures were considered to be the most critical features influencing the operator’s performance. Even in highly mechanized harvesting work, problems still exist despite extensive development of cabs. The best working conditions in terms of harvesting systems were provided by »harvester + forwarder« in cut-to-length harvesting and »feller buncher + grapple skidder« in full-tree harvesting. The traditional Russian tree-length harvesting done with cable skidders showed the worst results in terms of ergonomics. When a partially mechanized harvesting system is employed, use of cable skidders should be as limited as possible, because, as a whole, they do not comply with the present ergonomic requirements. Keywords: wood harvesting, ergonomics, harvester, forwarder, skidder, feller buncher

1. Introduction – Uvod The ergonomic design of harvesting machines has been the subject of continuous study. Ergonomic guidelines have been developed and successfully introduced to the manufacturers of the machines and to the forest industries. Manufacturers have implemented comprehensive ergonomic improvements. Operator workspace, visibility, lighting, operators’ seats, mounting and alighting, cab climate, and service of machines have been improved. Noise and vibration levels have been reduced (Hansson 1990, Harstela 1990). A very positive result of the mechanization of harvesting work is the drastic reduction of serious accidents and injuries (Axelsson 1998). Increasing mechanization is posing new problems, however. Operators of harvesting machinery are being afflicted by overload injuries to the neck, arms, and cervical spine. The main causes of these Croat. j. for. eng. 30(2009)2

injuries are probably excessive periods of sitting, excessive work intensity during work in fixed, ergonomically inappropriate positions, and repetitive, short-cycle movement patterns. Advice regarding the ergonomic design of the forest machine and maintenance work is given in the Nordic ergonomic guidelines for forest machines (Frumerie 1999). In Russia, wood harvesting has been associated with high accident risk due to low level of mechanization especially with a lethal outcome; the latter has been estimated at 1.4 deaths per 1 million m3 cut (Gerasimov and Karjalainen 2008). Recently, special attention has been paid to safe working conditions in harvesting operations regarding corporate social responsibility. Moreover, comfortable working conditions will make harvesting activities more attractive to youth and employment in a harvesting company more popular (Syunev et al. 2008).

159

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Due to the ergonomic feasibility of harvesting operations being a critical element for the development of wood harvesting in Russia, the main objective of the study was to compare ergonomic performance to harvesting machine operators’ work and propose viable solutions to improve the work environment.

2. Materials and methods – Materijali i metode Hence, there is a need for a comprehensive approach towards the evaluation of ergonomic performance of harvesting operations and selection of the most appropriate technology for Russian conditions. To evaluate the efficiency of the harvesting methods currently used in Russia, the authors performed comprehensive field studies. The Republic of Karelia in north-west Russia was selected as a study region because its territory is very representative in terms of the wide range of used harvesting machinery and the fact that nearly all employed harvesting technologies in different natural conditions are typical for north-west Russia. The study was performed in 2007–2009 and involved 15 harvesting companies which provide approximately 40% of the total harvest in Karelia. The selected companies perform harvesting operations across the whole territory of the Republic of Karelia in different natural and production conditions, and apply all the mentioned technologies using both Russian and foreign machinery. A common approach was used for field data collection and processing. Different parameters that impact ergonomics and work conditions were measured directly at workplaces in the actual working conditions. The results were then compared with the effective norms and standards and the degree of compliance with the stipulated values was determined. The obtained estimates for the degree of compliance for all the measured parameters were integrated into one indicator – to so-called integral work severity rate. This permits a direct comparison of working conditions at different workplaces. A higher severity rate stands for harder working conditions. Depending on this value, the working conditions were categorized as comfortable, relatively comfortable, relatively uncomfortable or uncomfortable.

vesters, forwarders, feller buncher, cable and grapple skidders) were studied during the field measurements (Table 1). A time study of the working cycle was made by means of direct timekeeping using video recording. The total time during which the operator’s body was in an uncomfortable work posture, and the number of working position changes, were averaged. It was necessary to find out the time required for each operation, because some factors determine working conditions change from one operation to another. For example, a harvester operator is exposed to the highest vibration load when the machine is moving, while moving and delimbing/cross-cutting cause the highest noise load. This had to be taken into account when calculating the work-load on operators. Altogether, more than 120 ergonomic parameters listed in the effective Russian and Swedish ergonomic standards and norms were measured in the course of the study, including: Þ Geometrical characteristics such as comfort of the cab layout and seat, location of controls and the operator’s body position were measured using a drawing scale, a measuring tape and a goniometer. Three measurements per parameter were averaged. Þ Forces on hand and foot-operated controls were measured using a laboratory dynamometer. Five measurements per parameter were averaged.

Table 1 Studied harvesting machines Tablica 1. Istra`ivani {umski strojevi Technology Metoda izradbe

Cut-to-length Sortimentna

2.1 Collection and processing of field data Prikupljanje i obrada podataka

Full-tree Stablovna

Field research was carried out at 23 harvesting sites, the locations of which are shown in Fig. 1. Twenty-five harvesting machines of 13 models (har-

Tree-length and full-tree Deblovna i stablovna

160

Type of machine Vrsta stroja Harvester Harvester Harvester Harvester Harvester Forwarder Forwarder Forwarder Forwarder Feller buncher Skidder, grapple Skidder, cable Skidder, cable

Model Number Model Broj John Deere 1070D 2 John Deere 1270D 2 Volvo EC210BLC 1 Valmet 901.3 1 Valmet 911.3 1 Timberjack 1010D 3 John Deere 1110D 3 John Deere 1410D 2 Valmet 840.3 1 Timberjack 850 1 Timberjack 460D 3 TDT–55A 3 TLT–100A 2

Croat. j. for. eng. 30(2009)2

Ergonomic Characterization of Harvesting Work in Karelia (159â&#x20AC;&#x201C;170)

Y. Gerasimov and A. Sokolov

Fig. 1 Study area Slika 1. Podru~je istra`ivanja Croat. j. for. eng. 30(2009)2

161

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Þ Parameters of noise and whole-body vibration were measured separately on all operations of the working cycle using a vibrometer and a noise meter. 20 measurements per operation within the working cycle, and the weighting according to the operation’s share in the working cycle time, were averaged. Þ The degree of windshield cleaning was defined using photo images. The average share of work time during which the operator has to be in an uncomfortable work posture is another important factor that affects the overall comfort of operating the machine. After averaging of the repeated measurements the weighting of the measurements in different conditions is given by Eq. (1). n

∑x t x'=

i=1

i i

where: n xi ti T

(1)

number of different conditions parameter value in the ith condition operational time in the ith condition total time of the working cycle

2 1 n mi ∑ N i = 1 Mi

(2)

where: n number of continuous groups of elementary operations in the working cycle number of elementary operations in the ith mi group total number of elementary operations and Mi logical conditions in the ith group The standardized coefficient of work complexity is defined by Eq. : p log

2 1 q ri = ∑ N i = 1 Ri

(3)

where: q number of continuous groups of logical conditions in the working cycle number of logical conditions in the ith group ri total number of elementary operations and Ri logical conditions in the ith group

162

Ti N    Wi   Z = ∑   Ti − 2∑ Uj  i=1  j =1  Ti 

(4)

where: N number of algorithms number of addresses to controls in the ith Ti algorithm j step number of algorithms 0 −  if action by right hand; Uj =  1 − if action by left hand; Wi

frequency coefficient of the ith algorithm

2.2. Compliance with the effective standards and guidelines – Sukladnost s postoje}im normama i smjernicama The compliance of ergonomic characteristics with the effective standards and norms is defined according to Frumkin et al. (1999) by Eq. .  x − 0.5 ⋅ ( x max + x )  min  V = 1 − 0.69 ⋅  0.5 ⋅ ( x max + x min )   

The working cycle was analyzed according to Frumkin et al. (1999) and was defined by the coefficients of work repetitiveness and complexity. The standardized coefficient of work repetitiveness is defined by Eq. : p st =

The assignment of control activities to the hands is defined by Eq. :

(5a)

if standards determine the possible interval of requirement  x  V = 1 − 0.69 ⋅    x max 

(5b)

if standards determine the maximum possible value of requirement  x min   V = 1 − 0.69 ⋅   x 

(5c)

if standards determine the minimum possible value of requirement where: V degree of compliance of the requirement x measured value of the requirement xmin and xmax minimum and maximum possible value of the requirement according to the standards and norms Each degree can be valued from 0 to 1. The higher the value, the better compliance with the effective standards and norms. The following sources of ergonomic standards and guidelines are taken into account: Þ State standards of the Russian Federation (GOST R 51863–2002, GOST 12.2.102–89, GOST 12.1.012–90, GOST 12.1.003–83, GOST 12.2.120–88), Croat. j. for. eng. 30(2009)2

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Y. Gerasimov and A. Sokolov

Table 2 The results of a time study Tablica 2. Rezultati studija vremena Type of machine Vrsta stroja

Recording time Snimljeno vrijeme (PSH0)

Travel loaded Vo`nja s tovarom

Travel empty Vo`nja bez tovara

Loading and unloading Utovar i istovar

Felling Sje~a

Processing Izradba

Idling Zauzimanje polo`aja

Uncomfortable work postures Neudobni radni polo`aj

min Harvester Harvester Forwarder Forvarder Feller buncher Sje~no vozilo Skidder, grapple Skider s hvatalom Skidder, cable Skider s vitlom

50.6

–

2.0

–

8.1

26.8

13.7

4.0

125.0

20.0

10.0

91.2

–

3.8

28.7

20.4

–

6.8

–

1.8

11.8

–

51.6

23.2

20.1

6.2

–

2.1

16.0

121.5

34.0

46.2

18.2

–

23.1

30.4

Þ Ergonomic guidelines by VNIITE (1983), Þ Ergonomic guidelines by the Swedish National Institute for Working Life, The Forestry Research Institute of Sweden (SkogForsk) and the Swedish University of Agricultural Sciences (Frumerie 1999),

where: weight of the ith requirement out of m requai irements in the ergonomic group

2.3 Categorizing of working conditions Razvrstavanje radnih uvjeta

Each integrated indicator can be valued from 0 to 1. The higher the value, the better the degree of machine sophistication by this factor. Thus, the different machines can be compared using particular ergonomic requirements. The total grading of machine sophistication by ergonomics can be done using the work severity rate (Frumkin et al. 1999) shown in Eq. . n−1 6p min − 1    7(n − 1) − 6∑ p i  (7) I = 7 − 6p min + 6(n − 1)   i=1

The ergonomic characteristics were grouped as following. Þ Location and course of hand and foot-operated controls,

where: minimum value of the integrated indicator pmin n number of the integrated indicators

Þ Ergonomic guidelines by Peskov (2004) and Frumkin et al. (1999). State standards of the Russian Federation prevail in case of different requirements.

Þ Force required to operate the controls,

i=1

Þ Work posture of the operator, Þ Cab and seat position in the cab, Þ Repetitiveness and complexity of the work, Þ Visibility of working and moving directions and cleanliness of the windshield, Þ Noise, Þ Whole-body vibration. The grading of machine sophistication by the ergonomic group can be done using the integrated indicator shown in Eq. m

i=1

Croat. j. for. eng. 30(2009)2

the sum of the integrated indicator values with the exception of the minimum

Þ Operator’s seat,

p = ∑ Vi ⋅ ai

n−1

∑p

(6)

The work severity rate can be valued from 0 to 6. A higher value means a higher severity of conditions of work. Thus, the different machines can be compared using ergonomic factors.

3. Results – Rezultati 3.1 Machines for cut-to-length harvesting method – Strojevi pri sortimentnoj metodi izradbe 3.1.1 Harvesters – Harvesteri Observations on the work cycle of the harvesters, video filming and a time study (Table 2) showed the

163

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Fig. 2 Uncomfortable work postures when operating a harvester without a rotating cab Slika 2. Neudobni radni polo aji pri upravljanju harvesterom bez okretne kabine following distribution of the harvesters’ working cycle by the main elements from the ergonomic point of view: processing (delimbing and cross-cutting) – 53%; tree felling – 16%; travel (movement of the machine to a new position) – 4%; idling (orientation when motionless) – 27%. Regarding uncomfortable work postures, the harvester is a comfortable machine. Valmet and Volvo harvester operators worked almost completely without uncomfortable work postures in typical conditions. This is because these harvester models have a rotating cab and the operator can always observe the operation process looking directly ahead and without having to turn his head in large angles. John Deere harvester cabs were not rotating and, therefore, time spent in uncomfortable work postures was

about 8% (Table 2). The uncomfortable position mainly meant that the operator had to turn his head in significantly large angles in order to monitor cross-cutting and delimbing (Fig. 2). Table 3 shows the main integrated indicators of the working conditions for the surveyed harvester models. The indicators varied between 0 and 1. The higher the indicator was, the better the working conditions were. Valmet harvesters got lower scores in »location and course of controls«. This is mainly because Valmet harvester controls did not comply with three requirements of the Russian norms and standards, namely: the diameter of the control handle falls outside the recommended range (49 mm in comparison with the norm of 20–40 mm); the distance between pedals operated with the same foot

Fig. 3 Work severity rate on ergonomic performance for harvesting machines operator’s work Slika 3. Stupanj te`ine rada prema ergonomskim svojstvima za rukovatelja strojeva pri sje~i, izradbi i privla~enju drva 164

Croat. j. for. eng. 30(2009)2

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Y. Gerasimov and A. Sokolov

Fig. 4 Uncomfortable work postures of forwarder operator’s work Slika 4. Nedobni radni polo`aji pri upravljanju forvarderom was too small (40 mm in comparison with the norm of >50 mm); similarly, the pedal stroke distance was too small (50 mm in comparison with the norm of 70–100 mm). Lower scores in the »work postures« and »operator’s seat« indicators for Valmet were caused by the fact that Valmet’s cabs were considered relatively more cramped compared with John Deere’s cabs. This resulted in noncompliance with the Russian norms set for the longitudinal and vertical seat adjustment range and, consequently, a less comfortable body position (in terms of the angles at the body joints). Volvo’s seat had too narrow arm rests and no adjustable seat backrest. Noise and vibration parameters of the surveyed harvester models did not differ significantly. The »noise« integrated indicator values were close to 0.7, while »vibration« scored close to 1. Comparatively low visibility angle values for Valmet machines resulted from the fact that the vertical observation angle, which is of particular importance for harvesters, was at the lower limit of the range recommended by the Russian standards. The work severity rates for all analyzed harvesters based on the measured data were estimated at less than 3.4, namely 3.2–3.4. Thus, for operators of harvesters the working conditions can be considered to be »comfortable« (Fig. 3). Croat. j. for. eng. 30(2009)2

3.1.2 Forwarders – Forvarderi A time study (Table 2) showed the following distribution of the forwarder’s working cycle by the main work elements from the ergonomic point of view: loading and unloading – 73%; travel loaded (forwarding) – 16%; travel empty – 8%; idling (motionless when orientation) – 3%. According to the time study, forwarder operators spent a considerable time in uncomfortable work postures: 23% of the total work time on average. Uncomfortable postures involved turning the head and body by large angles during loading and movement of the machine (Fig. 4). Table 3 shows the main indicators describing working conditions for the analyzed forwarder models. The Valmet 840.3 forwarder gained lower scores for »location and course of controls« and »foot-operated controls (pedals)«. This can mainly be explained by the fact that, similarly to harvesters of the same brand, the distance between the pedals operated with the same foot and the pedal stroke did not comply with the recommended norms. »Work postures« and »operator’s seat« indicators were lower because the adjustability of the seat position was at the limits of the recommended range. »Visibility of the moving direction« was substantially higher in a John Deere 1010 forwarder, because it has a much shorter front

165

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Fig. 5 Russian skidders TLT–100 Slika 5. Skider TLT–100 ruske proizvodnje (a more compact engine room). »Visibility of the operation direction« was somewhat lower in a John Deere 1410D forwarder, mainly due to the overall large dimensions of this model. Thus, working conditions of the operator were considered to be »comfortable« (I=3.4) for the Timberjack 1110D fowarder, and »relatively uncomfortable« (the work severity rate I being between 3.4 and 4.5) for the rest of the models. Equally to harvesters, the difference in the work severity rate was not significant (Fig. 3).

3.2 Machines for full-tree and tree-length harvesting methods – Strojevi pri stablovnoj i deblovnoj metodi izradbe 3.2.1 Feller buncher – Sje~no vozilo Only one feller buncher model was analyzed in the course of the study, namely the Timberjack 850. A time study (Table 2) showed the following distribution of feller bunchers’ working cycle by the main work elements from the ergonomic point of view: processing (setting the felling head at the tree and

bunching) – 58%; felling – 9%; travel (movement of the machine to a new position) – 33%. This machine proved to be the best in terms of the majority of the evaluation indicators. Table 3 shows the results of the measurements. According to the measurement data, the working conditions of the operators of the Timberjack 850 feller buncher fell into the category of »relatively uncomfortable« due to the value of the work severity rate I being between 3.4 and 4.5, namely 3.5 (Fig. 3). 3.2.2 Skidders – Skideri Finally, two models of Russian-made tracked skidders – TDT–55A and TLT–100 (Fig. 5) manufactured by Onezhsky Tractor Plant – and one model of a wheeled grapple skidder – Timberjack 460D – were analyzed. A time study (Table 2) showed the following distribution of a Russian tracked skidder’s working cycle by the main work elements from the ergonomic point of view: travel loaded (skidding) – 28%; travel empty – 38%; loading – 15%; idling (motionless when trees hooking) – 19%. The average time during which the operator had to be in uncomfortable work postures was 25% of the total work time. Uncomfortable work postures here were more diverse than in the cases of the other machines (Fig. 6). A time study (Table 2) showed the following distribution of the Timberjack 460D grapple skidder’s operation time by the main work elements from the ergonomic point of view: travel loaded (skidding) – 45%; travel empty – 39%; loading – 12%; idling (motionless when orientation) – 4%. Due to the working methods used with the wheeled grapple skidders and the cab design of the analyzed skidder, the operator had to spend a considerable time in uncomfortable work postures, namely 31% of the work time. A typical uncomfortable work posture occurred when

Fig. 6 Uncomfortable work postures of cable skidder operator’s work Slika 6. Neudobni radni polo`aji kop~a{a pri radu skidera s vitlom 166

Croat. j. for. eng. 30(2009)2

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Y. Gerasimov and A. Sokolov

John Deere 1270D

Volvo EC210BLC

Valmet 911.3

John Deere 1010

Timberjack 1110D

John Deere 1410D

Valmet 840.3

Timberjack 850

Timberjack 460D

TDT–55A

TLT–100

Location and course of controls Polo`aj i smjer upravlja~kih komandi

0.86

0.87

0.75 0.75

0.89

0.82

0.84

0.70

0.90

0.73

0.68

0.84

Force required to operate controls Sila potrebna za pokretanje upravlja~kih komandi

1.00

0.98

0.99

1.00 1.00

0.90

1.00

0.99

1.00

0.98

0.71

0.70

Hand – operated controls Ru~no aktiviranje upravlja~kih komandi

0.89

0.88

0.81 0.81

0.87

0.86

0.84

0.90

0.50

0.55

Foot – operated controls (pedals) No`no aktiviranje upravlja~kih komandi (pedale)

0.90

0.89

0.91

0.81 0.81

0.87

0.89

0.77

0.98

0.72

0.80

0.94

Work postures – Radni polo`aji

0.89

0.78 0.78

0.90

0.89

0.75

0.91

0.89

0.87

0.84

Operator’s seat – Sjedalo rukovatelja

0.86

0.73

0.75 0.75

0.88

0.86

0.77

0.70

0.40

0.55

Cab and seat position – Kabina i polo`aj sjedala

0.74

0.72

0.71 0.71

0.65

0.71

0.72

0.65

0.75

0.54

0.47

0.66

Noise – Buka

0.75

0.74

0.70

0.76 0.71

0.61

0.62

0.70

0.64

0.60

0.33

0.19

0.32

Vibration – Vibracije

1.00

0.99

1.00 1.00

1.00

0.98

0.99

0.98

0.69

0.21

0.55

Visibility angles – Kutovi vidljivosti

0.81

1.00

0.48 0.48

0.99

0.97

0.98

0.63

0.99

0.97

Visibility in the operation direction Vidljivost u smjeru radnoga prostora

0.86

0.97

0.99

1.00 1.00

0.95

0.89

0.79

0.85

0.99

0.45

1.00

Visibility in the moving direction Vidljivost u smjeru kretanja

1.00

0.99

1.00

1.00 1.00

0.98

0.46

1.00

0.00

0.99

1.00

Cleanliness of the windshield ^isto}a vjetrobranih stakala

0.90

0.63

0.69 0.69

0.53

0.71

0.67

1.00

0.56

0.00

0.70

Repetitiveness – U~estalost radnih zahvata

1.00

1.00 1.00

0.99

0.81

0.31

1.00

Complexity of work – Slo`enost rada

0.91

0.91 0.91

0.98

1.00

0.91

Ergonomic characteristics Ergonomske zna~ajke

the operator had to turn his head and body in large angles to monitor loading and unloading processes, and also when moving the machine in order to monitor and adjust the grapple and bunch positions. Results for the skidders are shown in Table 3. For the TLT–100 skidder, most indicators were better than for the TDT–55A skidder. This is because the TLT–100 is a later model equipped with a more comfortable and spacious cab, a more comfortable spring mounted seat, and so on. This is why working environment indicators are two to three times better for the TLT–100 skidder. The main weaknesses of the Timberjack 460D were the following: confined cabin, substantially high noise level and lack of visibility (visibility of the moving direction does not comply with the recomCroat. j. for. eng. 30(2009)2

Valmet 901.3

John Deere 1070D

Table 3 Main integrated indicators of working conditions for harvesting machine operator’s work Tablica 3. Osnovni slo`eni pokazatelji radnih uvjeta za rad rukovatelja strojeva za sje~u, izradbu i privla~enje drva

mendations at all, because the forward ground visibility was more than 14 m). Also, a high level of repetitiveness should be noted. Thus, the working conditions of the TLT–100 skidder operators can be considered as »relatively uncomfortable« (I = 4.9, within 4.6–5.8), while with the TDT–55A skidder they were »uncomfortable« (I=5.9) (Fig. 3). The operators’ working conditions with the Timberjack 460D skidder can be considered as »relatively uncomfortable« (I = 4.7). However, there was a significant difference in the measurement-based and personnel survey-based severity rates of work (Sokolov et al. 2008). Naturally, in such conditions only operators who do not perceive the conditions to be uncomfortable, thanks to their good adaptation skills, stay in the job. Other

167

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

operators simply quit the work. This can be seen in the presented results, since for this study operators having substantial work experience with these machines were interviewed.

4. Conclusion and discussion – Zaklju~ci s diskusijom The latest models of John Deere and Volvo machines held the leading position regarding »comfortable« conditions (Fig. 3). For other machines used in cut-to-length harvesting, the results were almost similar; each of these machines was assessed as »relatively uncomfortable«. The Valmet 840.3 had somewhat lower results together with the Timberjack 850 feller buncher. These were followed by a significantly worse Timberjack 460D skidder and Russian TLT–100 skidder. They had similar work severity rates and were assigned to the »relatively uncomfortable« working condition category. The working conditions of the TDT–55A skidder turned out to be totally unacceptable with regard to the present requirements. The Timberjack 850 feller buncher proved to provide the most ergonomic controls. Altogether, almost all the machines had rather good values for this indicator; however, for the Valmet machines and the Timberjack 460D grapple skidder these values were somewhat lower than for the John Deere machines. Russian tracked skidders, especially the TDT–55A, demonstrated substantially lower levels of this integrated indicator. John Deere cut-to-length harvesting machines were the leaders based on the ergonomic indicators related to the work place: cab entrance, cab interior, operator’s seat and controls. For the Valmet and Timberjack 460D machines, these values were somewhat lower. The value of the work place indicators for the TLT–100 skidders follows them closely. For the TDT–55A these indicators were considerably lower, even compared to the TLT–100. The harvesters, forwarders and tracked skidders showed good results with regard to the repetitiveness and complexity of work indicators. The feller bunchers’ values were slightly lower, and the wheeled skidder’s even lower. In both cases this was due to the high level of repetitiveness (compared to the standards); in other words, the job was very monotonous. Visibility was one of the few indicators where Russian machines gained good results. The TLT–100 skidder even got the best score. However, the results were not unambiguous because visibility is affected by many factors, such as: dimensions of the cab and the whole machine, size of the windows, the opera-

168

tor’s eye position with regard to windows, and so on. The Timberjack 460D skidder had the lowest values in visibility due to its very long engine room, limiting visibility in front of the machine. The harvesters achieved better results regarding the noise and vibration characteristics, with forwarders following close behind. The Timberjack 460D skidder and TLT–100 skidder demonstrated poor results (mainly due to noise). The TDT–55A skidder was inferior regarding this indicator. A summary of the evaluation of the machines by ergonomic parameters revealed that the best working conditions in terms of ergonomics and occupational safety were provided by the »harvester + forwarder« system in cut-to-length harvesting. Within this combination, the John Deere machine system showed the best results, while the Volvo and Valmet machine systems had lower ergonomic indicators. The »harvester + forwarder« technology was closely followed by the »feller buncher + grapple skidder« in fully mechanized full-tree harvesting, the difference not being significant. The traditional Russian tree-length harvesting done with cable skidders showed the worst results in terms of ergonomics, work severity and occupational safety. When a partially mechanized harvesting system is used, use of the TDT–55A skidder should be as limited as possible, because, as a whole, they do not comply with present ergonomics requirements (the »relatively uncomfortable« working conditions score).

Acknowledgements – Zahvala This work has been carried out in the project »Wood harvesting and logistics in Russia – Focus on research and business opportunities«, funded by the Finnish Funding Agency for Technology and Innovation (TEKES).

5. References – Literatura Axelsson, S. A., 1998: The mechanization of logging operations in Sweden and its effect on occupational safety and health. International Journal of Forest Engineering 9(2): 25–31. Cabins and work places for operators of tractors, self – propelled road – construction machines, single – axial haulers, dump – trucks and self – propelled agricultural vehicles. General safety requirements (1988) State standard. GOST 12.2: 120–88. Frumerie, G. (ed.), 1999: Ergonomic guidelines for forest machines. SkogForsk, Uppsala: 1–88. Frumkin, A. A., Zinchenko, T. P., Vinokurov, L. V., 1999: Methods and means of ergonomics during design. Transport University, Saint-Petersburg: 1–178. Croat. j. for. eng. 30(2009)2

Ergonomic Characterization of Harvesting Work in Karelia (159–170)

Gerasimov, Y., Karjalainen, T., 2008: Development program for improving wood procurement in north – west Russia based on SWOT analysis. Baltic Forestry, 14(1): 87–92. Hansson, J. E., 1990: Ergonomic design of large forestry machines. International Journal of Industrial Ergonomics 5(3): 255–266.

Y. Gerasimov and A. Sokolov

Methods of noise measurement at work places (1986) State Standard, GOST 12.1.050–86. Noise. General safety requirements (1983) State standard, GOST 12.1.003–83. Peskov, V. I., 2004: The basics of ergonomics and design of cars. Technical University, Nizhnii Novgorod: 1–223.

Harstela, P., 1990: Work postures and strain of workers in Nordic forest work, a selective review. International Journal of Industrial Ergonomics 5(3): 219–226.

Sokolov, A., Syunev, V. S., Gerasimov, Y. Y., 2008: Comparison of skidders and forwarders in working conditions and work safety, Forest and business 1(41): 56–61.

Harvesting and floating machines, forestry and silvicultural tractors. Safety requirements, methods for control of safety requirements and occupational safety evaluation (1989) State standard, GOST 12.2 102–189.

Syunev, V. S., Sokolov, A. P., Konovalov, A. P., Katarov, V. K., Seliverstov, A. A. Gerasimov, Y., Karvinen, S., Valkky, E., 2008: Comparison of wood harvesting methods in logging companies of the Republic of Karelia. Finnish Forest Research Institute, Joensuu: 1–126.

Harvesting machines, forestry and silvicultural tractors. Safety requirements (2002) State standard, GOST R 51863–2002. Machine noise. Methods to determine noise characteristics. General requirements (2002) State standard, GOST 23941–2002.

Vibration safety. General requirements (1990) State standard, GOST 12.1.012–90. VNIITE, 1983: Ergonomics. Principles and recommendations: Methodology guidelines. All – Russian Research Institute of Technical Aesthetics, Moscow: 1–184.

Sa`etak

Ergonomsko ozna~ivanje radova na pridobivanju drva u Kareliji Svrha je rada usporedba ergonomskih svojstava strojeva pri razli~itim metodoma pridobivanja drva te njihov utjecaj na rukovatelje strojeva radi pobolj{anja uvjeta rada. Rukovatelji {umskih strojeva ~esto ozlje|uju vrat, ruke i zatiljne kralje{ke vjerojatno zbog pretjeranoga sjedenja, ergonomski neodgovaraju}ega poloàja i ~estih istih pokreta dijelova tijela u kratkom vremenu. Za ergonomsku procjenu {umskih strojeva i metoda pridobivanja drva autori su obavili sveobuhvatno terensko istraìvanje u Republici Kareliji (sjeverozapadni dio Rusije) na 23 razli~ita radili{ta (slika 1). Istraìvanje je obavljeno na 25 strojeva me|u kojima su: harvesteri, forvarderi, sje~na vozila te skideri s vitlom i hvatalom (tablica 1). Razli~iti parametri koji utje~u na ergonomiju i radne uvjete mjereni su neposredno i u trenuta~nim radnim uvjetima. Studij vremena radnoga ciklusa obavljen je neposrednim snimanjem vremena videokamerom. Rezultati studija vremena za pojedine vrste {umskih strojeva prikazani su u tablici 2. Vi{e od 120 razli~itih parametara koji se nalaze u vaè}im normama Rusije i [vedske mjereno je neposredno na mjestu rada i pri trenuta~nim uvjetima rada. Najva`niji su mjereni parametri: geometrijske zna~ajke kao {to su udobnost kabine i sjedala, poloàj upravlja~kih komandi i poloàj tijela radnika, sile na ru~nim i no`nim upravlja~kim komandama, vibracije na cijelom tijelu voza~a, buka u kabini i vidljivost iz kabine. Prosje~ni udio vremena rada tijekom kojega je radnik u neprikladnom radnom poloàju tako|er je vaàn ~imbenik koji utje~e na ukupnu udobnost upravljanja strojem. Preglednost i radni poloàji tijela radnika uzeti su u obzir kao presudni ~imbenici koji utje~u na radni u~inak radnika. Uzimana je prosje~na vrijednost ukupnoga vremena tijekom kojega je radnikovo tijelo bilo u nepovoljnom radnom poloàju te broj promjena radnoga poloàja. âk i kod visoko mehaniziranoga rada problemi uvijek postoje usprkos naprednomu razvoju kabina. Izmjerene vrijednosti ergonomskih svojstava upore|ene su s vaè}im normama te je iz navednoga odnosa izraèn sloèni pokazatelj radnih uvjeta za rad rukovatelja stroja (izraz 6). Pokazatelj moè poprimiti vrijednosti od 0 do 1. Ve}a vrijednost pokazatelja ukazuje na bolje uvjete rada (tablica 3). Ukupno vrednovanje ergonomskih svojstava pojedinoga stroja izraèno je stupnjem teìne rada, koji se izra~unava prema izrazu 7 te moè poprimiti vrijednosti od 0 do 6. Ve}a vrijednost ukazuje na teè uvjete rada sa strojem (slika 3). Opisanim pokazateljima mogu}e je usporediti radne uvjete na razli~itim radili{tima i pri primjeni razli~itih {umskih strojeva.

Croat. j. for. eng. 30(2009)2

169

Y. Gerasimov and A. Sokolov

Ergonomic Characterization of Harvesting Work in Karelia (159â&#x20AC;&#x201C;170)

Najbolji uvjeti rada s obzirom na sustav pridobivanja drva jest kombinacija harvestera i forvardera u sortimentnoj metodi te sje~no vozilo i skider s hvatalom u stablovnoj metodi. Tradicionalno pridobivanje drva deblovnom metodom uz pomo} skidera s vitlom pokazalo je najgore rezultate {to se ti~e ergonomskih svojstava. Kada se koristi djelomi~no mehanizirani sustav pridobivanja drva, kori{tenje skidera s vitlom trebalo bi se {to vi{e ograni~iti jer ne udovoljava trenuta~nim ergonomskim zahtjevima. Klju~ne rije~i: pridobivanje drva, ergonomija, harvester, forvarder, skider, sje~no vozilo

Authors' address â&#x20AC;&#x201C; Adresa autorĂ˘: Yuri Gerasimov, PhD. e-mail: yuri.gerasimov@metla.fi Finnish Forest Research Institute Joensuu Research Unit Yliopistokatu 6 80101 Joensuu FINLAND

Received (Primljeno): April 13, 2009 Accepted (Prihva}eno): November 11, 2009005

170

Assoc. Prof. Anton Sokolov, PhD. e-mail: a_sokolov@psu.karelia.ru Petrozavodsk State University Forest Engineering Faculty A. Nevsky av. 58 185030 Petrozavodsk Republic of Karelia RUSSIA Croat. j. for. eng. 30(2009)2

Original scientific paper – Izvorni znanstveni rad

Severity Analysis of Accidents in Forest Operations Igor Poto~nik, Tibor Pentek, Anton Poje Abstract – Nacrtak This paper deals with the severity of accidents in forest operations in Slovenian state forests in the period 1990–2005. A total of 846 accidents were analyzed. Most accidents happened in felling (68% and 20 lost days), and somewhat less during skidding (24% and 19 lost days). Most injuries were caused by direct contact of the tree parts (60% and 20 lost days) with stones, rocks and surface (24% and 21 days lost days). Power saw only caused 6% of injuries (17 lost days). Contusion was the most frequent type of injury in accidents due to stroke (56%), open sore (19%) and sprain with muscle strain (11%). Severity of injuries of extremities (20.5 lost days) seems to be different from face injuries (11.6 lost days). It can be concluded that the results are useful for implementation in everyday forest operations based on safety and accident prevention. Key words: forest operations, accident, severity, injury

1. Introduction – Uvod Forestry continues to be one of the most hazardous industrial sectors in most countries. Around the world, there are often discouraging trends of increasing accident rates and a high incidence of occupational diseases and early retirement among forestry workers. However, clear evidence shows that good safety and health performance in forestry is feasible (ILO 1998). The employer is responsible for ensuring that any method chosen to control risk is working (Code of Practice 2002). Monitoring and review is a very important aspect of OHS (Occupational Health and Safety) management and should be included in regular performance reporting to management. In this way, OHS risk management is an ongoing process. In addition, to make sure that a workplace stays safe and keeps abreast of change, an employer must redo the risk assessment and review any control measures whenever: Þ There is evidence that the risk assessment is no longer applicable; Þ An injury or illness occurs because of a hazard that the risk assessment addressed, or failed to address; Croat. j. for. eng. 30(2009)2

Þ A change is planned to the place of work, work practices, or work procedures that the risk assessment addressed. Three steps should be performed: identification, assessment and elimination or control for every health and safety issue that requires attention. For forest harvesting operations, this method provides a systematic way of working out effective action to control risks. According to recommendations (ILO 1996), occupational accidents, diseases and dangerous occurrences should be reported and analyzed accordingly at 3 levels: national level, level of enterprise and level of self-employed person. Also, minimum required information is defined for each level an accident/disease. Forestry work managers should comply with the requirements of the key tasks (Management of Health and Safety at Work 1999): Þ Use the information from the landowner to prepare an outline risk assessment for on-site work and for timber extraction; Þ Select competent contractors who have made adequate health and safety provisions; Þ Specify health and safety measures for on-site contractors and others who may be affected by the work activity;

171

I. Poto~nik et al.

Þ Liaise with the landowner; Þ Monitor on–site health and safety. There are many factors that influence severity of accidents in mining, which is a physically intensive industry just like forestry; for example injured body parts, injured person’s age, cause of accident, source of accident, injured person’s activity, work duration prior to accident and workplace (Hull et al. 1996). Severity of accidents has a direct influence on employer’s costs, due to worker’s input loss and due to interruption of working process (Klen 1988). Application of safety equipment, such as safety trousers, decreases the number of accidents with chain saw (Sullman et al. 1999), but on the other hand it increases the number of injuries of other unprotected body parts. By using personal safety equipment, the comprehension of danger alters, and this can lead to increased risk-taking and consequently higher work tempo (Salminen et al. 1999, Klen 1997). Frequency and severity of accidents are also decreased by payment method. Thus, safety increased because of transition from piece rate to time based wages (Sundström and Frisk 1984). Mechanization of cutting and skidding has influenced the decrease of accident frequency, redistribution of accidents according to the injured body parts, and type of injuries, caused the occurrence of new risks (RSI), and changed the place of accident occurrence from productive working hours to the unproductive ones, i.e. maintenance of means of work (Axellson 1998, Väyrynen 1982, Salaminen et al. 1999, Laflamme and Cloutier 1998, Lefort 2003). The proposed measures for improvement of work safety are: additional education, lower work intensity and tempo, and shorter working hours (Nieuwenhuis and Lyons 2002). Taking into account the above facts and experiences and available reports on accidents in Slovenian state forests, we decided to conduct the present study in order to point out the most crucial factors of severity of accidents in forest operations.

2. Research objectives – Ciljevi istra`ivanja The study is limited to the assessment of activity when the accident happened, source and consequence of accidents related to severity of accidents. Therefore two general objectives were set: Þ To assess the influence of work activity, source of accident, type of injury and injured body part on severity of accidents; Þ To assess what combination of work activity, source and type of injury (injured part of

172

Severity Analysis of Accidents in Forest Operations (171–184)

body) contribute the most to lost days due to accidents as a whole. Results of the set objectives bring practical meaning to work organization in forestry, mostly in modeling working technique, time schedule preparation, and right choice and development of forest worker personal safety equipment.

3. Working method and material Metode rada i kori{teni materijal The study is based on information of forest workers accidents, which are collected on an annual basis by the Department of Forestry and Renewable Forest Resources (University of Ljubljana, Biotechnical Faculty) in close cooperation with forest enterprises-concessionaires in Slovenian state forests. Each accident in forest operations is recorded using special questionnaires related to information on age and qualification of the person involved in the accident, time and location, activities, working phase, form and distribution of injuries, severity of accident, source and cause, and specific environmental data. The recorded accident is not limited by severity, meaning that the amount of lost days might also be 0. On the other hand, accidents with absence from work over the fiscal year result in lost days in the next year. Questionnaires are completed by professionals responsible for safety at work, specially trained for health and safety at work in forestry, but not necessarily foresters by education. In the period discussed in this paper, the main harvesting technology was motor–manual work with chain saw in bucking conifers to length, whereas deciduous tree bucking was made according to quality. Apart from tree felling, trimming and bucking, some operations were also made in accordance with forest regulations so as to provide forest protection (piling of branches) and maintenance of roads and watercourses (removing of cutting remains). Skidding was mostly done by tractors (adjusted farm tractors and skidders), and tower yarders or manually at short distances. All forest workers had standard protection equipment (helmet with ear and eye protection, cut-proof trousers, shoes and gloves). Analysis was made of accidents that occurred in the stand, skidding trail, forest road and auxiliary yard during the cutting, skidding and tending of juvenile stand. The analyzed accidents occurred to 13 concessionaires on a total area of 303,778 ha (data from 2005) of forest and 736,000 m3 of annual cut, i.e. 84% of all wood (estimate for 1999–2003 period) that was part of concession works. Therefore, we managed to collect detailed information of 846 accidents (Table 1), where accidents severity (measured in lost Croat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171–184)

I. Poto~nik et al.

Table 1 Characteristics of accident severity distribution Tablica 1. Obilje`ja raspodjele teìne nesre}a Measure Mjerna veli~ina Number of accidents – Broj nesre}a Mean – Aritmeti~ka sredina Max. – Najve}e opaànje Min. – Najmanje opaànje Median – Medijana Mode – Mod Sum – Zbroj 1.Quartil – Prvi kvartil 2.Quartil – Drugi kvartil 3.Quartil – Tre}i kvartil

Unit Mjerna jedinica Lost days/accident – Izgubljeni dani/nesre}a Lost days/accident – Izgubljeni dani/nesre}a Lost days/accident – Izgubljeni dani/nesre}a Lost days – Izgubljeni dani Lost days – Izgubljeni dani Lost days – Izgubljeni dani Lost days – Izgubljeni dani Lost days – Izgubljeni dani Lost days – Izgubljeni dani

All accidents Sve nesre}e 846 29.27 722 0 14.00 8 24,762 8.00 14.00 27.00

95 % of all accidents* 95 % svih nesre}a* 804 19.10 111 0 14.00 8 15,375 8.00 14.00 23.00

* without 5 % of accidents with highest severity – bez 5 % najte`ih nesre}a

days/accident) showed typical J distribution and adequate distribution of mode, median and mean. The overall sum of lost days was 24,726. By additionally excluding 5% of accidents (n=42) with the highest severity (average severity of 223.5 of lost days/ accident), the overall sum was reduced to 9,387 days. The purpose of excluding the most severe accidents was to estimate the variability of the calculated values of accident severity. Referring to the study objectives, 4 factors were introduced that were supposedly related to accident severity: work activity, source of injury, type of injury and injured body part. Work activities were divided into 3 categories: tending (ground preparation and area clearance for forestation, nursing of young trees and grounds), cutting and skidding (manually, with tractors and tower yarders). Sources of injury (defined as objects that caused injury by direct contact) are hand tools (wire rope, chain, cutting wedge, axe), manual machines (power saw), mobile working machinery (tracked tractor, skidder, tower yarder, adjusted agricultural tractor), tree parts (trunk, branch, stump), insects and surface. Types of injuries were categorized into 6 categories: open sores (cuts, laceration, abrasion), contusions and suffusions, sprains (sprains, dislocations, muscle ruptures), fractures, insect bites, and eye injury/loss. Injured body parts were categorized according to adapted Injury Severity Score-ISS (Linn 1995) into 5 groups: head and neck, face, chest and back, stomach and hip, and limbs. Due to many recorded values of accidents with numerous injuries and injured body parts, only those assessed as the most severe were included into the study. Apart from descriptive methods, Welch test for comparison of means was also applied as well as Croat. j. for. eng. 30(2009)2

Tamhane test T2 for post hoc analysis, regression for curve estimation and binary multiple logistic regression for estimating factors within the complex influence on accident severity.

4. Results – Rezultati Overall analysis of accidents (n=846) shows (Fig. 1) that accidents with the lowest severity are »missed« in theoretical J distribution; mostly accidents with up to 2 lost days/accident. According to theoretical J distribution (Figure 1) 734 accidents should be in this class, which means that 642 accidents are »missed«. The estimate of supposedly »missed« accident was achieved by applying the data on number of accidents in classes with accident severity from 5 to 100 days and class average means, to which the inversion function was adapted (frequency = 1887.7686/ class average means –21.5954, R2 = 0.97, p>0.000). According to the estimate, it is highly probable that the actual number of accidents is between 92 and 734 days for the class up to 5 lost days. The number of accidents with accident severity above 100 days equals 47, but the number contributes with considerable 40% to the total number of lost days. Accidents with such high severity score might distort the actual severity by particular factors so all further analyses are represented for all accidents, as well as for 95% of accidents – without accidents with severity higher than 111 lost days. Within the discussed working phases, most accidents happened in cutting (68%), somewhat less during skidding (24%) and the least in tending operations (8%). Severity of accidents within working phases was significantly different when taking into account all accidents (Welch, p<0.000) or 95% of

173

I. Poto~nik et al.

Severity Analysis of Accidents in Forest Operations (171â&#x20AC;&#x201C;184)

Fig. 1 Frequency and cumulative frequency distribution of accident by severity Slika 1. Raspodjela frekvencije i kumulativna frekvencija nesre}a prema njihovoj te`ini them (Welch, p95%=0.002). In both cases accident severity was significantly different only between cutting and tending operations (Tamhaneâ&#x20AC;&#x2122;s T2, p=0.000; p95%=0.002), and skidding and tending operations (Tamhaneâ&#x20AC;&#x2122;s T2, p=0.001; p95%=0.024). Difference in accident severity within working phases between all accidents and 95% accidents shows (Fig. 2) that the

Fig. 2 Number and severity of accidents by forest operations Slika 2. Broj i te`ina nesre}a prema {umskim operacijama 174

most serious accidents mostly happened in cutting (D=10.1 lost days/accident) and skidding (D=13.5 lost days/ accident). When studying sources of injuries, 20 accidents with the unknown source were eliminated from the analyses. The range of the studied accidents was therefore decreased to 826 and 785 accidents, respectively, without taking account of the most severe accidents. Most accidents occurred by direct contact of the tree parts (60%) with stones, rocks and surface (24%). Contrary to common expectation, chain saw only caused 6% of injuries (Fig. 3). Severity of accidents within a source of injury is statistically significant for both ranges of accidents (Welch, p<0,000, p95%<0,000). Despite the highest severity by contact with working machines, this source is not different from other sources (the only exception are injuries caused by insects in the case of 95% of accidents (Table 2). When comparing all accidents and 95% accidents, it can be concluded that accident severity due to tree parts (33.3 and 19.8 lost days/accident); stones, rocks and surface (26.7 and 21.0 lost days/accident) are significantly higher than the accident severity due to hand tools (13.6 and 11.4 lost days/accident) or insects (5 lost days/accident). With the total accident range, both sources with the highest accident severity are also higher than the chain saw accidents (15.3 lost days/accident), whereas within 95% of all accidents the injuries caused by insects are lower than all compared sources. Very sever accidents mostly occur in acciCroat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171â&#x20AC;&#x201C;184)

I. Poto~nik et al.

Fig. 3 Frequency and severity of accidents by sources of injuries Slika 3. U~estalost i te`ina nesre}a prema izvoru ozlje|ivanja dents with mobile working machinery, tree parts, and stones, rocks and surface. Out of 846 accidents, this type of injury analysis considers only 819 accidents and 777 in the case of 95% of accidents without the most sever accidents,

respectively. From all eliminated accidents, 26 were listed under the category Other, whereas 1 accident lacked the information on the type of injury. Contusion was the most frequent type of injury (Fig. 4) in accidents due to hit (56%), open sore (19%), and

Table 2 Significant differences (p<0.05) between sources of injuries (Tamhaneâ&#x20AC;&#x2122;s t2 post hoc test) Tablica 2. Statisti~ki zna~ajne razlike (p<0,05) izme|u razli~itih izvora ozlje|ivanja (Tamhanesov t2 post hoc test)

Croat. j. for. eng. 30(2009)2

175

I. Poto~nik et al.

Severity Analysis of Accidents in Forest Operations (171â&#x20AC;&#x201C;184)

Fig. 4 Frequency and severity of accidents by type of injuries Slika 4. U~estalost i te`ina nesre}e prema vrsti ozljede sprain with muscle strain (11%), where all differences were significant (Welch, p<0.000, p95%<0.000). In both cases severity was higher (except with eye injuries) in case of fractures (123.9 and 50.3 lost days/ accident) and lower in case of insect bites (9.0 lost days/accident) comparing to all other injuries (Table

3). Rare and very sever accidents occur mostly due to bone fractures and contusions due to hit. Out of all accidents, 13 accidents were eliminated from the analysis of injured body parts due to listing of body part injury under the category Other and 5 accidents due to missing data on the injured

Table 3 Significant differences (p<0.05) between type of injuries (tamhaneâ&#x20AC;&#x2122;s t2 post hoc test) Tablica 3. Statisti~ki zna~ajne razlike (p<0,05) izme|u razli~itih vrsta ozljeda (Tamhanesov t2 post hoc test)

176

Croat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171–184)

I. Poto~nik et al.

Fig. 5 Frequency and severity of accidents by injured body part Slika 5. U~estalost i te`ina nesre}e prema ozlije|enom dijelu tijela body part. Thus, the analysis of injured body parts includes 828 (95%, 786) accidents. 66% of accidents caused injuries of extremities, i.e. arms and legs (Fig. 5). Differences among injured body part are different in both extents of accidents (Welch, p<0.000, p95%<0.000). Severity of injuries of extremities (30.7 and 20.5 lost days/accident) is significantly higher than face injuries (14.2 and 11.6 lost days/accident). Regarding the data without the most severe acci-

dents, injuries of thorax and back (20.7 lost days/accident) were significantly higher than those of face injuries (Tamhane’s T2, p95%=0.001). The most severe accidents mostly prevailed in injuries of extremities, although their biggest influence is also noted in stomach and hip injuries. The above analyses show that all studied factors affect the severity of accidents (Welch test, p<0.05), and that this influence is statistically significant only

Table 4 Impact factors on accidents severity by use of all accidents and binary logistics regression Tablica 4. Utjecajni ~imbenici na te`inu ozljede pri analizi svih nesre}a binarnom logisti~kom regresijom Factors/category ^imbenik/kategorija Injured body part – Ozlije|eni dio tijela Head and neck – Glava i vrat Chest and back – Prsni ko{ i le|a Stomach and hip – Trbuh i kuk Limbs – Udovi Source of injury – Uzrok nesre}a Chain saw – Motorna pila Mobile working machinery – Mobilni radni strojevi Tree parts – Dijelovi stabala Insects – Kukci Rocks stones and ground Stijene, kamenje i zemlja Constant – Konstanta

Croat. j. for. eng. 30(2009)2

B Procijenjeni parametar

S.E. Standardna pogre{ka

Wald Statistika

df Stupnjevi slobode 4 1 1 1 1 5 1 1 1 1

0,000 0.000 0.000 0.000 0.000 0.001 0.037 0.003 0.000 0.999

2.335 11.674 3.194 0.000

Sig. p

Exp(B) Exp(B)

1.525 1.219 2.285 1.425

0.390 0.310 0.633 0.245

0.848 2.457 1.161 –20.113

0.406 0.833 0.304 15987.291

37,919 15.307 15.475 13.020 33.758 19.681 4.361 8.704 14.636 0.000

1.239

0.323

14.695

0.000

3.451

–2.367

0.366

41.881

0.000

0.094

4.594 3.384 9.822 4.156

177

I. Poto~nik et al.

Severity Analysis of Accidents in Forest Operations (171â&#x20AC;&#x201C;184)

Fig. 6 Prevalent combination of injuries by injured body parts, source of injury and work activities Slika 6. Prevladavaju}e kombinacije ozljeda prema ozlije|enomu dijelu tijela, izvoru ozljeda i radnoj aktivnosti for some categories of specific factor (Tamhaneâ&#x20AC;&#x2122;s T2 test, p<0.05). The binary logistic regression was used for estimating simultaneous influence of factors on accident severity. We established that in the total accident range two factors significantly affected the accident severity, accident source and injured body part (Table 4). A more detailed analysis of accidents

showed that the likelihood for accidents with severity higher than 14 days in chain saw accidents is 2.3 time higher, in mobile working machinery 11.7 time higher, in tree parts accidents 3.2 time higher, and in accidents caused by rocks, stone and ground 3.5 time higher than in accidents caused by hand tools. At the same time, the likelihood of head and neck injuries is

Fig. 7 Prevalent combination of injuries by sort and source of injury and work activities Slika 7. Prevladavaju}e kombinacije ozljeda prema vrsti i izvoru te radnoj aktivnosti 178

Croat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171â&#x20AC;&#x201C;184)

4.5 time higher, chest and back 3.9 times higher, stomach and hip 9.8 times higher and limbs 4.2 higher than face injuries. According to the research aims, we analyzed the combination of the studied factors that represent the highest ratio of lost days. For the purpose of transparency, the combination of 3 factors is shown in two schemes (Fig. 6, Fig. 7), but only the combination that had the highest accident severity and whose contribution to the sum is more than 75% of all lost days. Out of 90 possible combinations (3x6x5), seven combinations between injured body parts, injury source, and working phase contribute to 76% of lost days (Fig. 6). Regardless of the working phase, cutting and skidding, the sources of accidents comprise mostly tree parts and rocks, stone and ground, but as a consequence of these accidents, mostly injuries of arms and legs. Considerable share of cutting accidents with tree parts as accident sources is also noted on other body parts, chest and back, head and neck, and face. With regard to the combination of type and source of accident, and working phase, the combination of 8 factors contributes to 75% of all lost days (Fig. 7). Contusions and suffusions and bone fractures are a consequence of injury sources, such as tree parts and rocks, stone and ground during cutting and tree parts during skidding operations. Regarding cutting operations, a considerable part of lost days is contributed by open sores due to tree parts as accident source, and also sprains and muscle strains as a consequence of rocks, stone and ground.

5. Discussion â&#x20AC;&#x201C; Rasprava There are only few comparable contemporary research works with the subject similar to our research, because countries with the most developed forestry conduct cutting and skidding mainly with machinery. In countries, where these operations are still mainly performed by chain saw, the social environment is probably less favorable for such research projects. With regard to transitional position in terms of forestry and technology, such research is necessary for Slovenia and similar countries (e.g. Croatia). Research of accidents is primarily intended for planning measures for future accident prevention on the basis of past accident analyses. The validity of these analyses and their subsequent measures is thus directly connected to quality and quantity of data used in such analyses. Quality and quantity of the recorded data depends mostly on the required data (country regulations, comparisons of companies), Croat. j. for. eng. 30(2009)2

I. Poto~nik et al.

evaluator and injured person (Johnson, Employment Outlook 1989). The alleged underestimation of accident number (under reporting), which is also present in the majority of other accident research works due to various reasons, is particularly evident in our research in minor accidents with the severity up to 2 days. The main reason can be ascribed to the fact that the control of selected codes on accidents could be performed through questionnaires required by the Slovenian Labour Inspectorate, which gathers information on accidents for severity of more than 2 days. Other, but not insignificant reasons regard workers, who due to economic or other interests do not report such accidents, and also companies, which apart from regularity do not see advantages (but mainly disadvantages) of such accident analyses. The data confirmed the alleged underestimation of accident number with absence from work of less than 5 days. The share of similar forestry accidents in New Zealand was 42% (Gaskin and Parker 1993), whereas in our research the share of these accidents is 10.8%. To provide the best possible records of accidents at work, it would be necessary to raise awareness of companies and put them under obligation to carry out measures for improving safety at work on the basis of all accident analyses (not only those with fatal and most severe results). Inspection services should require such analyses and should not be satisfied only with informative data on the number of accidents. Workers should be encouraged to report dangerous events to their superiors, because such information is of key importance to future accident prevention. Accident research showed that out of all accidents, 68% accidents occurred during cutting, 24% during skidding, and 8% during tending operations, and these results are slightly different from the results obtained in Sweden (60%, 20%, 20%, EngsĂĄs 1995). The difference can certainly be ascribed to ownership differences (state vs. private), working technology, and working conditions. Severity established during cutting (29.9 lost days/accident) and skidding (32.5 lost days/accident) greatly surpasses research results in New Zealand (13.6 and 11.1 lost days/accident; Bentley et. al 2005, Bentley et al. 2002). High accident severity during cutting rises from accident due to tree parts, whereas apart from tree parts, skidding accidents mainly occur due to mobile working machinery. Confirmed differences in severity between cutting and skidding compared to tending operations are a consequence of lower severity of accidents due to tree parts, hand tools, and rocks, stone and ground. There are relatively few chain saw accidents (6%). Moreover their share is comparable to the data (8%)

179

I. Poto~nik et al.

stated by Sullman et al. (1999). The share of chain saw accidents in Slovenian state forests is decreasing (Jereb 2009) similarly to the findings of other research projects (Lefort et al. 2003), but in case of Slovenia the reason does not lie in work mechanization (introduction of cut-to-length technology only since 2005), but allegedly due to constant application of safety means and education (awareness) of workers. The established accident severity in chain saw injuries (15.3 lost days/accident) is somewhat higher than that stated by Bentley et al. (2002), but it was only recorded during skidding (11.2 lost days/ accident). Accident share for all means of work (15%) is comparable with shares established in other research works (Lefort et al. 2003) and 14–20% (KWF). Despite satisfactory decreasing of share of accidents with chain saw, the share of accidents due to tree parts (60%), and due to rocks, stone and ground (24%) is increasing. Considering these accidents, the accident severity is the highest (33.3 and 26.7 lost days/accident) and also higher compared to accident severity caused by means of work, and especially higher due to insect bites. The increased share of accidents due to tree parts and rocks, stone and grounds (Jereb 2009), and high severity values of these accidents are logical, if we consider that accidents due to tree parts are especially difficult to prevent with personal safety equipment, and that the cause of such accidents can lie in improper work tempo. Arduous work contributes to poor concentration and incaution, and as such increases accident risk. The following types of injuries prevail: contusions (56%), injuries with open sore (19%), sprains and muscle strains (11%), and fracture bones (6%). The data for German state forests (KWF) show the same values for open sores, lower for contusions (41%), and higher for sprains (22%) and bone fractures (9%). The reason for these differences could be due to greater mechanization of cutting and skidding operations in German forests. During mechanized cutting, accidents mainly happen during maintenance works, whereas with regard to the types of injury the sprains (35%) and contusions (30%) prevail (Väyrynen 1982). The share of bone fractures is also increased (Lefort et al. 2003). Accident severity by types of injuries was higher in all cases compared to the research of Gaskin and Parker (1993), but the main reason is unknown. Neither mechanization nor working conditions have such an influence on differences in the number of lost days, e.g. due to bone fracture. Accident severity is significantly higher with bone fractures (123.9 lost days/accident), and lower with infections due to insect bites (9 lost days/accident) compared to other types of accidents. Our re-

180

Severity Analysis of Accidents in Forest Operations (171–184)

sults show that according to further mechanization of forestry, the accident frequency in Slovenia will decrease, whereas the accident severity will increase. Similar results were also established in other research works (Lefort et al. 2003). The majority of accidents by injured body parts occurred to arms and legs (66%) and is hence the same as in other research works (64%, KWF; 51%, Wang 2003; 50%, Lefort et al. 2003). Surprisingly, there were numerous face injuries (15%), especially during tending operations (33%), half of which pertain to eye injuries. This means, that all workers should use eye protection during all tending operations. Mechanization also influences the altered share of injured body parts (Väyrynen 1982). It was established that during cut-to-length operations there were numerous back injuries (20%), whereas our research shows a significantly lower share (6%). The highest severity occurred in stomach and back injuries (93.3 lost days/accident), and in head and neck accidents (38.5 lost days/accident), which is logical, if we consider possible consequences of such accidents. This research shows that accident severity depends on all studied factors and mainly depends on injured body part and injury source when all factors are included. Practically it means that we have to protect or prevent accidents of specific body parts, especially stomach and back injuries, by applying mobile working means regardless of the working phase, if we are to decrease accident severity. Furthermore, the safety measures for the highest decrease of accident severity need to be directed to decrease of injuries due to tree parts, the consequences of which are contusions and bone fractures mostly in extremity injuries in both working phases, and other body parts during cutting, and decrease injuries of strokes, sprains and bone fractures due to ground, stones and rock.

6. References – Literatura Anonym. 1999: Management of health and safety at work. Management of Health and Safety at Work Regulations 1999: Approved Code of Practice L21 HSE Books 2000. Anonym. 2002: Code of Practice – Safety in Forest Harvesting Operations. New South Wales Government Gazette No. 178, pp. 8952–9017. Axelsson, S. A., 1998: The mechanization of logging operations in Sweden and its effect on occupational safety and health. Journal of Forest Engineering 9(2): 25–31. Bentley, T. A., Parker, R. J., Ashby, L., 2005: Understanding felling safety in the New Zealand forest industry. Applied Ergonomics 36(2): 165–175. Croat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171–184)

Bentley, T. A., Parker, R. J., Ashby, L., Moore, D. J., Tappin, D. C., 2002: The role of the New Zealand forest industry injury surveillance system in a strategic Ergonomics, Safety and Health Research Programme. Applied Ergonomics 33(5): 395–403. Employment Outlook 1989. Occupational Accidents in OECD Countries. Organisation for Economic Co-operation and Development, Chapter 4: 133-159 <http://www. oecd. org/dataoecd/63/54/3888265.pdf >(Accessed 24 November 2009) Engsás, J., 1995: Accidents in the private woodlot sector. Small Scale Forestry 23(1): 15–22. Gaskin, J. E., Parker, R. J., 1993: Accidents in forestry and logging operations in New Zealand. Unasylva, 44, 172: 19–24 <http://www.fao.org/docrep/u8520e/u8520e05. htm> (Accessed 26 November 2009) Hull, B. P., Leigh, J., Driscoll, T. R., Mandryk, J., 1996: Factors associated with occupational injury severity in the New South Wales underground coal mining industry. Safety Science 21(3): 191–204. ILO (1996). Recording and notification of occupational accidents and disease. An ILO code of practice Geneva, International Labour Office, 94 pp. ILO (1998). Safety and health in forestry work: An ILO code of practice. Geneva, International Labour Office, 132 pp. Jereb, P., 2009: Nezgode pri delu gozdarskih podjetij pred in po letu 1993 (Work accidents in Slovenian forestry companies before and after year 1993. Bsc. Thesis, University of Ljubljana – Biotechenical Faculty, Department of Forestry and Forest Resources, 79 pp. Wang, J., Bell, J. L., Grushecky, S. T., 2003: Logging injuries for a 10-year period in Jilin Province of the People’s Republic of China. Journal of Safety Research 34(3): 273–279.

I. Poto~nik et al.

Johnson, C. W., 2003: Failure in Safety–Critical Systems: A Handbook of Accident and Incident Reporting. University of Glasgow Press, Glasgow, Scotland, 986 pp. Klen, T., 1989: Factors affecting accident costs to employers, employees and public administration in forestry. Journal of Occupational Accidents 11(2): 131–147. Klen, T., 1997: Personal protectors and working behaviour of loggers. Safety Science 25(1–3): 89–103. KWF. Unfallstatistik für den Staatswald 1999–2007, <http:// www.kwf-online.org/unfallstatistik.html>(Accessed 27 November 2009) Laflamme, L., Cloutier, E., 1988: Mechanization and risk of occupational accidents in the logging industry. Journal of Occupational Accidents 10(3): 191–198. Lefort, A. J. Jr., De Hoop, C. F., Pine, J. C., Marx, B. D., 2003: Characteristics of Injuries in the Logging Industry of Louisiana, USA: 1986 to 1998. Int. Journal of Forest Engineering 14(2): 75–89. Nieuwenhuis, M., Lyons, M., 2002: Health and Safety Issues and Perceptions of Forest Harvesting Contractors in Ireland. Int. Journal of Forest Engineering 13(2): 69–76. Salminen, S., Klen, T., Ojanen, K., 1999: Risk taking and accident frequency among Finnish forestry workers. Safety Science 33(3): 143–153. Sullman, M. J. M., Kirk, P. M., Parker, R. J., Gaskin J. E., 1999: New Zealand Logging Industry Accident Reporting Scheme: Focus for a Human Factors Research Programme. Journal of Safety Research 30(2): 123–131. Sundström-Frisk, C., 1984: Behavioural control through piece-rate wages. Journal of Occupational Accidents 6(1–3): 49–59. Väyrynen, S., 1982: Occupational accidents in the maintenance of heavy forest machinery. Journal of Occupational Accidents 4(2–4): 175–175.

Sa`etak

Analiza teìne nesre}a pri {umskim radovima [umarstvo je i dalje, u mnogim zemljama, jedna od najopasnijih gospodarskih grana. Stoga se nastoji smanjiti broj i teìnu nesre}a koje se doga|aju pri {umskim radovima, ~este profesionalne bolesti i prijevremeno umirovljenje {umskih radnika. Pri tome je va`no provo|enje sloènih mjera i postupaka u sigurnosti i za{titi na radu (Occupational Health and Safety – OHS management). Pra}enje i biljeènje vrlo su va`ne sastavnice cjelokupnoga procesa upravljanja sigurnosti i za{tite radnika u {umarstvu. Sukladno preporukama ILO-a (1996) nesre}e, bolesti i opasne doga|aje treba evidentirati i analizirati na razini dràve, poduze}a i pojedinoga zaposlenika. U rudarstvu (Hull i dr. 1996), koje je kao i {umarstvo za radnike fizi~ki vrlo zahtjevna djelatnost, postoji mnogo ~imbenika koji utje~u na teìnu nesre}a pri radu. Neki od najva`nijih su: ozlije|eni dio tijela, dob ozlije|ene osobe, na~in nastanka nesre}e, izvor nesre}e, tjelesna sprema radnika, trajanje rada prije same nesre}e, radno mjesto. Teìna nesre}e ima izravan utjecaj na tro{kove poslodavca zbog gubitka prinosa ozlije|enoga radnika te zbog zastoja radnoga procesa (Klen 1988). Mehaniziranje radova na sje~i, izradbi i privla~enju drva utjecalo je na

Croat. j. for. eng. 30(2009)2

181

I. Poto~nik et al.

Severity Analysis of Accidents in Forest Operations (171–184)

smanjenje u~estalosti nesre}a, preraspodjelu nesre}a prema ozlije|enomu dijelu tijela i vrsti ozljede, pojavu novih opasnosti i pomicanje teì{ta vjerojatnosti pojavljivanja nesre}a s proizvodnih na neproizvodne radne sate – odràvanje i popravak sredstava za rad (Axellson 1998, Väyrynen 1982, Salaminen i dr. 1999, Laflamme i Cloutier 1998, Lefort 2003). Nieuwenhuis i Lyons (2002) preporu~uju sljede}e mjere za unaprje|ivanje sigurnosti na radu: dodatno obrazovanje, manji radni intenzitet i ritam te kra}i radni dan. Uzev{i u obzir sve do sada navedeno, a koriste}i dostupna izvje{}a o nesre}ama pri {umskim radovima u dràvnim {umama u Sloveniji, napisan je ovaj rad kojim se èljelo osvijetliti najva`nije utjecajne ~imbenike na teìnu nesre}a pri {umskim radovima. Istraìvano je kako su okolnosti, izvor i posljedice same nesre}e povezane s njezinom teìnom. Definirana su dva temeljna cilja istraìvanja: Þpronalaènje utjecaja radnoga postupka, izvora nesre}e, vrste ozljede i ozlije|enoga dijela tijela na teìnu nesre}e, Þdefiniranje koja kombinacija radnoga postupka, izvora i vrste ozljede (ozlije|enoga dijela tijela) uzrokuje najve}i gubitak radnih dana zbog nesre}e koja se dogodila. Rezultati istraìvanja trebali bi pomo}i pri organizaciji rada u {umarstvu (najvi{e pri modeliranju radnih tehnika), pripremi i planiranju posla te dobrom odabiru postoje}ih i daljnjem razvoju osobnih za{titnih sredstava i opreme {umskih radnika. Studija se temelji na postoje}oj bazi podataka o nesre}ama u {umarstvu koju je, na godi{njoj razini, skupio Odjel za {umarstvo i obnovljive {umske izvore Biotehni~koga fakulteta Sveu~ili{ta u Ljubljani u suradnji sa zakupcima slovenskih dràvnih {uma. Svaka je nesre}a pri {umskom radu evidentirana primjenom posebnoga upitnika koji je sadràvao ove informacije: dob i stru~na sprema osobe uklju~ene u nesre}u, mjesto i vrijeme nesre}e, vrsta posla, radna operacija, vrsta i intenzitet ozljeda, teìna nesre}e, izvor i uzrok nesre}e te ostali specifi~ni podaci. Sve su nesre}e, bez obzira na njihovu teìnu, evidentirale stru~ne osobe zaduène za provedbu za{tite na radu. Tijekom istraìvanja sje~a i izrada obavljane su ru~no–strojno uz primjenu motorne pile. Uz sje~u i izradu obavljani su i ostali propisani poslovi (hrpanje grana te ~i{}enje prometnica i vodotoka od materijala koji je ostao nakon obavljenih navedenih radova). Pri privla~enju drva prete`no su kori{teni prilago|eni poljoprivredni i zglobni traktori i stupne ì~are, dok se na kra}im udaljenostima privla~ilo manualno. Svi su se {umski radnici koristili propisanom za{titnom opremom ({ljem sa za{titom o~iju i u{iju, za{titno radno odijelo, za{titne rukavice i cipele). Analizirane su sve nesre}e koje su se dogodile u sastojini, na traktorskim putovima i vlakama, na {umskim cestama i pomo}nim stovari{tima za vrijeme pridobivanja drva (sje~a, izrada, privla~enje) i njege. Istraìvanje je provedeno na ukupnoj povr{ini od 303 778 ha i godi{njim etatom od 766 000 m3. Skupljeni su podaci o 846 nesre}a (tablica 1), a teìna nesre}a (iskazana u izgubljenim radnim danima po nesre}i) pokazuje tipi~nu J distribuciju i adekvatnu distribuciju moda, medijane i srednje vrijednosti. Ukupan je broj izgubljenih radnih dana za sve nesre}e 24 726. Dodatnim eliminiranjem 5 % najteìh nesre}a (42 nesre}e) s prosje~nim gubitkom od 223,5 radnih dana/nesre}i ukupan je broj izgubljenih radnih dana smanjen na 9387. Sukladno ciljevima istraìvanja definirana su ~etiri ~imbenika koja pretpostavljano imaju utjecaj na teìnu nesre}e: vrsta posla (radna operacija), izvor ozljede, vrsta ozljede i ozlije|eni dio tijela. Radne operacije podijeljene su u 3 skupine: njega (priprema i ~i{}enje stani{ta za po{umljavanje, po{umljavanje i njega mladih biljaka), sje~a i privla~enje (ru~no, traktorom i stupnom ì~arom). Izvor ozljede, definiran kao objekt koji je izazvao nesre}u, moè biti: ru~ni alat (ì~no uè ili uè vitla, ru~na pila, klin, sjekira), ru~ni strojevi (motorna pila), mobilni strojevi (gusjeni~ar ili polugusjeni~ar, zglobni traktor, stupna ì~ara, prilago|eni poljoprivredni traktor), dijelovi stabla (deblo, grana, panj), kukci i povr{inski objekti. Vrste su ozljeda razdijeljene u 6 kategorija: otvorene rane (porezotina, razderotina, oguljotina), nagnje~enja i udarci, istegnu}a (istegnu}a, uganu}a, i{~a{enja, puknu}e mi{i}a), prijelomi, ubodi kukaca, ozljede/gubitak oka. Ozlije|eni dio tijela moè biti: vrat i glava, lice, prsni ko{ i le|a, trbuh i kuk, udovi. Uz opisne metode pri statisti~koj su analizi podataka kori{teni Welchov test, Tamhanesov T2 post hoc test i binarna multipla logisti~ka regresija. Najve}i se broj nesre}a (slika 2) dogodio za vrijeme sje~e (68 %), tijekom privla~enja zabiljeèno je 24 % nesre}a, dok je najmanji broj nesre}a utvr|en tijekom njege (8 %). Teìna je nesre}a prema radnim operacijama zna~ajno razli~ita bez obzira radi li se o analizi svih nesre}a ili analizi 95 % nesre}a (izuzeto je 5 % najteìh nesre}a). Ra{~lamba razlike u teìni nesre}a pri uzimanju u obzir svih ili samo 95 % nesre}a, za pojedinu radnu operaciju, pokazuje da su se najteè nesre}e doga|ale pri sje~i (D=10,1 izgubljeni dan/nesre}a) i privla~enju drva (D=13,5 izgubljeni dan/nesre}a).

182

Croat. j. for. eng. 30(2009)2

Severity Analysis of Accidents in Forest Operations (171–184)

I. Poto~nik et al.

Pri analizi izvora ozljede eliminirano je 20 nesre}a s nepoznatim izvorom pa je ukupan broj svih nesre}a bio 826, odnosno 785 nesre}a bez onih najteìh 5 %. Naj~e{}i uzrok ozljeda bili su razli~iti dijelovi stabla (60 %) te povr{inski objekti (stijene, kamenje i ostale povr{ine) s udjelom od 24 %. Suprotno uobi~ajenomu o~ekivanju motorna je pila uzrokovala tek 6 % ozljeda (slika 3). Teìna je nesre}a, unutar pojedinoga izvora ozljeda, statisti~ki zna~ajna za obje analize, sve nesre}e i 95 % nesre}a (tablica 2). Usporedbom svih nesre}a s 95 % nesre}a moèmo re}i da je teìna nesre}a uzrokovanih razli~itim dijelovima stabla (33,3 i 19,8 izgubljeni dan/nesre}a) i povr{inskim objektima (26,7 i 21,0 izgubljeni dan/nesre}a) zna~ajno vi{a od nesre}a uzrokovanih ru~nim alatima (13,6 i 11,4 izgubljeni dan/nesre}a) ili kukcima (5,0 izgubljeni dan/nesre}a). Unutar analize svih nesre}a oba izvora ozljeda s najve}im udjelom ve}a su od nesre}a uzrokovanih motornom pilom (15,3 izgubljeni dan/nesre}a). Najteè su nesre}e uzrokovane mobilnim strojevima, dijelovima stabla i povr{inskim objektima. Od 846 nesre}a, pri razmatranju vrste ozljeda, radilo se o 819, tj. 777 nesre}a. Za sve eliminirane nesre}e zabiljeèna je vrsta ozljeda u kategoriji »ostalo«, a za jednu je nesre}u nedostajao podatak o vrsti ozljede. Kod ve}ine je nesre}a (slika 4) vrsta ozljede bila nagnje~enje i udarac (56 %), otvorena rana (19 %), uganu}e i i{~a{enje (11 %), a sve su razlike bile statisti~ki zna~ajne (za analizu svih i za analizi 95 % nesre}a). Teìna je ozljeda bila u objema analizama (ne uzev{i u obzir ozljedu oka) najvi{a u slu~aju lomova (129,9 i 50,3 izgubljeni dan/nesre}a), a najnià u slu~aju uboda kukca (9,0 izgubljeni dan/nesre}a) u usporedbi sa svim ostalim vrstama ozljeda (tablica 3). Rijetke i vrlo te{ke nesre}e rezultat su prijeloma kostiju i kontuzija zbog udaraca. Iz svih je nesre}a, pri analizi ozlije|enoga dijela tijela, izuzeto 13 nesre}a jer je kao ozlije|eni dio tijela bila navedena kategorija »ostalo«, a za 5 je nesre}a nedostajala informacija o ozlije|enom dijelu tijela pa su i te nesre}e izuzete iz predmetne analize. Dakle je ra{~lanjeno 828 nesre}a (786 pri 95 % nesre}a), a 66 % ozljeda odnosilo se na ruke i noge (slika 5). Teìna nesre}a s ozljedama ekstremiteta (30,7 i 20,5 izgubljeni dan/nesre}a) statisti~ki je zna~ajno vi{a od teìne nesre}a s ozljedom lica (14,2 i 11,6 izgubljeni dan/nesre}a). Najteè su nesre}e povezane s ozljedama udova, iako je uo~en i velik udio te{kih nesre}a uzrokovanih ozljedama trbuha i kuka. Binarnom multiplom logisti~kom regresijom obavljena je procjena istodobnoga utjecaja razli~itih ~imbenika na teìnu nesre}e. Utvr|eno je da za sve nesre}e dva ~imbenika imaju statisti~ki zna~ajan utjecaj na teìnu nesre}e. To su izvor ozljede i ozlije|eni dio tijela (tablica 4). Radi preglednosti dobivenih rezultata na slici 6 i na slici 7 prikazane su kombinacije po 3 utjecajna ~imbenika ~ija kombinacija daje najve}u teìnu nesre}e i koji u zbroju izgubljenih radnih dana sudjeluju s vi{e od 75 %. Od 90 mogu}ih kombinacija (3 x 6 x 5) izme|u ozlije|enoga dijela tijela, izvora ozljede i radne operacije, njih 7 utje~e na 76 % izgubljenih radnih dana (slika 6). Pri kombinaciji vrste ozljede, izvora ozljede i radnoj aktivnosti 8 kombinacija utje~e na 7 % izgubljenih radnih dana (slika 7). Rezultati su analize pokazali da se 68 % nesre}a dogodi pri sje~i i izradi stabala, 24 % tijekom privla~enja drva, a samo 6 % za vrijeme njege; to je pone{to druga~ije nego u istraìvanjima provedenima u [vedskoj (60 %, 20 %, 20 %; Engsás 1995). Razlike su uvjetovane druga~ijom vlasni~kom strukturom {uma u kojima su istraìvanja provedena (dràvne {ume/privatne {ume), prije spomenutom tehnologijom rada i radnim uvjetima. Utvr|eni gubitak radnih dana po nesre}i pri sje~i i izradi stabala (29,9) i privla~enju drva (32,5) bitno odstupa od rezultata istraìvanja u Novom Zelandu (13,6 i 11,1 izgubljeni dan/nesre}a; Bentley i dr. 2002, 2005). Zabiljeèno je relativno malo ozljeda uzrokovanih motornom pilom (6 %), {to je usporedivo s istraìvanjima Sullmana i dr. (1999) na Novom Zelandu. Udio je ozljeda uzrokovanih motornom pilom u slovenskim dràvnim {umama u padu (Jereb 2009). Sli~no pokazuju i rezultati ostalih usporedivih projekata (Lefort i dr. 2003). Me|utim, u Sloveniji razlog ne leì u mehaniziranju sje~e i izradbe drva, ve} u stalnoj primjeni za{titnih sredstava i obrazovanju radnika. Udio nesre}a uzrokovanih svim radnim sredstvima (15 %) usporediv je s rezultatima KWF-a u njema~kim dràvnim {umama (14 – 20 %). Unato~ zadovoljavaju}emu padu udjela nesre}a u {umarstvu uzrokovanih motormom pilom, udio nesre}a ~iji su uzrok razli~iti dijelovi stabla i razli~iti povr{inski objekti u porastu je. Prema Jerebu (2009) takav je trend, uz visoku teìnu tih ozljeda, logi~an ako su nam poznate ~injenice da se ozljede od dijelova stabala mogu vrlo te{ko sprije~iti primjenom za{titne opreme i da razlog njihova nastajanja naj~e{}e leì u dana{njem prebrzom tempu rada. Prema vrsti ozljede zamije}eno je 56 % kontuzija, 19 % otvorenih rana, 11 % uganu}a i i{~a{enja te 6 % prijeloma kostiju. Podaci KWF-a (u njema~kim dràvnim {umama) imaju istu vrijednost za otvorene rane, manju vrijednost za kontuzije (41 %), a ve}e vrijednosti za uganu}a i i{~a{enja (22 %) i za lomove (9 %). Razlog tomu mogao bi biti ve}i stupanj mehaniziranosti sje~e, izrade i privla~enja u njema~kom {umarstvu. Tijekom mehanizirane sje~e i izradbe (Väyrynen 1982) glavnina se ozljeda doga|a pri popravku i odràvanju stroja, a onda su to naj~e{}e uganu}a i i{~a{enja (35 %) te nagnje~enja i udarci (30 %). Prema ozlije|enomu dijelu tijela naj~e{}e su ozljede ruku i nogu (66 %), a sli~no je i u istraìvanjima KWF-a (64 %), Wanga (2003) – 51 % i Leforta (2003) – 50 %. Iznena|uju}e je visok bio postotak ozljeda lica (15 %) posebno pri njezi (33 %), od ~ega je polovica ozljeda oka. Zaklju~ujemo da pri njezi svi radnici trebaju nositi za{titne nao~ale.

Croat. j. for. eng. 30(2009)2

183

I. Poto~nik et al.

Severity Analysis of Accidents in Forest Operations (171–184)

Istraìvanje je pokazalo da teìna nesre}e ovisi o svim odabranim ~imbenicima, a najvi{e o ozlije|enom dijelu tijela i izvoru ozljede kada su u analizu uklju~eni svi ~imbenici. U praksi to zna~i da se nesre}e pojedinoga dijela tijela, poglavito trbuha i le|a, mogu izbje}i ili sprije~iti primjenom mobilnih sredstava za rad u pojedinoj radnoj operaciji. Nadalje, mjere sigurnosti i za{tite usmjerene smanjenju teìne nesre}a u {umarstvu trebale bi voditi smanjenju ozljeda uzrokovanih razli~itim dijelovima stabla posljedice ~ega su kontuzije i lomovi ekstremiteta (te drugih dijelova tijela tijekom sje~e) te povr{inskim objektima posljedice ~ega su uganu}a, i{~a{enja i lomovi. Rezultate ovoga istraìvanja mogu}e je usporediti s vrlo malim brojem sli~nih, novijih istraìvanja jer, s jedne strane, razvijenije {umarske zemlje ({to se ti~e tehnolo{kih dostignu}a) u kojima se sli~na istraìvanja provode obavljaju sje~u, izradbu i koranje stabala mehanizirano pa usporedba rezultata nije mogu}a, dok se, s druge strane, u zemljama koje primjenjuju podudarne na~ine pridobivanja drva i s kojima bi usporedba rezultata bila mogu}a, sli~na istraìvanja nisu prioritetna. Ovakva istraìvanja sluè, ponajprije, da na temelju analiza nesre}a u pro{losti planiramo mjere i postupke usmjerene prevenciji budu}ih nesre}a pri {umskim radovima. Vrijednost rezultata istraìvanja i temeljem toga preporu~enih aktivnosti u izravnoj je vezi s kakvo}om i koli~inom ulaznih podataka. Klju~ne rije~i: {umski radovi, nesre}a, teìna, ozljeda

Authors' address – Adresa autorâ: Assoc. Prof. Igor Poto~nik, PhD. e-mail: igor.potocnik@bf.uni-lj.si Anton Poje, MSc. e-mail: anton.poje@bf.uni-lj.si University of Ljubljana, Biotechnical Faculty Department of Forestry and Renewable Forest Sources Ve~na pot 83 SI 1000 Ljubljana SLOVENIA

Received (Primljeno): March 8, 2009 Accepted (Prihva}eno): November 15, 2009

184

Assoc. Prof. Tibor Pentek, PhD. e-mail:pentek@sumfak.hr Department of Forest Engineering Forestry Faculty of Zagreb University Sveto{imunska 25 HR–10000 Zagreb CROATIA Croat. j. for. eng. 30(2009)2

Preliminary note – Prethodno priop}enje

Influence of Selected Meteorological Phenomena on Work Injury Frequency in Timber Harvesting Process Jozef Suchomel, Katarína Belanová Abstract – Nacrtak The aim of this article is to analyze the impact of the chosen meteorological phenomenon on the frequency of work accidents in the logging process in the Slovak forest sector. The article further evaluates the number of specific types of injuries, variations and sources of work accidents. Rain was the most frequently recorded meteorological phenomenon in the occurrence of accidents. Besides rain most injuries and wrenches were recorded on slippery surface and fractures occurred mostly during snowfall or in the presence of snow cover. Slips and falls were mostly recorded during rain as well as on slippery surface without the presence of any meteorological phenomena. The methods of analysis, synthesis and comparison were used In preparing this paper. Key words: forestry, forestry work, work safety, injuries, weather, meteorological phenomena

1. Introduction – Uvod The significance of weather and climate can be recognized for both biotic and abiotic organisms on Earth including humans. Weather and climate are important factors that influence labor productivity and quality, but also provide a certain comfort in performing work activities. Organism and environment form a biological unit. All atmospheric environmental components, physical, chemical and biological, affect the human body. Weather and climate are characterized by meteorological factors. Weather is characterized by the sum of meteorological inputs that have a short-term character. Climate is a sum of meteorological inputs with a long-term character. Weather and its changes characteristic for a certain area constitute a climate or a palate in that particular area. Weather or climate can directly influence human beings by cold, warmth, rainfall, wind; in case of fieldwork, weather can affect the changes of field production conditions and, by affecting the worker’s physical and psychic state, it is likely to influence the work performance. Croat. j. for. eng. 30(2009)2

2. Air temperature – Temperatura zraka Air temperature is closely linked with the human body performance, which decreases with higher temperatures. Physical activity suffers more than the psychic one. The rise in temperature for example by 4°C, from 20°C to 24°C, will cause the fall of physical activity of about 15%. The optimum temperature for medium-hard labor is the temperature range between 15° and 17°C. A properly dressed and protected person can bear and work in extreme outdoor temperatures from –50°C up to +100°C but its own body only tolerates variations of up to 4°C of the body core temperature without adversely affecting his/her labor and physical efficiency. The lower temperature limit, where an isolated cell is damaged by the creation of water crystals, is –1°C. On the other hand at +45°C the cell proteins coagulate. The cell can bear the temperature above +41°C only for a short time. Perspiration is a significant element for balancing between the loss and production of warmth. Warmth losses can be influenced by atmospheric humidity. In a very warm climate, where cooling by evaporation is the main cause of warmth losses because of

185

J. Suchomel and K. Belanová

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

high production of perspiration, it can easily lead to restrictions of evaporation caused by inappropriate clothing or high environment humidity. The parameter that characterizes the isolation capacity of clothing is called thermal isolating clothing resistance whose unit is clo. One clo is the isolating resistance of a common man’s clothing worn in the countries of middle climate and holds the thermal equilibrium of a person sitting standstill in an environment where the air temperature is 21°C, relative humidity is less than 50% and the velocity is not higher than 360 meters per hour. Under these conditions the thermal production is approximately 210 kJ m² per hour and the thermal loss must also equal 210 kJ m² per hour so that the person can feel thermal comfort. For ensuring appropriate clothing isolating features, it is necessary to change clothing for various types of hard physical labor and wear several layers of cloths. Wearing of clothing that can be zipped and unzipped enabling the circulation of air was well acquitted. The air heats itself by the body surface and rises by thermal convection. The ventilation can be reached by opening the top parts of the clothing. The main problem still remains of the thermal isolation of extremities/limbs especially fingers, which have become the limiting factor for extreme frost tolerance. Feet are the limiting factor for sitting or standing individuals. Wearing shoes with very good isolation and adequate thermal isolation of other body parts enable the individual to bear the foot skin pane at –20°C for approximately 140 minutes and at –40°C for approximately 70 minutes (Matou{ek 1987).

3. Rainfall – Oborina Rainfall influences directly a person working in the forest – just by its presence where the performance of certain operations can become more difficult, more exhausting and often impossible. A far greater impact of rainfall can be observed in combination with lower temperatures in winter when the prevailing amount of timber production in the Slovak forestry is carried out. Frozen soil surface or snow cover under which a lot of brushwood is hidden becomes a site of many forest accidents. Due to specific microclimatic forest soil conditions, there is an all-year-round risk of slips and falls. It would be interesting to analyze the impact of other weather factors for example humidity, atmospherics pres-

sure, air circulation, solar radiation and others on the occurrence of work accidents. The effects or concurrence of multiple meteorological factors with one figure are presented by various parameters for example effective temperature, equivalently-effective temperature, cooling-down temperature (refrigeration) ([amaj et al. 1994).

4. Material and methods – Materijal i metode The aim of the article is to analyze the impact of chosen weather factors on the occurrence of work accidents in a group of forest workers. Due to problem complexity, for the analysis of this issue it was necessary to choose branches1 typical of the location in lowland and highland areas (in terms of the current climatic classification of the Slovak territory). These criteria were met by Branches Liptovský Hrádok, Námestovo and Kriváò. Geographical location of the branch Liptovský Hrádok is in the Liptovská fold. The territory is characterized by great relief diversity; it has different altitudes and is therefore climatically diverse with annual average temperatures ranging between 0.2 and 6.3°C. The average annual rainfall is in the interval between 690 and 1810 mm. In this area the snow cover lasts for about 80 to 222 days (Hribik et al. 2008). The area and location of the branch Námestovo belongs to the geographical region of Orava. This area also has different altitudes, and it is therefore also climatically considerably heterogeneous with annual average temperatures ranging between 0.4 and 4.7°C. The average annual rainfall is in the interval between 720 and 1800 mm. In this area the snow cover lasts for about 100 to 220 days (Hribik et al. 2008). The branch Kriváò is geographically located in the Pol'ana mountains. Thank to its great orographic diversity this area is also climatically heterogeneous with the annual average temperatures ranging between 3.7 and 6.2°C. The average annual rainfall is in the interval between 720 and 1200 mm. In this area the snow cover lasts for about 80 to 180 days (Hribik et al. 2008). The primary data for the frequency evaluation of weather-based work accidents were the data on the weather that characterized a certain day or hour when the work accident occurred as well as data on a specific location where the meteorological and pre-

Regional branches of the state-owned enterprise Lesy SR, š.p.

186

Croat. j. for. eng. 30(2009)2

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

cipitation-gage station was installed. We used the following data: Þ air temperature [°C], Þ rainfall [mm], Þ snow cover [cm], Þ soil surface. The altitude of the occurrence of work accidents was determined by temperature setting. The basic methods used for temperature setting were interpolation or regressive theorem, which enabled us to assign other missing data by using the known temperatures. The data about rainfall and snow cover were collected from precipitation-gage stations enabling a considerably precise condition setting, because the density of the precipitation-gage stations network is wider than the density of meteorological stations. The frequency of work accidents can be influenced by the soil surface and by various weather effects. These are listed in the meteorological stations in the form of a code, as for example: I = dew, P = coherent snow cover of 1 cm and more, M = fog, ice fog, ground fog, Soil surface conditions: 0 = dry soil surface, 1 = wet soil surface, 3 = bare soil surface and frozen.

J. Suchomel and K. Belanová

The data on work accidents were taken from the existing state-owned enterprise Lesy SR work accidents database (Suchomel et al. 2008). The following data were used from the database: the source of work injury, type of injury and injured body part (the criteria were used according to the Ministry of Labor, Social Affairs and Family Regulation No. 500/2006, which specifies the pattern for recording a registered injury). The data were processed with Microsoft Excel. Afterwards different types of charts were created and evaluated.

5. Results – Rezultati The total sum of work injuries listed in the branches Liptovský Hrádok, Námestovo and Kriváò in the research period in the production-transport process was 209. Out of all accidents, 36% occurred during rain, ice rain, or rain shower. In 24 work-accident cases it was snowing or snow occurred and also in 24 cases the snow cover of not less than 1 cm was recorded. In 13% of days (27 cases) when the work accidents occurred, dew was recorded in the research area, which is equal, in case of forest work, to all other above mentioned meteorological factors in maintaining work safety in the sense of work accidents origin group IV – work, or road traffic area as a source of workers’ falls, where 29% of injuries were recorded in timber harvesting/logging process in the Slovak forestry sector (Suchomel et al. 2008). Figure 2 shows the share of injuries by occurrence of specific meteorological factors. In order to sim-

Fig. 1 The most occurring meteorological factors by work injuries Slika 1. Naju~estaliji meteorolo{ki ~imbenici prema ozljedama na radu Croat. j. for. eng. 30(2009)2

187

J. Suchomel and K. Belanová

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

plify the interpretation, injuries with the highest share were aggregated into three groups: wounds, fractures and the third group were wrenches and dislocations mostly concerning lower limbs. Besides the occurring meteorological factors, »dry surface« and »slippery surface« categories were also evaluated by the analysis of accident frequency. In the category »slippery surface«, the following soil surface conditions were considered: wet, soggy and frozen surface. Clearly, the accidents occurred most frequently during rain (72 cases) of which 65% were injuries (surface and open wounds), 21% fractures and 9% wrenches and dislocations. In addition to the graphically illustrated results, injuries were also recorded that required subsequent amputation (for example fingers). These injuries occurred during rain or directly under its influence. With 15 injuries, the work place with slippery surface was recorded as the second most risky work place. The same share of injuries (14 cases) was registered during snow fall, or snow showers and coherent snow cover. Most fractures happened during rain, as already stated above. The total number of 10 fractured body parts was recorded while it was snowing, or with the snow cover and it is interesting that the same number of work injuries with fractures occurred on dry work surface. Besides rain, wrenches and dislocations were most frequently recorded on slippery surface (5 cases) and during dew (4 cases). Registering work injuries with the ESAW (European statistics on accidents at work) methodology

requires the entry of variation category (Fig. 3). The variation describes unusual events as for example partial or whole loss of vehicle control or a fall on something or from something. The most recorded variation was the loss of control, command, whole or partial of the machine, vehicle or manipulating facility, hand gear (tools), object, animal – in 65 cases. We can also consider the negative impact of unfavorable weather conditions on the risk of occurrence of work accidents in these cases but the primary records were not sufficient for a precise analysis. Similarly, only hypothetical statements could be made in the case of variations »uncoordinated workers movements« and »material factor fall, collapse«, where again reliable assessments of variation by unfavorable weather conditions were not possible. In almost 38% of all analyzed work accidents, slip or bad pack – without fall (22 cases) and the worker’s slip with subsequent fall (57 cases) as variation or unusual event were recorded. Suchomel et al. 2008 analyzed in detail the frequency of worker’s »slips« and »falls« depending on unfavorable weather (snow and/or rain) and the ground slope. For examination and evaluation of this matter the statistical method of contingent table was used. The result of the analysis shows that the statement: Slip and fall accidents at work do not depend on changes of terrain slope and bad weather occurrence, cannot be rejected with a 95% reliability. By evaluating the sources of work injuries in the relevant data files, it was found out that working or

Fig. 2 Injury type and share by occurring meteorological phenomena Slika 2. Vrste i broj ozljeda prema vremenskim prilikama 188

Croat. j. for. eng. 30(2009)2

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

J. Suchomel and K. Belanová

Fig. 3 Variations occurred by meteorological phenomena Slika 3. U~estalost uzroka ozljede prema vremenskim prilikama road traffic places as the sources of workers’ fall were recorded in 53 cases. The highest amount of accidents was recorded on slippery work surface (without rainfall). This phenomenon was also frequent in the source group V – material, burdens, objects and I – means of transport, where workers slips often occurred in entering the vehicle, or in leaving it. Ap-

proximately 10% of all accidents happened when snow cover higher than 1 cm was present. As stated above, clearly the most frequently registered accidents occurred during rain and the most frequent accident source during this weather was group III – machines, where a transportable chain saw is also included, as well as group II – raisers and transport-

Fig. 4 Sources of injuries occurred by meteorological phenomena Slika 4. U~estalost izvora ozljeda prema vremenskim prilikama Croat. j. for. eng. 30(2009)2

189

J. Suchomel and K. Belanová

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

ers, raising and transporting tools. During rain, any activity in the forest is dangerous, not to speak about performing difficult and risky work activities such as logging.

5. Conclusion – Zaklju~ak It can be concluded that out of 209 work accidents in 190 cases the occurrence of some metrological phenomena was recorded. The most frequent phenomenon was rain (36%), in 13 percent of days (27 cases), where an accident occurred, dew was recorded in that location, in 24 cases it was snowing and similarly in 24 cases of work accidents, snow cover higher than 1 cm was recorded. Besides rain, the most frequent injuries, wrenches and dislocations were recorded on slippery surface, and fractures occurred mostly as a result of snow fall or snow cover. Slips and falls were mostly recorded during rain and also on slippery surface without the occurrence of any meteorological phenomena. The share of slippery surface in the cases of work accidents also corresponds to 25% of source accident group IV – work places. The occurrence of unfavorable weather cannot always be absolutely predicted. The elimination of work accident risks in these conditions is possible, especially by providing the proper protection at work, enhancing safety measures and especially by complying with the prescribed technological procedures.

6. Acknowledgements – Zahvala This article arose thanks to the Ministry of Education of the Slovak Republic within the grant KEGA 3/6429/08 – »Integration of content and structure of classes in field of ergonomics, work safety and protection of health at work, in study programs of Technical University in Zvolen.«

7. References – Literatura Matou{ek, J., 1987: Po~así, podnebí a ~lovìk. Avicenum, Praha, 293 pp. Suchomel, J. a kol., 2008: Analýza pracovných úrazov v Lesoch SR, {.p. TUZVO, ISBN 978–80–228–1979–4, 135 pp [amaj, F., Pro{ek, P., ^abajová, Z., 1994: Agrometeorológia a bioklimatológia. Vysoko{kolská skripta. Univerzita Komenského, Bratislava: p. 306 [kvarenina, J., Mindá{, J., 2001: Klíma. In: Bublinec, E., Pichler, V. et al. : Slovenské pralesy – diverzita a ochrana. ÚEL SAV Zvolen: 200 pp. Vyhlá{ka 500/2006 Z. z., ktorou sa ustanovuje vzor záznamu o registrovanom pracovnom úraze. Hríbik, M., [kvarenina, J., Kyselová, D. 2008: Hydrofyzikálne vlastnosti snehovej pokrývky v horských ekosystémoch Po'any, Nízkych a Západných Tatier v zimách rokov 2005/06-2007/08. In: Hydrologie malého povodí 2008 / ed. Miloslav [ír, Miroslav Tesaø, ¼ubomír Lichner. – Praha : Ústav pro hydrodynamiku AV ^R, 2008. ISBN 978–80–87117–03–3. p. 341–348.

Sa`etak

Utjecaj vremenskih prilika na u~estalost ozlje|ivanja u postupcima pridobivanja drva Vrijeme i klima zna~ajno utje~u na u~inkovitost i kvalitetu rada, ali i na sigurnost pri obavljanju {umskih radova. Sve atmosferske sastavnice okoli{a, fizi~ke, kemijske i biolo{ke, djeluju na ljudski organizam koji zajedno s okoli{em tvori jednu zajedni~ku biolo{ku cjelinu. Klima i vrijeme pritom su odre|eni meteorolo{kim ~imbenicima, vrijeme kao zbroj meteorolo{kih ~imbenika koji imaju kratkoro~an karakter, a klima kao zbroj meteorolo{kih ~imbenika s dugoro~nim djelovanjem. Vrijeme i vremenske promjene karakteristi~ne za odre|eno podru~je tvore klimu toga podru~ja. Vrijeme ili klima izravno utje~u na ~ovjeka preko hladno}e, topline, oborine, vjetra i dr. Pri radu na otvorenom vrijeme moè djelovati na promjenu uvjeta rada i na obavljanje posla utjecajem na psihi~ko i fizi~ko stanje radnika. Temperatura je zraka sna`no povezana s izvedbom ljudskoga organizma koja opada s vi{im temperaturama. Psihi~ka aktivnost pritom trpi vi{e od fizi~ke. Porast temperature za npr. 4 °C, s 20 °C na 24 °C, uzrokovat }e smanjenje psihi~ke aktivnosti za pribli`no 15 %. Optimalna temperatura za srednje teàk rad je temperatura izme|u 15 i 17 °C. Oborina izravno utje~e na radnika pri radu u {umi. Samom pojavom oborine obavljanje {umskih radova postaje teè, iscrpljuju}e i ~esto nemogu}e. Mnogo ve}i utjecaj oborina moè imati zajedno s niskom temperaturom i pojavom snijega zimi. Smrznuta povr{ina tla i snje`ni pokriva~ skrivaju mnoge opasnosti i postaju mjesta brojnih nesre}a u {umarstvu. Cilj je ovoga rada bio da se istraì utjecaj odabranih meteorolo{kih prilika na u~estalost pojavljivanja ozljeda pri {umskim radovima u slova~kom {umarstvu. U ~lanku se tako|er analiziraju brojnost i vrste ozljeda te uzroci i

190

Croat. j. for. eng. 30(2009)2

Influence of Selected Meteorological Phenomena on Work Injury ... (185–191)

J. Suchomel and K. Belanová

izvori ozljeda na radu. Istraìvanja su provedena u tri regionalne podru`nice dràvnoga {umarskoga poduze}a Lesy SR. Podru`nice Liptovský Hrádok, Námestovo i Kriváò odabrane su, s obzirom na klimatsku klasifikaciju Slova~ke, kao predstavnici tipi~nih lokacija nizinskih i gorskih podru~ja. Podaci o ozljedama na radu preuzeti su iz slu`benih evidencija dràvnoga {umarskoga poduze}a. Kao osnovni ulazi za procjenu u~estalosti ozljeda uzrokovanih vremenskim prilikama uzeti su podaci o vremenu tijekom odre|enoga dana ili sata u kojem se dogodila nesre}a i podaci o to~noj lokaciji nesre}e odnosno mjerenja najbliìh meteorolo{kih postaja. Ukupno je na pridobivanju drva u promatranom razdoblju zabiljeèno 209 ozljeda na radu. Od ukupnoga broja nesre}a 36 % ih se dogodilo za vrijeme ki{e, tu~e ili pljuskova. U 24 slu~aja je snijeìlo i isto tako u 24 slu~aja ozljede su se dogodile pri snje`nom pokriva~u ve}em od 1 cm. U danima kada su se dogodile ozljede na radu u 13 % njih (27 nesre}a) na radili{tima je zabiljeèna rosa. Najvi{e se nesre}a (72 ozljede) dogodilo prilikom padanja ki{e, a od toga je 65 % povr{inskih i otvorenih rana, 21 % prijeloma i 9 % uganu}a i i{~a{enja. Oko 10 prijeloma zabiljeèno je za padanja snijega ili pri snijegom pokrivenom tlu. Jednak broj ozljeda s prijelomima se dogodio i na suhoj radnoj povr{ini. Osim na ki{i uganu}a i i{~a{enja naj~e{}e su zabiljeèna na skliskoj povr{ini (5 slu~ajeva) i pri rosi (4 slu~aja). S obzirom na mjesto nastanka ozljeda radni je prostor kao izvor padova pri radu na pridobivanju drva zabiljeèn u 29 % nesre}a. Skliska povr{ina (bez ki{e) s 15 ozljeda drugi je najzna~ajniji izvor nesre}a. Kao naj~e{}i uzrok ozljeda u 65 slu~ajeva naveden je gubitak kontrole i mo}i upravljanja nad radnim sredstvom – strojem, vozilom, ru~nim alatom i predmetom rada. Gotovo u 38 % svih analiziranih nesre}a poskliznu}e bez pada (22 slu~aja) i poskliznu}e s padom radnika (57 slu~ajeva) zabiljeèni su kao uzrok ozljeda. Pojava nepovoljnih vremenskih prilika ne moè se uvijek to~no predvidjeti. Otklanjanje rizika od nesre}a u takvim uvjetima ipak je mogu}e, i to odgovaraju}om za{titnom opremom, pove}anom pa`njom pri radu i osobito primjenom pravilnih tehnolo{kih postupaka. Klju~ne rije~i: {umarstvo, {umski rad, sigurnost pri radu, ozljede, vrijeme, meteorolo{ke prilike

Authors' address – Adresa autorâ: Asst. Prof. Jozef Suchomel, PhD. e-mail: suchomel@vsld.tuzvo.sk Department of Fo0rest Exploitation and Mechanisation Forestry Faculty Technical University in Zvolen T. G. Masaryka 24 960 53 Zvolen

Received (Primljeno): June 4, 2009 Accepted (Prihva}eno): November 15, 2009 Croat. j. for. eng. 30(2009)2

Katarína Belanová, PhD. e-mail: belanova@vsld.tuzvo.sk Forest Enterprise Technical University in Zvolen [tudentská 20

191

Issue 2

2009

ISSN 1845-5719

9 771845 571000