Crojfe 35 1 14 by zpandur

2014

Original scientific paper

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption and Product Quality Carla Nati, Lars Eliasson, Raffaele Spinelli Abstract The study determined the time consumption, fuel consumption and chip size obtained with two different industrial chippers, working with logging residues (tops and branches), thinning material and pulpwood. Specific time consumption per oven dry tons (odt) was 83% higher for the less powerful disc chipper, and chipping forest residues resulted in a 35% increase in specific time consumption compared to chipping thinning material. What is more, the interaction between the two factors pointed at a different suitability of the two machines to chip different materials, since the difference in specific time consumption between the drum and the disc chipper was larger when chipping forest residues rather than thinning. Specific time and fuel consumption of the more powerful drum chipper increased by 30% and 39%, respectively, when working with dull blades compared to working with sharp blades. The best product quality was obtained when applying the disc chipper to pulpwood material. However, the same machine produced more fines when fed with forest residues. Keywords: chipping, disc, drum, fuel consumption

1. Introduction Nordic countries are innovators in the use of wood biomass for energy purposes. In Sweden, the utilization of forest residues started in the early 70s after the first oil crisis and increased dramatically since the late 90s (Mälkki and Virtanen 2003). In 2009 bioenergy accounted for 28.4% of the Swedish energy use (110.3 TWh) (Anon. 2010) and most of this came from woody biomass. Swedish heating and CHP plants generated 26.6 TWh of heat and electricity from wood chips and 5.8 TWh from wood pellets and briquettes in 2009, which increased to 30.0 and 6.6 TWh, respectively, in 2010 (Anon. 2011). Further amounts were used by the forest industry itself, to generate process heat (Björheden 2011). Chips are obtained from sawmill residues (mostly saw dust and bark) and from forest operations. Final cuts generate large amounts of logging residues, while pre commercial thinning operations offer small trees, unsuitable for other uses (Ranta 2005). Both sources of biomass explain the strong connection between the Croat. j. for. eng. 35(2014)1

energy sector, forestry and forest products industry (Hillring 2006). Since most biomass used in Sweden originates from the forests, forestry and forest industry represent key sectors for the Swedish biofuel market (Ericsson and Nilsson 2004). An increasing demand for solid biofuels requires increased efficiency in the supply chain, in order to avoid increased fuel costs (Björheden 2011). The cost of the whole chipping or comminution system represents a significant component of the overall supply chain expense. In particular, fuel costs account for a large share of the overall cost incurred by chipping contractors (Granlund 2011). Comminution also represents about 30% of the total sulfur dioxide, total suspended particles and carbon monoxide (SO2, TSP and CO) emissions generated by the forest energy supply chain (Mälkki and Virtanen 2003). In order to reduce emissions and contain supply cost, it is crucial to increase the productivity or reduce the fuel consumption of chipping operations. This can be done by manipulating several variables, and especially machine

C. Nati et al.

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7)

selection, feedstock type and knife sharpness (Nati et al. 2010). Chipping is the most efficient when performed at the plant or at a terminal using a large stationary machine, which explains the keen interest towards terminal based logistics (Kärha 2011). However, chipping increases the density and homogeneity of forest residues, which justifies its application early in the supply chain (Björheden 2008). Chip trucks have a higher payload than trucks for loose residues, which allows substantial savings on transportation cost. Thus, chipping at the roadside landing results in the lowest total costs of chipping and transportation, when the chips are moved over medium and long distances. That is well known to Swedish fuel suppliers, who in 2009 chipped 80% of the logging residues at roadside landings, 15% at plants or terminals after transportation in loose form, and only 5% at plants or at terminals after transportation as bundles (Brunberg 2011). Currently, roadside chipping is the dominant chipping technique in Sweden as well as in other European countries. That also accounts for Italy, where terrain and roadside chipping are prevalent because of increased efficiency of transportation to the plant (Spinelli and Hartsough 2001). That explains the large popularity of mobile chippers, despite the superior chipping performance of stationary units (Spinelli and Magagnotti 2010a).

2. Materials and Methods The study tested two different chippers used for roadside chipping. The drum chipper used was a Jenz HEM 561, powered by a 246 kW Claas Xerion, equipped with a crane and a grapple for grabbing the material to be chipped. The drum was equipped with 20 disposable micro knives. A 80 x 80 mm screen was placed between the drum and the evacuation system, in order to reduce the amount of oversize particles (slivers). The produced chips were blown directly into 40 m3 containers that were set out on the landing by the container truck or a tractor. Trials with this chipper were carried out at two different locations in south western Sweden. Logging residues of mainly birch and spruce from a final felling were chipped at Skultorp (N 58 20.268 E 13 51.267), and thinning material was chipped near Tibro (N 58 25.216 E 14 04.980). The thinning material consisted of 5 m long tree sections of mainly deciduous tree species (aspen, alder and birch). As the availability of residues was good, chipping of logging residues were run with used blades in good conditions (henceforth called »good«) and artificially dulled blades in order to test the effect of blade wear. The procedure was realized by the chipper op-

erator by means of an angle grinder, in order to reproduce the effect of several working hours. During the tests, carried out in October 2010, 14 full 40 m3 containers were produced – 5 using logging residues and sharp knives, 3 using logging residues and blunt knives, and 6 using tree sections from thinning. These corresponded to 84 and 58 green tons of chips (40% m. c.), respectively. The disc chipper was a TS 1200 powered by a 147 kW independent engine and mounted on a John Deere 810D forwarder. The forwarder also carried a dumping bin, with a capacity of 13 m3 loose. This setup gives the machine an increased off road mobility and the contractor used it to some extent for chipping in the forest stands. The disc chipper comminuted mixed hardwood and spruce logging residues at a landing near Mariestad (N 58 35.873 E 13 42.658), and mixed birch/pine pulpwood from a thinning at the biomass terminal in Götene (N 58 31.351 E 13 29.071). During the tests 6 containers of chips were produced, 3 using forest residues from a final felling and 3 using pulpwood from thinning. These corresponded to 23 and 22 green tons of chips (40% m. c.), respectively. As 36 m3 containers were used, only 2 bins were dumped in each container. During transport between the chipping site for logging residues and the place where the containers could be set out, the chipper engine was turned off. The tests were run with sharp blades and, for organizational reasons, no artificially wore blades were used. There was no significant difference in the moisture content of different material types, since all had been left to dry over the summer, which is common practice in Nordic countries (Suadicani and Gamborg 1999). The material was all in reach of the feeding boom, and the machine was sitting at a single location until the container was filled up. Time studies were carried out at the cycle level in order to measure time consumption and calculate productivity. Both productive and delay time were measured, but the analysis was conducted on productive time only. This was partly done to avoid the confounding effect of delay time, which is typically erratic (Spinelli and Visser 2009), but also as the studies were far too short to record representative delay times. Time was recorded with Allegro hand held computers, equipped with Skogforsk SDI software. Both chippers discharged chips into containers, and a full chip container was then assumed as a single cycle, and considered as a replicate. Due to the amount of available material and a somewhat limited machine availability for the disc chipper, the number of containers produced differs between the trials. Chip output was measured by taking all containers to a certified weighbridge, where both the filled and Croat. j. for. eng. 35(2014)1

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7)

empty weight of each container was recorded. Each container was identified with appropriate labels, in order to match its weight to the chipping time. A 10 dm3 sample of chips was taken from each container for determining moisture content and particle size distribution. Moisture content determination was conducted on subsamples, collected in sealed bags and weighed fresh and after drying at 105° C to constant mass (i.e. according to SS-EN 14774-2). Moreover, wood chip quality was assessed by sieving the wood chips according to the SIS-CEN/TS 15149-1 standard. Five sieves were used to separate the six following chip length classes: > 63 mm (oversize particles), ≤ 63 – 45 mm (large-size chips), ≤ 45 – 16 mm (medium size chips), ≤ 16 – 8 and ≤ 8 – 3 mm (small size chips), < 3 mm (fines). Each fraction was then weighed with a precision scale. Chipper fuel consumption was measured for each container, by starting with a full diesel tank and refilling it every time a cycle had been completed and a container load had been produced. To this end, the filling pump was equipped with a fuel reader, with an accuracy of 0.01 dm3. The evaluation of fuel consumption concerned the motive power for both of the chipping systems, the Claas Xerion on one hand and the independent engine of TS1200 on the other hand. The studies were conducted on commercial operations and not under controlled conditions. The analysis was divided in two parts. The first part consisted of comparing the two machines equipped with new and good blades on two different feedstock types, namely: residues and thinning material. The second part consisted of an analysis of the effect of different levels of blade wear on chipper productivity and fuel consumption. Material availability was limited to the landing were the Jenz chipper worked on logging residues. Data were analyzed with the SAS advanced statistics software, in order to check the statistical significance of the possible differences between treatments (SAS 1999). In particular, analysis of variance was used to determine the effect of machine type, feedstock type and blade wear levels on specific time and fuel con-

C. Nati et al.

sumption. χ2 tests were used in the comparisons of particle size distribution. The assumed significance level was 5%.

3. Results Table 1 shows the effects of machine and feedstock type on specific time consumption in minutes per oven dry ton (odt). These figures refer to actual chipping time, excluding the time taken by other activities, such as accessory work (moving of the loads, load transfer, etc.) and delays. In particular, chipping time accounted for 46.1% and 30.6% of total worksite time for the drum and disc chipper, respectively. Table 1 Total effective chipping time indicated as cmin odt-1 for different machines and materials (standard deviation in parenthesis) Material

Machine

Residues

Thinning

Jenz 561

336.9 (19.5)

300.9 (17.5)*

Jenz 561, dull knives

390.3 (2.3)

–

701.1 (167.8)

466.3 (24.4)**

TS1200

* Chipping of tree sections ** Chipping of small diameter pulpwood

All the analyzed factors had a significant effect on the specific time consumption absorbed by chipping (Table 2). The specific time consumption per odt was 83% higher for the less powerful TS chipper, and chipping forest residues from final cuts resulted in a 35% increase in specific time consumption compared to chipping small trees from thinning material. What is more, the statistical significance of the interaction factor shows that there is a significantly larger difference between the two chippers when chipping logging residues than when chipping thinning material (Table 2). The same was not verified for specific fuel consumption, where the only significant difference could

Table 2 Relationship between machine and material on total effective chipping time expressed as cmin odt–1 Source

Type III SS

Mean Square

F value

Pr > F

Machine

271 382.0715

58.23

<.0001

Material

70 989.1853

15.23

0.0018

Machine * Material

38 271.7114

8.21

0.0133

Croat. j. for. eng. 35(2014)1

C. Nati et al.

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7)

be attributed to machine type (Table 4). The less powerful TS chipper used 28% less fuel per oven dry ton, compared to the more powerful Jenz (Table 3). In the second part of the test, analysis of the Jenz results when chipping logging residues showed that

Table 3 Fuel consumption expressed as dm3 odt–1 for different chippers and materials (standard deviation in parenthesis). Thinning material is tree sections for Jenz 561 and small diameter pulpwood for TS1200 Material

Machine

Residues

Thinning

Jenz 561

2.43 (0.12)

2.26 (0.17)*

Jenz 561, dull knives

3.12 (0.15)

–

TS1200

1.68 (0.23)

1.66 (0.17)**

time consumption increased by 30% (p = 0.004) and fuel consumption by 39% (p = 0.001) when working with dull blades, compared to working with sharp blades. When chipping thinning material (pulpwood), the disc chipper produced significantly more chips in the 16 – 45 mm category and significantly less fines (< 3 mm) and oversized chips (> 100 mm) than was produced with any other chipper and material combination. However, the same machine gave significantly higher amounts of oversized material than any other combination when fed with logging residues (Fig. 1) resulting in a product unsuitable for non-industrial biomass systems. Also for the drum chipper the thinning material produced significantly more chips in the 16 – 45 mm category and significantly less fines (< 3 mm) and oversized chips (> 100 mm) than when logging residues were chipped.

4. Discussion

* Chipping of tree sections ** Chipping of small diameter pulpwood

The first part of the study confirms previous knowledge and offers interesting hints at new aspects, which should receive specific attention in the future. Other studies have already shown that specific time (Spinelli and Magagnotti 2010b) and fuel (Van Belle 2006) consumption are inversely proportional to machine power and piece size. In the present study there was no significant effect of material type on fuel consumption per odt and however the effect of the chipper used was significant. This confirms that disc chippers are more fuel efficient than drum chippers (Spinelli et al 2013) and that the fuel consumption per hour is proportional to the productivity. Furthermore, not all studies agree on the effect of piece size on specific fuel consumption (Spinelli et al. 2011a). Spinelli also reports of a higher fuel consumption at around 3.2 l odt–1 than the one in this study. Such difference is likely related to different data collection methods, since the study by Spinelli et al. (2011a) refers to pure chipping time, excluding all time when the chipping unit was not working. In that case, fuel consumption was recorded with a flow meter, and only when the drum was engaging the wood and the engine was un-

Fig. 1 Classification of woodchip according to standard

Table 4 Relationship between machine and material on fuel consumption (l odt–1) Source

Type III SS

Mean Square

F Value

Pr > F

Machine

1.65462675

55.97

<.0001

Material

0.03543882

1.20

0.2950

Machine * Material

0.02269759

0.77

0.3981

Croat. j. for. eng. 35(2014)1

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7)

der a workload. In the current study, fuel consumption was recorded by refilling the tank after each container load, so that the average fuel consumption figures accounted also for all inevitable short reductions of engine load, such as when the loader was handling wood and the chipper was running idle, waiting for new material to be fed. Each cycle lasted at least half an hour, so that the accumulated micro pauses were likely to have a significant impact on fuel consumption, and produce a different figure from those reported by Spinelli et al. (2011a). In this respect, it should be taken into consideration that the figures in this study refer to chipping time only, excluding all accessory work time and delays (Björheden et al. 1995). These figures are ideally suited for comparisons of different chippers and work methods, but do not reflect long term productivity levels per scheduled work hour. In particular, delays represent a significant proportion of a chipper’s scheduled work time, and may occupy up to 50% of the total worksite time (Spinelli and Visser 2009). In actual operations, the effect of delays may blur the eventual differences related to the characteristics of machine, material and blade conditions. The comparison between chipper types shows that the disc chipper offers excellent results with thinning material, but is not well suited to handle forest residues as it produces far too much oversized chips for a non-industrial use (UNI EN 14961-4:2001). This is something that many practitioners have stated but it has not been confirmed in earlier studies (cf. Spinelli et al. 2013). It may be argued that the thinning material was not exactly the same for both machines, since the Jenz handled tree sections while the TS was fed with delimbed pulpwood. However, lengths and diameters were almost the same, and the amount of branches on the tree sections was limited, so that this difference was unlikely to introduce a significant bias. The data on fuel consumption indicate that the smaller disc chipper used significantly less fuel per oven dry ton, compared to the larger drum chipper. That goes against the basic tenets of scale economy, which seem to be verified also for the specific fuel consumption of disc chippers (Marchi et al. 2011), but confirms the results of Spinelli et al. (2013) that disc chippers are more fuel efficient than drum chippers. The lower specific fuel consumption of the disc chipper was at least partly related to a different power transmission and to a different use of the available engine power. The drum chipper was powered by a tractor PTO and through a belt transmission and may have suffered higher power train losses than the disc Croat. j. for. eng. 35(2014)1

C. Nati et al.

chipper, which was powered directly through a belt transmission. More importantly, the engine powering the drum chipper was also used to run the loader while the independent engine mounted on the disc chipper powered the chipper only. On the disc chipper, the loader was powered by the engine of the forwarder. The fuel consumption of the forwarder was not recorded, because the forwarder was also used for moving the loads to the load transfer site, and for lifting and tilting the chip container during the load dumping phase. Hence, the recording of forwarder consumption while using the loader would have been rather complicated, due to the need for separating the consumption incurred during the load transfer phases. For this reason it was excluded from the measurements. Hence, it is not possible to state that the lower specific fuel consumption of the disc chipper was caused by the disc chipping mechanism only. While the results of the current comparison between disc and drum chipper cannot be assumed as conclusive due to the above-mentioned limitations, they certainly hint at very interesting trends, which is worth exploring with further research on the different performance of disc and drum chippers. The data obtained in the second part of the study closely match the results presented by Nati et al. (2010), who conducted a similar research with a drum chipper. Their study reported increases in specific time and fuel consumption of 50% and 22%, respectively. The corresponding figures in this study are 39% and 30%, instead. The difference is indeed minor, considering the variability of differences of material chipped, different chipper models and operator work techniques. In particular, the chipper studied by Nati et al. (2010) was equipped with two large re-usable single piece knives, whereas the chipper used for the present study used multiple disposable micro knives. The two different types of knives and knife set-ups may have had different wear patterns, so that one type may have lost its efficiency faster and steeper than the other. Operator effect could also be a main source of variability (Purfürst and Erler 2006), since it may account for productivity differences up to 77% in harvester work (Harstela 1988). Feeding a chipper is a simpler job than felling and processing trees with a harvester, and therefore differences may not be as large as reported in harvester studies. Nevertheless, operator technique may well explain part of the differences found between the two studies. What is more, none of the studies included a quantitative measure of knife wear, as it could be indicated by measuring the sharpness angle of the knives or other similar parameters. Blades

C. Nati et al.

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7)

were considered dull when the respective operators reputed they could not effectively work much longer. Such a subjective criterion is likely to introduce substantial differences between the studies, hence the importance of their general agreement on a common order of magnitude. These figures can be used to calculate a rough breakeven point, beyond which the savings inherent to the extended use of worn chipper blades are lower than the additional cost caused by blade wear. Knives should be replaced when this point is reached. This study only provides the starting (new blades) and arrival (dull blades) points for blade wear, and does not allow this calculation to be conducted. Further research should address this point, as well as the actual difference between the drum and disc chipping mechanisms in terms of productivity, fuel consumption and product quality.

Acknowledgements The study was made with the support of the ESSprogramme »Efficient Forest Fuel Systems« funded by the Swedish Energy Authority and the Swedish Forest Sector, and by the EU COST Action 0902 »Development and harmonization of new operational research and assessment procedures for sustainable forest biomass supply«, which provided the funding for personnel exchange between institutions.

5. References Anon. 2011: Fuels. Deliveries and consumption of fuels during 4th quarter 2010 and year 2010. Sveriges officiella statistik, Statistiska meddelanden no. EN 31 SM 1101, SCB, Statistics, Sweden. Anon. 2010: Bioenergi Sveriges största energikälla. Bioenergy Swedens largest source of energy. The Swedish bioenergy association SVEBIO. Björheden, R., Apel, K., Shiba, M., Thompson, M. A., 1995: IUFRO Forest work study nomenclature. Swedish University of Agricultural Science, Department of Operational Efficiency, Garpenberg. 16 p. Björheden, R., 2008: Optimal point of comminution in the biomass supply chain. Proceedings of the Nordic Baltic Conference on Forest Operations, Copenhagen 23 – 25 September 2008. Danish Forest and Lanscape, Copenhagen, Denmark. Björheden, R., 2011: Growing energy – The development of forest energy in Sweden. In: Thorsén, Å., Björheden, R., Eliasson, L.: Efficient forest fuel supply systems. Composite report from a four year. R&D programme 2007 – 2010. Skogforsk ISBN 978-91-977649-4-0.

Brunberg, T., 2011: Forest fuel survey. In: Thorsén Å., Björheden, R., Eliasson L.: Efficient forest fuel supply systems. Composite report from a four year. R&D programme 2007 – 2010. Skogforsk ISBN 978-91-977649-4-0. Harstela, P., 1988: Principle of comparative time studies in mechanized forest work. Scandinavian Journal of Forest Research 3 (1–4): 253–257. Hillring, B., 2006: World trade in forest products and wood fuel. Biomass and Bioenergy vol. 30: 815–825. Kärhä, K., 2011: Industrial supply chains and production machinery of forest chips in Finland. Biomass and Bioenergy 35(8): 3404–3413. Mälkki, H., Virtanen, Y., 2003: Selected emissions and efficiencies of energy systems based on logging and sawmill residues. Biomass and Bioenergy 24 (4–5): 321–327. Marchi, E., Magagnotti, N., Berretti, L., Neri, F., Spinelli, R., 2011: Comparing terrain and roadside chipping in Mediterranean pine salvage cuts. Croatian Journal of Forest Engineering 32(1): 587–599. Nati, C., Spinelli, R., Fabbri, P. G., 2010: Wood chips size distribution in relation to blade wear and screen use. Biomass and Bioenergy 34(5): 583–587. Purfürst, F. T., Erler, J., 2006: The precision of productivity models for the harvester – do we forget the human factor? In: Precision Forestry in Plantations, Semi-Natural and Natural Forests. Proceedings of the International Precision Forestry Symposium. Stellenbosch University, South Africa, 5 – 10 March 2006: 465–475. Ranta, T., 2005: Logging residues from regeneration fellings for biofuel production – a GIS-based availability analysis in Finland. Biomass and Bioenergy 28(2): 171–182. Spinelli, R., Visser, R. J. M., 2009: Analyzing and Estimating Delays in Wood Chipping Operations. Biomass and Bioenergy 33(3): 429–433. Spinelli, R., Magagnotti, N., 2010a: Comparison of two harvesting systems for the production of forest biomass from the thinning of Picea Abies plantations. Scandinavian Journal of Forest Research 25(1): 69–77. Spinelli, R., Magagnotti, N., 2010b: A tool for productivity and cost forecasting of decentralized wood chipping. Forest Policy and Economics 12(3): 194–198. Spinelli, R., Nati, C., Sozzi, L., Magagnotti, N., Picchi, G., 2011: Physical Characterization of Commercial Woodchips on the Italian Energy Market. Fuel 90(6): 2198–2202. Spinelli, R., Magagnotti, N., Paletto, G., Preti, C., 2011a: Determining the impact of some wood characteristics on the performance of a mobile chipper. Silva Fennica 45(1): 85–95. Spinelli, R., Cavallo, E., Eliasson, L., Facello, A., 2013: Comparing the efficiency of drum and disc chippers. Silva Fennica 47(2): 1–11. Suadicani, K., Gamborg, C., 1999: Fuel quality of whole-tree chips from freshly felled and summer dried Norway spruce Croat. j. for. eng. 35(2014)1

Effect of Chipper Type, Biomass Type and Blade Wear on Productivity, Fuel Consumption ... (1–7) on a poor sandy soil and a rich loamy soil. Biomass and Bioenergy 17(3): 199–208. UNI EN 14961-4:2011 Standards: Solid biofuels Fuel specifications and classes Part 4: Wood chips for non-industrial use.

C. Nati et al.

Van Belle, J F., 2006: A Model to Estimate Fossil CO2 Emissions During the Harvesting of Forest Residues for Energy – with an Application on the Case of Chipping. Biomass and Bioenergy 30(12): 1067–1075.

Authors’ address:

Received: October 26, 2012 Accepted: March 18, 2013 Croat. j. for. eng. 35(2014)1

Carla Nati, PhD.* e-mail: nati@ivalsa.cnr.it Raffaele Spinelli, PhD. e-mail: spinelli@ivalsa.cnr.it CNR – Ivalsa via Madonna del Piano 10 50019 Sesto Fiorentino ITALY Lars Eliasson, PhD. e-mail: Lars.Eliasson@skogforsk.se Skogforsk Uppsala Science Park 751 83 Uppsala SWEDEN * Corresponding author

Original scientific paper

Fuel Chip Supply System with Low Price Mobile Chippers Mika Yoshida, Hideo Sakai Abstract High price of chippers and small scale of forestry management area seemed to create a standstill situation of chip utilization system in Japan. By combining the benefits of small scale forestry and low price chippers, a simple and low-cost chip supply model could be established, creating new employments opportunities. The object of this study is to propose the benefits of a chip supply system in small scale forestry using multiple low price chippers at forest landings. The volume of chip material was ranged up to 56 000 m3 in a model area. The results showed that the annual labor cost of the proposed system was higher than that of the conventional system because the utilization ratio of chipper of the proposed system was also higher. The chip supply cost of the proposed system became lower than that of the conventional system because of its proper productivity and machine price. The utilization ratio of machine used in this study will be an important indice for creating employments when establishing chip supply chain. The benefits of the proposed system using multiple chippers are not only creating employments in local community but also bringing flexibility to a stable chip supply system and avoiding the risk of machine workouts and troubles. Therefore, the proposed system that introduces multiple lines using low price chippers will be useful for cost efficient management of small scale forestry and for providing sustainable bioenergy business. Keywords: chip supply, employment, fuel chip, mobile chipper, small scale forestry

1. Introduction One of the problems of Japanese forestry is that the demand of wood has been decreasing with lowering of wood price. Making new demand of wood in domestic market is urgently required. One of the remarkable new wood products is chip for energy use. There have been many studies about managing chipping and transportation (chip supply chain management) intended to optimize the large-scale operation and to reduce the cost (Young and Ostermeier 1991, Spinelli and Hartsough 2001, Stampfer and Kanzian 2006, Tolosana et al. 2011, Rรถser et al. 2011). Moreover, chipping at forest roadside was clarified to be the cheapest system and became common (Allen et al. 1998, Laitila 2008, Lindroos et al. 2011, Ghaffariyan et al. 2013). However, there are few studies from the aspect of chip supply chain management in Japan. The main reason is the limited research opportunity of chip supply chain system in actual operations. The increase of inCroat. j. for. eng. 35(2014)1

troduced chippers has been behind compared with other forestry machines (Fig. 1 Forestry Agency 2011). It can be said that one of the major reasons for the standstill in chip utilization in Japan is due to the unreasonably high price of mobile chippers compared to their performance (Yoshida and Sakai 2013) although the price of machines is usually confidential (Nordfjell et al. 2010). To solve such situations, we introduced a low price disk type mobile chipper to the actual Japanese forestry from Korea. This low price chipper had already been investigated in detail (Yoshida and Sakai 2013) and the results have been announced. The chipping costs were lower than shown by other results in Japan, even though not lower than those of overseas chippers. The benefit of a low price chipper was clarified to be suitable for small scale forestry because it did not require a large amount of material to achieve low chipping cost compared to high price chippers.

M. Yoshida and H. Sakai

Fuel Chip Supply System with Low Price Mobile Chippers (9–14)

2. Materials and methods 2.1 Materials

Fig. 1 Forestry machines used in Japan (Forestry Agency 2011) The benefit of small scale forestry is the flexible management system, which can be changed according to the change of social conditions and economy. Moreover, forestry is one of the major businesses in rural area, so that the development of a stable business market in forestry sector can secure better employment opportunities. In Japan, 75% of forest owners own less than 5 ha of forest and 13% own 5 – 10 ha of forest. Because of such a situation, it is necessary to introduce small scale forestry management. Furthermore, not only small scale forest owners, but also low quality broad-leaved forests in Satoyama area (a forested area near the city) all over the country and young conifer plantation forests cannot be ignored for a full utilization of forest resources. By combining the benefits of small scale forestry and low price chippers, it is possible to establish a simple and low-cost chip supply model and to create new employment opportunities. The object of this study is to propose the benefits of a small scale chip supply model by using one or multiple low price mobile chippers at forest landings (proposed system), accompanied by new employment opportunities. At first, we made a chip supply model of chip production from landing to storage by estimating the volume of byproduct and calculating chipping and chip transportation costs. Then, the chip supply model was applied to an actual forest management in Japan conducted by a forest owners’ cooperative and compared it with the system that used a conventional high price chipper.

The proposed system assumed the use of a mobile chipper made by »Yulim Machinery« in Korea as a representative example of low price (€ 34 546) with reasonable performance. The precise data was investigated and summarized (Yoshida and Sakai 2013). A system using hummer mill chipper/grinder, with the price of € 371 471 (Yoshioka et al. 2006), was assumed here as the conventional system. The grinder was considered one of the most productive chippers in Japan compared to the other chippers used in Japan, as summarized by Moriguchi et al. (2004). The main specifications such as price, productivity, and fuel consumption are shown in Table 1. Both chippers required a grapple loader for preparing and loading materials. In this study, the difference of quality of products was not considered, on the assumption that all products could be fully utilized as fuel. Table 1 Specifications of the investigated chipper of low cost and of the reference hummer mill chipper/grinder YM-400C

HD-9 Industrial Tub grinder

Disc chipper (four blades)

Tub grinder (hummer mill)

34 546

371 471

Engine power, kW

150

205.1

Width, m

1.95

2.39

Length, m

7.72

Height, m

2.7

2.62

Name Type Price, €

Conveyor size, m

Length 1.6 Width 0.55 Height 0.5

Slot size, m

Width 0.4 3

No conveyor Depth 1.02 Top diameter 2.9 Bottom diameter 2.29

Productivity, m /h

23.7

Fuel consumption, l/h

14.6

71.4

The specifications of tub grinder were cited from Yoshioka et al. 2006 The currency ratio was 1€ = 134.6 yen, at November 15, 2013

In this model, the chip material was assumed to be the byproduct of tree length or whole tree harvesting carried out at forest landing. Therefore, the cost for Croat. j. for. eng. 35(2014)1

Fuel Chip Supply System with Low Price Mobile Chippers (9–14)

harvesting was not considered here. Produced chip was directly put into the truck container with the capacity of 8 m3, because larger sized containers were not always available for small scale forestry due to terrain conditions and width restrictions of the current forest road regulations in Japan. The container filled up with chips was transported from the forest to the storage facility by truck. In this system, a single worker carried out all chipping operations. The chip supply model was simulated by Touma town forest owners’ cooperative, Hokkaido Island, Japan, which had a high potential of regional fuel chip utilization, and the data was investigated by interviews of persons in charge. The setting area was 5 000 ha of private plantation forests. The main tree species were Japanese larch (Larix kaempferi) and Sakhalin fir (Abies Sachalinensis). Although they owned other 2 600 ha of broad-leaved forests, these were not considered in this study because they were not used regularly. At first, the annual available volume of chip material was estimated, and the chip production cost was calculated based on that volume. In this study, the term m3 means 1 cubic meter of chips. The utilization ratio of machine to scheduled machine hours (SMH) was also calculated to evaluate whether the machine worked within the given capacity or not. The currency ratio was 1€ = 134.6 yen (November 15, 2013).

2.2 Methods The annual volume of chip material Vr (m3/year) and the utilization ratio of chipper/grinder to SMH s were calculated by the following equations.

Vr = A × G × g × (1 − u) × k

s = (Vr / Ec ) / T

(1) (2)

Where: A setting area, ha; G annual forest growth, solid m3/ha; γ ratio of byproduct (not including branches); u ratio of unavailable volume of material because of dirt or ecological reasons; k density of chip volume to solid volume, m3/solid m3; Ec chipping productivity, m3/h; T was annual scheduled machine hours. The chip supply cost C (€/m3) was calculated based on Miyata (1980) and Forestry Agency (2012) by the equations from (3) to (5) including depreciation cost, labour cost, maintenance and repair cost, tax, and supplemental grapple loader and truck transportation costs. Croat. j. for. eng. 35(2014)1

M. Yoshida and H. Sakai

Cma =

(1 − S) × Pma × 1 + a + 1 − S Y + 1 × P / 2Y × b ( ) (( )( ) ma ) Y

Cchip =

(∑C + ∑(E

fuel

(3)

× Cfuel × Vr / Ec ) + Clabor × Vr / Ec × n

)

(4)

(5)

C = Cchip / Vr + Ctrans

Where: Cma was annual machinery cost including depreciation cost, maintenance and repair cost, and tax, €; S salvage ratio of depreciation; Y depreciation years, years; Pma machine price, €; α coefficient of maintenance and repair cost; β coefficient of tax; Cchip total cost of chipping operation, €/year; Efuel machine fuel consumption, l/h; Cfuel fuel price, €/L; Clabour labour cost per worker, €/h; n number of workers for chipping operation; Ctrans was transportation cost, €/m3. The annual forest growth was estimated to be 5 m3/ha in this area. The ratio of byproduct and unavailable volume were 0.3 and 0.2, respectively, in this community. As well as other forest owners’ cooperatives, this cooperative was surrounded by forests and it managed a forest area where the furthest point was at the distance of 10 km. The average transportation cost (round trip) was 7.43 €/m3. These data were obtained from interview investigations. The density of chip volume to solid volume was assumed to be 2.8 m3/solid m3 according to Helle et al. (2002). To evaluate the effect of annual volume size on the chip supply cost, the ratio of byproduct was changed in increments from 0.1 to 1.0 in 0.1, where the value of 1.0 meant full utilization of tree length as chip material. The price of low price chipper and grinder were 34 546 € and 371 471 €, respectively, as shown in Table 1. The depreciation years of chipper were 5 years (Miyata 1980) and the salvage ratio of depreciation was 0.1 according to Japanese law. The coefficient of maintenance and repair of low price chipper was assumed to be 1, and of the grinder it was 0.5, because the low price chipper might often be broken down. The fuel consumption of the chipper and grinder was 14.6 L/h and 71.4 L/h, respectively. The price of the grapple loader was 99 554 € and depreciation years of the grapple loader were 8.5 years (Forestry Agency 2012). The salvage ratio of the grapple loader was also 0.1. The fuel consumption of the grapple loader was 5 l/h ac-

M. Yoshida and H. Sakai

Fuel Chip Supply System with Low Price Mobile Chippers (9–14)

cording to data obtained by interview. The labour cost was 14.8 €/hour. These cost calculation data of each system were summarized in Table 2. Table 2 Machinery data of both proposed and conventional systems for cost calculation System

Proposed system

Conventional system

Grapple loader

Price, €

99 554

*Depreciation, year

8.5

Salvage ratio, %

0.1

Maintenance ratio, %

0.5

Tax, %

0.05

Fuel consumption, l/h

Fuel cost, €/l

Chipper

Disk chipper

Tub grinder

Price, €

***34 546

****371 471

**Depreciation, year Salvage ratio, % Maintenance ratio, % Tax, % Fuel consumption, l/h Fuel cost, €/l Hourly labour cost, €/hour

0.1

0.5

0.05

***14.6

****71.4

15.1

Fig. 2 The changes of utilization ratio of chipper/grinder

Based on Forestry Agency; Ministry of Agriculture, Forestry and Fisheries, Japan 2012 **Based on Miyata 1980 ***Based on Yoshida and Sakai 2013 ****Based on Yoshioka et al. 2006 The currency ratio was 1€ = 134.6 yen, at November 15, 2013

3. Results – Rezultati From the equation (1), the annually produced chip volume ranged from 5 600 m3 up to 56 000 m3 according to the ratio of byproduct in a model area. Fig. 2 shows the changes of utilization ratio of chipper/ grinder in each system, according to the ratio of byproduct. When using the proposed one-line system, the utilization ratio of chipper exceeded 1.0 around the ratio of byproduct of 0.8. Therefore, it was necessary to introduce one more chipper, should the ratio of byproduct rise over 0.8. The utilization ratios of the proposed one-line and multiple (two) lines systems were always higher than that of the conventional system. Fig. 3 shows the changes of chip supply cost. The costs of the proposed one-line and multiple lines systems were always lower than that of the conventional system, while the difference between the costs of the

Fig. 3 Changes of chip supply cost proposed multiple lines system and that of the conventional system got close according to the increase of the ratio of byproduct. The ratio of byproduct in this area was 0.3 and the chip supply costs of the proposed one-line, multiple lines, and conventional systems of were 10.7 €/m3, 13.4 €/m3, and 16.6 €/m3, respectively. As the ratio of byproduct increased, the annual labor cost for chipping increased, while the costs of the Croat. j. for. eng. 35(2014)1

Fuel Chip Supply System with Low Price Mobile Chippers (9–14)

Fig. 4 Changes of annual labor cost

proposed one-line and multiple lines systems were the same and always higher than that of the conventional system (Fig. 4). The annual labor cost of the proposed one-line system has reached the ceiling at the ratio of byproduct of 0.8 as shown in Fig. 2, while that of the proposed multiple lines system could be increased until the ratio of byproduct of 1.0.

4. Discussion The machine utilization ratio in a system can be one of the important indices of employment. The increase of the machine utilization ratio in a system means that the net working hours of the worker become longer, and the total labor costs become higher as shown in Fig. 4. Therefore, the proposed systems were more effective for creating employments than the conventional system as shown in the results of chip supply model in Touma town. As far as they produce chips within the volume of forest growth, it is possible to introduce two lines of the proposed chip supply system, while retaining cost efficiency and securing employment for two workers. Productivity is indeed important to reduce costs, but the small scale forestry usually does not have enough material to achieve low cost by increasing chipping productivity. Therefore, paradoxically, the low efficiency of a low price chipper can create job opportunities and assure lower production cost provided that the quantities of the material in the region are limited. Local communities strongly Croat. j. for. eng. 35(2014)1

M. Yoshida and H. Sakai

expect the economic effect of forestry, so that not only the production cost but also the machine utilization ratio is important when introducing chip supply chain to small scale forestry. One of the difficulties of low productivity is the long response time for sudden chip demand caused by seasonal fluctuation. There are some solutions such as making storage terminal (Gunnarsson et al. 2004) and buying chip from other communities. In this study, we proposed introducing multiple lines system of chip supply, which can secure the production of a higher amount of chips. As mechanization of forest biomass for energy production is crucial in order to make forest operations economically sustainable (Röser et al. 2012), it is important to be prepared for machine troubles and workouts, which are inherent in mechanized systems. By dividing the chip production system into multiple lines, such risks will be dispersed securing a stable chip supply. This is another benefit of introducing multiple lines into chip production, and it will also make chip supply system more flexible. Multiple lines of chip production may decrease the utilization ratio of machines but the lifetime of small chippers can be lengthened in exchange. As seen through the analysis of the chip supply cost, the two typical systems considered in this study might be specific examples of the global standard because the productivity of both chipper and grinder was lower than in forestry developed countries such as Finland and Austria (Kärhä 2012, Stampfer and Kanzian 2006). More precise data about other chippers and chip supply chain management should be investigated to evaluate proper productivity and system according to the increase of chip demand.

5. Conclusions Lowering the cost of the chipper is possible both for increasing employment opportunities in small scale forestry and for ensuring stable chip supply by providing cost efficiency. The utilization ratio of machine will be useful as an indice of employment. It should be considered, for instance, when applying subsidies to a chip supply system. Cost efficiency and new employment opportunities are usually in conflict with each other and cost efficiency is often given priority in such situations. However, energy business is different from the existing forestry businesses, since the stability of fuel supply is widely recognized as the most important. Therefore, the positive characteristics of the proposed system should be taken into account when establishing chip supply chain management in small scale forestry with limited resources.

M. Yoshida and H. Sakai

Fuel Chip Supply System with Low Price Mobile Chippers (9–14)

6. References Allen, J., Browne, M., Hunter, A., Boyd, J., Palmer, H., 1988: Logistics management and costs of biomass fuel supply. Int. J. of Phys. Distribution & Logistics Manag. 28(6): 763–477. Forestry Agency, Ministry of Agriculture, Forestry and Fisheries, Japan. 2011: The situation of forestry machineries in Japan. World Wide Web Site http://www.rinya.maff.go.jp/j/kaihatu/ kikai/daisuu.html (Accessed on 18 February 2014). In Japanese. Forestry Agency, Ministry of Agriculture, Forestry and Fisheries, Japan 2012: Kouseinou-ringyo-kikairiyou-koudokamanual (Manual for efficient utilization of high productive forestry machineries). Tokyo. 124p. In Japanese. Ghaffariyan, M.R., Sessions, J., Brown, M., 2013: Roadside chipping in a first thinning operation for radiata pine in south Australia. Croat. J. For. Eng. 34(1): 91–101. Gunnarsson, H., Rönnqvist, M., Lundgren, J.T., 2004: Supply chain modelling of forest fuel. Eur. J. Oper. Res. 158(1): 103– 123. Helle, S., Falster, H., Gamborg, C., Gundersen, P., Hansen, L., Heding, N., Jakobsen, H.H., Kofman, P., Nikolaisen, L., Thomsen, I.M., 2002: Wood for Energy Production, Technology Environment Economy, Second revised edition. COFORD, Dublin. 69 p. Kärhä, K., 2012: Comminution productivity of forest chips during last 30 years in Finland. Abstracts. Latvian state forest research institute »SILAVA«. In: Special issue of scientific proceedings »Mežzinātne« 25(58): 99–100. Lindroos, O., Nilsson, B., Sowlati, T., 2011: Costs, CO2 emissions, and energy balances of applying Nordic slash recovery methods in British Columbia. West. J. Appl. For. 26 (1): 30–36. Miyata, E.S., 1980: Determining fixed and operating costs of logging equipment. General Technical Report NC-55. St.

Paul, MN: U.S. Dept. of Agriculture, Forest Service, North Central Forest Experiment Station. Moriguchi, K., Suzuki, Y., Gotou, J., Inatsuki, H., Yamaguchi, T., Shiraishi, Y., Ohara, T., 2004: Cost of comminution and transportation in the case of using logging residue as woody biofuel. J. Jpn. For. Soc. 86(2): 121–128. In Japanese. Nordfjell, T., Björheden, R., Thor, M., Wästerlund, I., 2010: Changes in technical performance, mechanical availability and prices of machines used in forest operations in Sweden from 1985 to 2010. Scand. J. For. Res. 25(4): 382–389. Röser, D., Mola-Yudego, B., Prinz, R., Emer, B., Sikanen, L., 2012: Chipping operations and efficiency in different operational environments. Silva Fenn. 46(2): 275–286. Spinelli, R., Hartsough, B., 2001: A survey of Italian chipping operations. Biomass and Bioenergy 21(6): 433–444. Stampfer, K., Kanzian, C., 2006: Current state and development possibilities of wood chip supply chains in Austria. Croat. J. For. Eng. 27(2): 135–145. Tolosana, E., Laina, R., Martínez-Ferrari, R., Ambrosio, Y., 2011: Recovering of forest biomass from Spanish hybrid poplar plantations. Biomass and Bioenergy (35)7: 2570–2580. Yoshida, M., Sakai, H., 2013: Importance of capital cost reduction of chippers and their required productivity. J. For. Res. DOI 10.1007/s10310-013-0426-x. Yoshioka, T., Sakurai, R., Aruga, K., Nitami, T., Sakai, H., Kobayashi, H., 2006: Chipping of logging residues with a tub grinder: Calculation of productivity and procurement cost of wood chips. Croat. J. For. Eng. 27 (2): 103–114. Young, T.M., Ostermeier, D.M., 1991: The economic availability of woody biomass for the Southeastern United States. Bioresource Technology 37(1): 7–15.

Authors’ address:

Received: July 02, 2013 Accepted: August 15, 2013

Mika Yoshida, MSc.* e-mail: yoshida@fr.a.u-tokyo.ac.jp Prof. Hideo Sakai, PhD. e-mail: sakaih@fr.a.u-tokyo.ac.jp University Tokyo Graduate School of Agricultural and Life Sciences Department of Forest Sciences Bunkyo-ku Yayoi 1-1-1 113-8657 Tokyo JAPAN * Corresponding author Croat. j. for. eng. 35(2014)1

Original scientific paper

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood Drying in Logwood Piles Gernot Erber, Johanna Routa, Marja Kolström, Christian Kanzian, Lauri Sikanen, Karl Stampfer Abstract Moisture management is a key element to improving the cost-efficiency of energy wood supply, through the whole supply chain. Numerous studies of natural drying of forest biomass have been carried out based on traditional sampling of piles or weighing. The latest methodology for monitoring moisture changes has been continuous weighing of biomass in racks built on load cells. The aim of this study was to develop accurate drying models in Austria and Finland for small diameter logs and test the exchangeability of the developed models between countries. Overall drying periods were December 2009 to February 2011 for Austria and March 2012 to June 2013 for Finland. Moisture content dropped from 50.1% to 32.2% (Austria) and from 62.2% to 38.6% (Finland) during the drying periods. Drying performance was evaluated for the period April to October. Two different types of models were developed and the results were cross validated. It proved to be possible to fit satisfactory accurate drying models within the target deviation of ± 5% using both approaches. Whereas the Austrian approach is based on a more basic set of variables, the Finnish approach combines the variables within one. Both approaches are justified depending on the available data. Keywords: fuel wood drying, drying modeling, logwood piles, meteorological models

1. Introduction Biomass fuel quality is often defined by the calorific value. Lower moisture content results in increasing calorific value (Hartmann and Kaltschmidt 2001, Stokes et al. 1987). Drying in piles can help to decrease significantly the moisture content of energy wood within a short period of time (Erber et al. 2012). Depending on conditions during the drying period, whole trees and logwood are likely to lose 20% to 30% in moisture content within 5 to 6 months (Nurmi 1995, Suadicani and Gamborg 1999, Gigler et al. 2000, Nurmi and Hillebrand 2007, Röser et al. 2010). An advantage of drying logwood is low dry matter loss compared to drying logging residues or forest chips. Golser et al. (2005) report 2% dry matter losses per year for Norway spruce (Picea abies L.) and Scots pine (Pinus sylvestris L.). Dry matter losses can be caused either by microbial activity, most commonly Croat. j. for. eng. 35(2014)1

fungal attacks (biological), or spillage of material during handling and storage (technical)(Pettersson and Nordfjell 2007). Drying of logwood for energy purposes is also economically beneficial. Erber et al. (2012) reported a gain in income of 14.40 € per air dry ton compared to yielding interest after having sold the material without drying. Drying in windrows by convection is a process governed by temperature, relative humidity, wind speed and rainfall (Kröll 1978). Kofman and Kent (2009) commented that wind and sun exposure are the most important factors for drying. Stokes et al. (1993) list a large variety of drying techniques, including transpirational air drying and foliage on un-delimbed logs, which can be used to improve drying performance. The idea of using drying models to predict moisture content alteration first appeared in the 1980.

G. Erber et al.

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

Stokes et al. (1987) presented drying models for soft and hardwoods, for loblolly pine (Pinus taeda L.), oak (Quercus spp.), sweetgum (Liquidambar spp. L.) and red maple (Acer rubrum L.) bundles in south eastern USA. Their goal was to model weight reduction through drying using non-linear models. These included weather data like the total daily precipitation and the average daily air temperature. Days since felling, diameter at breast height as well as species and further variables were also considered. Different equations were provided for each season of the year. Depending on the species, different variables (days since felling, total rainfall, original weight of the bundle) were found to be the best predictors. Liang et al. (1996) developed a model for Leucanea (Leucaena leucocephala (Lam.) de Wit) including days since drying, initial moisture content, cumulative precipitation and potential evaporation. Gigler et al. (2000) chose a different approach: a model, considering a willow (Salix viminalis L.) log as a »non-shrinking, infinite long cylinder of homogenous wood material surrounded by bark«, where any radial water transport depended on different diffusivities of wood and bark. Murphy et al. (2012) stored Sitka spruce (Picea sitchensis (Bong.) Carr.) logwood and energy wood in Ireland and developed drying models for off-forest storage. Moisture content loss over a 10 day period was related to the moisture content at the start of the interval, to cumulative precipitation and evapotranspiration for the period, woody biomass type and type of cover. Filbakk et al. (2011) developed a model for whole tree drying in piles, focusing on explanatory variables like days of storage, harvesting season, location, climatic conditions and position in the pile, tree species and relative crown length. Model by Erber et al. (2012) for pine logwood predicted daily alteration in moisture content based on mean daily temperature and relative air humidity and the daily sum of precipitation. Relative

humidity was found to be the most important factor for drying. Similar to Murphy et al. (2012), Dong Wook and Murphy (2013) developed drying models for Douglas Fir (Pseudotsuga menziesii (Mirb.) Franco) and hybrid poplar (Populus spp.) in Oregon using linear mixed effects models. Again a 10 day period was chosen to predict moisture content alteration depending on cumulative precipitation and evapotranspiration. Material size was a further variable. It was concluded that, due to the logical variation of drying with the climatic pattern of a region, these models can be extended to other regions of Oregon, too. Based on previous studies, the specific research objectives of this study were: 1) to develop climate based drying models for two different piles of pine logwood in Austria and Finland in the spring to autumn period, using two different approaches developed in Austria and Finland based on data collected during two former experiments; 2) to investigate the exchangeability of the developed models between the countries.

2. Materials The data for this comparison of modeling approaches was derived from two recently completed drying experiments. Details on the Austrian study are given in Erber et al. (2012). In this paper moisture content is expressed on wet basis, as the ratio (in percent) of the water weight and the total weight of the woody biomass.

2.1 Study sites The two experimental sites in this study were located in Austria and Finland. The Austrian site (47°17’N, 15°58’E; 350 m above the Adriatic) was in Hartberg, the province of Styria. The Finnish site was in Ilomantsi (62°46’N, 30°58’E; 150 m above Helsinki

Fig. 1 Study sites in Finland (left side) and Austria (right side) showing the experimental design

Croat. j. for. eng. 35(2014)1

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

level). Both sites were off forest study sites. The Austrian site was a timber yard, the Finnish site a storage area at a research station (Fig. 1).

2.2 Materials To monitor the change in moisture content through the change in weight, metal racks on load cells were used at both sites. Pine logs with average diameters of 15 cm and lengths of 4 m to 5 m were investigated. Details on the material are given in Table 1. Disc samples (three per stem) were taken by chainsaw in the beginning to measure the initial moisture content in the laboratory. Table 1 Parameters of the material and experimental site Parameter Total number of logs

Austria

Finland

208

~ 150*

Average length, m

4.72 m ± 0.50 ~ 4 m ± 0.4*

Average diameter, cm

15.2 cm ± 5.3 ~ 15 cm ± 5*

Number of sample logs Initial moisture content (analysis), % Initial total load, kg Elevation of the first layer above ground level, cm Ground material

47.2

61.5

16 670

11 710

soil and gravel

gravel

* Estimated parameters

2.3 Modeling data Weather data was recorded at both sites. Average daily moisture content (MC, %) of the piles, wind speed (WS, m s-1), relative humidity (RH, %) and air temperature (TC, C°) and the daily sums of precipitation (P, mm), and solar radiation (R, W m-2) were selected from the data pool. In order to calculate reference evapotranspiration (ET0) according to the universal standard of the FAO Penman-Monteith method (Allen et al. 1998), further data on daily minimum and maximum air temperature and dew point temperature were provided. Net evaporation (net, mm) was calculated by subtracting precipitation from the reference evapotranspiration. Whereas Austrian data on precipitation and solar radiation originated from the study site, Finnish data was partly provided by weather radar (precipitation) and a grid based model (10 km x 10 km, solar radiation). Air temperature data was converted to Kelvin (TK) to avoid ambiguous effects for temperatures around 0 °C. The analysis was carried out in the period April 1 to October 31, because during this period there was no Croat. j. for. eng. 35(2014)1

G. Erber et al.

snow at Finnish site. Snow cover on the pile affects the weight, causing confusion in the determination of the moisture content of the material. The Austrian dataset was developed in 2010, and the Finnish in 2012.

2.4 Modeling approaches The Austrian approach can be considered as a »cumulated sum approach«. The cumulative alteration in moisture content is calculated by a multiple linear regression using cumulative sums of daily means of air temperature, wind speed, relative humidity and daily sums of precipitation. Therefore, the model delivers moisture content alteration over a period in daily steps. In contrast, the Finnish model only uses the daily sum of net evaporation as input variable in its linear regression model. Daily alteration in moisture content is the dependent variable. These values can be cumulated afterwards for a specified period of time. The recorded dataset was split into half using random numbers. One half was assigned to be the analysis set, whereas the other was assigned to be the testing set for the developed model.

2.5 Model comparison and cross validation The main criteria for the accurateness of the models were their mean deviation, respective standard deviation and median deviation from the observed curve. A deviation of moisture content of up to ± 5% was considered a reasonable model. The models were applied to the full dataset of the other trial to investigate the validity for other regions.

2.6 Valid range The valid range of the models for any further use depends on the data the models were developed from. The limits applied were the 10% and 90% quantile for all variables. The models shall only be used for small diameter (10 cm to 20 cm) and 4 m to 5 m long pine logwood during the period April to October. Respective limits for daily averages and sums of the input variables are given in Table 2.

3. Results 3.1 Austrian approach »Cumulative sume« approach proved to work well for both datasets. Mean deviations of moisture content from the observed curve were 0.07% ± 0.49% (Austrian data) and –0.02% ± 0.46% (Finnish data), respectively. The median deviations were both 0.10%. Residual standard errors were 0.51% and 0.44%. Both R² ad-

G. Erber et al.

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

Table 2 Limits for the valid range models based on Austrian and Finnish data: 10% and 90% quantile on daily basis Basis

TC, °C

RH, %

P, mm WS, m s-1 Net, mm

10% Austrian based

7.02

34.45

0.00

0.26

90% Austrian based

21.83

89.84

9.03

10% Finnish based

0.43

57.00

90% Finnish based

16.90

95.25

Coefficients

Estimate

Std. Error

t value

p value

–5.73

Intercept

0.062

0.013

4.931

<0.001

0.79

4.08

net

0.020

0.002

10.137

<0.001

0.00

1.38

–4.14

8.44

4.34

4.67

Table 3 Parameters estimate, Student’s t-test and summarized test statistics for the Austrian data based cumulative sum model Coefficients Intercept WS

Estimate

Std. Error

t value

p value

Intercept

0.039

0.013

2.86

4.648 x 10-3

net

0.062

0.003

19.42

<0.001

11.40

< 0.001

–5.452 x 10–2 1.977 x 10–2

–2.76

0.0063

18.52

< 0.001

–14.10

< 0.001

–1.04% ± 1.43% (Austrian data) and 0.52% ± 0.87% (Finnish data), respectively. The median deviations were 0.32% and –0.03%. (Fig. 2).

3.23

0.0014

3.3 Cross validating

1.332

1.168 x 10

–3

–4

2.821 x 10

–1.283 x 10–3 9.101 x 10–5 –3

4.471 x 10

–3

1.381 x 10

Residual standard error: 0.507 on 209 degrees of freedom Multiple R squared: 0.991, Adjusted R squared: 0.991 F statistic: 5 788 on 4 and 209 DF, p value: < 0.001

Table 4 Parameters estimate, Student’s t test and summarized test statistics for the Finnish data based cumulative sum model Estimate

Std. Error –2

1.130

9.576 x 10

t value

p value

11.80

< 0.001

6.851 x 10–3 4.262 x 10–3

1.61

0.110

8.585 x 10–3 1.177 x 10–4

72.93

< 0.001

–44.82

< 0.001

12.54

< 0.001

–3

–5

6.561 x 10

–2.940 x 10

1.896 x 10–2 1.512 x 10–3

Residual standard error: 0.444 on 209 degrees of freedom Multiple R squared: 0.994, Adjusted R squared: 0.994 F statistic: 8 962 on 4 and 209 DF, p value: < 0.001

justed were 0.99. Contrary to the Austrian data based model, wind speed was not found significant in the Finnish data based model (Tables 3 and 4; Fig. 2).

3.2 Finnish approach The net evaporation approach, given in Table 5 and 6, provided a satisfactory outcome, but was not as accurate as the Austrian approach. Mean deviations of moisture content from the observed curve were

Estimate

p value

Intercept

Coefficients

t value

5.224 x 10

Coefficients

Table 6 Parameters estimate, Student’s t test and summarized test statistics for the Finnish data based net evaporation model

–1

Table 5 Parameters estimate, Student’s t test and summarized test statistics for the Austrian data based net evaporation model

Applying the derived curves to each other’s dataset did work well for three of the four modeled curves. For the use of the Austrian approach model, derived from Finnish data, on the Austrian data, a mean deviation of moisture content was –24.24% ± 13.09%. The Austrian approach model, derived from Austrian data, underestimated the drying performance of the Finnish pile (mean deviation of 1.27% ± 1.02%). The models based on net evaporation performed well – the Austrian data-based model underestimated the drying performance of the Finnish pile (mean deviation 2.32% ± 1.63%), and the Finnish data-based model varied strongly in its spring and summer estimates (mean deviation –0.68% ± 1.97%) (Fig. 3).

4. Discussion Similar to other studies (Stokes et al. 1987, Liang et al. 1996, Murphy et al. 2012, Filbakk et al. 2011, Erber et al. 2012, Dong-Wook and Murphy 2013), depending on the modeling approach, climatic conditions such as wind speed, precipitation, relative humidity and air temperature or precipitation and evapotranspiration were found to govern the drying process. Three of the developed models are plausible. The model developed from Finnish data using the Austrian approach assigns a – illogical – positive prefix to wind speed. Cross validation clearly showed that the use of these coefficients produces an absolutely unacceptable output. Therefore, it has to be concluded that this model is totally wrong Croat. j. for. eng. 35(2014)1

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

G. Erber et al.

Fig. 2 Above: observed (solid lines) and modeled moisture content decrease curves (dashed and dashed dotted lines) for Austrian and Finnish data. Below: respective deviations from the observed curves with a target deviation of ± 5% (solid lines)

and should not be used by any means. Wind speed and air temperature govern drying, whereas precipitation and relative humidity govern rewetting. Some details have to be considered when trying to evaluate these models. The Finnish data had three different origins – the meteorological station at Mekrijärvi research station, the weather radar for precipitation and a grid based model for solar radiation. The Austrian Croat. j. for. eng. 35(2014)1

data originated almost exclusively from the drying site. Only solar radiation was obtained from a nearby staterun meteorological station. Longwave and shortwave radiation for the Penman-Monteith equation were derived from this measurement. This is probably the explanation for a greater variation of deviation of the Austrian data based model around the observed curve. The Finnish data based model fit its pile better.

G. Erber et al.

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15â&#x20AC;&#x201C;22)

Fig. 3 Above: observed (solid lines) and modeled moisture content decrease curves (dashed lines) for Finnish data applying Austrian databased models. Below: observed (solid lines) and modeled moisture content decrease curves (dashed-dotted lines) for Austrian data applying Finnish data-based models

Similarity in experimental design can be a point of concern. Though both experiments used metal frames on load cells, the experimental design differed in some details. At the Finnish site, a paper cover was used at the sides and the bottom to simulate natural drying conditions. Such a cover was not used at the Austrian site. The limitations in wind exposure were probably

the reason why the wind speed was not insignificant in the Austrian approach and Finnish data based model. The use of similar material is crucial, too. Though the average diameter and length are alike, different wood densities could affect the drying performance. No analysis of wood density was carried out at the Finnish site. Hence the comparison was not possible. Croat. j. for. eng. 35(2014)1

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

Applying and trying to cross validate the models on each other’s dataset proved to be satisfactory. Only one of the four models did not reasonably fit. The different valid range (Table 2) of the models can be a major factor here. The Austrian approach model based on Austrian data works well in the Finnish dataset because it covers all the range of temperature and precipitation and almost all of relative humidity range. Only Finnish wind speed data is completely out of range. The Finnish data based model does not cover all of the Austrian temperature and relative humidity range. Especially low relative humidity and high temperature conditions – best fitted for drying – are not covered by this model. Finally, wind speed is completely out of range. Furthermore, wind speed was not significant within the Finnish model. When looking at the net evaporation approach models, a wider range of the Austrian data based model can be observed. However, satisfying accuracy was achieved for both models. The Finnish data based model showed weaknesses in prediction during spring and autumn. Approximated longwave and shortwave radiation in the Austrian data can be considered the reason for the less accurate prediction.

5. Conclusions It proved to be possible to fit accurate drying models within the target deviation of ± 5% using both approaches. Whereas the Austrian approach is focused on a more basic set of variables, the Finnish approach combines the variables within one. Both approaches are justified – depending on the available data. When applying drying models to other regions of the earth, accuracy of prediction can be affected by material and storing properties. Models can only be valid for the conditions under which they were developed. This study showed that, if variable values are out of the model range, a reasonable prediction is not possible. In order to ensure reasonable results, the conditions under which the models have been developed, have to be specified in terms of climatic conditions, storing technique and material type.

Acknowledgments The authors would like to thank University of Eastern Finland, Mekrijärvi Research Station. This work has been partly funded by LAAVA-project and INFRES-project (European Union Seventh Framework Programme (FP7/2012-2015) under grant agreement n°311881). Croat. j. for. eng. 35(2014)1

G. Erber et al.

6. References Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998: Crop evapotranspiration – Guidelines for computing crop water requirements. Irrigation and drainage paper no. 56. FAO, Rome: 1–15. Dong-Wook, K., Murphy, G., 2013: Forecasting air-drying rates of small Douglas-fir and hybrid poplar stacked logs in Oregon, USA. International Journal of Forest Engineering 24(2): 137–141. Erber, G., Kanzian, C., Stampfer, K., 2012: Predicting moisture content in a pine logwood pile for energy purposes. Silva Fennica 46(4): 555–567. Filbakk, T., Hoibo, O., Nurmi, J., 2011: Modelling natural drying efficiency in covered and uncovered piles of whole broadleaf trees for energy use. Biomass and Bioenergy 35(1): 454–463. Gigler, J.K., van Loon, W.K.P., van den Berg, J.V., Sonneveld, C., Meerdink, G., 2000: Natural wind drying of willow stems. Biomass and Bioenergy 19(3): 153–163. Golser, M., Pichler, W., Hader, F., 2005: Energieholztrocknung. Holzforschung, Wien, Österreich, 138 p. Hartmann, H., Kaltschmitt, M., 2001: Energie aus Biomasse: Grundlagen, Techniken und Verfahren. Springer Verlag. Berlin–Heidelberg–New York. 770 p. Kofman, P.D., Kent, T., 2009: Long term storage and seasoning of conifer energy wood. CoforD connects – Harvesting/ Transportation 20: 1–4. Kröll, K., 1978: Trocknungstechnik: Trockner und Trocknungsverfahren. Springer Verlag. Berlin–Heidelberg–New York. 654 p. Liang, T., Khan, M.A., Meng, Q., 1996: Spatial and temporal effects in drying biomass for energy. Biomass and Bioenergy 10 (5/6): 353–360. Murphy, G., Kent, T., Kofman, P.D., 2012: Modeling air drying of Sitka spruce (Picea sitchensis) biomass in off-forest storage yards in Ireland. Forest Products Journal 62(6): 443–449. Nurmi, J., 1995: The effect of whole tree storage on the fuelwood properties of short rotation Salix crops. Biomass and Bioenergy 8(4): 245–249. Nurmi, J., Hillebrand, K., 2007: The characteristics of wholetree fuel stocks from silvicultural cleanings and thinnings. Biomass and Bioenergy 31(6): 381–392. Pettersson, M., Nordfjell, T., 2007: Fuel quality changes during seasonal storage of compacted logging residues and young trees. Biomass and Bioenergy 31(11–12): 782–792. Röser, D., Erkkilä, A., Mola-Yudego, B., Sikanen, L., Prinz, R., Heikkinen, A., Kaipainen, H., Oravainen, H., Hillebrand, K., Emer, B., Väätäinen, K., 2010: Natural drying methods to promote fuel quality enhancement of small energywood stems. Working Papers of the Finnish Forest Research Institute 186: 1–60.

G. Erber et al.

Comparing Two Different Approaches in Modeling Small Diameter Energy Wood ... (15–22)

Stokes, B.J., Watson, W.F., Miller, D.E., 1987: Transpirational drying of energywood. ASAE paper No. 87-1530. St. Joseph, MI, USA. American Society of Agricultural Engineers, 13 p. Stokes, B.J., McDonald, T.P., Kelley, T., 1993: Transpirational drying and costs for transporting wood biomass: a prelimi-

nary review. Proceedings of IEA/BA Task IX, Activity 6: Transport and Handling. Aberdeen University, UK: 76–91. Suadicani, K., Gamborg, C., 1999: Fuel quality of whole-tree chips from freshly felled and summer dried Norway spruce on a poor sandy soil and a rich loamy soil. Biomass and Bioenergy 17(3): 199–208.

Authors’ address: Gernot Erber, MSc.* e-mail: gernot.erber@boku.ac.at Christian Kanzian, MSc. e-mail: christian.kanzian@boku.ac.at Prof. Karl Stampfer, PhD. e-mail: karl.stampfer@boku.ac.at University of Natural Resources and Life Sciences Peter Jordan Strasse 82 1190 Wien AUSTRIA Johanna Routa, PhD. e-mail: johanna.routa@metla.fi Prof. Lauri Sikanen, PhD. e-mail: lauri.sikanen@metla.fi Finnish Forest Reseach Institute Yliopistokatu 6 80101 Joensuu FINLAND Marja Kolström, PhD. e-mail: marja.kolstrom@uef.fi University of Eastern Finland Mekrijärvi Research Station Yliopistontie 4 82900 Ilomantsi FINLAND Received: June 17, 2013 Accepted: February 14, 2014

* Corresponding author Croat. j. for. eng. 35(2014)1

Original scientific paper

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets Simo Paukkunen Abstract This article presents the opportunities to use small size round wood as a raw material for wood pellets. Article is aimed to give an insight to results of the study and to initiate discussion related on topical questions among pellet industry: what is the quality of pellets produced of undebarked young thinning wood (pine and pine-birch mixture) or debarked young thinning wood. Other topics of the study are to find out: has felling timing or growing habitat any influence to the chemical composition of pellets. Fuel quality indicates that high quality pellets can be produced of alternative raw materials. Key findings of this study are that there is a big opportunity to use undebarked small diameter pine and undebarked small diameter pine pinebirch mixture as a raw material for wood pellets. According chemical analysis small diameter thinning wood with bark is useful raw material for EN 14961-2 (2011) A1 wood pellets. In Finland traditionally small diameter wood has been used as a raw material for wood chips and for pulp and paper industry. Due to the changes in of pulp and paper industry new production opportunities for using small diameter wood should be found. In the future using small size thinning wood as a raw material for pellets can boost the demand of thinning wood and so help to manage young forests in Finland. Keywords: Wood pellets, thinning, bioenergy, combustion, biofuel

1. Introduction In Finland chemical and mechanical forest industry is under big changes and that will have a strong influence to the markets of thinning wood. In Finland it is very typical to manage forests using thinning harvesting. Thinning harvesting means that extra trees are taken out from forest and only those trees which has opportunities to grow to stems are left to forest. In practice that means that after first thinning (done when trees are about 30–40 years old) only 900–1 100 of trees/ha are left to grow and after second thinning (done when trees are 50–60 years old) only 500–600 trees/ha are left to forest. Thinning harvesting technique will secure more space, nutrients, water and sunlight to those trees which have the best opportunities to grow for best quality stems valuable for sawing industry. Using thinning harvesting forest owner can get profit when selling thinning wood to pulp industry (Kellomäki 1991, Metsätalouden kehittämiskeskus Tapio 2006, Metsäntutkimuslaitos 2010). Big international pulp and paper companies are moving from Croat. j. for. eng. 35(2014)1

Finland to countries where economic factors of making pulp and paper are better. In the future situation could be that there will be no strong demand for thinning wood in Finland and lower demand for thinning harvesting will have an influence to the quality and number of mature trees for sawing industry. There is a real fear that in the future there will be lower demand of young thinning wood in Finland than it has been recently years. One opportunity is to use young thinning wood as raw material for wood pellets. Other thing is the sufficiency and price of the traditional raw material (dry shavings from planning mills and wet saw dust from saw mills) for wood pellets. Availability of an important fuel cannot be connected only with strongly fluctuating saw milling industry. Pelletizing woody raw material is known quite well (Kytö and Äijälä 1981a, Kytö and Äijälä 1981b, Kallio and Kallio 2004, Obenberger and Thek 2010, Kallio 2011). General opinion is that round wood with bark is not good raw material for EN 14961-2 (2011)

S. Paukkunen

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33)

2. Materials and methods

A1 pellets because of fairly high amount of ash and low melting point of ash (Alakangas 2000, Lehtikangas 2001, Obenberger and Thek 2010). Different growing sites have different chemical characteristics and these might have influence to the chemical characteristics of the trees. The use of wood from first thinning (debarked round wood and round wood with bark) as a raw material for wood pellets has been interesting issue. Most challenging factors have been ash behavior (e.g. low ash melting temperature and amount of ash) and economical profitability. Other interesting question is possibilities to use hard wood as a raw material for wood pellets (Kytö and Äijälä 1981a, Kytö and Äijälä 1981b, Okkonen et al. 2009, Obenberger and Thek 2010, Paukkunen et al. 2010). The aim of this study is to find out: Þ If it is possible to make EN 14961-2 (2011) A1 pellets using young pine with bark as a raw material; Þ If there are opportunities to use birch (Betula pendula Roth) as a raw material for wood pellets; Þ The influence of bark when using young Scots pine (Pinus sylvestris L.) as a raw material for wood pellets? Þ If harvesting dates any influence to the chemical composition of wood pellets; Þ If growing site has any influence to the chemical composition of wood pellets.

The starting point of the study was to find two forest plots which needs first thinning harvesting. Also growth habitat of forests should be different. Stora Enso Ltd did find four forest plots for the purpose of study. Forest plots were selected and invented 30.5.2011– 10.6.2011 and plots were harvested 13.6.2011–15.6.2011 and 5.9.2011. Forests are situated in Easter-Finland: Þ Area VT: N = 62°34.126', E = 31°6.944' Þ Area MT: N = 62°34.654', E = 31°8.67' Both plots were inventored using line method. There were two lines 50 m from each other and both lines contained 5 basal area sample areas. Distance between sample areas was 50 m. At the end from both forests 10 basal area sample areas were inventored. Forest stand information before thinning is presented in table 1. Number of stems series from basal area sample plots were calculated using formula (1.27324*q)/d2, were q = basal area gauge factor and d = diameter (measured at 1.3 m height) of chosen wood in meters (Kangas and Päivinen 2000). Every harvested tree were measured and inventory data of harvested trees is presented in tables 2 and 3. Every harvested tree were used for raw material of pellets.

Table 1 Forest stand inventory data. Data is collected before harvesting Area VT Median trees

Plot

Avg.

Max.

Min.

Diameter, cm

16.5

12.5

15.5

14.5

15.15

12.5

Height, m

15.5

13.5

12.5

13.5

14.5

12.5

11.5

13.35

15.5

11.5

Age, year

32.4

2 300

1 411

1 717

664

1 011

3 108

1 578

1 749

2 094

1 048

1 668

3 108

664

Number of stems per ha

Area MT Median trees

Plot

Avg.

Max.

Min.

17.5

14.5

12.5

15.55

17.5

12.5

Height, m

15.5

12.5

14.5

11.5

14.5

13.5

14.5

15.5

13.85

15.5

11.5

Age, year

28.4

2 050

1 174

1 166

1 007

1 488

1 236

1 098

1 406

1 153

1 376

1 315

2 050

1 007

Diameter, cm

Number of stems per ha

Croat. j. for. eng. 35(2014)1

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23â&#x20AC;&#x201C;33)

S. Paukkunen

Table 2 Inventory data on harvested trees, VT Area VT, summer pine With bark n=36

Area VT, summer pine Without bark n=49

Area VT, autumn pine With bark n=10

Area VT, autumn pine Without bark n=11

12.1

12.4

13.0

22.5

18.5

19.5

Diameter, min.

7.5

9.5

Thicknes of bark, avg.

4.3

N/A

5.4

5.5

N/A

13.5

14.8

14.5

16.2

Height, min.

8.5

9.9

Age, avg.

N/A

0.084

0.0928

0.0977

0.0958

0.1889

0.3128

0.2112

0.2473

0.0234

0.015

0.0424

0.472

Area MT, summer pine with bark n=35

Area MT, summer pine without bark n=48

Area MT, summer birch with bark n=15

Area MT, summer birch without bark. n=12

12.3

13.1

11.4

10.7

19.5

25.5

Diameter, min.

8.5

7.5

5.5

Thicknes of bark, avg.

4.6

N/A

13.1

13.3

N/A

16.5

16.0

N/A

Height, min.

8.5

8.4

N/A

Age, avg.

N/A

0.0872

0.1028

N/A

0.2780

0.2343

N/A

0.0155

0.0131

N/A

Unit Diameter, avg. Diameter, max.

Thicknes of bark, max.

Thicknes of bark, min. Height, avg. Height, max.

Age, max.

year

Age, min. Volume, avg. Volume, max.

Volume, min.

Table 3 Inventory data on harvested trees, MT Unit Diameter, avg. Diameter, max.

Thicknes of bark, max.

Thicknes of bark, min. Height, avg. Height, max.

Age, max.

year

Age, min. Volume, avg. Volume, max. Volume, min.

Croat. j. for. eng. 35(2014)1

S. Paukkunen

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33)

Heights of trees (pine with bark) were measured using hypsometer (Suunto PM5/1520). Height of trees (pine without bark, both cases) were calculated using Näslund`s formula (Näslund 1936, Kangas and Päivinen 2000). Näslund`s formula for calculating height of tree: h=

1.3 + d 2 ( a + b * d)2

1974, Mälkönen 1979). It is normal that MT growing habitat has higher pH value and mineral amounts are higher than in VT growing habitat (Urvas and Erviö 1974, Mälkönen 1979). At June 2011 harvesting was operated using two machines. Wood without bark was harvested using harvester which was equipped with barking felling head and extra attention was put to avoid the contamination of harvested wood. Wood with bark was harvested using Klapimaster, which is very new model of harvester. Klapimaster is made by HaVel Ltd. Klapimaster is harvesting trees straight to the container as chopped wood (HaVel 2012). This method avoids the risk of contamination. Chopped wood was packed straight from the container to big sacks waiting for the transportation. Harvested and chopped wood was transported to Mekrijärvi research center 15.6–16.6.2011 using medium size lorry car. In September harvesting was done manually using motor saw. Stems were carried near the road, transported to Mekrijärvi using tractor and forest trailer. Stems were secured from contamination (sand and dust from road) using covering plastic, and then debarked using debarking tool. Chopped wood was chipped using disc chipper driven by a tractor to the dryer trailer. Wood chips were dried using trailer based batch dryer. Loaded trailers (volume 3 loose cubic meter) were situated to sea container and heated air (temperatures max 65°C) was blew through wood chips batch. Drying from 55% moisture content to 10% moisture content (on wet basis) took about 64 hours per each load. Extra heat derived from district heating using heat exchanger to avoid contamination during drying (Öhman et al. 2004).

(1)

where: h height of tree, meters; d diameter (in centimeters) of tree measured from 1.3 meters height; a, b variables, calculated from measured trees using linear regression method. Statistics for calculation were taken from measured trees (pine with bark). Volumes of pines (every cases) were calculated using Laasasenaho`s formula number 2 (Laasasenaho 1982, Kangas and Päivinen 2000). Ground samples from growing sites were collected 15.9.2011 and samples were sent for analysis 16.9.2011 to Viljavuuspalvelu LTD, Mikkeli, Finland. Samples were collected from 10 different places from both forests. Humus layer and mineral soil layer were separated from samples. From collected samples mixed samples were taken resulting in 4 samples which were sent to Viljavuuspalvelu LTD: Humus and mineral samples from Area VT and humus and mineral samples from Area MT. Information from growing site is presented in table 4. Soil mineral and pH consistence of growth habitats were typical compare to literature (Urvas and Erviö

Table 4 Information from forest plots Humus layer Growth habitat

Ash

Grade of organic matter

Soil type

g/kg

mg/l

Cd mg/kg

Low

Sandy till

5.2

1.66

10.3

11.5

10.2

21.9

4.3

41.3

9 740

1.4

Low

Fine sandy till

5.4

2.15

8.2

12.6

12.9

31.5

5.6

27.8

7 330

1.7

N/A

Mineral soil layer Grade of organic matter

Growth habitat

Soil type

mg/l

–

Low

Sandy till

5.2

2.3

130

<20

N/A

–

Low

Sandy till

5.4

3.1

180

N/A

Croat. j. for. eng. 35(2014)1

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33)

Dried raw material was milled to suitable particle size using Miller 20 hammer mill equipped with 6 mm sieve was used. Raw material was pelletized with SPC PP300 pelletizer equipped with 50 mm long press channels die. For pelletizing any extra binding materials, water or steam were not used. Standards which has been followed in this study: Þ Determination of mechanical durability of pellets and briquettes EN 15210-1(2010); Þ Moisture content of raw material and pellets: ENEMEN 002 (Savonia University of Applied Sciences), EN 14774-1(2010), 14774-2(2010) and 14774-3(2010); Þ Sampling: EN 14778-1(2005), EN 14778-2(2005); Þ Calorific values EN 14918(2010); Þ Ash content SFS-EN 14775.

For pine the share of bark (dry basis) should be about 10.5% (Hakkila et. al. 1995, Alakangas 2000,) so in this study share of bark was little smaller than average. Chemical analysis was done by University of Eastern Finland, Joensuu and Savonia University of Applied Sciences, Varkaus. Chemical analysis was done in University Of Eastern Finland, Joensuu using Inductive coupled plasma mass spectrometry (ICP). Results of the chemical analysis (ICP) are presented in table 6. Amounts of Ca, Mg, Fe, Zn and Si are similar that earlier reported in literature (Alakangas 2000, Sippula 2007, Okkonen et al. 2009), but there are some differences e.g. in amounts of P and S witch were 10 times lower in this study (Lehtikangas 2001, Sippula et al. 2007). Also amount of K was about 0.5 times lower compared to the average level. The different amounts of minerals in wood are presented in literature (Obenberger and Thek 2010, Fillbakk et al. 2011). It is interesting to notice that the amount of Na is higher in the pellets which were made from autumn harvested trees. According the literature normal amount of Na should be about 0.02–0.08 g/kg (Sippula 2007, Okkonen et al 2009). After all, results from this study are quite similar to former study (PELLETime-project) concerning round wood with bark. In former study pellets were made from undebarked pine and undebarked pine-birch mixture. Results from the fuel analyses indicate that both pellet assortments have low ash contents (0.3%, analyzed as ignition residue at 550°C) (Okkonen et al. 2009, Paukkunen et al. 2010).

3. Results and discussion In table 5 are presented shares of bark from undebarked raw materials. Table 5 Shares of bark from undebarked raw materials Bark %, dry basis Area VT, pine with bark

8.6

Area VT, pine with bark autumn

8.2

Area MT, pine with bark

12.6

Area MT, birch with bark

18.9

S. Paukkunen

Table 6 Results of chemical analysis of pellets Results (dry basis) of chemical analysis Ash

Kjeldahl

Area V, pine without bark

0.26

0.071

10.3

1.65

546

0.804

31.4

321

Area VT, pine with bark

0.35

0.087

27.3

1.34

750

0.782

25.0

Area MT, pine without bark

0.25

0.075

6.8

1.03

556

0.768

Area MT, pine with bark

0.31

0.087

32.2

1.35

763

Area MT, pine without bark 45%, birch without bark 55%

0.27

0.085

1.8

1.18

Area MT, pine with bark 66%, birch with bark 33%

0.33

0.108

22.5

Area VT, pine without bark. autumn 0.26

0.055

Area VT, pine with bark. autumn

0.082

Croat. j. for. eng. 35(2014)1

0.32

123.6

13.6

34.3

12.0

8.3

396

145.4

13.9

57.5

7.7

11.0

15.8

368

137.4

14.0

44.4

4.6

7.5

0.789

17.3

474

147.3

13.6

71.3

5.8

8.7

591

0.791

16.2

377

146.0

15.6

60.8

4.9

16.9

1.40

772

0.826

17.0

476

146.3

19.0

74.6

5.2

15.0

<LQD

1.16

531

0.857

15.1

319

122.8

60.0

37.5

7.3

5.2

56.3

1.49

799

0.993

12.8

412

139.4

106

54.2

60.8

9.3

7.0

mg/kg

S. Paukkunen

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33)

The elements Ca, Mg, Si and K are the main ash forming elements in wood. The concentrations of Ca, Mg, K and Na in the ash influence the melting behavior, which is directly related to the reliability of the plant. Ca and Mg usually raise the ash melting point whereas K and Na lower it. A low ash melting point can lead to slagging and deposit formation in furnace and boiler. Si also influences the ash melting behavior main as low melting K silicates may be formed. A high concentration of K prompts the formation of aerosols during combustion, which not only raises the emission of fine particulate matter but also increases fouling of the boiler. Na behaves in a very similar way as K (Obenberger and Thek 2010). There are other studies where ash contents of hard woods are detected to be higher than CEN standard requirements (e.g. Oben-

berger and Thek 2010). Results from this study are slightly different. Assumption is there might be possibility to use hard wood as a raw material to CEN standard A1 pellets at least mixture to soft wood. Normal stem wood without any contamination should contain very limited amounts of Si and thus, it is less prone to form sticky silicates at typical temperatures in residential pellet burners (1000–1100°C), whereas contamination leads to the formation of sticky silicates and slagging problems (Öhman et al. 2004). After taking samples for analysis remained pellets were given for heating purposes. Pellets were used as a fuel to heat domestic house where gasifying burner (type name Pyro-Man) was used. Tables 7 and 8 presents characteristics of chipped raw material and made pellets.

Table 7 Characteristics of chipped raw material and produced pellets Moisture of chips, wet basis

Moisture of dried and milled raw material, wet basis

Durability of pellets

kg/ loose m3

% Area VT, pine without bark

Bulk density of pellets

55.2

9.2

95.2

565

14.7

91.8

540

Area MT, pine without bark

56.3

10.5

90.8

592

Area MT, pine with bark

59.1

9.8

93.8

619

Area MT, pine without bark 45%, birch without bark 55%

n.n

10.9

95.2

671

Area MT, pine with bark 66%, birch with bark 33%

n.n

8.7

96.1

736

Area VT, pine without bark, autumn

11.9

95.7

593

54.8

12.0

94.3

617

Area VT, pine with bark

Area VT, pine with bark, autumn

Table 8 Characteristics of chipped raw material and made pellets High heating value of Low heating value of LHV NCT of pellets pellets – HHV pellets – LHV as reseaved

Moisture of pellets

MJ/kg

Wet basis

Ash content of pellets, dry basis

Area VT, pine without bark

20.83

19.5

17.46

9.3

0.2

Area VT, pine with bark

20.8

19.47

17.21

10.3

0.3

Area MT, pine without bark

20.66

19.33

17.45

8.7

0.2

Area MT, pine with bark

20.64

19.31

17.66

7.6

0.3

Area MT, pine without bark 45%, birch without bark 55%

20.45

19.12

17.17

9.1

0.2

Area MT, pine with bark 66%, birch with bark 33%

20.6

19.27

17.63

7.6

0.3

Area VT, pine without bark, autumn

20.92

19.59

17.5

9.5

0.4

Area VT, pine with bark, autumn

21.16

19.83

17.56

10.2

0.6

Croat. j. for. eng. 35(2014)1

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23â&#x20AC;&#x201C;33)

S. Paukkunen

Fig. 1 Amount of ash %, mechanical durability, and ratios of K/Ca, K/ Si and CA/Si, measured from made pellets. Cases at the horizontal axis are: 1 = Area VT, pine without bark; 2 = Area VT, pine with bark; 3 = Area MT, pine without bark; 4 = Area MT, pine with bark; 5 = Area MT, pine without bark 45%, birch without bark 55%; 6 = Area MT, pine with bark 66%, birch with bark 33%; 7 = Area VT, pine without bark, autumn; 8 = Area VT, pine with bark, autumn. Ratio K/ Ca is higher when raw material has not bark and amount of ash is higher when raw material had bark Croat. j. for. eng. 35(2014)1

S. Paukkunen

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23â&#x20AC;&#x201C;33)

Fig. 2 Amounts of Al, Ca, Cu, K and Mg, measured from produced pellets. Amounts of Al, Ca and K are higher when raw material has bark

Croat. j. for. eng. 35(2014)1

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23â&#x20AC;&#x201C;33)

S. Paukkunen

Fig. 3 Amounts of Na, P, S, N and Zn, measured from produced pellets. Amount of Na is higher when trees are harvested in autumn. Amounts of P, S and N are higher when raw material has bark Croat. j. for. eng. 35(2014)1

S. Paukkunen

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33)

In this study durability of pellets did not reach the demands of quality of EN 14961-2 A1 pellets. There are many factors which influence the durability of pellets (e.g. particle size of raw material, amount of bark, moisture of raw material, used pelletizing technology and settings, time between harvesting and drying) (Lehtikangas 2001, Kallio and Kallio 2004, Obenberger and Thek 2010, Paukkunen et al. 2010). Durability demands of CEN 14961-2 A1 should be reached when using longer press tunnel (e.g. 55 or 60 mm) and longer aging of the raw material. Using hot steam as a pretreatment method just before pelletizing might also increase the durability of pellets (Kytö and Äijälä 1981a, Kytö and Äijälä 1981b, Obenberger and Thek 2010). Significance of growth habitat, presence of bark and time of harvesting were tested using one way ANOVA. Growth habitat has significance to the amount of N (p = 0.042), K (p = 0.03), Mg (p = 0.018), Mn (p = 0.007), Na (p = 0.017), P (p = 0.032), Si (p = 0.000), Zn (p = 0.034) and to ratio of K/Si (p = 0.000), Ca/Si (p = 0.000) and to ratio of K/Ca (p = 0.000). Presence of bark has significance to the amount of ash (higher, p = 0.000), N (higher, p = 0.004), Al (higher, p = 0.001), Ca (higher, p = 0.000), K (higher, p = 0.000), Mg (higher, p = 0.011), P (higher, p = 0.001), S (higher, p = 0.000) and to ratio of K/Ca (lower, p = 0.035). Harvesting time has significance to the amount of N (lower in autumn, p = 0.049), Al (higher in autumn, p = 0.001), Cu (higher in autumn, p = 0.000), Na (higher in autumn, p = 0.000), and Zn (lower in autumn, p = 0.021). Tree species has significance to the amount of N (higher in mixed raw material, p = 0.015), P (higher in mixed raw material, p = 0.043), Zn (higher in mixed raw material, p = 0.000), to the ratio of K/Si (higher in mixed raw material, p = 0.016) and Ca/Si (higher in mixed raw material, p = 0.021).

material for wood pellets in Finland and using birch could greatly wide the useful raw material base. From the chemical point of view there are no limits to use pine-birch mix pellets as a fuel to the small scale pellet heating units (e.g. households). In this study mechanical durability did not exceed the value 97.5% or more so making durable pellets from pine – birch mix needs more research if the goal is to make EN 14961-2 A1 pellets. The influence of growth habitat and harvesting time need also more research. In this study there were no big differences in chemical composition of mineral and humus layer between two growing habitats. In the future it might be interesting to compare chemical composition of pellets when raw material is harvested from very different growth habitats. Harvesting time might have an influence to the amount of Na and that has significance to the combustion process. It can be so that late harvesting time (autumn and winter) increases the amount of Na in wood. Increased amount on Na might decrease the ash melting temperature so summer might be the better harvesting time when the purpose is to collect raw material for EN 14961-2 A1 pellets. In Finland traditionally small diameter wood (energy and thinning wood) has been used as a raw material for wood chips and for pulp and paper industry. Small diameter wood has been harvested because of demand of raw material and also because of the need of forest management (e.g. stems will grow faster to logs). Forest owner can get quicker more profit from the forest which has been managed using thinning method harvesting. In the future using small scale thinning wood as a raw material for pellets can boost the demand of thinning wood and so help to manage young forests in Finland.

4. Conclusions

This study was done with the economical and mentally support from William and Ester Otsakorpi Foundation (www.otsakorpi.fi) and Stora Enso LTD, which donated the woody raw material.

It seems that small diameter pine can be used both without bark and with bark as a raw material for EN 14961-2 A1 pellets when only the chemical composition of pellets is taken into account. Drying techniques, costs from drying the raw material and the price of the raw material are the key factors when the economics of pellet industry is evaluated (e.g Flykman 2001, Zakrisson et al. 2002, Obenberger and Thek 2010). It is also sure that contamination of pellet raw material should be avoided in every way. This study points out that pellet market has big opportunity to use birch as a raw material for wood pellets. Broad-leaved trees are very poorly used as raw

Acknowledgements

5. References Fillbakk, T., Jirjis, R., Nurmi, J., Höibö, O., 2011: The effect of bark content on quality parameters of Scots pine (Pinus sylvestris L.) pellets. Biomass and Bioenergy 35(8): 3342–3349. Flyktman, M., 2001: Pellettien kuivauskustannukset eri laitoskytkennöillä. VTT Energia. 18 p. Hakkila, P., Kalaja, H., 1983: Puu- ja kuorituhkan palauttamisen tekniikka. Folia Forestalia 552. 37 p. Croat. j. for. eng. 35(2014)1

Opportunities to Use Thinning Wood as Raw Material for Wood Pellets (23–33) Hakkila, P., Kalaja, H., Saranpää, P., 1995: Etelä-Suomen ensiharvennusmänniköt kuitu- ja energialähteenä Metsäntutkimuslaitoksen tiedonantoja 582. 100 p. Hakkila, P., 1989: Utilization of residual forest biomass. New York. Springer. 568 p. HaVel 2012: http://www.havel.fi/index.php?470 Kallio, M., Kallio, E., 2004: Puumateriaalien pelletöinti. Projektiraportti pro2/p6012/04. VTT. 61 p. Kallio, M., 2011: Critical review on the pelletizing technology. IEE/09/758/SI2.558286 - MixBioPells. VTT. 54 p. Kangas, A., Päivinen, R., 2000: Metsän mittaus. Silva Carelica 35. Joensuun yliopisto. 205 p. Kellomäki, S., 1991: Metsänhoito. Silva Carelica 8. 501 p. Kytö, M., Äijälä, M., 1981a: Metsäenergian käyttö ja jalostus. Osa 3. Biomassan pelletöinnin laitetekniikka. Espoo. VTT. Tiedonanto 38. 46 p. Kytö, M., Äijälä, M., 1981b: Metsäenergian käyttö ja jalostus. Osa 4. Puun pelletöinnin kokeellinen tutkimus. Espoo. VTT. Tiedonanto 41. 45 p. Laasasenaho, J., 1982: Taper curve and volume functions for pine, spruce, and birch. Communicationes Instituti Forestalis Feanniae 108. 74 p. Lehtikangas, P., 2001: Quality of pelletised sawdust, logging residues and bark. Biomass and bioenergy 20(5): 351–360. Metsäntutkimuslaitos., 2010: Metsätilastollinen vuosikirja (Finnish Forest Institute, Finnish Statistical Yearbook of Forestry). Vammalan kirjapaino Oy. Sastamala 2010. 472 p. Metsätalouden kehittämiskeskus Tapio., 2006: Hyvän metsänhoidon suositukset. Metsäkustannus. 100 p. Mälkönen, E., 1979: Metsämaatieteen perusteita. Helsingin yliopisto. Tiedonantoja 19. 107 p. Nurmi, J., 1993: Heating values of the above ground biomass of small-sized trees (Pienikokoisten puiden maan-

S. Paukkunen

päällisen biomassan lämpöarvot). Acta Forestalia Fennica 236. 30 p. Nurmi, J., 2000: Characteristics and storage of whole-tree biomass for energy. Metsäntutkimuslaitoksen tiedonantoja 758. 42 p. Näslund, M., 1936: Skogsförsöksanstaltens gallringsförsök i tallskog. Meddelanden från Statens Skogsförsöksanstalt 29. 169 p. Okkonen, L., Paukkunen, S., Lamberg, H., Sippula, O., Tissari, J., Jokiniemi, J., 2009: PELLETime investigates alternative raw materials of pellet production. Bioenergy 2009 – Book of Proceeding: 755–759. Paukkunen, S., Sikanen, L., Vilppo, T., Lamberg, H., 2010: Energy pellets in the future – markets and raw materials. Forest Bioenergy 2010 – Book of Proceedings. 305–313. Sippula, O., Hytönen, K., Tissari, J., Raunemaa, T., Jokiniemi, J., 2007: Effect of wood fuel on the emissions from a top-feed pellet stove. Energy & Fuels 21: 1151–1160. Urvas, L., Erviö, R., 1974: Influence of the soil type and the chemical properties of soil on the determining of the forest type. Journal of the Scientific Agricultural Society of Finland. Vol 3: 307–319. Voipio, R., Laakso, T., 1992: Pienikokoisten puiden maanpäällisen biomassan kemiallinen koostumus (Chemical composition of the above ground biomass of small-sized trees ). Folia Forestalia 789. 22 p. Zakrisson, M., 2002: Internationell jämförelse av produktionskostnader vid pelletstillverkning. Swedish University of Agricultural Scioences, Department of Forest Management and Products. Examensarbeten nr 39. 66 p. Öhman, M., Nordin, A., Hedman, H., Jiris, R., 2004: Reasons for slagging during stemwood pellet combustion and measures for prevention. Biomass & Bionergy 27(6):597–205.

Author’s address:

Received: May 06, 2013 Accepted: December 31, 2013 Croat. j. for. eng. 35(2014)1

Simo Paukkunen e-mail: paukkune@student.uef.fi University of Eastern Finland School of Forest Sciences Yliopistokatu 7 80101 JOENSUU Karelia-University of Applied Sciences Centre for Bioeconomy Sirkkalantie 12 A 80100 JOENSUU FINLAND *Corresponding author

Original scientific paper

Use of Marketing Tools in the Slovakian Forest Biomass Trade Daniel Halaj, Yvonne Brodrechtova Abstracts The fast growing renewable energy market offers opportunities for the traditional forest sector both in Europe in general and Slovakia in particular. The reasoning behind this is twofold. First, in last decade the renewable energy business has gained significant attention. Among other reasons, this is due to the goal of a reduction of greenhouse gas emissions. Second, changes in downstream markets due to financial crises have put pressure on forest enterprises to redefine their sales portfolios. As marketing can help to realize new opportunities for forestry, the aim of this research study is to shed more light on how forestry enterprises in Slovakia use marketing tools in the trade of forest biomass, the main renewable energy source available to Slovakia. Due to its explorative nature, a case study research design with qualitative methodology has been applied. The capacity for use of marketing tools by (state and private) forest enterprises was explored through in-depth personal interviews analyzed with the help of content analysis. By presenting four detailed cases, the results show that in the forest biomass trade, the forest enterprises all used a marketing mix of »The 5Ps« (product, price, place, promotion, and people) and consequently applied a set of tools different from those observed in other industry sectors. Keywords: marketing tools, renewable energy, forest biomass, Slovakia, qualitative inquiry

1. Introduction The current national and international energy argument is mainly preoccupied with the renewable energy business. Especially, forest biomass1 is identified in many research studies as a most promising source for generating energy (e.g. Stupak et al. 2007, Stidham and Simon-Brown 2011, Schwarzbauer et al. 2013). Although the use of biomass has a long history, only in the last decade has it experienced revival as a more carbon neutral and local source of energy. In

Woody biomass from forests and/or tree plantations (FAO 2004). The forest biomass, which is part of harvested raw wood, is set for use in the energy sector because it is of no further use in the wood processing industry. Forest biomass could include dimensionally untreated (i.e. firewood, handling waste, individual waste after mechanical processing) or treated biomass (i.e. chips from pruning, from energy plantations, from stumps or roots, pellets, briquettes). Different sources of biomass exist: biomass coming from thin or large branch-wood, from juvenile thinning or thinning, or from the opening up of roads (Trenčiansky et al. 2007).

Croat. j. for. eng. 35(2014)1

comparison to a demand for forest products driven by economic developments, the demand for forest biomass used for generating energy is mainly driven by policies (Schwarzbauer et al. 2013). Accordingly, biomass utilization for energy purposes is proposed in many EU documents (e.g. Biomass Action Plan 2005, Renewable Heating, Action Plan for Europe 2007, The Forest Sector in the Green Economy 2009, Directive 2009/28/EC on the promotion of the use of energy from renewable sources, Report from the Commission to the council and the European Parliament on sustainability requirements for the use of solid and gaseous biomass sources in electricity, heating and cooling 2010). By now, many national programs have been implemented (Stupak et al. 2007, Schwarzbauer et. al. 2013) including several in Slovakia (Trenčiansky et al. 2007, Halaj and Ilavský 2009, Lieskovský et al. 2009). In Slovakia these encouraging conditions (e.g. Concept for use of renewable sources of energy 2003, Program for rural development 2007 – 2013, Action plan on biomass use for years 2008 – 2013, National action plan

D. Halaj and Y. Brodrechtova

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

for energy from renewable resources 2010) supported newcomers in the renewable energy business, such as: 18 producers of briquettes, 14 producers of wood pellets and the construction of many heat and energy production plants (Lagaňa and Réh 2008). At present, there are 35 heating and power plants with installed outputs greater than 0.2 MWth (Jankovský 2012) and these capacities are expected to grow in coming years (Haluza 2011). Excluding byproducts of the wood processing industry and other sources, forest biomass represented 54% of 0.5 million tons of all biomass sources for heat production in 2011 (The Ministry of Agriculture and Rural Development of the SR 2012). Moreover, the amount of forest biomass (from forest land) used for energy purposes almost tripled between 1990 and 2010 (Oravec et al. 2012). Furthermore, scenarios for EU members show that energy production from renewable energy will double from 103 Mtoe (Million Tonnes of Oil Equivalent) in the year 2005 to 217 Mtoe in the year 2020 (final brutto energy demand) and it is expected that the main renewable energy source will be biomass (The European Commission 2011). This scenario is especially applicable to Slovakia where, in the coming 5 to 10 years, biomass will offer the highest technical potential among renewable energy resources. Regarding the energy potential of biomass volume, forest biomass (from forest land) submits a 9.7% share and it has one of the highest energy potentials, amounting to 26.8 PJ (The Ministry of Agriculture and Rural Development of the SR 2008). However, higher utilization of forest biomass for fuel has been limited by, among other things, current legislation on the production of wood on fallow agricultural land and the creation of energy forests (Oravec et al. 2012, The Ministry of Agriculture and Rural Development of the SR 2012). The production of fuel forest biomass on non-forest land had reached only 0.1 tons by 2011 (The Ministry of Agriculture and Rural Development of the SR 2012). Domestic consumption in 2011 was 1.3 million tons of forest biomass originating from forest land, yet the potential for the use of forest biomass from forest land was set to 2.5 million tons in the same year (The Ministry of Agriculture and Rural Development of the SR 2012) and is expected to reach 2.85 million tons by 2025 (Oravec et al. 2012). Although the use of forest biomass is based mainly on energy reasoning, the emergent renewable energy business has offered opportunities (e.g. new source of revenues, creation of new jobs in rural areas, solving forest management problems) for the traditional forestry sector throughout the EU in general and Slovakia in particular. The forestry sector in Slovakia

has been undergoing a long-term developmental crisis as a result of institutional changes that have taken place in the last 20 years (Greppel et al. 2009, Novotný 2011). Moreover, this unconstructive situation is influenced by the increasing intensity of natural disasters (Lieskovský et al. 2009, Suchomel and Gejdoš 2009; 2011). Also, with the economic crisis in downstream markets, the forest enterprises have been forced to look for new opportunities to redefine their sales portfolios. The new prospect to reinvigorate their profits has been mainly seen in the forest biomass. Nonetheless, the new market segment requires a new application of marketing tools. As a consequence, questions arose as to how forest enterprises tailor marketing tools to the needs and wants of customers in a new target segment?

2. Aim of the study Marketing can help to ensure new opportunities for the forestry sector (Ok 2005). For instance, »some services deemed as non-marketable forest goods and services in forestry will turn out to be marketable« (Ok 2005: 499). However, the use of marketing for forest products within forest products enterprises is very low, although the potentials were identified as high (Šulek 2004, Smith et al. 2009, 2010). On the other hand, most general marketing concepts observed in the literature were prompted some time ago; however, more research on basic issues such as understanding target customers wants and needs is still needed (Wagner and Hansen 2004). This also applies to the forestry sector as expectations toward forestry and forests vary over time (Ok 2005). Furthermore, »…if forestry is investigated historically, it is seen that the forestry product mix (FPM) is changing with time« (Ok 2001: 7). Since the current demand of society consists of a new product – forest biomass, exploring the use of marketing mix in forestry enterprises is needed. Yet, there are only a few studies on how marketing tools are defined, especially for renewable energy sources (e.g. Menegaki 2012). Most of the literature focus is either on the marketing of forest products in general (e.g. Rich 1970, Sinclair 1992, Ok 2005, Elyakime and Cabanettes 2009, Becker 2012) or on the marketing strategies of forest products enterprises in particular (e.g. Schadendorf 1994, Borowski 1996, Piest 1999, Brodrechtova 2009, Hansen and Juslin 2011). So far, the biomass business has been mainly explored from a technological point of view (Stupak et al. 2007, Pätäri et al. 2011). Since over the decade the demand for a new product – forest biomass, is increasing, surveying the marketing mix of forestry enterprises toward the needs and wants of Croat. j. for. eng. 35(2014)1

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

the target segment is desired. To fill the research gap, an exploratory case study on the use of marketing tools targeting forest biomass was proposed. Coming from a marketing perspective, the concept of marketing mix was tested on four decisively selected forest enterprises in Slovakia.

3. Theoretical concept Unique problems in forestry such as not easily sold forest outputs, societal pressure for more nonmonetary goods, long production periods and uncertainty (Klemperer 2003), among other things, form a special challenge for strategic planning in forest enterprises. Strategic planning used to be too focused on competitors, while often neglecting the customers (Webster 1988). However, today marketing integrates the different functions of a company, thus connecting the firm to its customers and its investors (Hansen and Juslin 2011). Accordingly, the objective of marketing is to identify the wants and needs of customers, and/or to satisfy those needs with a special focus on profit instead of sales volume (Webster 1994, Slater and Narver 1998, Tadajewski 2010). This idea also applies to environmentally sound goods (Ottman et al. 2010, Menegaki 2012) like forest biomass (Stupak et al. 2007). At the core of marketing is marketing mix (Kotler 1984). »A marketing mix is a developing process that aims to develop an internally coherent action plan with mutual support of its constituent elements. A marketing mix for renewable energies is simultaneously regarded as a de-marketing mix for fossil fuels« (Menegaki 2012: 34). The origin of marketing mix can be traced to the description of a marketing executive as a »mixer of ingredients« someone who is permanently engaged in creating a mix of various elements generating profit for the firm (Culliton 1948). Based on this idea, the term »marketing mix« was introduced via teaching and in the publications of Borden (1964). He defines marketing mix as a set of 12 elements, which cover the principal areas of marketing activities: product planning, pricing, branding, channels of distribution, personal selling, advertising, promotions, packaging, display, servicing, physical handling, fact handling and analysis. Reducing these 12 elements to four, the four element framework (product, price, promotion, and place) of marketing mix was introduced (McCarthy 1960). Marketing mix is perceived as a conceptual framework rather than a scientific theory (Goi 2009). Due to its practical framework, it found wide acceptance among field marketers (Constantinides 2006). On the other hand, many marketing models are also based on »The 4Ps« (Hansen and Juslin 2011). Croat. j. for. eng. 35(2014)1

D. Halaj and Y. Brodrechtova

However, some weaknesses of »The 4Ps« such as the ignorance of human factors, the lack of strategic dimensions, an offensive posture and a lack of interactivity were identified (Constantinides 2006). As a consequence, various types of marketing mixes evolved. For instance, in the case of marketing mix for service, an additional 3 Ps (participants, physical evidence and process) were added to the original concept of »The 4Ps« (Booms and Bitner 1981). This framework of »The 7Ps« was further applied to the marketing mix of consumer goods (Rafiq and Ahmed 1995). Since forest biomass could be considered a consumer good, the »The 7Ps« framework formed the basis for the theoretical concept of this study. In general, the product tool is defined by the combination of goods and services the company offers to the target market (Kotler and Armstrong 2010). The amount that the consumer must exchange to obtain the offering is defined under the price tool (Solomon 2008). Under the place tool are understood to be all activities necessary for making the product or service available to target consumers (Kotler and Armstrong 2010). Steps taken to inform consumers about products and to encourage them to buy these products are known as the promotion tool (Solomon 2008). The marketing tool »people« is defined as »all human actors who play a part in service delivery and thus influence the buyer’s perceptions; namely, the firm’s personnel, the customer, and other customers in the service environment« (Zeithaml et al. 2008). Overall, the process tool consists of actual procedures, mechanisms, and the flow of actions necessary for the delivery of service. Furthermore, physical evidence describes the surroundings where the service is delivered and interaction takes place. Also, physical evidence is anything that assists with performance or communication of the service (Zeithaml et al. 2008). Finally, the specific features of one marketing tool can often be considered to be a part of other tools.

4. Methodology 4.1 Case study approach Given the limited attention paid to marketing tools in the research context of the forest biomass trade, the testing of marketing perspective has had little contextual basis. For that reason, a case study with linear analytic structure (Yin 2003) and with a qualitative research approach was used. The main argument behind a qualitative research approach was found in the possibility to explore and to capture the complexity of a new research area – marketing tools targeting forest biomass. Furthermore, deductive logic was applied as

D. Halaj and Y. Brodrechtova

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

Table 1 General profile of interviewed forest enterprises in 2011 Indicators/Forest enterprises Ownership type Forest land in use, ha 3

Timber felling, m /year Biomass production, t/year

Forest Enterprise Biomass Levice

Forest Enterprise Kriváň

Urban Forests Kremnica Ltd.

University Forest Enterprise Zvolen

State

Private

State

58 440

48 222

9 701.5

9 964

163 300

185 400

60 000

42 500

5 075

8 700

1 000

145

Source: Annual reports of interviewed forest enterprises

we moved from general marketing tools to relatively more specific instances of forest enterprises. The databank of the forest enterprises was drawn on existing databanks of the National Forest Center in Zvolen, the Ministry of Agriculture and Rural Development of the Slovak Republic, the Technical University of Zvolen and the Statistical Office of the Slovak Republic. The target population consisted of 26 state forest enterprises (independent enterprises of the state forest enterprise LESY SR, š.p.), four military forest enterprises, one university forest enterprise, and 1 238 private forest enterprises with ≥ 50 ha. Subsequently, the forest enterprises for inquiry were identified through targeted selection based on a prepared strategy (Halaj 2012). In other words, selection was not random, but was adjusted to the focus of the exploratory research study (Lamnek 1993a; 1993b). The prepared selection strategy included forest enterprises with various ownership structures and sizes in order to capture typical actors in Slovakian forestry. The four purposely chosen forest enterprises were addressed via telephone and email with the explanation of the research target. All enterprises agreed to participate in the study (Table 1).

4.2 Methods applied Particularly in depth semi structured interviews (Krott and Suda 2001) were conducted with managers of selected forest enterprises and recorded in digital form or by taking notes. The one to two hour long interviews took place either at the actual forest enterprise or at the Technical University of Zvolen and between May and July of 2011. Transcribed conversations were analyzed with the help of content analysis, specifically the technique of text structuring by content was applied (Mayring 2003). In other words, the text was split by content and assigned to categories deductively derived from the theoretical framework. This art of coding is called deductive application of categories (Mayring 2003). Table 2 demonstrates an example of

the coding system for the category »Product«. Subsequently, a review of relevant categories was done by the summarization technique (Mayring 2003). These steps were repeated twice to prevent incorrect text allocation to the categories. Coding conducted by two independent researchers guaranteed validation of text allocation. This was all done with the help of MAXqDA software for qualitative analysis. Overall, the general focus of content analysis with a qualitative approach was on identifying various meanings of the text (Kollárik and Sollárová 2004). Table 2 Example of coding system for category »Product« Category

Subcategory Product assortment

Value Untreated forest biomass Treated forest biomass Moisture of forest chips

Product Product characteristics

Propositions of forest chips Share of thin and large branch-wood in forest chips

5. Results of the study 5.1 Forest enterprises and their marketing mix The marketing mix for forest biomass consisted of »The 6Ps« (Table 3); however, only »The 5Ps« were common for all four forest enterprises. In the following, the marketing mix for each of the interviewed enterprises is summarized. The entrepreneurial activities of Forest Enterprise Biomass Levice (one of the 26 forestry enterprises of the state owned enterprise Lesy SR š.p.) were focused on the biomass business. Specifically, Forest Enterprise Biomass Levice provided at its seven regional centers, specific services »product«. It bought and processed Croat. j. for. eng. 35(2014)1

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

biomass (i.e. firewood, residues after felling: thin or large branch wood, handling waste, calamity wood) from the other forest enterprises of Lesy SR š.p. (the biomass volume was appointed by Lesy SR š.p.) and also from private forest owners. The desired outputs were forest chips for which the processing »price« depended mainly upon the truck hauling distance, and upon the cost of the skidding of biomass from forest stand to roadside and decking it in piles. The »place« of processing was usually a forest roadside. Forest chips were further stored at the roadside, in meadows or pastures, or in rented former storage facilities of agricultural cooperatives. Transport of forest chips to customers was organized via outsourcing. Consumer sales promotion and personal selling »promotion« were used. In processing forest biomass, the enterprise created employment opportunities in rural areas. It also tested chipping machines and subsequently gave advice to its machine producers »people«. Within »process« activities, Forest Enterprise Biomass Levice monitored financing possibilities from EU funds, the payment discipline of its customers and the quality of the chipping process. Forest Enterprise Kriváň (one of the 26 forestry enterprises of Lesy SR š.p.) focused mostly on the production and sale of sawn logs. Business with forest biomass was only 6.5% of total sales. This enterprise produced and sold biomass »product« such as: firewood, residues after felling (thin or large branch wood), and handling waste. However, based on directives from Lesy SR š.p., certain volumes of biomass were sold to Forest Enterprise Biomass Levice for an asked price. Generally, the »price« depended on the cost of concentration of biomass to designated areas, the truck hauling distance and the tree species. Additionally, the firm offered locals a promotion sale price of one cent for residues the locals themselves collect after felling. The timber felling was outsourced to private companies, which also hauled biomass to the roadside, meadows, pastures or former storage facilities of agricultural cooperatives »place«. It used only consumer sales promotion and personal selling »promotion«. Due to the outsourcing of the felling, employment opportunities in rural areas were created »people«. Forest enterprise Kriváň monitored »process« financing possibilities by EU funds, evaluated the payment discipline of its customers and ensured the quality of the chipping process by investing in new chipping machines. The forestry enterprise Urban Forests Kremnica Ltd. focused mostly on the production and sale of sawn logs. The forest biomass business was only a byproduct. This private enterprise produced and sold Croat. j. for. eng. 35(2014)1

D. Halaj and Y. Brodrechtova

biomass »product« such as firewood, one meter wood logs and handling waste. »Price« was generally based on costs; however, a promotion sale price of one cent was used for collecting residues after felling. Forest biomass was hauled to the roadside »place« by forest biomass customers. Firewood was sold and delivered directly to the locals »place«. Only consumer sales promotion and personal selling »promotion« were done. Hauling of one meter logs created employment opportunities in rural areas and on the other hand, customer service offered a social program in the form of delivering these one meter logs directly to local customers »people«. The task of the »process« tool was in the overview of EU structural funds for possible investments in new chipping machines. An additional task included the identification of customer needs via SWOT analysis. The main purpose of University Forest Enterprise Zvolen was scientific (e.g. silviculture, felling, or hunting practices among others). However, the production and sale of sawn logs and byproducts such as biomass were done on a commercial basis. The focus was on the production and selling of biomass »product« such as firewood, residues after felling (thin or large branch wood), one meter wood logs and handling waste. It offered a promotion sale price for the residues after felling. Forest biomass was hauled to the roadside, meadows, or pastures »place«. The forest enterprise used consumer sales promotion and personal selling »promotion«. Commercial utilization of biomass created employment opportunities »people«.

5.2 Marketing tools and their features used in the Slovakian forest biomass trade Despite diverse company characteristics, the marketing tools showed some similarities. In the following, »The 6Ps« (Table 3) are summarized and described based on respondents’ rate (percentage of respondents who used specific tool and specific feature in forest biomass trade). The marketing tool »product« was defined via »product assortment« and »product characteristics«. In the case of »product assortment« the forest enterprises produced forest biomass (100%) and forest chips (75%). In other words, some enterprises subsequently processed the biomass into forest chips. Particularly, the technology for processing coniferous (thickness ≤ 45 cm and length ≥ 2 m) and broadleaved biomass (thickness ≤ 35 cm and length ≥ 2 m) was used. On the other hand, »product characteristics« defined only forest chips parameters and moisture. The »price« was described by respondents via different pricing methods such as: »cost based pricing«

D. Halaj and Y. Brodrechtova

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

Table 3 Used marketing tools and their features identified by interviewed forestry enterprises Product

Price

Place

Promotion

People

Process*

Product assortment

Cost based pricing Competition oriented pricing Differentiated pricing

Suppliers

Sales promotion

Employees

Monitoring

Distribution channels

Personal selling

Customer service

Quality assurance

Product characteristics

Consumers

Consulting

Location Transportation Storage * This feature was identified only in three of four interviewed enterprises

(100%), »competition oriented pricing« (25%) and »differentiated pricing« (25%). »Competition oriented pricing« was used only in the case of heat stations with lower forest chip consumption and many suppliers. Due to the various locations of biomass sale, respondents also used the »differentiated pricing« method. Generally, the main costs for forest enterprises were costs associated with biomass skidding and hauling from forest, and biomass decking at the roadside. All respondents stated that the only profitable trade was with forest chips originating from biomass after deforestation of meadows or pastures. The marketing tool »place« was described via features such as »suppliers« (25%), »distribution channels« (100%), »consumers« (100%), »location« (100%), »transport« (100%) and »storage« (100%). Only one forest enterprise used »suppliers«, local saw mills, to buy cuttings. This represented just 1% of biomass offered by this forest enterprise. Forest biomass and forest chips were »directly« distributed »distribution channels«. In the case of biomass they were sold to private »consumers« collecting residues after felling in the forest »location«. On the other hand, forest chips were sold to the heat stations and power plants »consumers« from the roadside »location«. The physical distribution of forest chips was organized by forest enterprises, but it was outsourced to the truck companies chosen via public tenders. Optimal truck hauling distance was specified to be in the range of 30 – 40 km. Skidding of biomass »transport« to the roadside »location« was done by tractors (100%) with some companies also using horses for this task (50%). Generally, forest biomass and forest chips were stored at former facilities of agricultural cooperatives, roadsides, meadows or pastures »storage«). »Promotion« was interpreted only via »customer sales promotion« (100%) and »personal selling« (100%). In the former case, »promotion sale price« of one cent for the collection of residues after felling was

offered to the locals. All interviewed companies stated that the sale of biomass (residues after felling) to the locals was based on »verbal agreement«. In the latter case, »personal selling« was perceived as the most important feature in the biomass trade. Specifically, written one year contracts were preferred in the case of forest chips. The minimum volume of contracted forest chips was set at 100 tons. Overall, »personal selling« (including »internal selling« to existing customers, »sales calls« to potential purchasers and »email correspondence« among others) was done by the business managers of the particular forest enterprises. The use of the marketing tool »people« was explained via features such as: »employees« (100%), »customer service« (25%) and »consulting« (25%). Due to forest biomass processing new »positions«, such as jobs for handling one meter wood or for chipping forest biomass, were created. »Customer service« was recognized through its positive effects of social programs supporting local consumption of energy wood. This program consisted of delivering one meter wood (firewood) to the local consumers and its unloading at their courtyards. »Consulting« activity was done via feedback to the manufactures of forest chipping machines, which brought technological improvement of chipping machines. The marketing tool »process« consisted of »monitoring« (75%) and »quality assurance« (75%). Particularly, »monitoring« of EU structural funds to support investment activities, »monitoring« current trends in felling and chipping technologies, »monitoring« of customers’ payment discipline and »monitoring« of stock turnover of forest chips were carried out. Via purchasing new forest chipping machines (due to ending of their depreciation period), the forest enterprises tried to »assure quality« of chipping production. Forest enterprises used either their own financial resources from a so called intra plant bank, which was established within the Lesy SR š.p. or EU structural funds. Croat. j. for. eng. 35(2014)1

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

6. Discussion and conclusion Developing new sources of income through innovation and entrepreneurship offers the possibility for economic renewal in the forestry sector (Rametsteiner et al. 2006). Marketing in general and its tools in particular can help to ensure new opportunities for forestry and save forestry from certain social and political dilemmas (Ok 2005). »It must never be over looked that marketing is the income generating activity of the firm« (Smith et al. 2010: 1). Since the rapidly growing renewable energy market brought a new opportunity to forestry (Halaj 2012, Stidham and Simon-Brown 2011, Schwarzbauer et al. 2013), a new target segment – forest biomass, required the appointment of unique marketing tools. The results of the exploratory research study showed that in the business of forest biomass, forest enterprises in Slovakia used a marketing mix of »The 6Ps«: »product«, »price«, »place«, »promotion«, »people« and »process«. Additionally, it was found that »The 5Ps« (excluding »process«) were universal for all interviewed companies despite their major differences in ownership structure or strategic planning. Consequently, based on the results, three major conclusions could be drawn. First, the testing of the proposed marketing mix »The 7Ps« showed that for a special segment such as the forest biomass market, the marketing tool »physical evidence« is not applicable. Similar results were also observed in theoretical analysis of marketing mix for forest biomass (Halaj and Ilavský 2009). The explanation behind this finding lies in the characteristics of forest biomass, which is a commodity product on one hand but a very specific product on the other (Šulek 2004, Greppel et al. 2007). Thus, the results added to the findings of other authors (e.g. Hesková 2001, Reimann 2010) that specific industry sectors require a specific combination of marketing tools tailored to their customer needs and wants. Consequently, the identification of the marketing mix »The 6Ps« for the forest biomass segment could be seen as a major contribution to the existing models of marketing mixes of other industry sectors. Second, the results of in depth interviews revealed general use of »The 5Ps« such as »product«, »price«, »place«, »promotion« and »people« for all forest enterprises. The reasoning behind the commonality of »The 5Ps« is mainly explained by the product characteristics, and natural conditions in which the enterprises operated. Forest biomass was defined as a seasonal commodity »product«, for which production and costs were mainly challenged by hauling and concentrating at designated areas »place«, and lengthy Croat. j. for. eng. 35(2014)1

D. Halaj and Y. Brodrechtova

and quality storing »place« (Halaj 2012, Ruiz et al. 2013, Shabani et al. 2013). Moreover, distinguished natural conditions (e.g. high diversity of terrain; inaccessibility of forest in some months) made the whole logistics of forest biomass complex »place« and often unprofitable »price«. »Promotion« was identified as the most effective marketing tool, particularly its feature »personal selling« (Smith et al. 2009). This form of selling industrial goods was observed in other industry sectors as it plays an essential role before the consumer makes the purchase, during the purchase and after the purchase. This showed that the forest biomass sector was more customer than product oriented, the current trend also observed in other forest products industries (Brodrechtova 2009, Tadajewski 2010, Juslin and Hansen 2011). The main argument behind the tool »people was the creation of jobs in rural areas, which in turn lead to rural community revitalization (Halaj 2012, Stidham and Simon-Brown 2011). Generally, the common use of »The 5Ps« revealed that with the new target segment of forest biomass, there was an associated high degree of uncertainty due to high logistics costs, market instability, natural conditions and policy changes (Shabani et al. 2013). Although the application of marketing tools offered an opportunity for the economic restoration of forest enterprises through the renewable energy trade, it remained unknown how effective the forest biomass trade will be for forest enterprises in the long term. Finally, it was found that a case study approach allowed for testing marketing tools of only four forest enterprises; the fact that the proposed marketing mix concept can be readily measured and tested serves to overcome this limitation. Further research on marketing tools targeting forest biomass trade could supplement the results of this exploratory study. This could be done either by considering a representative sample of forest enterprises in Slovakia or internationally. Additionally, as time progresses, the marketing tools used in the biomass trade might change. Therefore, periodic research would further extent current findings as the use of forest biomass will potentially intensify in Slovakia as well as globally.

Acknowledgements This research received support from VEGA 1/1099/12 »The Economic Effectiveness of Timber Trade: A Transaction Cost Perspective« and it was presented briefly as oral presentation at the IUFRO conference: FORTECHNENVI held in 2013 in Brno, Czech Republic. The authors would also like to thank Charles Kennedy for proof-reading.

D. Halaj and Y. Brodrechtova

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

7. References Becker, M., 2012: Absatzmarketing. In: Management von Forstbetrieben, Band III: Leistungssystem, Zusammenfassung und Ausblick (Oesten, G., Roeder, A., Eds.). Institut für Forstökonomie der Albert-Ludwigs Universität Freiburg, p. 19–58. Booms, B.H., Bitner, M.J., 1981: Marketing strategies and organization structures for service firms. In: Marketing of Services. American Marketing Association (Donnelly, J.H., George, W.R., Eds.). Chicago, IL, p. 47–51. Borden, N.H., 1964: The concept of the Marketing Mix. Journal of Advertising Research, Classics, Volume II (September 1984): 7–12. Borowski, S., 1996: Marketing-Strategien von Forstbetrieben. Schriften aus dem Institut für Forstökonomie der Universität Freiburg. Band 7, 166 p. Brodrechtova, Y., 2009: Export marketing strategy and its shaping factors in the context of transition: Theoretical perspectives and empirical testing in the forest products industries of Slovakia. Freiburger Schriften zur Forst und Umweltpolitik. Verlag Dr. Kessel, Remagen-Oberwinter. Band 21, 159 p.

Haluza, I., 2011: Ošiaľ okolo obnoviteľných zdrojov sa skončil: Slnečné i veterné elektrárne štát už brzdí a zásoby biomasy nie sú neobmedzené (The folly around renewable resources is over: sun and wind power plants are held back by the state and biomass resources are limited). Trend holding s. r. o., 18: 54–55. Hansen, E., Juslin, H., 2011: Strategic Marketing in the Global Forest Industries. 2nd Edition, Corvallis, Oregon, 327 p. Hesková, M., 2001: Marketingová komunikace součást marketingového mixu (Marketing communication as a part of marketing mix). 1. vydanie Praha. Vysoká škola ekonomická, Praha, 106 p. Jankovský, J., 2012: Využitie biomasy v rámci centrálnej výroby energie (The biomass utilization within the central energy production). PhD dissertation, Technická Univerzita vo Zvolene, 148 p. Klemperer, W.D., 2003: Forest Resource Economics and Finance. McGraw-Hill Science/Engineering/Math, 551 p. Kollárik, T., Sollárová, E., 2004: Metódy sociálno psychologickej praxe (Methods of socio psychological practice). Bratislava. Pegas, 264 p. Kotler, P., 1984: Marketing Management: Analysis, Planning and Control. 5th Edition. New Jersey: Prentice-Hall., 792 p.

Constantinides, E., 2006: The Marketing Mix Revisited: Towards the 21st Century Marketing. Journal of Marketing Management 22(3–4): 407–438.

Kotler, P., Armstrong, G., 2010: Marketing: An Introduction. 10th Edition. New Jersey: Prentice Hall, 648 p.

Culliton, J.W., 1948: The Management of Marketing Costs. Division of Research, Graduate School of Business Administration, Boston, MA: Harvard University, 166 p.

Krott, M., Suda, M., 2001: Befragung als Methode der Sozialforschung in der Forstwissenschaft. J.D. Sauerlaender’s Verlag, Frankfurt am Main, 160 p.

Elyakime, B., Cabanettes, A., 2009: How to improve the marketing of timber in France? Journal for Forest Economics and Policy 11(3): 169–173.

Lagaňa, R., Réh, R., 2008: Wood material flows currently present in the Slovak Republic. In the proceedings: Cost Action E44 Final Conference in Milan on A European wood processing strategy: Future resources matching products and innovations. Ghent University, p. 87–92.

FAO 2004: Unified Bioenergy Terminology UBET. Food and Agriculture Organization of United Nations December 2004, 58 p. Goi, CH. L., 2009: A review of marketing mix: 4Ps or more? International Journal of Marketing Studies 1(1): 2–15. Greppel, E., Šulek, R., Drličková, E., Paluš, H., 2007: Kvalita dreva a obchod s drevom (Wood Quality and Wood Trade). Národné lesnícke centrum, Zvolen, 181 p. Greppel, E., Paluš, H., Chudovský, D., Šulek, R., 2009: Zhodnotenie drevnej hmoty a marketingové riadenie (Evaluation of wood mass and marketing management). Národné lesnícke centrum, Zvolen, 146 p. Halaj, D., 2012: Marketing tools supporting the utilization of wood as a renewable energy source by chosen forest enterprises. In the proceedings of the international scientific conference: Management of Organization in Real and Virtual Environment: Opportunities and Challenges IV. Poprad, 2012, 120 p. Halaj, D., Ilavský, J., 2009: Policies and their implementation tools enhancing the energy wood market: a comparative case study of Finland and Slovakia. Finnish Forest Research Institute. Working Papers of the Finnish Forest Research Institute, 87 p.

Lamnek, S., 1993a: Qualitative Sozialforschung, Band 1: Methodologie. 2. ueberarb. Auflage, Beltz, Psychologie Verlags Union, Weinheim, 308 p. Lamnek, S., 1993b: Qualitative Sozialforschung, Band 2: Methoden und Techniken. 2. ueberarb. Auflage, Beltz, Psychologie Verlags Union, Weinheim, 438 p. Lieskovský, M., Suchomel, J., Gejdoš, M., 2009: Energetický potenciál vybraných druhov disponibilnej biomasy lesa (The energy potential of selected types of forest biomass). Technická Univerzita vo Zvolene. Vydavateľstvo TU Zvolen, 73 p. Mayring, P., 2003: Qualitative Inhaltsanalyse. Beltz Verlag, Weinheim und Basel, 144 p. McCarthy, E.J., 1960: Basic Marketing, a Managerial Approach. Richard D. Irwin, Inc. Homewood, Ill. 770 p. Menegaki, A., 2012: A social marketing mix for renewable energy in Europe based on consumer stated preference surveys. Renewable Energy 39(1): 30–39. Novotný, J., 2011: Kam kráčaš slovenské lesníctvo? ... alebo, ako zabezpečiť funkčnosť lesníctva na Slovensku v dlhodobom horizonte? (Slovak forestry, where are you heading? … Croat. j. for. eng. 35(2014)1

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44) or, how to secure the functionality of forestry in Slovakia in the long-term?). Les & Letokruhy. Bratislava: Lesmedium 67 (9–10): 6–10. Ok, K., 2001: Evolution of traditional product mix in forestry marketing. 55th Forest Product Society Annual Meeting Maryland, USA. Ok, K., 2005: Idea marketing in forestry: some implications from the Turkish forestry experience. Journal for Forest Policy and Economics 7(4): 493–500. Oravec, M., Bartko, M., Slamka, M., 2012: Postupy intenzifikácie produkcie drevnej biomasy na energetické využitie (Steps towards intensification of forest biomass production for energy use). Národné lesnícke centrum, Zvolen, 65 p. Ottman, J.A., Stafford, E.R., Hartman, C.L., 2010: Avoiding green marketing myopia: Ways to improve consumer appeal for environmentally preferable products. Environment: Science and Policy for Sustainable Development 48(5): 22–36. Pätäri, S., Puumalainen, K., Jantunen, A., Sandström, J., 2011: The interface of the energy and forest sectors – Potential players in the bioenergy business. International Journal of Production Economics 131(1): 322–332. Piest, E., 1999: Grundlagen für ein Marketingkonzept – Buchenstammholz: fuer die Niedersaechsische Landesforstverwaltung. PhD dissertation. Georg August Universitaet Goettingen, 173 p. Rafiq, M., Ahmed, P.K., 1995: Using the 7Ps as a generic marketing mix: an exploratory survey of UK and European marketing academics. Marketing Intelligence and Planning 13(9): 4–15. Rametsteiner, E., Hansen, E., Niskanen, A., 2006: Introduction to the special issue on innovation and entrepreneurship in the forest sector. Forest Policy and Economics 8 (7): 669– 673. Reimann, M., Schilke, O., Thomas, J.S., 2010: Toward an understanding of industry commoditization: Its nature and role in evolving marketing competition. International Journal of Research in Marketing 27: 188–197. Rich, S.U., 1970: Marketing of Forest Products: Text and Cases. McGraw-Hill, Inc. New York, 712 p. Ruiz, J.A., Juárez, M.C., Morales, M.P., Munoz, P., Mendívil, M.A., 2013: Biomass logistics: Financial & environmental costs. Case study: 2 MW electrical power plants. Biomass and Bioenergy 56: 260–267. Sinclair, S. A., 1992: Forest Products Marketing. McGrawHill, Inc., 403 p. Schadendorf, C., 1994: Strategisches Marketing im Forstbetrieb: ein Konzept für Nadelstammholz in der Landesfortverwaltung Schleswig-Holstein. PhD dissertation. Universitaet für Bodenkultur Wien, 204 p. Schwarzbauer, P., Weinfurter, S., Stern, T., Koch, S., 2013: Economic crisis: Impacts on the forest-based sector and woodbased energy use in Austria. Forest Policy and Economics 27: 13–22. Shabani, N., Akhtari, S., Sowlati, T., 2013: Value chain optimization of forest biomass for bioenergy production: A reCroat. j. for. eng. 35(2014)1

D. Halaj and Y. Brodrechtova

view. Renewable and Sustainable Energy Reviews 23: 299– 311. Slater, S.F., Narver, J.C., 1998: Customer-led and marketoriented: Let’s not confuse the two. Strategic Management Journal 19(10): 1001–1006. Smith, B., Hansen, E., Olah, D., 2009: Personal selling for forests products industry. Virginia Cooperative Extention. Produced by Communications and Marketing, College of Agriculture and Life Sciences, Virginia Polytechnic Institute and State University. 12 p. Available: http://pubs.ext.vt. edu/420/420-146/420-146_pdf.pdf [Accessed 15 Juni 2013]. Smith, B., Hansen, E., Olah, D., 2010: Marketing for wood products companies. Virginia Cooperative Extention. Produced by Communications and Marketing, College of Agriculture and Life Sciences, Virginia Polytechnic Institute and State University. 7 p. Available: http://pubs.ext.vt.edu/420/420145/420-145_pdf.pdf [Accessed 15 Juni 2013]. Solomon, M.R., 2008: Marketing: Real People, Real Decisions. Prentice Hall. 6th Edition, 640 p. Stidham, M., Simon-Brown, V., 2011: Stakeholder perspectives on converting forest biomass to energy in Oregon, USA. Biomass and Bioenergy 35(1): 203–213. Stupak, I., Asikainen, A., Jonsell, M., Karltun, E., Lunnan, A., Mizaraite, D., et al., 2007: Sustainable utilization of forest biomass for energy-possibilities and problems: policy, legislation, certification, and recommendations and guidelines in the Nordic, Baltic, and other European countries. Biomass & Bioenergy 31(10): 666–84. Suchomel, J., Gejdoš, M., 2009: Analýza vývoja cien vybraných sortimentov surového dreva a výrobkov z dreva (Analysis of price developments of selected solid wood assortments and products from wood). In the proceedings of the international scientific conference: Financovanie 2009 Lesy-Drevo, Zvolen, 14 p. Suchomel, J., Gejdoš, M., 2011: Vplyv náhodných ťažieb na lesníctvo Slovenska. Progresívne postupy spracovania náhodných ťažieb (The influence of accidental felling on Slovak forestry. Progressive practices for the processing of accidental felling). In the proceedings of the international scientific conference: Financovanie 2011 Lesy – Drevo, Zvolen, p. 7–22. Šulek, R., 2004: Marketingové riadenie lesných podnikov (Marketing management of forest enterprises). Vedecké štúdie 5/2004/A. Technická Univerzita vo Zvolene, 59 p. Tadajewski, M., 2010: Eventalizing the marketing concept. Journal of Marketing Management 25 (1–2): 191–217. The European Commission, 30. January 2011: Communication from the Commission to the European Parliament and the Council. Renewable Energy: Progressing towards the 2020 target. Brussels, 31.01.2011. COM (2011) 31 final, 16 p. Available: http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=COM:2011:0031:FIN:EN:PDF [Accessed 20 Juni 2013]. The Ministry of Agriculture and Rural Development of the SR, 11 November 2008: Akčný plán využívania biomasy na roky 2008-2013 (The action plan on biomass use for years 2008-2013). Available: http://www.mpsr.sk/sk/index.php?sta

D. Halaj and Y. Brodrechtova

Use of Marketing Tools in the Slovakian Forest Biomass Trade (35–44)

rt&language=sk&navID=2&navID2=2&sID=26&id=1214 [Accessed 20 Juni 2013]. The Ministry of Agriculture and Rural Development of the SR, November 2012: Správa o lesnom hospodárstve v Slovenskej Republike za rok 2011: Zelená správa (Report on forest management of Slovak Republic for 2011: Green Report). Bratislava, 67 p. Trenčiansky, M., Lieskovský, M., Oravec, M., 2007: Energetické zhodnotenie biomasy (Energy extraction from biomass). Národné lesnícke centrum, Zvolen 2007, 147 p. Yin, R.K., 2003: Case Study Research. Design and Methods. SAGE Publications, California. 3rd Edition, 181 p. Wagner, E.R., Hansen, E.N., 2004: A method for identifying and assessing key customer group needs. Industrial Marketing Management 33: 643–655. Webster, Jr.F.E., 1988: The rediscovery of the marketing concept. Business Horizons 31(3): 29–39.

Interviewed forestry enterprises Forest Enterprise Biomass Levice (OZ Odštepný závod Biomasa Levice). Available: http://www.lesy.sk/files/sprava2012/ vs-lesy-2012-web-sj.pdf [Accessed 10 January 2013]. Forest Enterprise Kriváň (OZ Odštepný závod Kriváň). Available: http://www.lesy.sk/showdoc.do?docid=223&id=18 [Accessed 10 January 2013]. Urban Forests Kremnica, Ltd. (Mestské Lesy Kremnica, s.r.o.). Available: http://www.mslkca.sk/index.html [Accessed 10 January 2013]. University Forest Enterprise in Zvolen (Vyskoškolský lesnícky podnik Technickej Univerzity vo Zvolene). Available: http://www.tuzvo.sk/sk/organizacna_struktura/dalsie_organizacne_sucasti/vysokoskolsky-lesnicky-podnik/o-nas/novaweb-stranka.html [Accessed 10 January 2013].

Webster, Jr.F.E., 1994: Executing the new marketing concept. Marketing Management, 3(1): 9–16. Zeithaml, V.A., Bitner, M.J., Gremler, D., 2008: Services Marketing. McGraw-Hill/Irwin; 5th Edition, 736 p.

Authors’ address:

Received: September 18, 2013 Accepted: December 09, 2013

Daniel Halaj, PhD. e-mail: halaj@tuzvo.sk Yvonne Brodrechtova, PhD.* E-mail: brodrechtova@tuzvo.sk Technical University in Zvolen Faculty of Forestry Department of Economics and Management of Forestry T. G. Masaryka 24 96053 Zvolen SLOVAKIA * Corresponding author Croat. j. for. eng. 35(2014)1

Original scientific paper

Examining the Optimal Bucking Method to Maximize Profits in Commercial Thinning Operations in Nasunogahara Area, Tochigi Prefecture, Japan Chikara Nakahata, Kazuhiro Aruga, Masashi Saito Abstract Optimal bucking methods were applied to two operational sites of the Nasu-machi Forest Ownersâ&#x20AC;&#x2122; Co-operative to maximize profits, with and without taking log size into consideration, and to maximize revenues considering a new subsidy system and higher unit prices of smallsized logs. Corresponding optimum extraction rates and small-sized log prices were examined. Extraction rates from stands using the optimal bucking method to maximize profits considering log sizes were similar to the actual values, unlike the estimations obtained using other methods, e.g. without considering log sizes or maximizing revenues. However, the differences in extracted volumes and economic balances among these estimations were small and can be said to be within the error limits traditionally seen for forests and forestry conditions in the stands. The stands were about 50 years old, with an average diameter at breast height (DBH) of about 20 cm. Differences in extraction rates from stems with a DBH exceeding 20 cm were small. However, extraction rates from stems with a DBH less than 20 cm were significantly different. Therefore, the optimal bucking method to maximize profits considering log sizes could help determine the optimal extraction rates of younger stands with smaller DBHs. Possible effects of the new subsidy system and different unit prices for small-sized logs were also discussed. Both the new subsidy system and the unit prices with feed-in tariff contributed to an increase in extracted volumes of small-sized logs. Keywords: Optimal bucking, maximum profit, extraction rate, economic balance, small-sized logs, subsidy

1. Introduction Facilities such as biomass power plants, chip production factories, and pellet plants in Tochigi prefecture require woody biomass resources. These resource requirements are mainly fulfilled by sawmill residues and construction waste woods. However, there are concerns about the future adequacy of supplies of these materials, because of the increased number of entrepreneurs who have set up biomass power plants as a measure against climate change and to improve energy security. Furthermore, the Feed-In Tariff (FIT) scheme was introduced in Japan starting July 1, 2012. In FIT, the purchase price (without tax) of electricity Croat. j. for. eng. 35(2014)1

generated with unused materials such as small-sized logs and logging residue, general materials such as sawmill residue, and recycled material such as construcÂ tion waste wood are 0.32 USD/kWh, 0.24 USD/kWh, and 0.13 USD/kWh, respectively (Agency for Natural Resources and Energy 2012). Power generation with unused materials was incentivized. Therefore, the use of small-sized logs and logging residue must be promoted in the near future. Tochigi prefecture measures 640 785 ha, of which about 54.5% is covered by forests (Tochigi Prefectural Government 2009). About 45% of these forests are man-made, constituting a volume of 42 255 000 m3

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

(62.4%). Given the delay in thinning, which was a serious problem for man-made forests throughout Japan, the Tochigi Prefectural Government introduced a new regulation in April 2008, along with subsidies for thinning operations, aimed at making forests healthy (Tochigi Prefectural Government 2010a). As a result, thinning operations were conducted on 2 663 ha of forest in 2009. However, large amounts of thinned wood were left in the forests, because nearly all the thinning operations conducted were pre-commercial for small-sized trees and affected profitability. The Nasu-machi Forest Owners’ Co-operative extracts smaller diameter logs for pellets or pulp in addition to larger diameter logs for saw timber or laminated lumber. However, the extraction of smaller diameter logs would increase revenue, but also increase costs, and subsequently, decrease profitability. Numerous studies have examined optimal bucking for increasing revenues (Akay et al. 2010, Haynes and Visser 2004, Nagumo et al. 1981, Nakajima et al. 2008, Nakajima et al. 2009, Olsen et al. 1991, Sessions et al. 1989a, Wang et al. 2009, Yoshida and Imada 1989). However, the bucking methods affected the efficiencies of the bucking and extracting operations, thereby increasing costs and lowering profitability. Therefore, it is essential to conduct optimal bucking with a consideration of costs and profitability as well as revenues. The optimum bucking problems can be categorized into three levels: (1) stem-level problems to determine the optimum bucking for each stem in a way that maximizes the total stem value; (2) stand-level problems to determine the best possible bucking results with maximum aggregate production value; and (3) forest-level problems to maximize the global profit considering demand constraints, merchandising restrictions, and forest-estate (Laroze 1999). Network analysis techniques (Sessions 1988) and DP (Nasberg 1985) have been effectively used to solve stem-level optimum bucking problems. At stand-level, optimum bucking problems have been generally formulated as two-level optimization problems utilizing both LP and DP (Sessions et al. 1989b, Laroze and Greber 1997). The optimization procedures involving heuristic techniques such as Tabu Search (Laroze 1999) and Genetic Algorithm (Kivinen 2004) have been used to solve forest-level bucking problems. Uusitalo (2007) developed a Genetic Algorithm based method to integrate transportation cost and product values into forestlevel optimum bucking problem. This study integrated forwarding costs into stemlevel optimum bucking problem. Nakahata et al. (2013) investigated commercial thinning operations around Nasunogahara area, where the Nasu-machi

Forest Owners’ Co-operative is located. They analyzed the relationships between log sizes and operational costs of processing as well as forwarding, and developed equations to estimate operational costs according to log sizes. The operation costs estimated with the equations were increased according to the extraction of smaller diameter logs. Thus, we determined the optimal bucking methods to maximize profits and the optimum extraction rates of smallsized logs. Here, we examine optimum extraction rates using different optimal bucking methods and small-sized log prices.

Table 1 Study sites A

Japanese

cedar

cypress

cedar

5.25

1.87

6.70

Slope angle, °

DBH*, cm

19.9

18.7

22.2

Stem volume, m /stem

0.80

0.46

0.62

Stand density, stem/ha

937

1 169

1 054

749.60

537.74

653.48

Thinned wood tree height, m

18.5

17.7

20.1

Thinned wood volume, m3/stem

0.49

0.27

0.36

Thinning rate of stem, %

27.6

35.4

33.3

Thinning rate of volume, %

17.1

20.5

19.3

668.27

207.75

844.23

Species Stand age, year Area, ha

Tree height, m 3

Stock, m3/ha Thinned wood DBH, cm

Thinned volume, m

606.97

311.08

Extracted rate, %

69.3

36.9

Road density, m/ha

300.3

229.7

Extracted volume, m

Rate of bunching area, %**

83.0

88.2

43.9

Forwarding distance, m***

120.0

247.6

111.9

* DBH: Diameter at Breast Height ** Rate of bunching area to stand area. Bunching area was assumed to be within 20 m from the roads *** Forwarding distances were estimated from landing to each stand by the Dijkstra method

Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

2. Optimal bucking method to maximize profits The optimal bucking method to maximize profits was determined as follows: 1) estimate thinning volumes, 2) determine the taper curve formula, 3) estimate extracted volumes, 4) estimate revenues, 5) estimate expenses, 6) estimate economic balances, and 7) determine the optimal bucking method to maximize profits.

2.1 Study sites The optimal bucking methods were applied to two commercial thinning operation sites where small-sized logs were extracted, in addition to saw logs and logs for laminated lumber (Table 1, Fig. 1). Forestry operations included chainsaw felling, grapple-loader bunching, processor processing, and forwarder forwarding, with a high-density road network, more than 200 m/ha of relatively steep terrain, around 20° (Fig. 2). Road width was 3.5 m. Before 2008, smaller-sized machines were used, and the road width was 2.0 m. Before 2008, the forestry operation system included chainsaw felling and processing, mini grapple-loader bunching, and mini-forwarder forwarding. Using larger machines after 2008, operational efficiencies improved from 2.3–3.8 m3/person-day to 3.9–5.3 m3/person-day, and direct operation costs were reduced from 59–76 USD/m3 to 37–44 USD/m3 (Aruga et al. 2013). Site A included 55-year old Japanese cedar and Japanese cypress stands, with an area of 7.12 ha and a slope angle of 19°. Commercial thinning operations were conducted by the Nasu-machi Forest Owners’ Cooperative between January 28 and March 9, 2010. The forest road

C. Nakahata et al.

was 268 m long. The strip road was 3.5 m wide and 1 870 m long. Therefore, the road density was 300.3 m/ha. Grapple-loader bunching operations were conducted within 20 m from the roads, and thinned woods beyond 20 m from the roads were left in the forest (pre-commercial thinning operations were conducted in the forest beyond 20 m from the roads). Therefore, the extraction is assumed to be within 20 m from the roads. The rates of the bunching areas to the study site area were 83.0% for Japanese cedar and 88.2% for Japanese cypress. Site B was a 52-year old Japanese cedar stand, with an area of 6.70 ha and a slope angle of 23°. Commercial thinning operations were conducted by the Nasumachi Forest Owners’ Co-operative between November 12 and December 10, 2010. The forest road was 587 m long. The strip road was 3.5 m wide and 952 m long. Therefore, road density was 229.7 m/ha. Grapple-loader bunching operations were conducted within 20 m from the strip roads only. Therefore, the rate of the bunching area to the study site area was only 43.9%.

2.2 Results of the study In order to estimate thinning volumes, the diameter at breast height (DBH) distributions of the thinned woods were estimated to apply the Weibull distribution (Kinashi 1978) to the results of the plot data (Fig. 3). The tree heights H (m) were estimated as follows. Stand A Japanese cedar: H = 5.342D0.382 (R2 = 0.53) (1) Stand A Japanese cypress: H = 8.817D0.239 (R2 = 0.56) (2) Stand B Japanese cedar: H = 5.065D0.46 (R2 = 0.56) (3)

Fig. 1 Study sites (Left: Stand A, Right: Stand B Croat. j. for. eng. 35(2014)1

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

Thinned volumes in the stands VT (m3) were estimated as follows.

VT =

Dmax

∑

j = Dmin

VTj

(4)

where VTj denotes thinned volumes in the j DBH class (m3), and Dmax and Dmin are the maximum and minimum DBH class of thinned woods, respectively. VTj was estimated as follows.

VTj = Vnj × N j × A ×

RTj 100

(5)

where Vnj denotes stem volumes of thinned woods in the j DBH class (m3/stem), Nj refers to the number of thinned woods in the j DBH class (stem/ha), A is the stand area (ha), and RTj is the thinning rate of stems in the j DBH class (%).

2.3 Determination of the taper curve formula Stem diameter d (cm) at the height h (m) above the ground was estimated with the following taper curve formula (Inoue and Kurokawa 2001).

    1.2  h − 0.9 a + 1.8 ×  1.0 −  a  1.0 −      H H  d=  × D (6)     1.2  h a  1.0 −  − 0.9a + 1.8 ×  1.0 − H  H     Coefficient a was estimated as follows.

 21.6  7  18.0 − H  − 12.6 10 f  2.4   8.4  7  2.0 − H  −  0.7 − H  10 f

(7)

where f denotes the breast height form factor estimated as follows.

Fig. 2 Operational system with grapple-loader bunching (A), processor processing (B), and forwarder forwarding (C)

where D denotes DBH (cm). Stem volumes of thinned woods Vn (m3/stem) were estimated with the two-way volume equation (Forestry Agency of Japan 1970).

4Vn  HD 2p   10 000   

(8)

Average Root Mean Squared Errors (RMSEs) between the measured and estimated stem diameters were 1.1 cm for both Japanese cedar and Japanese cypress. Log diameters were usually rounded by 2 cm. Therefore, these RMSEs were within allowable ranges.

2.4 Estimation of extracted volumes Extracted logs were classified as saw logs, logs for laminated lumber, and small-sized logs (Table 2). Saw Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45â&#x20AC;&#x201C;61)

C. Nakahata et al.

Fig. 3 DBH distributions (Left: Survey results, Right: Weibull distribution) Croat. j. for. eng. 35(2014)1

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

logs were 3.00–6.00 m long, with the top-end diameter exceeding 10 cm. Logs for laminated lumber were 2.00 m long, with the top-end diameter exceeding 16 cm. Corresponding values for small-sized logs were 2.00 m and over 3 cm. Small-sized logs were sold to a pellet plant or a chip production factory. Log lengths were assumed to be 2.00 m, 3.00 m, 3.65 m, and 4.00 m for Japanese cedar and 2.00 m, 3.00 m, and 4.00 m for Japanese cypress (interview with the Nasu-machi Forest Owners’ Co-operative). Cutting height was assumed to be 20 cm above the ground (Iehara and Kurokawa 1990). Possible combinations of log length bucked from a thinned wood were estimated, and log volumes v (m3/log) were estimated with log length l (m) and top-end diameters of logs using estimated stem diameter d (cm).

d2l 10 000

Table 2 Log unit prices Species

Diameter

Stand A

Stand B

11–14

69.50

92.70

16–20

116.60

141.00

22–28

112.20

135.10

18–28

127.40

142.90

>30

127.40

150.40

10–14

75.90

120.10

16–20

107.90

125.70

22–28

127.50

133.20

>30

125.90

145.70

11–14

66.40

94.40

16–28

179.80

223.10

>30

170.00

260.40

10–14

94.00

128.40

16–20

197.30

237.50

22–28

225.00

257.30

>30

–

390.50

3.00

3.65

Japanese Cedar

4.00 Saw logs

(9)

Then, extracted volumes Vl (m3/stem) were estimated as follows.

Vl =

∑ v i =1

3.00 Japanese Cypress

(10)

4.00

where vi is the i-th log from the butt end (m3/stem), and n is the number of logs from a thinned wood. Extracted volumes were estimated with possible combinations of log length bucked from a thinned wood. Extracted volumes in the stands VE (m3) were estimated as follows.

VE =

Dmax

∑

j = Dmin

VEj

Logs for laminated lumber

(11) Small-sized logs

where VEj denotes extracted volumes in the j DBH class (m3). VEj was estimated as follows.

VEj = Vlj × N j × A ×

RTj 100

RS 100

Japanese Cedar

2.00

Japanese Cypress

2.00

Japanese Cedar

2.00

Japanese Cypress

16–18

50.00

>20

55.00

16–18

70.00

>20

90.00

30.00

(12)

where Vlj refers to extracted volumes of thinned woods in the j DBH class (m3/stem), and RS is the rate of bunching area to stand area (%).

2.5 Estimation of revenues Log prices p (m3/log) were estimated with log volume and unit prices (Table 2). Unit prices of saw logs were sourced from the unit prices in the Ootawara log market, where the Nasu-machi Forest Owners’ Cooperative sells saw logs. Unit prices of logs for laminated lumber and small-sized logs were sourced from the unit prices of a laminated lumber factory, the pellet

Unit prices, USD/m3

Length

plant, and the chip production factory, where the Nasu-machi Forest Owners’ Co-operative sells logs for laminated lumber and small-sized logs. Then, revenue from a thinned wood was estimated with the number of logs per thinned wood and each log price. Revenues in the stands S (USD) were estimated as follows.

Dmax

∑

j = Dmin

(13)

where Sj denotes revenues in the j DBH class (USD). Sj was estimated as follows. Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

Sj = n

∑

∑ i =1

Pij × N j × A ×

RTj 100

RS 100

Table 3 Standard unit costs for strip road construction (3.5 m wide) (14)

Pij is the log price in the j DBH class (USD/ where stem). i =1 In addition to revenues from log sales, subsidies were estimated with standard unit costs, areas, assessment coefficients, and the subsidy rate offered by the Tochigi Prefectural Government (2010b). Standard unit costs were determined by age class, thinning rates, and whether extraction occurred (Fig. 4). An extraction rate of at least 50% was required. The assessment coefficient and the subsidy rate were assumed to be 1.7 and 4/10, respectively. In Japan, subsidies for strip road establishment are also provided for stands with subsidized thinning operations. Standard unit costs for strip road establishment were determined using the average slope angle (°) and road width (Table 3). Subsidies were estimated with standard unit costs, length, assessment coefficients, and the subsidy rate offered by the Tochigi Prefectural Government (2011). The assessment coefficient and the subsidy rate were assumed to be 1.7 and 4/10, respectively.

2.6 Estimation of revenues Costs in the stands C (USD) were estimated as follows.

Dmax

∑

j = Dmin

OEj + OC

(15)

Fig. 4 Standard unit costs for subsidy for 30% stem thinning rate Croat. j. for. eng. 35(2014)1

C. Nakahata et al.

Slope angle, °

Standard unit costs, USD/m

35 –

43.97

30 –

12.50

25 –

7.02

20 –

4.06

15 –

3.38

10 –

2.81

5–

2.34

where OEj denotes direct expenses in the j DBH class (USD), and OC refers to other expenses (USD). OEj was estimated as follows.

(

)

OEj = OEF + OES × RS VTj + (OEP + OEE ) VEj (16) where OEF denotes chainsaw felling costs (USD/m3); OES, grapple-loader bunching costs (USD/m3); OEP, processing costs (USD/m3); and OEE, forwarding costs (USD/m3). Costs of chainsaw felling and grapple-loader bunching were estimated using the equations in Nakahata et al. (2011). These equations were developed without considering log sizes, because these operations were conducted with whole trees before processing operations, and the costs were not affected by log sizes (Table 4). Costs of processor processing and forwarder forwarding were estimated using the equations developed in Nakahata et al. (2013), which considered log sizes. RMSEs between measured values and estimations using the equations in Nakahata et al. (2013) that considered log sizes, estimations using the equations in Nakahata et al. (2011) and Ishikawa et al. (2008) that did not consider log sizes, were compared. For processing operations, the RMSE of 1.29 USD/m3 obtained from the equations in Nakahata et al. (2013) was lower than that obtained from the equations in Nakahata et al. (2011) (2.14 USD/m3) and Ishikawa et al. (2008) (6.12 USD/m3) (Fig. 5). For forwarding operations, the RMSE of 0.85 USD/m3 using the equations in Nakahata et al. (2013) was also lower than that using the equations in Nakahata et al. (2011) (2.31 USD/m3) (Fig. 6). RMSEs in Nakahata et al. (2013) were the smallest of all estimations. In addition to these direct expenses, strip road expenses, truck transportation expenses, machine transportation expenses, insurance costs, handling fees of the Forest Owners’ Co-operative and the log market,

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

Table 4 Direct expenses, USD/m3 Nakahata et al. (2011)

Nakahata et al. (2013)

Chainsaw felling

Whole tree

0.91/Vn + 1.12

Grapple-loader bunching

Whole tree

12.41

Processor processing

Log

2.88/Vl + 2.66

{(2.42Vla + 0.22) n + 1.90}/Vla · n

Forwarder forwarding

Log

0.0060LF + 7.08

{(-34.30Vla + 16.77) + 0.012 LF}/(1.74Vla + 1.71)

Vn – Stem volume, m3/stem; Vl – Extracted volume, m3/stem; Vla – Average log volume, m3/log; n – Number of logs per stem, logs/stem; LF – Forwarding distance, m.

Fig. 6 Forwarding distances, m and operational costs, USD/m3 backhoe, a grapple-loader, a processor, and two forwarders. Insurance costs were estimated to be 18.4% of direct expenses. Handling fees of the Forest Owners’ Co-operative and the log market were estimated to be 5% and 5%, respectively. Piling fees in the log market were estimated to be 7.00 USD/m3.

2.7 Estimation of economic balances Economic balances were estimated with revenues and costs of possible combinations of log length bucked from a thinned wood. Then, bucking method and extraction rate with maximum profits were determined as the optimal bucking method and the optimal extraction rate, respectively. Extraction rate from stems RES was estimated as follows. 3

Fig. 5 Extracted volumes, m /stem and operational costs, USD/m

and piling fees in the log market were also estimated. Strip road expenses were estimated assuming labor expenses of 25.50 USD/h, backhoe machinery expenses of 46.39 USD/h, and strip road construction times of 117 and 33 hours for Stand A and B, respectively. Truck transportation expenses were estimated to be 13.00 USD/m3 for saw logs, 0 USD/m3 for logs for laminated lumber (because of landing sales), and 15.00 USD/m3 for small-sized logs. Machine transportation expenses were estimated to be 50.00 USD/machine. There were five machines to be transported: a

RES =

Vl × 100 Vn

(17)

Extraction rate on the stands RE was estimated as follows.

RE =

VE × 100 VT

(18)

3. Results and discussion 3.1 Optimal bucking method to maximize profits By the optimum bucking method to maximize profits, maximum four to five logs were made for Stand A and B, respectively (Tables 5, 6, and 7). The Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

C. Nakahata et al.

Table 5 Results of the optimal bucking method to maximize profits in Stand A of Japanese cedar Thinned trees

DBH

Height

Stem volumes

Extracted volumes

Stem/ha

0.210

0.176

0.270

Direct expenses

Economic balance

No. of logs

Revenues

83.6

91.69

43.80

47.89

0.218

80.7

97.40

39.64

57.77

0.320

0.257

80.3

114.49

36.98

77.51

0.400

0.349

87.3

107.05

34.37

72.68

0.490

0.426

87.0

117.41

32.03

85.38

0.550

0.490

89.0

121.33

30.60

90.73

0.660

0.578

87.6

121.49

28.92

92.56

0.740

0.652

88.1

121.97

27.80

94.17

0.870

0.749

86.1

123.82

26.53

97.29

0.960

0.841

87.6

122.67

25.53

97.14

1.050

0.962

91.6

126.57

24.42

102.15

1.210

1.082

89.4

126.67

23.39

103.28

1.330

1.206

90.7

125.86

22.48

103.37

259*

25.9

18.5

0.495

0.429

86.6

3.7

116.91

31.22

85.69

Direct

Economic

m3/stem

Rate %

USD/m3

*Total value. The other values refer to averages

Table 6 Results of the optimal bucking method to maximize profits in Stand A of Japanese cypress Thinned trees

DBH

Height

Stem

Extracted

volumes

volumes 3

m /stem

Rate %

No. of logs

Revenues

expenses

balance 3

USD/m

Stem/ha

0.100

0.000

0.0

–

0.130

0.000

0.0

–

0.180

0.127

70.4

93.38

48.27

45.11

0.220

0.160

72.7

159.81

43.09

116.73

0.290

0.232

79.9

154.66

35.59

116.08

0.340

0.295

86.8

179.67

35.20

144.47

0.430

0.350

81.3

180.35

32.84

147.51

0.490

0.448

91.5

193.73

31.57

162.16

0.560

0.523

93.4

198.89

30.00

168.89

0.670

0.574

85.6

200.30

28.97

171.33

0.770

0.677

87.9

212.11

27.45

184.66

414*

18.8

17.5

0.266

0.195

73.3

2.1

168.04

36.67

131.37

*Total value. The other values refer to averages

Croat. j. for. eng. 35(2014)1

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

Table 7 Results of the optimal bucking method to maximize profits in Stand B of Japanese cedar Thinned trees

DBH

Height

Stem

Extracted

volumes

volumes m3/stem

Rate %

No.

Revenues

of logs

Direct

Economic

expenses

balance

USD/m3

Stem/ha

0.100

0.000

0.0

–

0.140

0.098

69.7

120.49

55.62

64.87

0.190

0.136

71.6

120.00

46.63

73.37

0.250

0.204

81.4

127.55

40.85

86.70

0.320

0.263

82.1

136.56

36.72

99.84

0.410

0.373

91.0

134.32

33.65

100.67

0.500

0.453

90.5

139.97

31.45

108.53

0.610

0.568

93.1

140.02

30.25

109.76

0.690

0.685

99.3

140.71

28.61

112.10

0.820

0.772

94.2

141.16

27.50

113.66

0.940

0.885

94.2

141.31

26.35

114.96

1.090

1.019

93.5

142.57

25.19

117.38

417*

19.0

19.5

0.312

0.262

83.9

2.9

134.35

35.58

98.77

*Total value. The other values refer to averages

average number of logs was 3.7 and 2.1 for the Japanese cedar and the Japanese cypress of Stand A, and 2.9 for Stand B. The number of logs increased according to the DBH. However, thinned trees with DBH values between 12 and 14 cm for Stand A and of 12 cm for Stand B were not bucked into logs (i.e., they were left in the forests), because these thinned trees with small DBH were not profitable with the optimum bucking method for maximizing profits. The average extraction rates from stems were 86.6% and 73.3% for Japanese cedar and Japanese cypress of Stand A, and 83.9% for Stand B. The average extraction rates from stems also increased according to the DBH (Fig. 7). Revenues increased and direct expenses decreased according to the DBH (Tables 5, 6, and 7 and Fig. 8). Therefore, economic balances increased according to the DBH. Despite smaller log volumes and lower extraction rates, the average revenues for the Japanese cypress of Stand A were higher than those for the Japanese cedar of Stands A and B, because of the former’s higher unit prices. The average direct expenses for the Japanese cypress of Stand A were a little higher than those for the Japanese cedar of Stands A and B, but almost same. Therefore, the average economic balances for the Japanese cypress of Stand A were higher than those for the Japanese cedar of Stands A and B.

Optimal extraction rates were 70.3% and 37.9% for Stand A and B, respectively. Extracted volumes with these rates were 86.48 m3/ha and 47.77 m3/ha, respectively (Fig. 9). Conversely, actual extraction rates and volumes were 69.3% and 36.9%, or 85.25 m3/ha and 46.43 m3/ha for Stand A and B, respectively. Estimated optimal extraction rates and volumes were similar to the actual values. However, small-sized log volumes differed. The actual values were 1.79 m3/ha and 1.76 m3/ha, although no small-sized logs were estimated with the optimal extraction rate, because the profitability of extracting small-sized logs was low. However, the Nasu-machi Forest Owners’ Co-operative had an annual contract with a pellet production factory to transport a certain log volume. Furthermore, estimated and actual log volumes for laminated lumber also differed. The actual values were larger than the estimations, because the Nasu-machi Forest Owners’ Co-operative also had an annual contract with a laminated lumber production factory at log prices, which were more stable than those in log markets. Fig. 10 shows revenues, costs, and economic balances for Stands A and B. Estimated revenues were larger than actual values, because estimated volumes of saw logs, which were relatively higher priced, were larger than the actual values, and log quality Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45â&#x20AC;&#x201C;61)

C. Nakahata et al.

Fig. 8 DBH, cm; and revenues, direct expenses, and economic balances, USD/m3 in Stand A of Japanese cypress

was not considered in this estimation. Estimated costs were smaller than actual costs, because actual loss times would be larger than 1/4 of productive time as assumed in the study (Kamiiizaka and Kanzaki 1990, Nakahata et al. 2013). Therefore, estimated economic balances were larger than actual values.

3.2 Comparison with the optimal bucking method without considering log sizes

Fig. 7 DBH, cm and extraction rates from stems, % Croat. j. for. eng. 35(2014)1

In the previous section, the optimal bucking method to maximize profits was determined using equations developed in Nakahata et al. (2013), which con-

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45â&#x20AC;&#x201C;61)

Fig. 9 Extracted volumes (M: Measurement, 1: The optimum bucking method, 2: Without considering log sizes, 3: With maximum revenues, 4: With the new subsidy, 5: Unit prices of small-sized logs, 40.80 USD/m3, 6: Unit prices of small-sized logs, 68.00 USD/m3)

Fig. 10 Revenues, costs, and economic balances per extracted volume (M: Measurement, 1: The optimum bucking method, 2: Without considering log sizes, 3: With maximum revenues, 4: With the new subsidy, 5: Unit prices of small-sized logs, 40.80 USD/m3, 6: Unit prices of small-sized logs, 68.00 USD/m3) sidered log sizes. This section determines the optimal bucking method to maximize profits without considering log sizes using equations developed in Nakahata et al. (2011) and compares the two optimal buck-

ing methods. The average extraction rates from stems without considering log sizes were 89.3% and 81.2% for the Japanese cedar and the Japanese cypress of Stand A, and 92.2% for Stand B. The average extraction Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

Fig. 11 DBH, cm and direct expenses, USD/m3 in Stand A of Japanese cypress

rates from stems without considering log sizes were higher. Notably, the differences became larger with smaller DBHs (Fig. 7), because direct expenses of small-sized logs did not increase when log sizes were not considered; subsequently, small-sized logs tended to be extracted although the differences in direct expenses were small between the two cases (Fig. 11). For example, only five logs could be bucked maximally from the thinned woods in the optimal bucking method considering log sizes. On the other hand, maximum seven logs were bucked from the thinned woods by the bucking method that did not consider log sizes. Optimal extraction rates without taking log size into consideration were 73.5% and 41.1% for Stand A and B, respectively (Fig. 9). The corresponding extracted volumes with optimal extraction rates were 90.44 m3/ha and 51.83 m3/ha. Small-sized log volumes were 4.03 m3/ha and 4.04 m3/ha for Stand A and B, respectively. These were larger than the actual values and estimations made by the optimal bucking method considering log sizes. However, revenues per m3 decreased, and subsequently, economic balances per m3 also decreased (Fig. 10).

3.3 Comparison with the optimal bucking method to maximize revenues The optimal bucking method usually maximizes revenues and is used to estimate revenues (Iehara and Kurokawa 1990). Here, the optimal bucking method Croat. j. for. eng. 35(2014)1

C. Nakahata et al.

to maximize revenues is determined and compared. The average extraction rates from stems with maximum revenues were 90.0% and 90.0% for the Japanese cedar and the Japanese cypress of Stand A, respectively, and 94.2% for Stand B. The average extraction rates from stems with maximum revenues increased compared to those from stems with maximum profits. Notably, the differences became larger for smaller DBHs (Fig. 7), because revenues increased according to extracted volumes. Furthermore, thinned trees with a DBH between 12 and 14 cm for Stand A, and of 12 cm for Stand B, were extracted with the optimal bucking method to maximize revenues. As with the optimum bucking method to maximize profits, these were not bucked into logs (i.e., they were left in the forests), because these thinned trees with small DBH were not profitable. Therefore, for the optimal bucking method to maximize revenues, direct expenses exceeded revenues, and economic balances suffered a deficit with small DBHs (Fig. 8). Extraction rates with maximum revenues were estimated to be 76.0% and 41.4% for Stand A and B, respectively (Fig. 9). Corresponding values for extracted volumes with maximum revenues were 93.50 m3/ha and 52.22 m3/ha. Small-sized log volumes were 6.14 m3/ha and 4.26 m3/ha for Stand A and B, respectively. These were larger than the actual values and estimations with maximum profits. Therefore, revenues per ha increased according to extracted volumes. However, costs per ha also increased according to extracted volumes. Subsequently, the estimated economic balances decreased. On the other hand, revenues per m3 decreased and subsequently, economic balances per m3 also decreased (Fig. 10).

3.4 Comparison with the optimal bucking method considering the new subsidy system The subsidy system was changed in 2011. Until 2010, standard unit costs (Tochigi Prefectural Government 2010b) were determined by age class, thinning rates, and extraction, if any (Fig. 4). An extraction rate of at least 50% was required. In the new subsidy system, standard unit costs (Tochigi Prefectural Government 2011) were determined by operation system, thinning rate, and extracted volumes (Fig. 12). Operation systems were classified into two types: ground-based and skyline system. The sites in this study use the ground-based operation system and have a thinning rate of 30%. In the new subsidy system, the subsidy was increased according to extracted volumes. The average extraction rates from stems with the new subsidy system were 89.3% and 79.1% for the

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45â&#x20AC;&#x201C;61)

Fig. 12 Standard unit costs for the new subsidy with a groundbased system Japanese cedar and the Japanese cypress of Stand A, and 88.3 % for Stand B. These rates were higher compared to those for the previous subsidy system, because the subsidy increased according to extracted volumes (Fig. 13). Extraction rates from stands with the new subsidy system in place were estimated to be 73.3% and 39.9% for Stand A and B, respectively (Fig. 9). Corresponding extracted volumes were 90.12 m3/ha and 50.29 m3/ha. Small-sized log volumes were 3.63 m3/ha and 2.51 m3/ha for Stand A and B, respectively. Smallsized log volumes with the new subsidy system increased compared to those in the previous subsidy system. Consequently, the economic balances increased with the increase in extracted volumes, especially in Stand B (Fig. 10).

3.5 Comparison of optimal bucking methods with higher unit prices for small-sized logs In addition to the unit prices for small-sized logs: 30.00 USD/m3, the following unit prices were examined assuming no subsidy for thinning operations and strip road establishments: 40.80 USD/m3 and 68.00 USD/m3. These prices would be converted using a bulk density of 0.68 ton/m3 (Mikamo Forest Ownersâ&#x20AC;&#x2122; Co-operative and Sumitomo Osaka Cement Company 2008) from 60.00 USD/ton with additional subsidies to small-sized log prices (Japanese Forestry Investigation Committee 2011b) and 100.00 USD/ton if FITs were introduced in Japan (Japanese Forestry Investigation Committee 2011a).

Fig. 13 DBH, cm and extraction rates from stems, % with the new subsidy and higher unit prices Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

The average extraction rates from stems with the unit price 40.80 USD/m3 were 86.7% and 73.3% for the Japanese cedar and the Japanese cypress of Stand A, and 83.9% for Stand B. The average extraction rates from stems with the unit price 40.80 USD/m3 were almost the same as those from stems with the unit price 30.00 USD/m3 excluding DBH higher than 38 cm for the Japanese cedar of Stand A (Fig. 13). Extraction rates from stands with the unit price 40.80 USD/m3 were estimated to be 70.4% and 37.9%, and corresponding extracted volumes were 86.60 m3/ha and 47.77 m3/ha. Small-sized log volumes were 0.12 m3/ha and 0.00 m3/ha for Stand A and B, respectively (Fig. 9). Small-sized log volumes increased marginally to 0.12 m3/ha with 40.80 USD/m3 for Stand A; subsequently, extraction rates from stands and extracted volumes also increased at almost the same rate. Revenues of 40.80 USD/m3 were relatively higher. However, economic balances decreased in the absence of a subsidy (Fig. 10). The average extraction rates from stems with the unit price 68.00 USD/m3 were 91.4% and 81.5% for the Japanese cedar and the Japanese cypress of Stand A, and 91.5% for Stand B. These rates are higher compared to those from stems with the unit price 30.00 USD/m3. Notably, the differences became larger for smaller DBHs (Fig. 13). Extraction rates from stands with the unit price 68.00 USD/m3 were estimated to be 74.5% and 40.4% for Stand A and B, respectively, and corresponding extracted volumes were 91.60 m3/ha and 50.95 m3/ha. Small-sized log volumes were 12.14 m3/ha and 3.17 m3/ha for Stand A and B, respectively (Fig. 9). These increased in line with unit price increases for small-sized logs. Notably, small-sized log volumes in Stand A were the largest of all estimations. However, economic balances decreased in the absence of a subsidy (Fig. 10).

C. Nakahata et al.

tions in the stands. Stand in this study were about 50 years old, with an average DBH of about 20 cm. Among the estimations, differences of extraction rates from stems with DBH exceeding 20 cm were small. However, extraction rates from stems with a DBH less than 20 cm were significantly different. Therefore, it is likely that the optimal bucking method to maximize profits while considering log sizes can help determine the optimal extraction rates of younger stands with smaller DBH. Additional studies for different forests and forestry conditions would be useful. Small-sized log volumes with the optimal bucking method were smaller than the actual values, because this model did not consider contracts with factories concerning annual supply obligations and stable prices. Furthermore, this model did not consider log quality, and bucking operations should be conducted also taking log quality into consideration. Low-quality logs were sold to chip or pellet factories rather than to log markets. Therefore, future studies should consider supply obligations and log quality, although the latter can be difficult to predict. Effects of the new subsidy system and higher unit prices for small-sized logs were also discussed. The new subsidy system and unit prices with FITs both contributed to increased extracted volumes of smallsized logs. However, it is likely that subsidies will be reduced in the future, due to the government’s budget constraints. Therefore, a low-cost harvesting system should be developed to establish forest road networks and improve forestry operation systems.

Acknowledgements We are grateful to the Nasu-machi Forest Owners’ Co-operative for their cooperation during the course of this research.

4. Conclusions Optimal bucking methods to maximize profits were applied to two operational sites of the Nasumachi Forest Owners’ Co-operative, and optimum extraction rates were examined with different optimal bucking methods and small-sized log unit prices. Extraction rates from stands with maximum profits were similar to the actual values, compared with the optimal bucking method that does not consider log sizes and the optimal bucking method to maximize revenues. However, differences in extracted volumes and economic balances among these estimations were small and are likely to fall within the error limits traditionally seen for forests and forestry condiCroat. j. for. eng. 35(2014)1

5. References Akay, A. E., Sessions, J., Serin, H., Pak, M., Yenilmez, N., 2010: Applying optimum bucking method in producing Taurus Fir (Abies cilicica) logs in Mediterranean region of Turkey. Baltic Forestry 16(2): 273–279. Aruga, K., Hiyamizu, G., Nakahata, C., Saito, M., 2013: Effects of aggregating forests, establishing forest road networks, and mechanization on operational efficiency and costs in a mountainous region in Japan. Journal of Forestry Research 24(4): 747–754. Agency for Natural Resources and Energy, 2012: Settlement of the details of the Feed-in Tariff scheme for renewable en-

C. Nakahata et al.

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61)

ergy, including purchase price and surcharge rates. http:// www.meti.go.jp/english/press/2012/0618_01.html (Accessed on June 18, 2012). Forestry Agency of Japan, 1970: Ryuboku Kanzaiseki Hyo – Higashi Nihon Hen (Stem volume table - East Japan). Japanese Forestry Investigation Committee, Tokyo.

Nakahata, C., Aruga, K., Takei, Y., Yamaguchi, R., Ito, K., Murakami, A., Saito, M., Tasaka, T., Kanetsuki, K., 2011: Improvement on operational efficiencies and costs of extracting thinned woods using a processor and a forwarder in Nasunogahara area (II) – Based on comparative analyses of current operations and mechanized operations. Bulletin of Utsunomiya University Forest 47: 27–34.

Haynes, H. J. G., Visser, R. J. M., 2004: An applied hardwood value recovery study in the Appalachian Region of Virginia and West Virginia. International Journal of Forest Engineering 15(1): 25–31.

Nakahata, C., Aruga, K., Takei, Y., Yamaguchi, R., Saito, M., Kanetsuki, K., 2013: Examining the optimum extraction rate of extracting thinned woods in Nasunogahara area. Journal of Japan Forest Engineering Socciety 28(1): 17–28.

Iehara, T., Kurokawa, Y., 1990: An economic evaluation for reforestation with Hinoki on low-productivity forest sites. Journal of Japan Forest Society 72(1): 34–45.

Nakajima, T., Matsumoto, M., Tatsuhara, S., 2008: Application of an algorithm to maximize stumpage price. Kanto Forest Research 59: 55–58.

Inoue, A., Kurokawa. Y., 2001: A new method for estimating relative stem profile equations: Application to system yield tables. Journal of Japan Forest Society 83(1): 1–4.

Nakajima, T., Matsumoto, M., Tatsuhara, S., 2009: Development and application of an algorithm to calculate crosscutting patterns to maximize stumpage price based on timber market and stand conditions: A case study of Sugi plantations in Gunma Prefecture. Japan Journal of Forest Planning 15: 21–27.

Ishikawa, T., Tsujibata, T., Matsushita, A., Itaya, A., Hamamoto, K., Tsujibata, T., 2008: Operation analysis and improvement of a logging system using high-performance forestry machines in a mature forest. Journal of Japan Forest Engineering Society 23(2): 53–62. Japanese Forestry Investigation Committee, 2011a: Feed-in Tariff (FIT) for wood chip for power generation could improve the unit price of logging residues to 20 yen/kWh and promote effective utilization of logging residues. Rinsei (Forest Policy) News 405: 7. Japanese Forestry Investigation Committee, 2011b: Tosa no mori system, in which thinned trees are traded with community currency for 6,000 yen/ton, begins in Chizu Town, Tottori Prefecture and the system has been successful. Rinsei (Forest Policy) News 409: 14. Kamiiizaka, M., Kanzaki, K., 1990: Shinrin Sagyo Gaku (Forest operation systems), Buneido Publishing, Tokyo. Kinashi, K., 1978: Shinrin Tyosa Shosetsu (Forest inventory). Norin (Agriculture and Forest) Publishing, Tokyo. Kivinen, V. P., 2004: A genetic algorithm approach to tree bucking optimization. Forest Science 50(5): 696–710. Laroze, A. J., Greber, B. J., 1997: Using tabu search to generate stand-level, rule-based bucking patterns. Forest Science 43(2): 367–379. Laroze, A. J., 1999: A linear programming, tabu search method for solving forest-level bucking optimization. Forest Science 45 (1): 108–116. Mikamo Forest Owners’ Co-operative and Sumitomo Osaka Cement Company, 2008: Research report on extracting forest biomass resources. Sano, Tochigi. Nagumo, H., Shiraishi, N., Tanaka, M., 1981: Systematic method for constructing a yield table for Sugi even-aged stand: A case study in experimental plots of the Tokyo University Forest in Chiba. Bulletin of Tokyo University Forest 71: 269–330.

Nasberg, M., 1985: Mathematical programming models for optimal log bucking. Linkoping Studies in Sci. and Tech. Dissertation No. 132. Dept. of Mathematics, Linkoping Univ., Sweden. 199 pp. Olsen, E., Pilkerton, S., Garland, J., Sessions, J., 1991: Computer-aided bucking on a mechanized harvester. Journal of Forest Engineering 2(2): 25–32. Sessions, J., 1988: Making better tree bucking decisions in the woods: an introduction to optimal bucking. Journal of Forestry 86(10): 43–45. Sessions, J., Garland, J., Olsen, E., 1989a: Testing computeraided bucking at the stump. Journal of Forestry 87(4): 43–46. Sessions, J., Olsen, E., Garland, J., 1989b: Tree bucking for optimal stand value with log allocation constraints. Forest Science 35(1): 271–276. Tochigi Prefectural Government, 2009: Forest and forestry statistics of fiscal year 2009 in Tochigi Prefecture. Utsunomiya, Tochigi. Tochigi Prefectural Government, 2010a: Tochigi no genkina moridukuri prefectural tax (to make healthy forests with subsidies for thinning operations) - Business evaluation report for fiscal year 2009. Utsunomiya, Tochigi. Tochigi Prefectural Government, 2010b: Forestation program - Standard unit cost table for fiscal year 2010. Utsunomiya, Tochigi. Tochigi Prefectural Government, 2011: Forestation program - Standard unit cost table for fiscal year 2011. Utsunomiya, Tochigi. Uusitalo, J., 2007: Forest-level bucking optimization including transportation cost, product demands and stand characteristics. The 3rd Forest Engineering Conference. October 1–4, in Mont-Tremblant, Quebec, Canada. Croat. j. for. eng. 35(2014)1

Examining the Optimal Bucking Method to Maximize Profits in Commercial Operations ... (45–61) Wang, J., Liu, J., LeDoux, C. B., 2009: A three-dimensional bucking system for optimal bucking of central Appalachian hardwoods. International Journal of Forest Engineering 20(2): 26–35.

C. Nakahata et al.

Yoshida, S., Imada, T., 1989: Stem cross-cutting algorithm based on log unit price. Bulletin of Faculty Agriculture, Kagoshima University 39: 319–329.

Authors’ address:

Received: July 26, 2013 Accepted: February 3, 2014 Croat. j. for. eng. 35(2014)1

Graduate Student, Chikara Nakahata, MSc. e-mail: c.nakahata0927@gmail.com Assoc. Prof., Kazuhiro Aruga, PhD.* e-mail: aruga@cc.utsunomiya-u.ac.jp Assist. Prof., Masashi Saito, PhD. e-mail: m_saito@shinshu-u.ac.jp Department of Forest Science, Faculty of Agriculture, Utsunomiya University, 350 Mine, Utsunomiya, Tochigi 321-8505, JAPAN *Corresponding author

Original scientific paper

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations Miroslav Kleibl, Radomir KlvaÄ?, Carolina Lombardini, Josef Porhaly, Raffaele Spinelli Abstract This study gauged the severity and permanence of soil compaction associated with mechanized clear felling of umbrella pine plantations. We tested three treatments: not harvested, harvested one year earlier and harvested six years earlier. Each treatment was replicated eight times in randomly distributed 0.5 ha plots, on the same soil type. Soil compaction was assessed by gauging soil bulk density, penetration resistance, deflection under impact and CO2 concentration. These parameters were measured with steel rings, penetrometer, deflectometer and soil air analyzer, respectively. Measurements were conducted on 8 clear cut blocks per treatment, which had been randomly distributed over the same forest, with identical soil and stand type. One year after clear fell, bulk density increased by 9%, penetration resistance by 50% and deflection by 60%. Porosity decreased by 10%, which entailed a parallel 30% increase of both soil moisture content and CO2 concentration in the soil air. Six years after clear fell, there was no sign of recovery for bulk density, deflection and moisture content. On the contrary, penetration resistance was significantly reduced, and CO2 concentration was back to the values recorded in plots that had not been harvested. Keywords: harvesting, disturbance, clear fell, impact

1. Introduction Rising labour cost and global competition have eroded the economic sustainability of traditional wood harvesting technology. Forest management is increasingly mechanized in all industrial countries, where animal power (Magagnotti and Spinelli 2011) or small scale forest technology (VusiÄ&#x2021; et al. 2013) are only profitable under specific circumstances. In turn, the rapid progress of mechanized harvesting has brought about an increased awareness of the potential site impact generated by industrial forest technology, and by forest operations in general. Many forest owners fear that the large size and heavy weight of modern machinery may determine a significant increase of stand and soil impacts, compared to traditional motor manual operations (Vokoun et al. 2006). Foresters feel especially uneasy about the difficulty in detecting and predicting soil compaction, whose occurrence may elude visual inspection, at least Croat. j. for. eng. 35(2014)1

initially and until its consequences become apparent (Horn et al. 2004). Traffic is particularly heavy in clear fell operations, due to the larger size of machinery and larger removal (Marchi et al. 2011). Soil compaction may cause physiological stress in the tree or seedling, reducing its ability to cope with adverse climatic conditions and/or to compete with other vegetation. Fortunately, physical and biological agents will eventually loosen up the soil, leading to recovery. However, this process may take considerable time, which is particularly worrying for intensely managed stands, entered several times during a rotation (Grigal 2000). The occurrence, severity and persistence of soil compaction are very difficult to predict, since they are the result of a complex interaction of harvest, soil and stand characteristics (Ampoorter et al. 2012). That prevents extrapolating the results of individual studies, as well as formulating generally valid guidelines. Relatively few compaction studies have been conducted in the Mediterranean region (Gondard et al. 2003),

M. Kleibl et al.

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

partly due to the slower progress of mechanized harvesting. Nevertheless, harvesting operations in Mediterranean forests are often associated with widespread soil disturbance (Spinelli et al. 2010) and compaction (Magagnotti et al. 2012). Matters become especially complex in coastal pine plantations under multifunctional management, with the multiple goals of timber production, extraction of non-timber products (pine nuts) and recreation (Carrasquinho et al. 2010). On one hand, the forest soil is impacted by more activities than just timber harvesting, which may confound the effects; on the other hand, the impact of harvesting is under severe scrutiny due to the intense public frequentation. Therefore, the goals of this study were: Þ to quantify the extent of soil compaction associated to the clear felling of coastal pine plantations; Þ to gauge soil recovery years after clear felling.

2. Materials A trial was carried out in the Regional Park of San Rossore, near Pisa, on the Tyrrhenian coast. The park encloses a surface of about 3 000 ha and is mostly covered by pine plantations (Spinelli et al. 2009). All forests inside the park are protected and under a special management regime. Careful exploitation is conducted according to sustainable management rules. All harvesting is completely mechanized, with the intent of enhancing worker safety and minimizing operation residence time, to the benefit of the intense recreational use. The trial was conducted in October 2012 on 24 plots, equally divided into the following three treatments: not harvested, clear felled one year earlier (2011) and clear felled six years earlier (2005). All plots represented mature umbrella pine (Pinus pinea L.) plantations, with an age of about 100 years. Stand density and stocking were in the range of 200 trees ha-1 and 320 tons ha-1, respectively. Umbrella pine trees grew on loamy sand, developed over a quaternary dune just a few kilometers from the present coastline. Under these conditions, soil drainage characteristics depend on the micro relief: the old dune tops drain very easily, whereas the small hollows between them tend to retain water and fill with clay. For this reason, pine is only planted on the dune tops, while the hollows are left to the natural regeneration of hygrophilous hardwood species. Test plot selection was done after consulting the local soil map, in order to probe plots growing on exactly the same soil type, in this case a Typic Udipsamment (USDA 1999). Soil texture was sandy (86% sand, 3% silt, 11% clay), with an organic matter

content of about 3%. Each plot represented one clear cut block, with a surface of about 5 000 m2. Plots were laid out in an alternate sequence because of the prescribed felling pattern, aimed at maintaining some form of lateral cover to mitigate visual impact and prevent windthrow in non-harvest areas (Spinelli et al. 2013). Control plots were represented by the blocks that had not been clear felled yet. All 16 cut plots were clear felled with the same method and technology, applied by the same company and the same machine operators. Trees were felled with a 27 tons tracked feller buncher, equipped with a high speed disc saw (hot saw). The feller buncher also performed a rough debranching and crosscutting, using a special articulated joint on the boom, which allowed turning and tilting the disc saw. Basal logs were crosscut in 4 to 5 m random lengths, and extracted to roadside using an 8 wheel drive forwarder, with a 14 tonne load capacity. Branches and tops where chipped inside the blocks using a forwarder mounted chipper, powered by a 350 kW independent engine. Chips were discharged into three axle silage trailers with a 10 tonne payload capacity, towed by 100 kW four wheel drive farm tractors, for extraction to the roadside. The total weight of loaded machines was: 27 tons for the tracked feller buncher, 30 tons for the loaded forwarder, 30 tons for the forwarder mounted chippers and 22 tons for the tractor and trailer chip shuttling units. All forwarders were equipped with 700 mm tires, inflated at a 450 kPa. Tractor trailers were equipped with 380 mm tires, also inflated at 450 kPa. At the time of data collection, the plots clear felled in 2005 had already been replanted with pine seedling, set 3.5 m apart. Plantation had been conducted manually, and consisted in the localized opening of a fissure for inserting the seedling. The soil had not been tilled or disturbed by machinery during planting operations, and it was invariably covered by a dense brush layer about 1 m tall. The plots clear felled in 2011 had not been replanted yet, but the brush was starting to come up, although not as densely as in the 2005 plots. In both cases, it was very difficult to detect any tracks, or other visible signs of machine traffic, except for occasional diffused wood chip spills, indicating the stations of the forwarder mounted chipper. Therefore, it was impossible to differentiate between inside track and outside track sample points, and we opted for a diffused systematic sample pattern.

3. Methods Soil compaction was determined concurrently with four different methods, with the goal of implementing Croat. j. for. eng. 35(2014)1

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

a robust experimental setup, capable of internal corroboration and multiple detection capacity, where the effects that may elude one method are captured by the others. Therefore, soil compaction was gauged through: Þ bulk density; Þ penetration resistance; Þ soil deflection; Þ soil CO2 concentration. Core samples were collected in rings of thin walled stainless steel tubing, with an internal diameter of 8 cm and a height of 5 cm, corresponding to a volume of 250 cm3. Rings were pushed into the soil down to a 10 cm depth, after removing the litter layer and the first 5 cm of soil, where litter elements could be mixed, which could bias the measurements. Besides, this most superficial layer could have been affected by tire slippage, which would loosen the soil rather than compact it, thus potentially masking the compaction caused by machine traffic. Rings were then removed from the soil, for trimming the sample and placing it into a sealed plastic bag. Bags were taken to the laboratory and weighed before and after oven drying at 105°C for 48 hours. Finally, samples were placed in a picnometer. The resulting figures were used to calculate: bulk density (BD), solid density (D), gravimetric water content and porosity, for each sample. A total of 240 cores were collected, i.e. 10 for each test plot. This method was only used to explore the first 10 cm soil layer, where the main impacts are often concentrated, at least in Mediterranean and sub Mediterranean soils (Makineci et al. 2007, Picchio et al. 2012). Penetration resistance was measured with a Eijkelkamp Penetrologger cone penetrometer (www.Eijkelkamp.com), on 50 points per plot, for a total of 1 200 measurements. The cone used for the tests had a 1 cm2 base area and 60° top angle. Penetration rate was approximately 2 cm per second – with equal pressure exerted onto both handles. The instrument automatically recalculated the penetration force from an inbuilt pressure gauge, and recorded data in MPa at one centimeter depth intervals. Penetration resistance was measured up to a 40 cm depth. After completing measurements on one plot, the cone was checked with calliper (provided by manufacturer) in order to determine if the abrasive effect of sand had reduced its base area below acceptable limits for reliable measurement. Worn cones were replaced before sampling the next plot. All penetration tests were conducted during several consecutive days, in order to minimize variations of soil moisture content, which could have biased the results. Soil deflection was measured with a portable falling weight deflectometer (PFWD) Loadman II (www. Croat. j. for. eng. 35(2014)1

M. Kleibl et al.

al-engineering.fi), on 20 points per plot. The Loadman II PFWD was specifically developed for measuring the rigidity of road pavements, but it was successfully tested for measuring soil bearing capacity and compaction (Klvac et al. 2010). Litter and the first 5 cm of soil were removed before measurement, in order to increase accuracy. Prior to the first measurement, the instrument was calibrated according to the size of the reaction base plate. The diameter of the reaction base plate was 132 mm and the calibration module of elasticity was chosen to be E 160, as advised by the manufacturer. The falling weight induced a non-destructive shock wave spreading in the soil, and evoking a reaction according to soil properties. Soil reaction was measured through accelerometers built into the reaction base plate of the instrument. Deflection was calculated in millimeters and increased with the degree of compaction. Softer uncompacted soil absorbed a larger portion of the energy released by the falling weight, so that deflection was lower. CO2 concentration was measured on 10 points per sample plot, using a portable Carbocap GM70 device (www.vaisala.com), fitted with Carbocap GMP221 sensors. Soil air was extracted from a depth of about 10 cm using a cylinder probe, and it was analyzed on the spot with the inbuilt GMP221 sensors. On each measurement point, readings were taken at 1 minute intervals after letting the instrument stabilize for 10 minutes. The 15 minute reading was accepted into the analysis as the reference. Sample points were evenly distributed across the test plots, using a fixed sampling distance calculated on the basis of the number of sample points and the size of each individual plot. Data were analyzed with the Statview advanced statistics software. In particular, the software was used for performing unpaired t tests and ANOVA post-hoc tests. Distribution histograms were drawn in order to check whether the distribution of experimental data met the normality assumption. Square root and LOG transformations were applied to normalize data sets that violated the normality assumption. The relation between penetration depth and penetration resistance was estimated with GraphPad Prism 5, after eliminating outliers with the ROUT method (Motulsky and Brown 2006).

4. Results The main results of the study are reported in Table 1. Soil characteristics in clear felled plots were significantly different from those in non-harvested plots. All measures indicated the presence of com-

M. Kleibl et al.

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63â&#x20AC;&#x201C;71)

Table 1 Main results of the study Not harvested

OBS N 3

Mean

Clear cutting in 2005

Clear cutting in 2011

Mean

Bulk density, g/cm

240

1.277

0.134

1.375

0.103

1.391

0.158

Porosity, %

240

50.9a

4.5

47.7b

3.3

46.0b

4.5

240

4.5

2.0

5.4

2.6

5.9

4.5

Penetration resistance at 10 cm depth, MPa

1 200

1.851a

0.772

2.152b

0.901

2.824c

1.128

Penetration resistance at 20 cm depth, MPa

1 200

2.676a

1.110

3.305b

1.284

4.028c

1.402

1 200

1.163

1.502

4.661c

1.391

1.270

Moisture content, %

Penetration resistance at 30 cm depth, MPa Penetration resistance at 40 cm depth, MPa Soil deflection, mm CO2 concentration, ppm

3.147

3.952

1 100

3.589

1.154

4.330

1.474

4.967

480

5.8a

3.6

9.6b

3.7

9.3b

240

2 375.1

691.0

2 152.1

622.4

3.7 b

3 133.0

1 312.5

Note: SD = Standard Deviation; different letters on the mean values indicate that differences between treatments (figures in the same rows) are statistically significant at the 5% level; OBS = observations; N = number of observations, which is equally distributed among the three treatments

paction. Compared to non-harvested plots, recently clear felled plots had a 9% higher density, a 50% higher penetration resistance and a 60% higher deflection. Porosity in recently clear felled plots was 10% lower than in non-harvested plots, and both soil moisture content and CO2 concentration in the soil air were 30% higher. Plots clear felled six years earlier showed the same differences with non-harvested plots for what concerned bulk density, deflection and moisture content. On the contrary, differences in penetration resistance were significantly smaller, although this parameter was still about 15 to 25% higher than in non-harvested plots. Differences in soil porosity were smaller than for recently clear felled plots, although the statistical significance of differences between recent and older clear fells was borderline (p = 0.066). What is most important, CO2 concentration was the same as in non-harvested plots. Table 2 shows the results of the analysis of variance conducted on the study data. The ANOVA shows the very high random variation in the data, which is consistent with a diffused sampling. Sampling points were likely to hit trafficked as well as intact areas within the same sample plot, which explains high random variation. At the same time, sampling intensity was high enough to capture significant differences between treatments, as shown by the very low p Values. Fig. 1 shows the relationship between penetration resistance and soil depth. Resistance was higher for the clear felled plots, and increased with depth, also due to increased probe friction. In the recently harvested plots, resistance at the 40 cm depth almost reached 6 MPa, which is the critical value indicating

Fig. 1 Relationship between penetration resistance and soil depth

compaction of sandy soil (Lhotsky 2000). Table 3 shows the main parameters for the three regression curves, which have a very good coefficient of determination, also due to the effective elimination of outliers through the ROUT method. Finally, soil moisture content was significantly higher in the clear felled plots than in the control plots. Higher water retention in the clear felled plots could be related to soil compaction, as well as to evapotranspiration reduction due to the absence of a mature stand. Croat. j. for. eng. 35(2014)1

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

M. Kleibl et al.

Table 2 ANOVA table for the main results of the study SS

F value

P value

Bulk density, g/cm Treatment Residual

0.587

0.13

16.502

<0.0001

230

4.088

0.87

–

504.415

0.20

14.625

<0.0001

1 983.201

0.80

–

Porosity, % Treatment Residual

115

Moisture content (%) square root transformed Treatment Residual

2.708

0.04

4.419

<0.0001

230

70.484

0.96

–

Penetration resistance at 10 cm depth (MPa) log transformed Treatment Residual

6.942

0.15

102.545

<0.0001

1 197

40.519

0.85

–

Penetration resistance at 20 cm depth (MPa) square root transformed Treatment

Residual

1 197

27.257

0.15

104.590

<0.0001

155.974

0.85

–

Penetration resistance at 30 cm depth, MPa Treatment Residual

452.136

0.17

122.413

<0.0001

1 197

2 182.880

0.83

–

Penetration resistance at 40 cm depth, MPa Treatment Residual

352.131

0.16

103.310

<0.0001

1 110

1 891.719

0.84

–

1 196.406

0.15

44.307

<0.0001

6 764.088

0.85

–

Soil deflection, mm Treatment Residual

501

CO2 concentration (ppm) log transformed Treatment Residual

0.935

0.18

26.074

<0.0001

231

4.143

0.82

–

Table 3 Main statistics for the regression models shown in Fig. 1 Model type: PR = a x Db Parameter

Coefficient

0.942

0.054

0.468

0.018

95% CI

Valid observations

Outliers

0.972

16 098

0.984

16 240

0.986

16 274

149

Clear cuttings in 2011 0.833 to 1.051 0.432 to 0.504 Clear cuttings in 2005 a

0.657

0.032

0.592 to 0.722

0.526

0.015

0.495 to 0.556 Not harvested

0.628

0.025

0.577 to 0.679

0.477

0.012

0.454 to 0.502

Note: PR = Penetration resistance (MPa); D = Depth, cm; SE = Standard error; CI = Confidence interval. The effect of the independent variable is significant at the 1% level.

Croat. j. for. eng. 35(2014)1

M. Kleibl et al.

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

5. Discussion In this study, we have used a spatial assumption to represent a time trend. In particular, we have used different plots to represent the state of forest soil at different times with respect to final clear fell. Of course, that is an approximation. The effect of local stand and soil variability could introduce a bias, capable of weakening the results of our study. The same can be said for weather conditions at the time of harvest, which may have differed between the 2005 and the 2011 events. However, we controlled these sources of error by choosing a large number of sample plots in an alternating sequence, with the purpose of spreading local variability equally over all treatments. Furthermore, soil and stand characteristics were very homogeneous, which may support the original assumption of a very limited local variability. Concerning weather at the time of harvest, this factor was controlled by choosing operations that were conducted during the same season, and by expanding the number of study plots over a wide area that would take many weeks to harvest. However, it should be taken into consideration that our definition of »recovery« is based on this assumption, and not on a long term study that recorded soil characteristics in the same plots, over the years. Furthermore, this study differs from most soil compaction studies (e.g. Sakai et al. 2008, Gerasimov and Katarov 2010, Majnounian and Jourgholami 2013) for its choice of a diffused sampling design. Other studies often adopt a localized sampling design, where paired samples are collected inside and outside machine tracks. However, these studies offer a very limited picture of the overall impact, unless they quantify the proportion of the total area covered by the tracks. Obtaining such information is especially difficult when sampling older cuts, where tracks have been cancelled by weather and re growth. In that case, there is a risk of sampling only the most severe examples of soil disturbance (i.e. those that are still visible after years) while missing lightly affected areas. As a result, these studies (e.g. Sakai et al. 2008, Gerasimov and Katarov 2010, Majnounian and Jourgholami 2013) may overestimate damage severity while underestimating the affected area. Furthermore, in the specific case of our study, operators did not follow a regular traffic pattern, which would have made it extremely difficult to estimate the total surface covered by tracks, if these were at all visible after 6 years. In fact, tracks were not visible in the 2011 plots, either. Therefore, we could not go for a paired sampling design, where samples would be collected separately inside and outside the ruts. Therefore, dif-

fused random sampling was the only available option. In general, diffused sampling is a sensible choice whenever older plots need to be sampled, unless researchers have clearly marked the position of the tracks at the time of harvesting – years earlier. It should be taken into consideration that our study presents an average level of compaction, derived from sampling both trafficked and non-trafficked soil. That may lead to underestimating the actual compaction in the tracks. In contrast, conventional localized sampling as applied in other studies (e.g. Sakai et al. 2008, Gerasimov and Katarov 2010, Majnounian and Jourgholami 2013) tends to overestimate the average compaction level of the site. Since we adopted the same sampling design for all three treatments on test, the value of our comparison remains unbiased, whereas comparisons with previous studies should be interpreted with caution. This said, the results of this study are consistent with those of other studies conducted in the same area, although under different silvicultural and technological conditions. In their study about mechanized thinning of pine plantations, Magagnotti et al. (2012) reported that the average bulk density of the forest soil was 1.2 g cm-3 and 1.3 g cm-3, respectively, before and after harvesting. Slightly lower figures were obtained from windthrow salvage operations in the Park, where pre harvest bulk density was about 1 g cm-3 and postharvest bulk density ranged from 1.1 to 1.2 g cm-3 (Spinelli et al. 2013). In general, the post-harvest bulk density increases recorded in the Park are lower than reported for Central Europe and North America (Froehlich et al. 1986, Ampoorter et al. 2010), where they often range between 15 and 30% of the original pre harvest-values. This difference is likely explained by the resistance of sandy soils to compaction (Wästerlund 1985) and by the different sampling design. High CO2 concentration is a good indicator of reduced soil conductivity (Ponder 2005), which is a common consequence of compaction (Schack-Kirchner et al. 2001). Compaction decreases soil porosity, especially for what concerns pores greater than 3 mm in diameter, which have the highest conductivity (Huang et al. 1996). When air exchange is restricted, the respiration of soil biota induces an increase of CO2 concentration (Von Wilpert and Schäffer 2006). If gas conductivity is severely reduced, CO2 levels may reach very high concentrations that will restrict further breathing. In that case, soil productivity is curtailed, and so is the ability for biological recovery (Dick et al. 1988). Previous studies have shown that biological recovery is expected when CO2 concentration is below 10 000 Croat. j. for. eng. 35(2014)1

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

ppm: beyond this threshold, biological activity is so constrained that biological recovery will take a very long time. A CO2 concentration above 20 000 ppm stalls almost all biological activity (Paul 2007): then recovery may only happen through physical agents, at times of prolonged drought or freezing (Magagnotti et al. 2012). CO2 concentration in the soil air in the recently clear felled study plots was much lower than recorded after pine thinning in the same area (3 000 vs. 8 000 ppm), which may depend on a number of factors, including the different data collection method (Magagnotti et al. 2012). In the quoted study, CO2 concentration readings were collected directly in the machine ruts, where compaction was highest. In this study, post-harvest readings were obtained by randomly sampling the whole clear cut area, so that our research averaged readings obtained inside and outside the ruts, which were invisible at the time of sampling. One of the main assets of this study is in the concurrent use of different methods for gauging soil compaction. The agreement of all methods makes our conclusions especially robust. So we can safely state that the clear felling operations implemented in the park do result in measurable soil compaction, as witnessed by the concurrent increase of soil bulk density, soil penetration resistance, soil deflection and soil CO2 concentration in the clear felled plots. However, the levels of compaction measured in this study are below critical values, which are estimated to 1.7 g cm–3 for bulk density (Heilman 1981), 3 to 6 MPa for penetration resistance (Lhotsky 2000, Whalley et al. 1995) and 7 000 to 10 000 ppm for CO2 concentration (Magagnotti et al. 2012, Qi et al. 1994). The same accounts for soil porosity, which was significantly lower in clear felled plots, but still above the estimated 38% critical value (Lhotsky 2000). It indicates the resistance of sandy soil to compaction, even in case of operation of heavy machinery traffic. Furthermore, long term studies indicate that pine trees are resistant to the effect of soil compaction, which has little consequence on early (Lacey and Ryan 2000) and mature growth (Sanchez et al. 2006), which may justify cautious optimism. What is more, this study suggests that recovery may already be visible six years after clear felling, especially for what concerns soil air conductivity. There was no significant difference in soil CO2 concentration between plots clear felled in 2005 and plots that were not harvested at all. Similarly, penetration resistance was significantly lower in the 2005 plots, compared to recently clear felled plots. Of course, complete regeneration is not achieved within such a short period, as Croat. j. for. eng. 35(2014)1

M. Kleibl et al.

indicated by the permanence of alterations in bulk density, porosity and deflection characteristics. Yet, recovery seems well under way, and the vigorous brush regeneration that exploded right after clear fell may have an important role in loosening up the soil structure, especially in a Mediterranean climate where soil freezing is not a factor. Such a rapid soil recovery is in contrast with other studies, which report much longer regeneration times, in the order of 15 to 40 years (Ampoorter 2010, Von Wilpert and Schäffer 2006). Probably, recovery time is proportional to impact severity, which may be overestimated in traditional studies probing only inside the wheel tracks (Seybold et al. 1999). A final and important remark must be made on the other causes of soil disturbance, besides mechanized clear-felling. It should be noted that we used the descriptive »non-harvested« instead of »undisturbed«. This choice is to acknowledge that areas that were not harvested did not necessarily represent pristine undisturbed sites. In the park there are many other sources of disturbance, including recreation, wildlife management and pine nut collection. Signs of these activities were especially visible in areas that were not harvested, while they had been cancelled in the clear felled areas. Furthermore, non-harvested areas may still bear the effects of previous thinning operations, if their impacts were particularly heavy and recovery incomplete. Therefore, our study quantified the additional impact of clear felling on areas that are already disturbed by a wide range of activities. Fortunately, the cumulative effect of these activities (including clear felling) does not seem to alter soil characteristics so much as to seriously threaten forest regeneration and further growth (Magagnotti et al. 2012).

6. Conclusions Mechanized clear felling of mature umbrella pine plantation produces significant alterations of soil physical characteristics, additional to those eventually caused by other management activities. However, the extent of these alterations may not exceed critical limits, partly due to the resistance of sandy soil to compaction. What is more, recovery seems to be relatively fast. Six years after clear felling, CO2 concentration is similar to that recorded in non-harvested plots, possibly indicating a recovery of soil conductivity, which is likely to accelerate the further restoration of original soil characteristics. These results may support the introduction of mechanized harvesting to coastal pine plantations established on sand dunes, which seem especially resilient to soil compaction.

M. Kleibl et al.

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71)

Acknowledgments This project was funded by the Regional Park Migliarino-San Rossore-Massaciuccoli. Special thanks are due to Dott. Francesca Logli for her support with experiment planning and organization. This study was also made possible thanks to funding received from the STSM programme of Action COST FP902 and from the Mendel University project OC10041.

7. References Ampoorter, E., Van Nevel, L., De Vos, B., Hermy, M., Verheyen, K., 2010: Assessing the effects of initial soil characteristics, machine mass and traffic intensity on forest soil compaction, Forest Ecology and Management 260(10): 1664–1676. Ampoorter, E., De Schrijver, A., Van Nevel, L., Hermy, M., Verheyen, K., 2012: Impact of mechanized harvesting on compaction of sandy and clayey forest soils: results of a meta-analysis, Annals of Forest Science 69: 533–542. Carrasquinho, I., Freire, J., Rodrigues, A., Tomé, M., 2010: Selection of Pinus pinea L. plus tree candidates for cone production, Annals of Forest Science 67(8): 814–823. Dick, R. P., Myrold, D., Kerle, E., 1988: Microbial biomass short-term effects that soil compaction might have on and soil enzyme activities in compacted and rehabilitated skid trail soils, Journal of the Soil Science Society of America 52: 512–516.

Huang, J. Lacey, S. T., Ryan, P.J., 1996: Impact of Forest Harvesting on the Hydraulic Properties of Surface, Soil Science 161, p. 79–86. Klvač, R., Vrána, P., Jiroušek, R., 2010: Possibilities of using the portable falling weight deflectometer to measure the bearing capacity and compaction of forest soils. Journal of Forest Science 56(3): 130–136. Lacey, S. T., Ryan, P. J., 2000: Cumulative management impacts on soil physical properties and early growth of Pinus radiata. Forest Ecology and Management 138(1 – 3): 321–333. Lhotský, J., 2000: Zhutňování půd a opatření proti němu. Ústav zemědělských a potravinářských informací, Praha. 62 pp. Magagnotti, N., Spinelli, R., 2011: Financial and energy cost of low-impact wood extraction in environmentally sensitive areas. Ecological Engineering 37(4): 601–606. Magagnotti, N., Spinelli, R., Güldner, O., Erler, J., 2012: Site impact after motor-manual and mechanised thinning in Mediterranean pine plantations. Biosystems Engineering 113(2): 140–147. Majnounian, B., Jourgholami, M., 2013: Effect of rubber-tired cable skidder on soil compaction in Hyrcanian forest. Croatian Journal of Forest Engineering 34(1): 123–135. Makineci, E., Demir, M., Comez, A., Yilmaz, E., 2007: Chemical characteristics of the surface soil, herbaceous cover and organic layer of a compacted skid road in a fir (Abies bornmulleriana Mattf.) forest. Transportation Research Part D 12: 453–459.

Paul, E. A., 2007: Soil Microbiology, Ecology and Biochemistry. III Edition. Elsevier Academic Press. Dordrecht, Netherlands. ISBN 9780125468077, 552 pp.

Marchi, E., Magagnotti, N., Berretti, L., Neri, F., Spinelli, R., 2011: Comparing terrain and roadside chipping in Mediterranean pine salvage cuts. Croatian Journal of Forest Engineering 32(2): 587–598.

Froehlich, H. A., Miles, D., Robbins, R., 1986: Growth of young Pinus ponderosa and Pinus contorta on compacted soil in central Washington. Forest Ecology and Management 15(4): 285–294.

McNabb, D. H., Startsev, A. D., Nguyen, H., 2001: Soil Wetness and Traffic Level Effects on Bulk Density and Air-Filled Porosity of Compacted Boreal Forest Soils. Journal of the Soil Science Society of America 65: 1238–1247.

Gerasimov, Y., Katarov, V., 2010: Effect of bogie track and slash reinforcement on sinkage and soil compaction in soft terrains. Croatian Journal of Forest Engineering 31(1): 35–45. Gondard, H., Romane, F., Aronson, J., Shater, Z., 2003: Impact of soil surface disturbances on functional group diversity after clear-cutting in Aleppo pine (Pinus halepensis) forests in southern France. Forest Ecology and Management 180(1 – 3): 165–174.

Motulsky, H. J., Brown, R., 2006: Detecting outliers when fitting data with nonlinear regression – a new method based on robust nonlinear regression and the false discovery rate. BMC Bioinformatics vol. 7: 123. Available at: http://www. ncbi.nlm.nih.gov/pubmed/16526949. Picchio, R., Neri, N., Petrini, E., Verani, S., Marchi, E., Certini, G., 2012: Machinery induced soil compaction in thinning two pine stands in central Italy. Forest Ecology and Management 285(1): 38–43.

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

Ponder, F. Jr., 2005: Effect of soil compaction and biomass removal on soil CO2 efflux in a Missouri forest. Soil Science and Plant Analysis 36(9 – 10): 1301–1311.

Heilman, P., 1981: Root penetration of Douglas-fir seedlings into compacted soil. Forest Science 27(4): 660–666.

Qi, J., Marshall, J. D., Mattson, K. G., 1994: High soil carbon dioxide concentrations inhibit root respiration of Douglas fir. New Phytologist 128(3): 435–442.

Horn, R., Vossbrink, J., Becker, S., 2004: Modern forestry vehicles and their impacts on soil physical properties. Soil and Tillage Research 79(2): 207–219.

Sakai, H., Nordfjell, T., Suadicani, K., Talbot, B., Bøllehus, E., 2008: Soil compaction on forest soils from different kinds of Croat. j. for. eng. 35(2014)1

Soil Compaction and Recovery after Mechanized Final Felling of Italian Coastal Pine Plantations (63–71) tyres and tracks and possibility of accurate estimate. Croatian Journal of Forest Engineering 29(1): 15–27. Sanchez, F., Scott, A., Ludovici, K., 2006: Negligible effects of severe organic matter removal and soil compaction on loblolly pine growth over 10 years. Forest Ecology and Management 227(1 – 2): 145–154. Schack-Kirchner, H., Gaertig, T., Von Wilpert, K., Hildebrand, E., 2001: A modified McIntyre and Phillip approach to measure top-soil gas diffusivity in-situ. Journal of Plant Nutrition and Soil Science 164(3): 253–258. Seybold, C. A., Herrick, J. E., Brejda, J. J., 1999: Soil Resilience: A Fundamental Component of Soil Quality. Soil Science 164(4): 224–234. Spinelli, R., Magagnotti, N., Picchi, G., 2009: Complete tree harvesting as an alternative to mulching in early thinnings. Forest Products Journal 59(6): 79–84. Spinelli, R., Magagnotti, N., Nati, C., 2010: Benchmarking the impact of traditional small-scale logging systems used in Mediterranean forestry. Forest Ecology and Management 260(11): 1997–2001. Spinelli, R., Lombardini, C., Magagnotti, N., 2013: Salvaging windthrown trees with animal and machine systems. Ecological Engineering 53: 61–67.

M. Kleibl et al.

USDA Soil Survey Staff, 1999: Soil taxonomy: A basic system of soil classification for making and interpreting soil surveys. 2nd edition. Natural Resources Conservation Service. U.S. Department of Agriculture Handbook 436. Downloaded from: http://soils.usda.gov/technical/classification/taxonomy/. Vokoun, M., Amacher, G., Wear, D., 2006: Scale of harvesting by non-industrial private forest owners. Journal of Forest Economics 11(4): 223–244. Von Wilpert, K., Schäffer, J. 2006. Ecological effects of soil compaction and initial recovery dynamics: A preliminary study. European Journal of Forest Research 125: 129–138. Vusić, D., Šušnjar, M., Marchi, E., Spina, R., Zečić, Z., Picchio, R., 2013: Skidding operations in thinning and final felling treatments in mixed stands – work productivity, energy inputs and emissions. Ecological Engineering – Accepted manuscript. Wästerlund, I., 1985: Compaction of till soils and growth tests with Norway spruce and scots pine. Forest Ecology and Management 11(3): 171–189. Whalley, W., Dumitru, E., Dexter, A. R., 1995: Biological effects of soil compaction. Soil and Tillage Research 35(1 – 2): 53–68.

Authors’ address:

Received: December 18, 2013 Accepted: January 14, 2014 Croat. j. for. eng. 35(2014)1

Miroslav Kleibl, PhD. e-mail: kleibl@mendelu.cz Prof. Radomir Klvač, PhD. e-mail: klvac@mendelu.cz Josef Porhaly, PhD. e-mail: porhaly@mendelu.cz Mendel University in Brno Zemedelska 3 61300 Brno CZECH REPUBLIC Carolina Lombardini e-mail: lombardini@ivalsa.cnr.it Raffaele Spinelli*, PhD. e-mail: spinelli@ivalsa.cnr.it CNR IVALSA Via Madonna del Piano 10 50019 Sesto Fiorentino (FI) ITALY * Corresponding author

Original scientific paper

Effects of Skidder Passes and Slope on Soil Disturbance in Two Soil Water Contents Ramin Naghdi, Ahmad Solgi Abstract Skidding operations induce changes in soil physical properties, which have the potential to impact soil sustainability and forest productivity. Our objective was to investigate the effects of traffic frequency, trail slope, and soil moisture content on soil compaction, total porosity and rut depth. Treatments included a combination of three different traffic intensities (3, 7, and 14 passes), three levels of slopes (< 10%, 10–20% and > 20%), and two levels of soil moisture content (18% and 32%). Soil bulk density and total porosity were measured as 0.75 g cm-3 and 71%, respectively, along the undisturbed area. The results show that bulk density, total porosity and rut depth on skid trails were significantly affected by traffic frequency, skid trail slope and soil moisture content. As the skidder passed, skid trail slope and soil moisture content increased, increasing significantly the average bulk density. Bulk density draws near the critical value after 7 and 14 passes, respectively, at higher and lower soil water content. At each moisture content, the increase of slope > 20% caused a significant increase of the average bulk density. Total porosity on the skid trail was measured from minimum 45% (14 passes and slope > 20%) to maximum 58% (3 passes and slope < 10%) at higher soil water content, and minimum 49% (14 passes and slope > 20%) to maximum 68% (3 passes and slope < 10%) at lower soil moisture content. Rut depth was recorded at 7 and 14 passes at high and low soil water content, respectively, and it increased with the slope. The results show that slope and moisture content had strong effects on soil disturbance. Keywords: bulk density, rutting, skidding, soil compaction, total porosity

1. Introduction Mechanized forest harvesting operations have induced changes in soil physical properties with the potential to negatively impact soil sustainability and forest productivity (Powers et al. 1990). The most significant changes have been shown to occur in soil surface layers, which can restrict the movement of air and water into soil layers (Botta et al. 2006). Undisturbed forest soils have high macroporosity and low soil bulk density and are easily compacted by logging machinery (Lacey and Ryan 2000). Soil compaction, often accompanied by rutting, is a typical process that may appear as a result of inappropriate use of heavy forest machinery. Rut formation may involve direct or indirect damage to the root system of trees and soil animals (Lindo and Visser 2004), and altered soil and climatic conditions may Croat. j. for. eng. 35(2014)1

increase plant diversity (Buckley et al. 2003). Soil compaction involves the compression of pores, which leads to decreased porosity and pore continuity (Berli et al. 2003, Teepe et al. 2004). As a result, there is often an increase in dry bulk density (BD) (e.g. Miller et al. 1996), defined as the dry mass of the soil to its volume. In addition, smaller pore sizes reduce hydraulic conductivity, leading to a slower water infiltration and increased runoff. In general, gas exchange is also hampered (Gaertig et al. 2002), possibly affecting growth and activity of roots and soil organisms (Bathke et al. 1992, Zhao et al. 2010), and leading to an alteration of chemical processes (Woodward 1996, Ballard 2000). As pores become smaller, soil strength increases (Shetron et al. 1988). Nugent et al. (2003) found a 30–50% increase of the penetration resistance (PR) with machine traffic, a measure for soil strength. As root tips have to overcome soil strength to be able to elongate, root

R. Naghdi and A. Solgi

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

growth may be hampered (Greacen and Sands 1980), depending on soil type, water regime and tree species (Heninger et al. 2002, Dexter 2004). The dimension of the impact varies according to many factors such as skidder passes, slope, site characteristics, harvesting machines, planning of skid roads and production season (Najafi et al. 2009, Jamshidi et al. 2008, Demir et al. 2007). The number of machine passes is a factor that significantly influences the degree of soil damage (Jun et al. 2004). Machine passes have an important influence on soil structural characteristics, soil aeration and the soil water balance, and may therefore considerably affect soil organisms and root development. The initial passes cause the highest increase in soil compaction in relation to subsequent passes but these may lead to further soil disturbance by deepening the ruts. During skidding on the steep terrain, a given load gets uneven weight balance on the axles (usually rear axle) and increases soil disturbance. Najafi et al. (2010) found that, during timber harvesting, terrain slope had a stronger effect on soil disturbance so that snig track related disturbance was greater on slopes > 20% than on slopes < 20%. The potential for soil compaction is greater on wet soils than on dry soils. Botta et al. (2006) found that more severe compaction occurs from traffic on saturated soils than on dry soils. The extent of severe disturbance from ground based harvesting systems varies depending on the slope and soil moisture content, although the effects of slope and soil moisture on soil disturbance have received less attention. The aim of this study was to characterize the effects of traffic intensity, skid trail slope and soil moisture content on bulk density, total porosity and rut depth in a northern mountainous forest of Iran.

in October, while the minimum rainfall of 25 mm occurs in August. The mean annual temperature is 15 ºC, with the lowest values in February. Soil texture, analyzed by using the Bouyoucos hydrometer method (Kalra and Maynard 1991), was clay loam along the trail. The skidder type used in this study was a rubber tired skidder, HSM 904 – 4 wheels, tire/chain dimension 600/60–30.5, tire inflation pressure 250 kPa and weighing 8.71 Mg without load in the proportion of 60% on the front axle to 40% on the rear axle.

2.2 Experimental design and data collection A skid trail 1 200 m long with downslope skidding direction was chosen for the experiments. In choosing the skid trail, attempts were made to select a trail that had different longitudinal slopes and no lateral slope. The longitudinal profile showed that the slopes of skid trail ranged from 0 to 36%. The factorial design was generalized to three factors and each factor was fixed. Three traffic frequencies (3, 7 and 14), three slope classes (0–10%, 10–20% and > 20%) and two soil moisture contents (18% and 32%) were applied with two replications or in total three test series (Jansson and Johansson 1998). In total, 54 plots (10 m × 4 m in size each) were set up in the study. The impacts of skidding on the surface soil layer (0–10 cm depth) were examined using bulk density, total porosity and rut depth in comparison to the undisturbed area. In a given plot, samples were taken along four randomized lines across the wheel track perpendicular to the direction of travel with 2 m buffer zone between lines to avoid interactions. At three different points of each line (left track LT, between track BT and right

2. Material and methods 2.1 Site description The research was started in April 2013 at Sorkhkola forest, Mazandaran province, North of Iran between 36°11′ N and 36°17′ N and 52°17′ E and 52°57′ E. The area is covered by Fagus orientalis and Carpinus betulus stands. Canopy cover is estimated to 80%, the average diameter is 29.72 cm, the average height is 22.94 m and stand density is 220 trees/ha. Elevation is approximately 700 m above sea level with a north aspect. The average annual rainfall, recorded at the closest national weather station, was 810 mm. The maximum mean monthly rainfall of 110 mm usually occurs

Fig. 1 Treatment set-up with sample lines within plots Croat. j. for. eng. 35(2014)1

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

track RT) one sample was taken from the 0–10 cm soil layer (Fig. 1). The soil samples were collected at the depth interval of 0–10 cm with a soil hammer and rings (5 cm in diameter, 10 cm in length). Samples were put in polyethylene bags and labeled. Collected samples, brought to the laboratory from the research area, were promptly weighed (soil samples). Soil samples were dried in an oven at 105 °C to constant mass (24 h). The moisture content in the soil samples was measured gravimetrically after drying in an oven (Kalra and Maynard 1991). Total porosity was calculated by Equation (1):

AP =

Db 2.65 VC

1−

Wd VC

3.1 Dry bulk density

(2)

Where: Wd weight of the dry soil, g.

Ruts, at least 5 cm deep from the top of the mineral soil surface and 2 m long, were sampled. Rut depth was measured using a profile meter consisting of a set of vertical metal rods (500 mm in length and 5 mm in diameter), spaced at 25 mm horizontal intervals, sliding through holes in a 1 m long iron bar. The bar was placed across full width of the wheel tracks (four meters) perpendicular to the direction of travel

Fig. 2 Illustration of the technique used for rut depth measurement (LT: left rack trail, RT: right rack trail) Croat. j. for. eng. 35(2014)1

3. Results Soil bulk density and total porosity were 0.75 g cm-3 and 71%, respectively, and soil texture was clay-loam along the undisturbed area.

Soil bulk density was calculated by Equation (2): Db =

and rods positioned to conform to the shape of the depression (Nugent et al. 2003). Rut depth was calculated as the average depth of 40 reads on 1 m bar (see Fig. 2). One-way and three-way ANOVA by SPSS software version 11.5 was used to assess the significant differences between average bulk density, total porosity and rut depth in different traffic levels, trail slope, and soil moisture content and their interaction effects. Duncan’s multiple range tests was used to determine the significant differences between average bulk density, total porosity and rut depth in different treatments.

(1)

Where: AP total apparent porosity, Db soil bulk density, 2.65 assumed particle density, VC volume of the soil cores (196.25 cm3).

R. Naghdi and A. Solgi

The results showed that skidder passes, skid trail slope, and soil moisture content had significant effect on soil bulk density of skid trail, and however the interaction between them was not significant. The average soil bulk density on the skid trail was minimum 1.09 g cm-3 to maximum 1.48 g cm-3 at higher soil water content, and minimum 0.83 g cm-3 to maximum 1.34 g cm-3 at lower soil moisture content (Table 1). Table 1 Effect of skidder passes on dry bulk density, g cm–3 Soil moisture content, %

Passes

Slope, %

0–10

1–20

(> 20)

0–10

10–20

> 20

Undisturbed

0.73 d

0.72 d

0.77 d

0.74 c

0.75 c

0.79 c

0.83 c

0.84 c

1.1 c

1.09 b

1.12 b

1.29 b

1.087 b

1.09 b

1.23 b

1.26 a

1.32 a

1.42 a

1.21 a

1.24 a

1.34 a

1.31 a

1.37 a

1.48 a

Bulk density clearly increased with the increasing of slope at each tested traffic frequency. There were significant differences (p < 0.05) between treatments at terrain slope under and over 20 % (Fig. 3). In all skidder passes and skid trail slope treatments, dry bulk density increased considerably with an increase in soil moisture content (Fig. 4).

R. Naghdi and A. Solgi

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

Fig. 3 Effect of skid trail slope on dry bulk density, g cm–3

Fig. 4 Average bulk density of skid trail with different soil moistures

Table 2 Effect of skidder passes on total porosity, %

changes were influenced significantly by the number of skidder passes (p < 0.05), skid trail slope (p < 0.05), and moisture content (p < 0.05), and however the interaction between them was not significant (p > 0.05).

Soil moisture content, %

Passes 0–10 Undisturbed

Slope, %

10–20

> 20

0–10

71.52 a 71.76 a 70.02 a

71.5 a

58.81 bc 58.96 bc 54.62 bc 52.34 c

54.27

52.83

49.35

58.80

68.54

58.62

68.06

50.18

3.3 Rut depth

10–20

> 20

71.29 a

69.94 a

51.29 b

50.83 c

47.45 c

56.43 48.00

45.708 c

This indicates that, during the field operations, the soil moisture content is an important factor influencing compactibility.

3.2 Total porosity Total porosity of the skid trail is considerably lower than the total porosity of the undisturbed area. The average total soil porosity was 71% on undisturbed area, and on the skid trail it was minimum 45% to maximum 58% at higher soil water content, and minimum 49% to maximum 68% at lower soil moisture content (Table 2). Total porosity decreased with skidder traffic frequency, skid trail slope and soil moisture content. Porosity is inversely related to bulk density, meaning that a decrease in mean porosity comes with an increase in mean bulk density after skidding. Total porosity

Rut depth increased significantly by the number of machine passes, skid trail slope and soil moisture content, and however it was not significantly affected by the interactions between them. The results showed that, at higher soil water content, rutting began after 7 passes, and at lower soil water content, it began after 14 passes (Table 3). In all soil water treatments, by increasing skid trail slope rut depth increased significantly (Fig. 5). Indeed, rut depths were significantly deeper for the steep slopes than the gentle slopes regardless of traffic intensity and soil water content. Table 3 Effect of skidder passes on rut depth (cm) Soil moisture content, % Slope

Passes

14 b

15.2 b

0–10

8.5

7.4

10–20

13.7 ab

11.1ab

24.6 ab

34.5 a

> 20

15.3

Croat. j. for. eng. 35(2014)1

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

R. Naghdi and A. Solgi

lubricating the particles thus allowing the particles to slide together and compact by squeezing out air. However, the bulk density decreases at higher soil water contents (after the maximum density is reached) because with further addition of water, soil has greater pore water pressures and becomes less compactible (Mosaddeghi et al. 2000). Relative bulk density between 1.40 and 1.55 g cm-3 is considered as the critical level at which plant roots cannot penetrate into soils with light and medium texture (Kozlowski 1999). Our results showed that bulk density draws near the critical value after 7 and 14 passes, respectively, at higher and lower soil water content (Table 1).

4.2 Total porosity

Fig. 5 Effect of slope on average rut depth, cm

4. Discussion 4.1 Dry bulk density Most of the compaction, expressed as bulk density increase, occurs during the first pass. As shown in Fig. 6, bulk density of the skid trail seriously increased already after three passes of the skidder. Our results are in accordance with the results of Ampoorter et al. (2007) and Brais and Camire´ (1998), who found that bulk density increases more gradually with 50% of the total impact occurring after three passes. When the number of machine passes increases, the additional bulk density increment is negligible (Ampoorter et al. 2007). Botta et al. (2006), Eliasson (2005), Raper (2005), and Labelle and Jaeger (2011) reported that dry bulk density increased with skidder traffic frequency. The average dry bulk density comes faster at higher trail slope level (Fig. 3). Our results are in accordance with the results of Najafi et al. (2009), who found that slope affected deep disturbance caused by skidder logging. The increasing of bulk density in the higher trail slope may be associated with the lower speed of skidder on steep slope trails. When the skidder passes slower on a steep slope, obviously the top soil vibrates more and consequently causes more disturbance than on gentle trail. The average dry bulk density comes faster at higher moisture content (Fig. 4). McNabb et al. (2001), and Startsev and McNabb (2000) reported that the winter harvested sites displayed a greater percent increase in bulk density than summer harvested sites. This occurs because the soil becomes wetter, water films weaken the interparticle bonds and reduce internal friction by Croat. j. for. eng. 35(2014)1

Total porosity of the skid trail is considerably lower than the total porosity of the undisturbed area. Total porosity decreased with skidder traffic frequency, skid trail slope and soil moisture content. Porosity is inversely related to bulk density, meaning that a decrease in mean porosity comes with an increase in mean bulk density after skidding. When soil is compacted, total porosity is reduced at the expense of large voids (Greacen and Sands 1980). Motavalli et al. (2003) found that surface compaction significantly decreased total porosity both at the depth of 0–10 cm and 10–20 cm. The major differences in pore volume between slope treatments could probably be explained by greater soil compaction in the steep trail, which indicated that during ground skidding slope had a strong impact on soil porosity (Fig. 7). Besides skidder traffic frequency

Fig. 6 Effect of traffic frequency on the increase of dry bulk density, %

R. Naghdi and A. Solgi

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

Fig. 7 Effect of slope on total porosity, % and skid trail slope, total porosity decreased with soil moisture content. As with all skidder passes and skid trail slope treatments, by increasing soil moisture, total porosity was significantly decreased. Xu et al. (2000) reported that on the wet-weather harvested sites, macroporosity and total porosity decreased by 44% and 8%, respectively. In contrast, the changes of these properties on the surface of dry-weather harvested sites were much smaller (in the same arrangement and direction): 3% and 1%.

4.3 Rut depth Rut depths were significantly (p < 0.05) correlated to traffic frequency, skid trail slope and soil moisture content. Positive correlation between rut depth and traffic frequency was in accordance with Eliasson (2005), Botta et al. (2006) and Najafi et al. (2009). The results showed that by increasing skid trail slope rut depth were significantly increased. Indeed rut depths were significantly deeper for the steep slopes than the gentle slopes regardless of traffic intensity and soil water content. Our result was accordance with the study of Najafi et al. (2009). The results showed that at higher soil water content rutting began after 7 passes and at lower soil water content it began after 14 passes. Rutting often occurs when traffic is applied to compacted soil (Startsev and McNabb 2000). Rollerson (1990) reported that rutting depths tended to be deeper in moist than in dry soil conditions, and the effect is usually higher with increasing traffic.

It can be concluded that soil disturbance is affected by several factors such as soil water content, wheel slip, vibration and number and duration of loading events. The decrease of total porosity and increase of bulk density and rut depth in the steep slope trail may be associated with the lower speed of skidder on steep slope trail. When skidder passes slower on steep slope, the top soil is obviously vibrated more and consequently it is exposed to more disturbances compared to gentle trails. Furthermore, when logs are pulled downhill, rear axle gets more loads, the radius of rear wheels decreases accordingly and barking slipping may occur because the rear compressed tires roll shorter radius than the front wheels (Najafi et al. 2009). This is in agreement with Davies et al. (1973) and Raghavan et al. (1977), who identified wheel slip on agricultural tractors. The wheel or track slip directly affected the soil structure and altered physical soil properties down to deeper depths. Spinning, digging and slipping may mix mineral soil and forest floor resulting in increased compaction and rutting and decreased total porosity. This study was conducted with the overall objective of characterizing the effects of skidder passes, skid trail slope and soil water content on bulk density, total porosity and rutting. Compaction of soil under the impact of skidding caused the increase in bulk density rates on the skid road (Demir et al. 2007). As compaction increased, the rates of total porosity decreased. When soil is compacted, total porosity is reduced at the expense of large voids. There is a positive relationship between soil compaction and skid trail slope and number of passes. Therefore, the hypothesis that skid trail slope and skidder passes affect dry bulk density has been supported. Soil moisture has significant effects on soil disturbance. One strategy to limit soil disturbances is to avoid traffic whenever the water content is higher. Skidding operations should be planned when soil conditions are dry so as to minimize rutting, but if skidding must be done under wet conditions, the operations should be stopped when machine traffic provides deep ruts. Within the limits of experimental conditions, the following conclusions can be drawn and therefore applied for proper harvesting and management of forest ecosystems: Þ Skidding should be limited to the slope < 20%; Þ Only when traffic level and soil moisture content are high, rutting can occur; Þ Soil compaction, total porosity and rut depth are significantly affected by traffic intensity, skid trail slope and soil moisture content. Croat. j. for. eng. 35(2014)1

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

Acknowledgements This paper has been written based on the results of a research titled »Assessing the effects of various systems of wood extraction from forests on physical, chemical and micromorphological regeneration in northern forest of Iran«, which was sponsored by Iran National Science Foundation – Depute of Science and Technology – Presidential Office. The authors are grateful to Iran National Science Foundation.

5. References Ampoorter, E., Goris, R., Cornelis, W. M., Verheyen, K., 2007: Impact of mechanized logging on compaction status of sandy forest soils. Forest Ecology and Management 241: 162– 174. Ballard, T. M., 2000: Impacts of forest management on northern forest soils. Forest Ecology and Management 133: 37–42. Bathke, G. R., Cassel, D. K., Hargrove, W. L., Porter, P. M., 1992: Modification of soil physical properties and root growth response. Soil Science 154: 316–329. Berli, M., Kulli, B., Attinger, W., Keller, M., Leuenberger, J., Flühler, H., Springman, S. M., Schulin, R., 2003: Compaction of Agricultural and forest soils by tracked heavy construction machinery. Soil and Tillage Research 75: 37–52. Botta, G. F., Jorajuria, D., Rosatto, H., Ferrero, C., 2006: Light tractor traffic frequency on soil compaction in the Rolling Pampa region of Argentina. Soil and Tillage Research 86: 9–14. Brais, S., Camire´, C., 1998: Soil compaction induced by careful logging in the claybelt region of northwestern Quebec (Canada). Canadian Journal of Soil Science 78: 197–206. Buckley, D. S., Crow, T. R., Nauertz, E. A., Schulz, K. E., 2003: Influence of skid trails and haul roads on understorey plant richness and composition in managed forest landscapes in Upper Michigan, USA. Forest Ecology and Management 175: 509–520. Davies, D. B., Finey, J. B., Richardson, S. J., 1973: Relative effects of tractor weight and wheel-slip in causing soil compaction. Journal of Soil Science 24: 401–409. Demir, M., Makineci, E., Yilmaz, E., 2007: Investigation of timber harvesting impacts on herbaceous cover, forest floor and surface soil properties on skid road in an oak (Quercus petrea L.) stand. Building and Environment 42: 1194–1199. Dexter, A. R., 2004: Soil physical quality. Part I. Theory, effects of soil texture, density and organic matter, and effects on root growth. Geoderma 120: 201–214.

R. Naghdi and A. Solgi

Greacen, E. L., Sands, R., 1980: Compaction of forest soils: a review. Australian Journal of Soil Research 18: 163–189. Heninger, R., Scott, W., Dobkowski, A., Miller, R., Anderson, H., Duke, S., 2002: Soil disturbance and 10-year growth response of coast Douglas-fir on nontilled and tilled skid trails in the Oregon Cascades. Canadian Journal of Forest Research 32: 233–246. Jamshidi, R., Jaeger, D., Raafatnia, N., Tabari, M., 2008: Influence of two ground-based skidding systems on soil compaction in different slope conditions. International Journal of Forest Engineering 19: 9–16. Jansson, K. J., Johansson, J., 1998: Soil change after traffic with a tracked and wheeled forest machine, a case study on a silt loam in Sweden. Forestry 71: 57–66. Jun, H. G., Way, T. R., Löfgren, B., Landström, M., Bailey, A. C., Burt, E. C., McDonald, T. P., 2004: Dynamic load and inflation pressure effects on contact pressures of a forestry forwarder tire. Journal of Terramechanics 41: 209–222. Kalra, Y. P, Maynard, D. G., 1991: Methods and manual for forest soil and plant analysis. Forestry Canada, Re NORX-319. Northern Forestry Center. 116 p. Kozlowski, T. T., 1999: Soil compaction and growth of woody plants. Scandinavian Journal of Forest Research 14: 596–619. Labelle, E.R., Jaeger, D., 2011: Soil compaction caused by cut-to-length forest operations and possible short-term natural rehabilitation of soil density. Soil Science Society of America Journal 75: 2314–2329. Lacey, S. T., Ryan, P. J., 2000: Cumulative management impacts on soil physical properties and early growth of Pinus radiata. Forest Ecology and Management 138: 321–333. Lindo, Z., Visser, S., 2004: Forest Floor microarthropod abundance and oribatid mite (Acari: Oribatida) composition following partial and clear-cut harvesting in the mixedwood boreal forest. Canadian Journal of Forest Research 34: 998– 1006. McNabb, D. H., Startsev, A. D., Nguyen, H., 2001: Soil wetness and traffic on bulk density and air-filled porosity of compacted boreal forest soils. Soil Science Society of American Journal 65: 1238–1247. Miller, R. E., Scott, W., Hazard, J. W., 1996: Soil compaction and conifer growth after tractor yarding at three coastal Washington locations. Canadian Journal of Forest Research 26: 225–236. Mosaddeghi, M. R., Hajabbasi, M. A., Hemmat, A., Afyuni, M., 2000: Soil compactibility as affected by soil moisture content and farmyard manure in central Iran. Soil and Tillage Research 55: 87–97.

Eliasson, L., 2005: Effects of forwarder tyre pressure on rut formation and soil compaction. Silva Fennica 39: 549–557.

Motavalli, P. P., Anderson, S. H., Pengthamkeerati, P., 2003: Surface compaction and poultry litter effects on corn growth, nitrogen availability, and physical properties of a claypan soil. Field Crops Research 84: 303–318.

Gaertig, T., Schack-Kirchner, H., Hildebrand, E. E., Von Wilpert, K., 2002: The impact of soil aeration on oak decline in southwestern Germany. Forest Ecology and Management 159: 15–25.

Najafi, A., Solgi, A., Sadeghi, S. H., 2009: Soil disturbance following four wheel rubber skidder logging on the steep trail in the north mountainous forest of Iran. Soil and Tillage Research 103: 165–169.

Croat. j. for. eng. 35(2014)1

R. Naghdi and A. Solgi

Effects of Skidder Passes and Slope on Soil Disturbance ... (73–80)

Najafi, A., Solgi, A., Sadeghi, S. H., 2010: Effects of skid trail slope and ground skidding on soil disturbance. Caspian Journal of Environmental Science 8: 13–23. Nugent, C., Kanali, C., Owende, P. M. O., Nieuwenhuis, M., Ward, S., 2003: Characteristic site disturbance due to harvesting and extraction machinery traffic on sensitive forest sites with peat soils. Forest Ecology and Management 180: 85–98. Powers, R. E, Alban, D. H. M., Miller, R. E., Tiarks, A. E., Wells, C. G., Avers, P. E, Cline, R. G., Fitzgerald, R. O., Loftus Jr, N. S., 1990: Sustaining site productivity in North American forests: problems and prospects. In: Gessel, S.P., Lacate, D.S., Weetman, G.F., Powers, R.F. (Eds.), Sustained Productivity of Forest Soils. University of British Columbia, BC, 49–79. Raghavan, G. S. V., McKyes, E., Beaulieu, B., 1977: Prediction of clay soil compaction. Journal of Terramecanics 14: 31–38. Raper, R. L., 2005: Agricultural traffic impacts on soil. Journal of Terramecanics 42: 259–280. Rollerson, T. P., 1990: Influence of wide-tire skidder operations on soils. International Journal of Forest Engineering 2: 23–30.

wheel track surface soil bulk density. Northern Journal Applied Forestry 5: 120–123. Startsev, A. D., McNabb, D. H., 2000: Effects of skidding on forest soil infiltration in west-central Alberta. Canadian Journal of Soil Science 80: 617–24. Teepe, R., Brumme, R., Beese, F., Ludig, B., 2004: Nitrous oxide emission and methane consumption following compaction of forest soils. Soil Science Society of American Journal 68: 605–611. Woodward, C. L., 1996: Soil compaction and topsoil removal effects on soil properties and seedling growth in Amazonian Ecuador. Forest Ecology and Management 82: 197–209. Xu, Y. J., Burger, J. A., Aust, W. M., Patterson, S. C., 2000: Responses of surface hydrology and early loblolly pine growth to soil disturbance and site preparation in a lower coastal plain wetland. New Zealand Journal of Forestry Science 30: 250–265. Zhao, Y., Krzic, M., Bulmer, C. E., Schmidt, M. G., Simard, S. W., 2010: Relative bulk density as a measure of compaction and its influence on tree height. Canadian Journal of Forest Research 40: 1724–1734.

Shetron, S. G., Sturos, J. A., Padley, E., Trettin, C., 1988: Forest soil compaction: effect of multiple passes and landings on

Authors’ address: Ramin Naghdi, PhD. e-mail: rnaghdi@yahoo.com Ahmad Solgi, MSc.* e-mail: solgi_ahmad231@yahoo.com University of Guilan Faculty of Natural Resources Department of Forestry Box 1144 Somehsara, Guilan Province IRAN Received: August 11, 2013 Accepted: January 26, 2014

* Corresponding author Croat. j. for. eng. 35(2014)1

Original scientific paper

Untreated Wood Ash as a Structural Stabilizing Material in Forest Roads Gerald Bohrn, Karl Stampfer Abstract Due to the euphoric use of »green« energy produced by biomass power plants, up to 350 000 tons of ash are accumulated as a waste product every year in Austria and the estimated costs for landfilling are 1.7 million € per year. For this reason, methods must be found for the utilization of wood ash. One solution is to use it as a stabilizing material in forest roads. The pozzolanic characteristic of ash is used to bind the gravel in the road base. Wood ash is expected to reduce the need for gravel on forest roads and at the same time to enhance the load bearing capacity of forest roads. Two different untreated wood ashes were used in two mixture ratios, each on a 50 meter long forest road section, to investigate the load bearing capacity. The ashes were selected by their different properties: high lime and low heavy metal content, production of ashes in Austrian biomass power plants with various furnace technologies and disposal costs. Mixing depth was 0.50 m and the road base was covered by a 0.10 m thick surface layer. Elastic moduli of these sections were measured before the application, and repeated monthly by using a light falling weight deflectometer. After the first vegetation period, the mean elastic modulus of the sections mixed with dry bed ash showed an improvement. The increase of the initial mean load bearing capacity of 32.0 MNm-2 was 65% for 15/85 mixture and 76% for 30/70 mixture. The results for the fluidized bed ash sections fell short of expectations. Only 95% of the initial value could be reached for both mixing values. Keywords: wood ash, utilization, forest road, load bearing capacity, stabilization

1. Introduction The energy generated by biomass has gained importance in recent years. It is an essential contribution to sustainable regional energy supply, especially in Austrian wooded areas. Wood ash, produced in these processes, was previously considered to be a »wasteproduct« (Stupaka et al. 2007). In the late 1990s the Austrian Advisory Board for Soil Fertility highlighted the properties of this secondary raw material. Recycling of minerals corresponds to the ecological principle of the closed biogeochemical cycles and helps to spare natural mineral resources (Holzner and Obernberger 2011). In the year 2007, the total ash production of biomass use in Austria was 350 000 tons; 295 000 tons in biomass power plants and 55 000 tons in small scale Croat. j. for. eng. 35(2014)1

burning facilities, where further use is unknown. 125 000 tons were recycled and the rest of 170 000 tons was landfilled (Environment Agency Austria 2009). With an average fee of 100 € per ton of deposited material, the total costs for the annual ash disposal is 1.7 million € just for landfilling. It is necessary to develop methods of wood ash utilization so as to prevent this nutrient rich material from being landfilled. Depending on the characteristics of different types of wood ash, various purposes can be selected. Recycling nutrients of wood ash by using it as fertilizer has been common for a very long time (Holzner 1999). Another method could be the reinforcement of unpaved lowvolume forest roads (Lahtinen 2001). In Austria, the standard of the secondary road network is very high. Almost all roads in this network are paved. Low-volume roads are mostly situated on agri-

G. Bohrn and K. Stampfer

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

cultural or forested areas and mainly used by the land owners. In other European countries public use of lowvolume roads is more frequent and their legal frameworks allow the use of alternative structural road materials like wood ashes. Due to mismanagement of ashes in large road constructions and introduction of stricter regulations, there had been some setbacks in the establishment of ash as binding material (Kärrman et al. 2004). However, results of research on ecological influence (Lahtinen 2001, Thurdin 2004), and technical requirements and quality criteria of wood ash (von Bahr et al. 2006) formed the basis for suitable guidelines of ash application (Munde et al. 2006). The main objective was to verify the suitability of untreated wood ash as a reinforcing structure material in the load bearing layer of low-volume roads based on the hypotheses that wood ash is self-hardening. Unlike the experiments in Scandinavia with pure fly ashes or in combination with other industrial by-products (Lahtinen 2001, Mácsik and Svedberg 2006), where the influence of the mixture properties is managed by changing the proportion of different components, in this study untreated dry bottom ash and fluidized bed ash were used. In the case of suitability, gravel could be replaced with wood ash resulting in a cost-effective use of wood ash in forest road construction.

2. Material and Methods 2.1 Road reconstruction The initial situation was an existing forest road in need of maintenance and reconstruction in some places. The geographical position of the research area was the Attergau (47° 56′ N, 13° 88′ E), where most of the parental material contains flysch. Flysch is a combination of clay, argillite, lime rock and sandstone. Under dry conditions, the subgrade has a high load bearing capacity, but with increasing moisture it loses strength. So the superstructure of a forest road in flysch areas has to provide a stable layer. In most cases this aim is reached by constructing a bulky bearing layer of compressible gravel. For an average subgrade elastic-modulus of 20 MNm-2, it is possible to reach 30 – 35 MNm-2 under dry conditions and with compaction. Based on this, a road surface elastic modulus of 80 MNm-2 with a 0.20 – 0.30 m thick load bearing layer made of gravel with a grain size distribution curve of 0 – 60 mm can be reached (Dietz et al. 1984). With a flysch subgrade, the elastic modulus often drops below 10 MNm-2. For this reason, 0.40 – 0.50 m gravel is used to construct the load bearing layer to reach the necessary load bearing capacity.

This structural element is covered by a surface layer of gravel with a grain size distribution curve of 0 – 32 mm. During the reconstruction of a forest road of a total length of 1 850 m, four 50 m sections (Fig. 1) were reinforced with two different types of wood ash, dry bottom ash (DBA) and fluidized bed ash (FBA) in two mixing ratios (15% ash to 85% original forest road material and 30% ash to 70% original forest road material). On a fifth section, the forest road received the same treatment without wood ash, which served as a control treatment, and a sixth section remained untreated to provide reference data of load bearing capacities. The elastic modulus was measured on these sections, and the data were compared. In the text below, these sections will be referred to as:

DBA–X for dry bottom ash, where X is the percentage of dry bottom ash mixed with original material, for example DBA-30 for 30% dry bottom ash to 70% original material;

FBA–X f or fluidized bed ash, where X is the per-

centage of fluidized bed ash mixed with original material; TWOA for treatment without ash (mixing, rolling and grading); REF for reference with no treatment.

The mixing ratios are related to the application of other binding materials like lime or Portland cement (Dietz et al. 1984). The effectiveness of the reinforcement are, on one side, dependent on the ash characteristics, which are influenced by the type of combustion, differing fuel types structure of grains and free lime content. On the other side, the main factors for a good performing load bearing layer are heterogeneous base material and the optimal water content. Earlier research showed that a small addition of lime caused a significant improvement of strength (Lahtinen 2001).

2.2 Bearing layer construction Depending on the mixing ratio, a 0.08 m (15/85) or 0.12 m (30/70) thick layer of wood ash was applied onto the road surface. After that the mixing procedure was done by a WR 2400 recycler, a product of the Wirtgen Company, with a mixing depth of 0.50 m and a working width of 2.40 m (Fig. 2 and Fig. 3). The machine hydraulic system guaranties a homogenous mixing result by keeping the rotation speed of the mill constant and varying the machine speed. Under dry soil conditions, an additional water tank supplies the recycler with water to reach the required moisture and prevents dust formation. After the mixing procedure, the modified load bearing layer is covered by a 0.10 m thick surface layer of gravel with a parCroat. j. for. eng. 35(2014)1

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81â&#x20AC;&#x201C;89)

G. Bohrn and K. Stampfer

Fig. 1 Layout overview

2.3 Falling weight deflectometer The load bearing capacity was measured with a falling weight deflectometer (FWD). It was originally used to evaluate the physical properties of pavements like the in situ base and subgrade moduli during construction. FWD data allows to estimate the pavement structural capacity for overlay design and to determine if a pavement is being overloaded.

Fig. 2 WR2400 recycler during mixing process ticle size of 0 â&#x20AC;&#x201C; 32 mm. The finalizing treatment included conventional grading and compacting with a vibrating roller.

In this test, a Light Falling Weight Deflectometer TerraTest 3000 GPS from the TerraTest Company was used (Fig. 4 and Fig. 5). The big advantage, in comparison to other measuring methods, is the fast and easy repetition of measurements on different sample plots (Tholen et al. 1985). The measurement is based on the impulse of the light falling weight, which drops the loading plate with a diameter of 0.30 m. This impulse generates a maximum force (Fmax) of 7.07 MN. This force is gauged

Fig. 3 Scheme of the reconstruction procedure Croat. j. for. eng. 35(2014)1

G. Bohrn and K. Stampfer

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

Table 1 Elastic modulus of all sectio DBA–30

DBA–15

FBA–15

FBA–30

TWOA

REF

Date

mean

06.05.2010

19.996

29.934

18.220

16.653

21.256

36.554

12.05.2010

18.481

30.161

22.076

21.371

19.506

44.569

27.05.2010

22.392

31.701

22.167

17.897

13.027

44.619

23.06.2010

31.740

39.931

25.758

19.696

21.466

42.924

22.07.2010

51.373

42.607

41.996

36.162

41.190

62.766

23.08.2010

47.853

53.294

28.276

19.696

39.327

50.216

22.09.2010

56.360

52.937

30.300

30.241

32.049

49.666

04.11.2010

55.817

52.821

31.858

28.500

28.817

50.133

during the calibration to ensure a normal tension of 0.1 MNm-2 under the plate while performing the tests. Deflection sensors mounted on the load plate measure the deformation of the pavement in response to the load. The deformability parameter of the soil caused by this impulse is the elastic modulus called Evd and is calculated by the Light Falling Weight Deflectometer TerraTest 3000 (TERRATEST 2009).

Fig. 4 Falling weight deflectometer TerraTest 3000

Evd = 1.5 r

s max smax

(1)

Where: smax m ean value of the displacements σ4max, σ5max, σ6max of 3 measurements after 3 preconsolidation measurements, m; r radius of the plate, m; σmax normal tension under the plate, MNm-2.

Fig. 5 TerraTest 3000 in use Croat. j. for. eng. 35(2014)1

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

G. Bohrn and K. Stampfer

road. For the load detection, a second camera was mounted facing the opposite direction. The pictures from the backside of the trucks were used to estimate their load. These estimations were based on sample loads recorded on delivery notes of the saw mill and summed up daily. A total of 3 570 tons was transported on the road during the observation.

2.5 Basic meteorological data Fig. 6 Layout of elastic-modulus measurement depending on the surface location The measurements were taken on the trackway and on the medial strip of the road (Fig. 6). The reason for this layout was to find different measurement parameters for the influence of the upcoming traffic. While the whole load of the upcoming traffic was an additional compacting force on the trackways, the medial strip stayed untreated after the road construction was completed.

2.4 Traffic observation and load calculation For the calculation of the bearing load [t], the traffic was observed during the whole growing season from 6th May to 11th November 2010 (Fig. 7) with scouting cameras, which have an integrated motion detector combined with an automatic release. Pictures were taken of every type of moving vehicle, truck with or without trailer, agricultural tractor or car on the forest

The climate in the research area around Weyregg am Attersee is warm and temperate. There is significant rainfall throughout the year, even in the driest month (81 mm). The region is described as Cfb in the Köppen-Geiger climate classification system. The average annual temperature is 8.2 °C and the precipitation approximately 1 318 mm per year (Fig. 8).

3. Results and discussion In total 1 251 elastic modulus measurements were taken on eight different days. The total elastic modulus ranged from 5.70 – 114.20 MNm-2 (Table 1). The mean elastic modulus of the whole forest road was measured before any treatment and had a mean value of 32.0 MNm-2. The elastic modulus of the treated sections was reduced by the disaggregation of the mixing process. It took approximately three months for each section to reach the initial value of the forest road again. The mean elastic-modulus of the reference section varied in the observed time between 36.55 and 62.78 MNm-2 and at the end it was 50.13 MNm-2.

Fig.7 Loads of the upcoming traffic Croat. j. for. eng. 35(2014)1

G. Bohrn and K. Stampfer

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

Fig. 8 Climograph

In the case of a total reconstruction, comparison with the initial situation is not possible, because the whole road needs stabilizing or conditioning to make sure that the road can be used for traffic. With the DBA, it is possible to increase the elastic modulus by more than 20 MNm-2, compared to mechanical stabilization. FBA was not suitable for reinforcing the road construction, neither 15 % nor 30 % (Fig. 10). The two reasons for this behaviour are the different content of free lime in both ash types and the grain structure and size (von Bahr et al. 2004). The difference between the influence of the mixing values and ash types was checked with ANOVA and the Duncan Test. The results of 5th May confirmed no significant difference of the elastic modulus between FBA-30, FBA-15, DBA-30 and TWOA. DBA-15 and REF exhibited a significantly higher elastic-modulus (Table 2). The observation on 4th November 2010 shows

Fig.9 Development of elastic moduli of all sections over growing season Over the 6-month period, the sections treated with dry bottom ash increased their elastic moduli significantly. An improvement of the elastic modulus of 5.4% (52.8 MNm-2) for the DBA–15 section and 11.3% (55.8 MNm-2) for the DBA–30 section was detected. The results for the fluidized bed ash sections fell short of expectations. Only 31.9 MNm-2 for the FBA–15 section and 28.5 MNm-2 for the FBA–30 section could be reached for both mixing values. This was nearly 95% of the initial value and 63.5% (FBA–15) and 56.8% (FBA–30) of the reference value (Fig. 9).

no significant difference between both FBA sections and the TWOA section. The DBA sections showed no significant difference from the REF section (Table 3). The hypothesis that forest roads can be reinforced by the application of wood ash could be confirmed for DBA with the mixing values of 15/85 and 30/70, but disproved for FBA with a similar mixing value. After one growing season the elastic-modulus of the DBA sections had surpassed the values of the REF section. The influence of the traffic did not show the expected results. The hypothesis that the additional comCroat. j. for. eng. 35(2014)1

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

G. Bohrn and K. Stampfer

Fig. 10 Comparison of elastic moduli of sections treated with and without ash

Table 3 ANOVA (Duncan Test) for results of 4 November 2010

Table 2 ANOVA (Duncan Test) for results of 6 May 201 Section

FBA–30

Subcategory for Alpha = 0.05. 1

16.65

–

Section

FBA–30

Subcategory for Alpha = 0.05. 1

28.50

–

FBA–15

18.22

–

TWOA

28.82

–

DBA–30

20.00

–

FBA–15

31.86

–

TWOA

21.26

–

REF

–

50.13

DBA–15

–

29.93

–

DBA–15

–

52.82

REF

–

36.55

DBA–30

–

55.82

0.19

1.00

Significance

–

0.46

0.20

Significance

paction due to traffic leads to higher elastic moduli on the trackways could not be verified. The mean elastic modulus (Mean), elastic modulus of the trackway (Track) and medial strip (Middle) of three different sections were measured and compared. On 23th of August 2010 the measurements on the bottom ash section showed higher elastic modulus at the medial strip than at the trackways. The same effect was measured on 26th of June and 22th of September for the section treated without ash (Fig. 11). The reference measurements show (Fig. 9) a variation over the whole growing season, which could be caused by the influence of weather conditions. For subsequent measurements, meteorological stations will provide exact weather data for closer consideration. Croat. j. for. eng. 35(2014)1

To reach the target elastic modulus of 45 MNm-2, the self-hardening process of wood ash, initiated by water contact, is important. The dryness of the used ashes must be guaranteed via logistical organization. The ash is transported directly from the power plant to the construction site, where the ash is mixed with the raw material. It starts hardening by water contact and it is sealed with the surface layer. Outdoor storage and the contact with air humidity will definitely influence the self-hardening process of ashes and should be prevented. If there is no possibility of subsequent treatment, the ashes have to be covered with water proof material (von Bahr et al. 2004). The limiting factor in forest road construction will be the cost in money. Long term observations will show the potential of the cost effectiveness of this reinforcing method.

G. Bohrn and K. Stampfer

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89)

Fig. 11 Mean elastic moduli depending on the surface location

4. Summary

5. Literature

For the evaluation of untreated wood ash as a structural stabilizing material, two different ash types were applied in two mixture ratios, each on a 50 meter long forest road section, to investigate the load bearing capacity. The ashes were selected by their different properties: high lime and low heavy metal content, their production in Austrian biomass power plants with various furnace technologies and disposal costs. The mixing depth was 0.50 m and the road base was covered by a 0.10 m thick surface layer. The elastic moduli of these sections were measured before the application, and repeated monthly by using a light falling weight deflectometer. After the first vegetation period, the mean elastic modulus of the sections mixed with dry bed ash showed an improvement. The increase of the initial mean load bearing capacity of 32.0 MNm-2 was 65% (52.821 MNm-2) for 15/85 mixture and 76% (55.817 MNm-2) for 30/70 mixture. The results for the fluidized bed ash sections fell short of expectations. Only 99% (31.858 MNm-2) for the 15/85 mixture and 89% (28.500 MNm-2) for the 30/70 mixture of the initial value could be reached. The results of the field trial show that dry bed ash is suitable as a structural stabilizing material. This could not be confirmed for the fluidized bed ash. In this case other utilization methods should be exploited.

Dietz, P., Knigge, W., Löffler, H., 1984: Walderschließung. Ein Lehrbuch für Studium und Praxis unter besonderer Berücksichtigung des Waldwegebaus. Parey, Hamburg und Berlin 1984, ISBN 3-490-02116-9. 436 p.

Environment Agency Austria 2009: Baseline study of Austrians waste management, Status Report 2008, Vienna. Holzner, H., Obernberger, I., 2011: Guideline for appropriate usage of biomass ashes on agricultural and forested areas, Advisory Board for Soil Fertility, Federal Ministry of Agriculture, Forestry, Environment and Water Management, 66 p. Holzner, H., 1999: Die Verwendung von Holzaschen aus Biomassefeuerungen zur Düngung von Acker und Grünland, Doctor Thesis, Division of Agronomy, University of Natural Resources and Life Sciences, Vienna, Austria, 126 p. Kärrman, E., Van Moeffaert, D., Bjurström, H., Berg, M., Svedberg, B., 2004: Prerequisites for an effective use of ashes in road construction, Värmeforsks askprogram, project Q4207, VÄRMEFORSK Service AB, 101 53 Stockholm, 85 p. ISSN 0282-3772. Lahtinen, P., 2001: Fly Ash Mixtures as Flexible Structural Materials for Low Volume Roads, Finnra Reports 70/2001, 95 p. + annexes 55 p. ISBN 951-726-826-2. Mácsik, J., 2006: Fly ash stabilisation of gravel roads, Värmeforsks askprogram, project Q4-229, VÄRMEFORSK Service AB, 101 53 Stockholm, 91 p. ISSN 1653-1248. Mácsik, J., Svedberg, B., 2006: Gravel road stabilisation of Ehnsjövägen, Hallstavik, Värmeforsks askprogram, project Croat. j. for. eng. 35(2014)1

Untreated Wood Ash as a Structural Stabilizing Material in Forest roads (81–89) Q4-285, VÄRMEFORSK Service AB, 101 53 Stockholm, 75 p. ISSN 1653-1248. Munde, H., Svedberg, B., Mácsik, J., Maijala, A., Lahtinen, P., Ekdahl, P., Néren, J., 2006: Fly ash in civil engineering, Gravel roads, Värmeforsks askprogram, project Q4-270, VÄRMEFORSK Service AB, 101 53 Stockholm, 90 p. ISSN 1653-1248. Stupaka, I., Asikainen, A., Jonsell, M., Karltun, E., Lunnan, A., Mizaraité, D., Pasanen, K., Pärn, H., Raulund-Rasmussen, K., Röser, D., Schroeder, M., Varnagiryté, I., Vilkriste, L., Callesen, I., Clarke, N., Gaitnieks, T., Ingerslev, M., Mandre, M., Ozolincius, R., Saarsalmi, A., Armolaitis, K., Helmisaari, H.S., Indriksons, A., Kairiukstis, L., Katzensteiner, K., Kukkola, M., Ots, K., Ravn, H.P., Tamminenj, P., 2007: Sustainable utilisation of forest biomass for energy – Possibilities and problems: Policy, legislation, certification, and recommendations and guidelines in the Nordic, Baltic, and other European countries, Biomass and Bioenergy 31 (10): 666–684.

G. Bohrn and K. Stampfer

Tholen, O., Sharma, J., Terrel, R.L., 1985: Comparison of Falling Weight Deflectometer with other Deflection Testing Devices, Transportation Research Record N1007, Nondestructive Pavement Evaluation and Overlay Design. p. 20–26. Thurdin, R. 2004: Environmental Impact of Bio Fuel Ash in Road Construction, Mid Sweden University Licentiate Thesis 9, 35 p. ISSN 1652-1064. von Bahr, B., Ekvall, A., Schouenborg, B., 2004: Quality criteria for bottom ashes for civil construction, Värmeforsks askprogram, project Q4-143, VÄRMEFORSK Service AB, 101 53 Stockholm, 107 p. ISSN 0282-3772. von Bahr, B., Loorents, K., Ekvall, A., Arvidsson, H., 2006: Quality criteria for bottom ashes for civil construction, Part II – Technical characteristics of bottom ashes, Värmeforsks askprogram, project Q4-282, VÄRMEFORSK Service AB, 101 53 Stockholm, 152 p. ISSN 1653-1248.

Authors’ address:

Received: October 16, 2013 Accepted: February 10, 2014 Croat. j. for. eng. 35(2014)1

Gerald Bohrn, MSc.* e-mail: gerald.bohrn@boku.ac.at Prof. Karl Stampfer, PhD. e-mail: karl.stampfer@boku.ac.at University of Natural Resources and Applied Life Sciences Vienna Department of Forest and Soil Sciences Institute of Forest Engineering Peter Jordan Straße 82/3 1190 Vienna AUSTRIA * Corresponding author

Original scientific paper

Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland Jukka Malinen, Ville Nousiainen, Kari Palojärvi, Teijo Palander Abstract The objective was to study how entrepreneurs taking care of the transport of wood perceive the current challenges in the operating environment and reflect these challenges in the changing climate. The data was collected in spring 2012 by mail questionnaire, which was answered by a total of 86 entrepreneurs giving a response rate of 19.1%. According to the respondents, the most debilitating factors in the changing infrastructure in future will be the challenges of the low volume road network, especially winter maintenance including removing snow and ice and preventing slippery conditions. The entrepreneurs were concerned about labour costs caused by a possible shortage of skilled drivers. The main problem concerning Information and Communication Technology (ICT) in timber trucking logistics is the inability to use one ICT system for two or more clients, which leads to problems in handling multiple clients. Almost all respondents agreed that timber trucking enterprises are too dependent on clients’ ICT systems. Keywords: Wood procurement, wood transport, ICT, business environment, profitability

1. Introduction In Finland, 76% of roundwood used by the forest industry is transported to the mill by 1 700 trucks (Finnish statistical... 2012). The remaining 24% is transported to the mill either by train or waterways, although this includes timber trucking at the beginning of the transportation chain, 50 km on average. In general, timber trucking is a flexible transportation method in circumstances where the starting point of transportation is continuously changing, and the extent of value added is low. Thus, timber trucking is the most important transportation method for the raw material supply of the Finnish forest industry. A typical roundwood truck in Finland operating in the Nordic cut-to-length system is a three-axle truck pulling a four-axle trailer. The maximum weight, including payload, of the vehicle was limited to 60 t, but on Oct. 1, 2013, the maximum permitted weight of the vehicles was increased to 76 t. Regardless of the high load, the trucks are designed to operate on public roads and to work under demanding conditions on Croat. j. for. eng. 35(2014)1

narrow and sometimes inadequate forest roads. The trucks are equipped with removable roundwood cranes and separate loaders are seldom used. Roundwood trucks are typically owned by family enterprises situated in the countryside, where the entrepreneur participates actively in production work. There are about 900 timber trucking entrepreneurs employing 2 600 truck drivers, and the average number of trucks per enterprise is less than two (Rieppo 2010). At the same time, customers are concentrated; the share of the three largest customers is about 90%. This, with specialised trucks incapable of competing in other markets, has produced strong dependency between timber trucking entrepreneurs and the Finnish forest industry. Timber trucking in Finnish conditions has been considered relatively competitive compared to other countries (Högnäs 2001). In 2000, about 2/3 of the timber trucking enterprises were considered to be profitable with a positive net income and over 15% return on invested capital (Högnäs 2001). In the period 2001

J. Malinen et al. Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100)

– 2006, timber trucking enterprises grew by about 10% per year. The profitability of the smallest companies with a turnover under € 1 million clearly benefited from the growth, whereas € 1 – 1.5 million seems to be the most efficient size for a timber trucking enterprise (Soirinsuo and Mäkinen 2010). The competitiveness of Finnish timber trucking is based on long term development work, investments in information and communication (ICT), optimisation systems and high net carrying loads. Although timber trucking has been a cost effective transportation method, there are some threats to be faced. The rising price of fuel, environmental impact, changing regulations and taxes and changes caused by climate change may have an effect on the operational environment of timber trucking. In the changing operating environment, the forest industry is attempting to improve its cost effectiveness by offering extended entrepreneurship agreements that increase the responsibilities of timber trucking entrepreneurs (Palander et al. 2006). In a typical regional entrepreneurship application, the wood procurement organisation signs haulage contracts with fewer entrepreneurs, each responsible for a larger area (Palander et al. 2012). These contracts include tasks, which are larger and more diverse including more responsibilities than in conventional contracts, and thus the requirements for entrepreneurs are growing, as well their degree of freedom for organising the work. Roundwood logistics in Finland is controlled by ICT systems owned by wood procurement companies. Each company has its own ICT system, which is delivered to the timber trucking entrepreneur. If the timber trucking entrepreneur has multiple clients, the onboard computer, or at least the hard drive, has to be changed. This has complicated operations with multiple clients since the optimisation of the routes and timetables is more difficult, preventing the chance of carrying payload during backhaul instead of driving empty (Palander and Väätäinen 2005). Additionally, drivers have to master several different ICT systems. However, recently an integrated ICT system, the LogForce, for timber trucking enterprises has been developed, and at least two of the three largest wood procurement companies are planning to use it. According to final report of the investigation of roundwood trade by the Finnish Ministry of Employment and Economy (Puumarkkinoiden… 2009), the extended use of integrated roundwood transportation logistics ICT systems will promote more efficient timber trucking, especially enhancing the operational environment and cost effectiveness of timber trucking enterprises. In the changing business environment of timber trucking entrepreneurs, the management of cost struc-

ture and price setting is becoming more demanding. The margins are low and small mistakes in pricing may steer the enterprise to a situation where the price does not cover expenses. For assigning logistic costs to roundwood assortment and lots, Nurminen et al. (2009) introduced an activity based cost (ABC) management system for cut-to-length roundwood harvesting and trucking. In the system, costs are traced to individual stands and to roundwood assortment lots from a stand. Palander et al. (2012) have also suggested planning systems based on the dynamic activity based costing (DABC), which can be used for the determination of certain costs and their sources in the comparative work studies of time dependent work processes. These systems would promote development of a more efficient and plausible ICT system for timber trucking organisations. Finnish industry, with the forest industry at the forefront, has been demanding higher loads for trucking. Although gross weights for trucks are already at the top of the European scale (Permissible… 2012), both Finland and Sweden are pushing towards higher gross weights. In Finland, the maximum weight of the trucks has been increased to 76 t for vehicles equipped with four axles on the truck and five on the trailer (Korpilahti 2013), and the first experimental permit for the 80 t truck has been given. In Sweden, research project ‘ETT, Modular System for Timber Transport’ has run vehicles with a gross weight of up to 90 t on public roads. In Sweden, the argument for an increase in gross weights has been supported by lower fuel consumption and therefore lower emissions of environmental contaminants and lower transport costs (ETT, Modular System… 2013). In Finland, the primary argument has been the global economy, especially the effect of the EU’s sulphur directive on Finnish industry. The sulphur directive restricts the sulphur content of fuel used in the shipping industry to 0.1% in the Baltic Sea, which is classified as a Sulphur Emission Control Area under the directive, meaning stricter limits than for waters in southern Europe. Furthermore, in Finland, the increase in gross weight has been justified by the smaller stress on the road caused by the increased number of axles and better road safety caused by the improved weight ratio between truck and trailer. Recently in Finland, some experiments have been done in Metsähallitus (Finnish state run enterprise managing state forests) with central tyre inflation (CTI) systems, which have been used more frequently, for example in Canada. In the CTI system, the vehicle operator can change the tyre inflation pressure while the vehicle is moving. The results have been promising. Unpublished findings correspond to the findings Croat. j. for. eng. 35(2014)1

Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100) J. Malinen et al.

of Woodrooffe and Burns (1998) and Vuorimies et al. (2009), who discovered that lower tyre inflation pressure reduces road wear and improves vehicle performance in traction, braking, ride quality and vehicle maintenance. However, return on investment in CTIsystems has been questioned by timber trucking entrepreneurs but, due to costs saved in road maintenance and heavier gross weights of trucks, it is more likely that the renewal of contracts demands roundwood trucks with CTI-systems. In Nordic countries such as Finland, weather conditions mean that the operating environment is often challenging. There are three major weather factors that affect road transport (Saarelainen and Makkonen 2007): temperature, rain and wind. Low (or high) temperature, rainstorms, increasing average rainfall and violent winds cause disturbance in road transport, damage road infrastructure and complicate road maintenance. Some weather conditions have direct implications for road maintenance and transportation. Roads must be cleared of snow and slippery ice needs anti-icing or gritting. The Finnish climate is characterised by large variation between years. On the basis of climate change models, the mean temperature is expected to rise and precipitation will increase, particularly in winter time. These changes will affect timber trucking. Some of the effects of climate change will be favourable, but some will be unfavourable (Tiehallinto 2009). Climate change will affect different parts of Finland differently. For instance, anti-icing and de-icing will decrease in Southern Finland, but increase in Central Finland and Northern Finland. Nevertheless, climate change is expected to increase the total costs of road maintenance to some extent. The aim of the study was to see how timber trucking entrepreneurs experience the operating environment of timber trucking and what they feel about the ICT systems of roundwood logistics used in operational management, and to consider the results in the future operating environment as climate changes. The infrastructure and logistic chain as a whole are under consideration.

2. Material and Methods The study data was collected in spring 2012 using a 5 page questionnaire sent by mail to Finnish timber trucking entrepreneurs, of which 410 belonged to the Finnish Association of Forest Road Carriers. This association represents companies offering road timber trucking services for forestry. The sample was complemented with 40 entrepreneurs not belonging to the asCroat. j. for. eng. 35(2014)1

sociation, making a total of 450. Answering the question naire was estimated to take approximately 20 minutes. In total, 86 entrepreneurs returned the questionnaire giving a response rate of 19.1%. The questionnaire was divided into five parts. The first part asked entrepreneurs about their current situation and obtained other background information. The second part considered the infrastructure of timber trucking and the availability of manpower. The third part considered opinions about the ICT systems of logistics used in operational management. The fourth part examined the overall operational environment of timber trucking and the last part examined the operations of particular enterprises. Most of the questions were composed using a five point Likert scale. The respondents were divided into three groups with the number of trucks as the indicator of the size of a respondent’s operations. Out of 86 respondents, 40 had only one truck (»small company«), 18 had two trucks (»medium company«) and 28 had more than two trucks (»large company«). The biggest company had 14 trucks. The average weekly number of working hours of entrepreneurs varied greatly between respondents. The average number was 62.5 hours per week, the least being 4 hours per week and the greatest 105 hours. There were no statistically significant differences in the number of weekly working hours between entrepreneurs of different sizes. The relationships between questionnaire variables were examined using principal component analysis (PCA). PCA is a statistical technique where correlated variables are transformed into a smaller set of uncorrelated, composite variables called principal components (PCs). The use of PCA with a dataset as small as this (n < 100) is questionable, but because of the item ratio (over 20:1), the critical use of PCA was enabled (Osborne and Costello 2004). PCA involves calculating the eigenvalue decomposition of a data covariance matrix and the results are usually discussed in terms of PC loadings. In this study, PC loadings smaller than 0.40 were considered insignificant, and they were removed from the PCs. The variables (n = 21) depicting the operational environment of timber trucking were reduced to three PCs. Both the Kaiser-Meyer-Olkin measure of sampling adequacy (0.763) and Barlett’s test of sphericity (< 0.00) show that the data is appropriate for this type of analysis. Since the PCs were correlated with many variables, the varimax rotated factor matrix was computed. By rotating the factor, we would like each factor to have non zero or significant loadings for only some variables. Rotation does not affect the communalities and the percentage of total variance explained.

J. Malinen et al. Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100)

In the analysis of the background information and the attitude results, non-parametric analysis of variance (the Kruskal-Wallis test) and the Mann-Whithey U-test were utilised. The tests were selected because the values of answers did not show a normal distribution. The statistical significance level of p < 0.05 was used for all results.

3. Results 3.1 Profitability of timber trucking enterprises Almost half of the respondents felt that in recent years the profitability of timber trucking in their enterprise had decreased greatly (22.6%) or to some extent (23.8%). One-third (33.3%) answered that the prof-

Table 1 Attitudes of entrepreneurs towards the infrastructure of timber trucking by the geographical location of their enterprise. 1 »Not at all influential«; 2 »Slightly influential«; 3 »Somewhat influential«; 4 »Very influential«; 5 »Extremely influential« Northern Finland (A)

Western Finland (B)

Eastern Finland (C)

Southern Finland (D)

Average

Winter maintenance (incl. removing of snow and ice and slip prevention)

4.6

0.7

4.3

1.0

4.4

0.9

4.4

0.6

4.44 0.85

Dimensions and condition of storage areas and intersections (B – C*)

4.0

0.8

4.3

0.8

3.9

0.7

4.2

0.7

4.05 0.80

Condition and sizing of passing points and turning places (C – D*)

4.1

0.9

4.1

0.9

3.7

0.8

4.3

0.8

4.01 0.89

Bearing capacity of roads (year – around)

3.9

0.7

4.1

0.8

4.0

1.0

4.0

1.0

3.99 0.86

Proportion of multi-source loads

3.7

1.2

4.0

1.2

3.8

0.9

4.2

0.9

3.89 1.06

Frost-damaged roads (incl. spring and autumn damages)

3.9

0.7

3.9

0.8

3.9

1.1

3.8

0.9

3.88 0.89

Condition of road surface

3.7

0.7

3.8

1.0

3.9

1.0

3.7

0.9

3.80 0.91

Grading of road surface

3.7

1.4

3.8

1.1

3.8

1.3

3.8

1.3

3.78 1.25

Terminal operations at pulp and paper mills

3.9

0.9

3.7

1.0

3.8

0.9

3.7

0.9

3.76 0.90

Terminal operations at sawmills

3.9

1.2

3.7

0.9

3.6

1.1

3.6

1.0

3.71 1.06

Roundwood stacks and loading conditions

3.7

1.1

3.7

1.1

3.3

1.1

3.6

0.6

3.56 1.03

Slopes and curves at forest roads

3.3

1.1

3.3

1.2

3.6

0.9

4.0

1.1

3.52 1.08

Condition of bridges, tunnels and pipe culverts (C – D*)

3.4

1.0

3.5

1.1

3.3

0.9

4.0

1.1

3.48 1.02

Proportion of driving empty (B – D**)

3.3

1.2

3.6

1.1

3.3

1.1

3.8

1.1

3.46 1.14

Particularly small roundwood lots

3.5

0.9

3.1

0.9

3.5

1.0

3.9

0.8

3.45 0.95

Other road maintenance (removing of stones, repairs of damages)

3.3

1.0

3.5

1.0

3.4

0.8

3.6

1.1

3.42 0.93

Sizing of terminals and loading points for short-term storing (A – D*, B – C*, C – D**) 3.5

0.9

3.0

1.2

3.8

1.0

2.6

1.1

3.37 1.10

Possibilities for intermediate storage of roundwood

3.4

1.1

3.1

1.2

3.4

1.0

3.0

0.8

3.25 1.04

Hot shot hauling of roundwood by truck to trailer (A – D*)

3.0

1.1

3.1

1.1

3.3

1.2

3.9

0.9

3.24 1.11

Clearing of roadside coppice (A – B*)

2.7

0.9

3.5

1.2

3.0

1.1

3.3

1.4

3.09 1.13

Parking and rest areas

3.0

1.1

2.5

1.2

3.0

1.1

2.8

1.2

2.84 1.15

Terminal operations for railway transportation (A – D**)

3.2

1.0

2.6

1.1

2.8

1.3

2.1

0.8

2.76 1.16

Other terminal operations (A – D*, B – D*, C – D*)

2.8

1.0

2.7

1.1

3.0

1.2

1.9

0.8

2.70 1.12

Roundwood rafting terminals (A – B*, A – D*, B – C**, C – D*)

2.2

1.3

1.6

1.1

2.5

1.2

0.6

2.01 1.21

* p < 0.05; ** p < 0.01; *** p < 0.001, Mann-Whitney U-test

Croat. j. for. eng. 35(2014)1

Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100) J. Malinen et al.

itability of their enterprise was unchanged, and the rest answered that their profitability had increased to some extent (16.7%) or greatly (3.6%). The size of the enterprise or geographical location did not have any statistical significance on profitability. The prospects for the future were similar. Four out of ten felt that their profitability had decreased, half of them greatly (20.2%) and other half (19.0%) felt that it had decreased to some extent and would continue to do so. The same number of respondent (39.3%) believed that their profitability would continue on a similar level, whereas the rest felt that they had increased profitability to some extent (20.2%) or greatly (1.2%).

3.2 Infrastructure of timber trucking The respondents’ opinion of the infrastructure of timber trucking was examined by 24 questions. The questions were prepared using the Likert-scale, where: 1 »Not at all influential«, 2 »Slightly influential«, 3 »Somewhat influential«, 4 »Very influential«, 5 »Extremely influential«. All variables within the twelve most important variables were considered to have a great effect on the operating environment of timber trucking (Table 1).

Table 2 Varimax rotated factor matrix for the PCA of the variables depicting the attitudes of entrepreneurs towards the infrastructure of timber trucking PC1

PC2

PC3

Winter maintenance (incl. removing of snow and ice and slippery prevention)

.728

–

Condition of bridges, tunnels and pipe culverts

.717

–

Dimensions and condition of storage areas and intersections

.707

–

Grading of road surface

.681

–

Condition of road surface

.662

–

Bearing capacity of roads (year-around)

.631

–

Other road maintenance (removing of stones, repairs of damages)

.617

–

Condition and sizing of passing points and turning places

.566

–

Slopes and curves at forest roads

.547

–

Frost-damaged roads (incl. spring and autumn damages)

.539

–

Roundwood rafting terminals

–

.765

–

Sizing of terminals and loading points for short-term storing

–

.716

–

Terminal operations for railway transportation

–

.681

–

.488

.638

–

Terminal operations at pulp and paper mills

–

.594

–

Terminal operations at sawmills

–

.586

–

.445

.566

–

Particularly small roundwood lots

–

.809

Hot shot hauling of roundwood by truck to trailer

–

.787

Proportion of multi-source loads

–

.606

Roundwood stacks and loading conditions

–

.586

Extraction sums of squared loadings

5.672

2.362

2.176

Coefficient of determination (%)

31.54

13.12

12.08

Cronbach’s alpha

0.83

0.839

0.746

Parking and rest areas

Possibilities for intermediate storage of roundwood

Croat. j. for. eng. 35(2014)1

J. Malinen et al. Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100)

The most influential infrastructure factor affecting timber trucking was winter maintenance, including removing snow and ice and anti-slip measures. There were no statistically significant differences between different geographical regions in Finland, although winter maintenance was most influential in Northern Finland where the winter is longest and the amount of snowfall the largest. The second influential infrastructure factor was dimensions and the condition of storage areas and intersections. This, as well as the third influential factor, the condition and sizing of passing points and turning places, was a greater problem in densely populated areas in Southern and Western Finland than in rural areas in Eastern and Northern Finland. The infrastructure of timber trucking was analysed by PCA, which converts a set of correlated variables into a set of linearly uncorrelated variables, PCs. In the study, 21 original variables were transformed into three PCs (Table 2), which were named according to variables included in the PCs. PC1 included variables depicting the condition and usability of the road network, hence PC1 was named »low volume roads«. The extraction sums of squared loadings was high (5.7), and PC1 explained almost one third of the variation. The internal consistency depicted by Cronbach’s alpha value (0.83) was also high. PC2 was named »Loading and terminal operations« due to variables included in PC2. The extraction sums of squared loadings were 2.4, and Cronbach’s alpha 0.84. PC3 included variables depicting load

specific factors and was named »Stand specific factors«. The extraction sum of squared loadings for the third PCI was 2.2, and Cronbach’s alpha 0.75.

3.3 Availability of manpower The attitudes of timber trucking entrepreneurs towards factors depicting the availability of manpower were studied by nine questions prepared using the Likert scale. The most important factor was a possible increase in drivers’ salaries, and consequently, an increase in timber trucking costs caused by a possible labour shortage (Table 3). This is mostly caused by the second important factor: most of the current drivers are retiring in the next few years. However, especially in large companies, the entrepreneurs felt that the drivers were committed to the enterprise and timber trucking business. At the same time, they felt that the means to prevent a labour shortage are not working properly: strengthened operations and larger enterprises will not prevent a labour shortage, young persons are not interested in timber trucking as a profession and the co-operation between timber trucking enterprises and learning institutes is not working properly.

3.4 ICT systems of logistics used in operational management The entrepreneurs responded to 18 claims concerning the ICT systems of logistics in operational management. Attitudes were neutral towards most of the claims, especially towards claims concerning the op-

Table 3 Attitudes of entrepreneurs towards variables describing the availability of manpower by the size of the company. 1 »Not at all influential«; 2 »Slightly influential«; 3 »Somewhat influential«; 4 »Very influential«; 5 »Extremely influential« Small company A

Medium company B

Large company C

Average

Possible labour shortage increases salary and transportation costs

4.2

0.7

3.9

1.1

4.0

1.0

4.1

0.9

Most of the current drivers are retiring in the next few years (A – C*)

4.1

0.9

3.6

1.1

3.4

1.2

3.7

1.1

Employees are committed to the enterprise and timber trucking (A – C**)

3.3

1.0

3.1

0.9

3.9

0.7

3.4

0.9

I’m aware of how the process of succession is conducted

3.2

1.5

3.4

1.5

2.8

1.5

3.1

1.5

Induction for new employees is a planned process in my enterprise

2.9

1.1

2.9

0.8

3.1

0.8

3.0

1.0

Succession will be topical in the coming years

2.8

1.7

2.7

1.7

2.4

1.4

2.6

1.6

Co-operation between enterprises and learning institutions is working well (A – -C*)

2.6

0.7

2.2

0.7

2.2

0.9

2.4

0.8

Young persons are interested in timber trucking

2.4

0.9

2.1

0.9

2.4

1.1

2.3

1.0

Strengthened operations and larger size enterprises will help labour shortage

2.3

0.9

2.2

1.0

2.2

0.9

2.2

0.9

* p < 0.05; ** p < 0.01; *** p < 0.001 Mann-Whitney U-test

Croat. j. for. eng. 35(2014)1

Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100) J. Malinen et al.

erational function of current ICT systems (Table 4). The strongest agreement was towards the claim »entrepreneurs are completely dependent on clients’ ICT systems«, and the strongest disagreement towards the claim »ICT systems are compatible with different operations«.

4. Discussion The main objective of the study was to investigate what timber trucking entrepreneurs feel about the operating environment of timber trucking and the ICT systems of roundwood logistics used in operational management. The main concern regarding the operating environment was the low volume roads. Winter maintenance including removing snow and ice and anti-slip measures especially caused concern. Winter maintenance has a large impact not only on the logistics of timber trucking, but also road safety. Although the dimensions and condition of low volume roads and seasonal problems such as year-around bearing

capacity and frost-damaged road were raised, winter maintenance was clearly a greater cause for concern than other problems in low-volume roads. Although the climate is changing due to increased CO2 levels caused by human actions, the importance of winter maintenance in timber trucking will not be reduced. According to the Finnish Road Administration (Tiehallinto 2009), increasingly mild winters are expected to mean fewer days of sub-zero temperatures in Southern Finland, reducing the need for winter maintenance in southern parts of the country. However, the need for slip prevention is likely to increase in Central Finland and Northern Finland, and the focus of winter road maintenance will shift further north. The likelihood of freezing rain, which creates treacherous road conditions, is expected to decrease in Southern and Central Finland. However, freezing rain is expected to become slightly more common in the northernmost parts of Finland, as temperatures around zero degrees become more common. Shortlived snowstorms, when a lot of snow falls at once

Table 4 Attitudes of entrepreneurs towards ICT systems of logistics used in operational management by the size of the company. 1 »Strongly disagree«, 2 »Disagree«, 3 »Neither agree or disagree«, 4 »Agree«, 5 »Strongly agree« Small company A

Medium company B

Large company C

Average

Entrepreneurs are completely dependent on clients’ ICT systems

4.2

1.0

4.2

0.8

4.0

1.0

4.2

1.0

ICT has improved possibilities to serve customers (B – C*)

3.8

1.0

3.5

0.9

4.1

0.9

3.8

1.0

ICT has improved the management and execution of timber trucking (B – C*)

3.6

0.8

3.2

0.9

3.8

1.0

3.5

0.9

Employees demand more education regarding ICT systems

3.6

0.9

3.2

1.1

3.4

1.0

3.5

1.0

Profitability of ICT system investments includes uncertainty

3.5

1.1

3.8

1.1

3.4

0.8

3.5

1.0

ICT systems overly dominating the drivers’ work

3.2

1.2

3.4

1.3

3.4

1.0

3.3

1.1

ICT supports extended responsibility of entrepreneurship

3.2

0.9

3.6

0.9

3.2

0.8

3.3

0.9

Current ICT systems are adequate for the required working efficiency

3.3

1.0

3.4

1.0

3.1

1.0

3.3

1.0

Helpdesk operations help me when needed

3.1

1.3

3.1

1.1

3.3

1.1

3.2

1.2

Management of operations works well with current ICT systems

3.2

1.1

3.1

1.2

3.0

1.0

3.1

1.1

Information exchange routes between prime- and subcontractors are sufficient

3.1

0.8

2.9

1.0

3.0

0.8

3.0

0.8

ICT has decreased the number of human mistakes (A – B*)

3.2

1.0

2.7

0.8

3.0

1.0

3.0

0.9

Current ICT systems are adequate for different clients

2.8

1.1

2.9

1.2

3.2

1.0

2.9

1.1

Tailored ICT systems are available

2.8

0.9

2.8

1.0

2.9

0.9

2.8

0.9

The ICT system used by my enterprise is insufficient

2.6

1.1

2.5

1.3

2.7

1.0

2.6

1.1

ICT has reduced the costs

2.7

1.2

2.4

1.1

2.6

0.9

2.6

1.1

I demand more ICT for my everyday routines

2.1

0.9

1.9

0.8

2.1

1.0

2.1

0.9

ICT systems are compatible with different operators

2.2

1.1

1.9

0.8

2.1

1.1

2.1

1.0

* p < 0.05; ** p < 0.01; *** p < 0.001 Mann-Whitney U-test

Croat. j. for. eng. 35(2014)1

J. Malinen et al. Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100)

causing disruptions to timber trucking, are likely to increase. All in all, the need for snow removal is likely to remain relatively unchanged or to decrease, because the period of snow cover will become shorter, and the need for slip prevention will shift further north (Tiehallinto 2009). This will not help timber trucking in Northern Finland, where winter maintenance is already a greater problem than in southern regions. The year-around bearing capacity was the fourth most important factor affecting timber trucking. The bearing capacity of the road is dependent on the subsoil, drainage and structure of the road framework. In wintertime, ground frost helps the bearing capacity and some of the roads are so called winter roads, which are designed to access harvesting sites in locations where it is not profitable to build a summer road, or the construction and use of a summer road is not feasible for reasons of conservation. Climate change may affect the likelihood of ground frost in two ways. On the one hand, higher average temperatures limit frost formation, and frost may not penetrate as deep into the ground as before. On the other hand, the thinner snowpack may counteract this: frost penetrates deeper into the ground if temperatures drop below freezing when there is no protective layer of snow. The likelihood of frost is nevertheless expected to decrease on the whole, and this will have negative implications on roundwood and energy wood harvesting and hauling since ground and forest roads will no longer be able to support the weight of forest machinery and roundwood trucks as well as before, so there will be problems for roundwood and energy wood logistics (Marttila et al. 2005, Tiehallinto 2009). At the same time, the timber trucking environment is going through two changes affecting to bearing capacity of roads. CTI systems will reduce the ground pressure allowing transportation on roads, which would not otherwise permit the load of roundwood trucks. On the other hand, realized higher gross weights in Finland from the current 60 tonnes to up to 76 tonnes demand more from the road network. In conclusion concerning the impact of climate change on Finnish timber trucking, Southern and Western Finland will have fewer problems with winter maintenance, but more problems with year-around bearing capacity and the condition of the road surface. The problems with winter maintenance will increase in Northern Finland, whereas the operating environment in Eastern Finland will remain the same. According to the respondents of the study, the sizing and condition of storage areas, intersections, passing points and turning places was a great problem especially in Western and Southern Finland. The un-

satisfactory sizing and condition of forest roads is a well-known problem for all parties in timber logistics. However, the problem is diverse and more connected to deserted rural areas. To reduce the problems of forest roads, co-operation between forest owners, government and forest industry must be enhanced. The results of the study emphasised concern over the availability of manpower in timber trucking. Consequently, timber trucking entrepreneurs were worried about labour costs caused by a shortage of skilled drivers. According to Solakivi et al. (2012), there is a need to ensure adequate supplies of labour, skills and expertise, and first rate training and research to sustain competitiveness in the hauling business. Although the problem of labour availability and commitment to timber trucking business was a larger problem in small companies, the respondents of the study did not feel that being larger helped the problem. The main problem concerning ICT in timber trucking logistics is the inability to use one ICT system with two or more clients causing problems in handling multiple clients. Almost all respondents agreed that timber trucking enterprises are too dependent on the clients’ ICT systems. However, recent development towards global transaction standards for paper and forest supply chains (PapiNet 2013) and an integrated ICT system (LogForce) for timber trucking enterprises may enhance the possibilities to handle multiple clients.

5. Conclusion The results of the present study indicated that timber trucking entrepreneurs felt that the changing environment, low-volume roads and the availability of skilled labour were major complications for their business. The condition and maintenance of low volume roads, especially in winter time, is one of the crucial factors in successful roundwood supply in Finland. However, due to tight financial circumstances, it is unlikely that the budget for road maintenance will increase as demand increases. To facilitate roundwood supply for the Finnish forest industry, research will need to develop new concepts for timber trucking. As a start, in spring 2013, Metsäteho Oy, a Finnish research company, awarded a €10 000 prize for the completion of ‘new innovations in timber trucking’. The winning entry included several innovations for adapting roundwood trucks according the hauling task. The timber trucking operating environment is changing due to enhancement in ICT technology, climate change and increasingly fierce global competition. To enhance the hauling environment, entrepreneurs are willing to participate in planning a timber Croat. j. for. eng. 35(2014)1

Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100) J. Malinen et al.

trucking logistic chain together with their clients. Entrepreneurs also want ICT systems that can be used with multiple clients. According to Pajuoja and Hämäläinen (2012), roundwood logistics is demanding new operating models for roundwood harvesting and hauling, wood procurement and the whole forest cluster. Development is towards co-operational models between timber trucking entrepreneurs, including joint venture and subcontracting as mechanisms for extending their responsibilities. To facilitate the development, of co-operation, the objectives for ICT systems should be determined quickly.

Acknowledgements The authors are grateful to the Metsämiesten säätiö foundation for funding. We also would like to thank all the entrepreneurs who participated in our study by responding to the questionnaire.

6. References ETT – Modular System for Timber Transport. Skogforsk project brochure 2013: 38 p. http://www.skogforsk.se/en/ Press/ETT---Modular-system-for-timber-haulage / (Accessed 13. 2. 2013) Finnish Statistical Yearbook of Forestry, 2012: Editor-inChief: E. Ylitalo. Finnish Forest Research Institute. 454 p. Högnäs, T., 2001: A Comparison of Timber Haulage in Great Britain and Finland. Forestry publication of Metsähallitus 39: 31 p. Korpilahti, A., 2013: Puutavara-autot mitta- ja massamuutoksen jälkeen. [Timber trucks according new height and weight restrcitions]. Metsätehon tuloskalvosarja 11/2013. http://www. metsateho.fi/files/metsateho/Tuloskalvosarja/Tuloskalvosarja_2013_11_Puutavara-autot_mitta_ja_massamuutoksen_jalkeen_ak.pdf (In Finnish) (Accessed 11. 2. 2014). Marttila, V., Granholm, H., Laanikari, J., Yrjölä, T., Aalto, A., Heikinheimo, P., Honkatuki, J., Järvinen, H., Liski, J., Merivirta, R., Paunio, M., 2005: Ilmastonmuutoksen kansallinen sopeutumisstrategia. [National Strategy for Adapting Climate Change]. Report of the Ministry of Agriculture and Forestry 1/2005. 276 p. (In Finnish) Nurminen, T., Korpunen, H., Uusitalo, J., 2009: Applying the activity-based costing to cut-to-length timber harvesting and trucking. Silva Fenn. 43(5): 847–870. Osborne, J.W., Costello, A.B., 2004: Sample size and subject to item ratio in principal components analysis. PARE 9(11). http://pareonline.net/getvn.asp?v=9&n=11 (Accessed 18. 2. 2013) Palander, T., Väätäinen, J., 2005: Impacts of inter-enterprise collaboration and backhauling on wood procurement in Finland. Scand. J. Forest Res. 20: 177–183. Croat. j. for. eng. 35(2014)1

Palander, T., Säynäjärvi, T., Högnäs, T., 2006: Puutavaran autokuljetuksen uudet organisointimallit. [New organising models of timber truck transportation]. Metsätieteen aikakauskirja 1: 5-22. (In Finnish.) Palander, T., Vainikka, M., Yletyinen, A., 2012: Potential Mechanisms for Co-operation Between Transportation Entrepreneurs and Customers: a Case Study of Regional Entrepreneurship in Finland. Cro. J. For. Eng. 33(1):89–103. PapiNET. 2013: http://www.papinet.org/ (Accessed 1. 3. 2013) Permissible Maximum Weights of Trucks in Europe. 2011: International transport forum. www.internationaltransportforum.org (Accessed 13. 2. 2013) Persson, G., Bärring, L., Kjellström, E., Strandberg, G., Rummukainen, M., 2007: Climate indices for vulnerability assessments. SMHI Reports Meteorology and Climatology. No. 111 Aug 2007. Swedish Meteorological and Hydrological Institute. h t t p : / / w w w. s m h i . s e / p o l o p o l y _ fs/1.2096!RMK111%5B1%5D.pdf (Accessed 1. 4. 2013) Puumarkkinoiden toimintaa koskeva selvitys – Työ ja elinkeinoministeriö. 2009. [Final Report of investigation of Roundwood trade by Ministry of Employment and Economy]. Pöyry Forest Industry Consulting Oy. 84 p. (In Finnish) Rieppo, K., Jouhiaho, A., Kettunen, A., 2008: Metsä – ja puualan pienyritysten toimialakatsauksesta perusteet tutkimus – ja kehitystoiminnalle. [Small enterprise report for R&D program in the forest and wood sector], TTS-institute (Work Efficiency Institute).6 p. (In Finnish) Saarelainen, S., Makkonen, L., 2007: Ilmastonmuutokseen sopeutuminen tienpidossa. Esiselvitys. [Adaptation to Climate Change in Road Maintenance. Preliminary report]. Report of Finnish Road Administration: 04/2007. Helsinki 2007. 53 p. (in Finnish) Soirinsuo, J., Mäkinen, P. 2010: Puunkorjuu – ja puunkuljetusyritykset kasvavat asiakkaan pyynnöstä. [Timber harvesting and transportation enterprises grows for customers request], In: Rieppo, K. (ed.). Kasvun eväät metsä – ja puualan pienyrityksille. [Seeds of growth for small enterprises in forestry and wood processing industries] TTS-institute: 3440. (In Finnish) Solakivi, T., Ojala, L., Lorentz, H., Laari, S., Töyli, J., 2012: Finland State of Logistics 2012. Publications of the Ministry of Transport and Communications 25/2012. 82 p. Tiehallinto, 2009: Ilmastonmuutoksen vaikutus tiestön hoitoon ja ylläpitoon. (The Affect of Climate Change to Road Maintenance). Report of Finnish Road Administration: 8/2009. 80 p. (In Finnish) Vuorimies, N., Matintupa, A., Luomala, H., 2009: CTI puutavara-autossa. Keuruun metsätien syksyn 2008 ja Vesilahden maantien kevään 2009 mittausten tulokset. [CTI in timber truck. Results from the measurement at Keuruu forest road in autumn 2008 and Vesilahti road in spring 2009. Metsäteho raportti 207. Metsäteho Oy, Helsinki. 50 p. (In Finnish)

J. Malinen et al. Prospects and Challenges of Timber Trucking in a Changing Operational Environment in Finland (91–100) Väkevä, J., Imponen, V., 2001: Puutavaran korjuu – ja kuljetusyritysten kannattavuus vuosina 1995 – 2000. [Profitability of harvesting and transportation enterprises in wood procurement during 1995 – 2000] Metsätehon raportti 125. Metsäteho Oy, Helsinki. 67 p. (In Finnish)

Woodrooffe, J., Burns, N., 1998: Effects of Tire Inflations Pressure and CTI on Road Life and Vehicle Stability. Proceedings of 5th International Symposium on Heavy Vehicle Weights and Dimension, Maroonchydore, Queensland, Australia, Part 1: 203–221.

Authors’ address: Jukka Malinen, PhD.* e-mail: jukka.malinen@uef.fi Ville Nousiainen, MSc. e-mail: villenou@gmail.com Prof. Teijo Palander, PhD. e-mail: teijo.s.palander@uef.fi University of Eastern Finland, School of Forestry P.O.Box 111 80101 Joensuu FINLAND

Received: October 09, 2013 Accepted: January 13, 2014

100

Kari Palojärvi, MSc. e-mail: kari.palojarvi@skal.fi Association of Forest Road Carriers Nuijamiestentie 7 00400 Helsinki FINLAND * Corresponding author Croat. j. for. eng. 35(2014)1

Original scientific paper

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species, Using Satellite Imagery (Kastamonu Central Forest Directorate Sample) Burak Aricak, Omer Kucuk, Korhan Enez Abstract Becoming successful in fighting forest fires is not only a matter of taking the required measures into consideration and efficiently and economically using the resources, but also employing the cutting edge science and technology in every aspect of the process. Determining the potential fire prone regions within forest stands, plays an important role in the success rate of forest fire prevention and firefighting. Various techniques are used in the determination of fire potential; especially high resolution satellite imagery can provide very sensitive and detailed information regarding the conditions of regional topography and fuel material (fuel) accumulation. Fuel material models have long been effectively used in fire management, fire behavioral estimates and determination of fire danger risks. Additionally, fuel material maps, prepared based on fuel material features, can help in the determination of fire potential. Fire potential maps include fuel material types and their distribution in the field. In this particular study, an October, 2011 dated ÂťGeoEyeÂŤ image, encompassing 24 320 ha of Kastamonu Central Forest Directorate area, 15 685 ha of which is forested, was used. The image was classified based on fuel material features, such as tree species, mixture, crown closure, age classes, etc. Acreages and distributions of the potential fire prone areas were determined, and where these areas were concentrated, possible fire suppressing precautionary methods were discussed. As long as the satellite image acquisition is periodically supplied, fire potential map can be updated depending on fuel material features. Keywords: forest fire, remote sensing, GIS, Turkey

1. Introduction Forest fires, while devouring thousands ha of productive forests every year and costing hundreds of thousands of Euros of lost resource revenues and firefighting expenditures, are jeopardizing the well-being of many assets like water management, soil reclamation, climate regulation, social well-being, nature protection, natural defense, aesthetic, recreation, science, Croat. j. for. eng. 35(2014)1

research, etc. (Eraslan 1982, Kourtz 1984). Fuel material is defined as any type of easily ignitable fuel material or mixture, which can be found in, over and even above the soil (Robertson 1971). Determination of the amount of fuel material plays an important role when it is highly important to decide whether or not the precautionary measures are needed to lessen the possible effects of wildfires. The amount of such fuel material, which is in direct relationship with the fire

101

B. Aricak et al.

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101â&#x20AC;&#x201C;108)

behavior (Rothermel 1983), dictates the spread of the fire and the amount of energy released from it (Stinson and Wright 1969, Whelan 1995). The spread of the fire depends on area topography and meteorological conditions. Similarly, while very intensive wildfires occur in some particular fuel types (coppice and rather young plantations, etc.), comparably less intensive ones occur in mature to old stands. When this situation is taken into consideration, it is very logical to assume that the distribution of fuel material will greatly affect the spread of the fire (Bilgili 1998). The main principle of creating land cover maps by remote sensing techniques is the classification of data in the zone image. Image classification is a process performed to generate topical maps from multispectral images (Yomralioglu 2002). Fuel material features such as height, crown length, crown diameter and ÂťdbhÂŤ can be obtained both from terrestrial surveying measurements and from very detailed aerial photography or high resolution satellite imagery, and they play an important role in biomass estimations (Alemdag 1986, Ottomar and Vihnanek 1998). Especially, in young plantations, the establishment of stand and tree features (tree type, age, and crown) based on the information acquired from aerial photography or satellite imagery can simplify the biomass estimation (Oladi 1996). Remote sensing, artificial intelligence, geographic information systems and forest fire management decision support systems are widely used in any type of fire management administrations worldwide (Bilgili and Kucuk 2002). Factors affecting the accidental fire ignition in forests are the proximity to nearby settlements, vegetation moisture, slope, aspect, distance from roads and elevation. These markers are then assigned weight values to set up the fire sensitive areas to compare with the hot spots derived from MODIS satellite imagery (Adab et al. 2013). Fuel type maps include the conditions of fuel material in localized regions; give detailed information about the fire potential and its possible effects on other resources. Since forests are dynamically structured, fuel type maps must be periodically updated. These maps provide the forest managers the information as to where, in what amount, and in what ratio and variety these dangerous fuel materials are accumulated. Using these maps, the current potential effects of fuel material conditions on fire behavior and determination of fire damage and cost can also be determined (Kucuk 2005). In a study conducted by Eva and Lambin (2000), remote sensing was used to investigate the relation-

102

ship between forest fires and land cover change. Fuel type maps, fire activity maps, fire intensity and burned area maps can be produced, using remotely sensed data and GIS in fire and fuel material management (Schaaf 1996, Conard et al. 2001, Loveland 2001, Congaltion 2001). Due to GIS flexibility, it was suitable for making very simple and effective digital maps as well as very complex models and analyses. It could be efficiently used in every stage and hence it was an ideal tool for decision makers. The basis of this important contribution comes from the fact that GIS can excellently match the attribute values to graphic representations. Since GIS proved to be so suitable, it is widely used in making and compiling fire databases and fire danger ratios (Bilgili et al. 2001), similarly as in other environmental science disciplines. The aim of this study is to identify the types of flammable material in the forest areas with respect to prominent features related to the combustion of flammable materials (type, closure, stand age). This will be done by means of high-resolution satellite images, by taking into account changes of positions and areas in small locations. Fuel material distribution maps, used to determine the potential fire risk regions, will provide great help in minimizing the danger of fire, in determining where the possible fire may commence and in what regions it may spread more easily, as well as in planning fire management and suppression efforts.

2. Materials and methods High resolution, colored, panchromatic, GeoEye satellite imagery, acquired on September 3, 2011, was used for determining, regionally and spatially, the potential fire risk regions based on the fuel features determined in the study area. ArcGIS 9.3 and Erdas 9.1 were used for establishing and evaluating the database and satellite image processing, respectively. Digital terrain model (DTM) was generated by using the elevation data obtained from the digital topographic map of the study area. In this particular study, where potential fire risk regions are characterized by the fuel features of Kastamonu central forest directorate, the flow chart was created as shown in Fig. 1. In gathering preliminary information and designing the database, 1/25000 quad maps containing boundaries of Kastamonu Central Forest Directorate, forest management map and GeoEye imagery, were obtained. All types of GIS functions like transferring, storing, processing, various questioning and analyses Croat. j. for. eng. 35(2014)1

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

Fig. 1 Work flow of the study

of different raw data types and presenting the produced maps and attributes were performed using ArcGIS 9.3. Individual layers of the GeoEye imagery, acquired as panchromatic image, were first layer stacked in Erdas 9.1. As for geometric rectification, UTM coordinate system and available quad maps were used. For geometric rectification, 32 ground control points were used consisting of road intersects, bridges and houses. The nearest neighbor transformation type was used. The boundaries of the study area were set and they coincided with the map of Kastamonu central forest directorate. Then, the GeoEye imagery was delineated and cut, using the vector boundary polygon in Erdas 9.1. GeoEye imagery delineated for the study area. The study area was categorized according to tree species, crown closure and stand age into 10 fire risk groups (Kucuk 2004), number 1, being the most fire prone group (Table 1). Supervised classification was applied to GeoEye image, employing ERDAS, based on 10 potential fire risk groups formulated according to fuel material features in Table 1. Supervised classification of all 10 potential fire risk groups was done. Maximum likelihood algorithm was applied in the classification. Classification accuracy assessment was done, making a comparison between the classified image and a reference data set (maps or GPS measurements), Croat. j. for. eng. 35(2014)1

B. Aricak et al.

whose accuracy had previously been validated. For this reason, random pixels from the classified image were compared to the corresponding pixels from a known source. Random selection of these pixels can abate the analyst’s former knowledge about the region/surrounding (Musaoglu 2000). As a result of such a comparison, the possibilities of categorizing the randomly selected pixels into accurate classes can be presented in a classification error matrix and using a Kappa coefficient. In this way, statistical analyses can be made (Sunar and Musaoglu 1998). The GeoEye image was classified according to stand age, crown closure and tree species. Using the digitized stand map, stand ages were categorized into 4 classes: »a-ab-b«, »bc«, »c-cd-d« and all other areas, which do not belong to a stand age (forest soil, residential, agriculture, etc.). In terms of the stand progress, the ones ranging from 1.30 to 7.9 cm in diameter are considered as »Juvenile and density = a«, from 8 to 19.9 cm as »Pole and mast = b«, from 20 to 35.9 cm as »Thin wood = c«, from 36 to larger diameters as »Thick wood = d«. Crown closure was categorized into three groups; »0–1«, »2«, »3«. Crown closure of less than 10% = »0«, 11% to 40% = »1«, 41% to 70% = »2«, 71% to 100%= »3«.

Table 1 Regional type groups based on tree species, crown closure and stand development age Fire Risk Group

Tree Species

Crown Closure

Stand Age

»a«,»ab«, »b«

0 or 1

»a«, »ab«, »b«

»bc«

DCP

–

1, 2 or 3

»c«, »cd«, »d«

Other species

–

A, FS or R areas

–

CP: Corsican Pine DCP: Degraded Corsican Pine

a: Youth and Density b: Pole and Mast

FS: Forest Soil

c:Thin Wood

A: Agriculture

d: Middle Woodland

103

B. Aricak et al.

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

Tree species were categorized into four groups; »Corsican Pine«, »Degraded Corsican Pine«, »Other species« and all other areas (forest soil, residential, agriculture, etc.) At least 200 training areas were picked from each category of each group and the images were classified. General classification accuracies and Kappa statistics were calculated. All original band pixel values in polygonal train areas were used for the classification. Boundary pixel was eliminated. The classified images were converted into vector format. 3 different vector layers obtained from the classifications were intersected in ArgGIS 9 database and a whole new layer was attained; from this new layer, through a new questioning, 10 fire risk groups were regrouped according to their features presented in Table 1. In this way, a new map was developed (Fig. 3d).

Each image, whose supervised classification had been applied, was subjected to accuracy assessment through ERDAS, and the classification accuracy of images was obtained. Besides, on the vector map of fire risk groups, prepared based on the stand groups presented in Table 1, 10 random sampling areas of »20 × 20 m« (400 m2) were determined from each group. Control measurements were done on 100 different sampling areas (10 groups × 10 sampling areas). Through the vector map of potential fire risk groups, 100 sample areas, whose coordinates had already been established, were located, using a handheld GPS throughout the study area. Classification accuracy of each sample group found in the study area, according to the groups in Table 1, was checked. Thus, both spatial and classification accuracy of the sample areas was validated through field measurements.

Fig. 2 The study area in Turkey

104

Croat. j. for. eng. 35(2014)1

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

3. Area of study

B. Aricak et al.

Table 4 Classification error matrix by stand closur

Central Forest District of Kastamonu Regional Forest Directory encompassing an area of 24 320 ha of extensive Corsican Pine (Pinus nigra) forest, which is relatively susceptible to fire, was selected as the study area. The location of the study area positioned in the map of Kastamonu, as well in Turkey, can be viewed in Fig. 2. Pure and mixed stands predominantly composed of high fire prone Corsican pine trees have been extensively observed in the mountainous area. Besides, other stand types consisting of Fir (Abies bornmulleriana), Scotch Pine (Pinus sylvestris) and Oak (Quercus patreae) can also be seen in the study area. The elevation range of themountainous study area is between 745 m and 1 510 m.

Classes

1. 0-I Close

2. II Close

3. III Close

1. 0-I Close

2. II Close

3. III Close

Table 5 General classification accuracy by stand closure Classes

Reference Classified Number of Producer’s User’s Total Total Corrects Accuracy Accuracy

1. 0-I Close

97.56

100.00

2. II Close

84.21

80.00

3. III Close

80.49

82.50

4. Results of the study By taking the tree species, crown closure and stand age into consideration, supervised classification was applied to the GeoEye image of the study area, using Maximum Likelihood Algorithm. Table 2 illustrates the image classification error matrix, depending on Table 2 Classification error matrix by tree species 1. CP*

2. DCP*

3. Other Species

4. All other areas

1. CP

–

2. DCP

–

3. Other Species

–

4. All other areas

–

Classes

*CP: Corsican Pine DCP: Degraded Corsican Pine

tree species. Table 3 illustrates the general image classification accuracy based on tree species. General classification accuracy of the GeoEye satellite image was 97.22% and Kappa statistic was 0.9630. Fig. 3a illustrates the satellite image classified based on tree species. Table 4 illustrates the classification error matrix based on stand closure, and Table 5 illustrates the overall accuracy by stand closure. General classification accuracy of the GeoEye satellite image was 87.5% and Kappa statistic was 0.8125. Fig. 3b illustrates the satellite image classified based on tree species stand closure. Table 6 illustrates the classification error matrix based on stand age, and Table 7 illustrates the general classification accuracy by stand age. General classification accuracy of the GeoEye satellite image was 85% and Kappa statistic was 0.8000. Table 6 Classification error matrix by stand age

Table 3 Illustration of the general classification accuracy based on tree species Classes

Reference Classified Number of Producer’s User’s Total Total corrects Accuracy Accuracy

a-ab-b

c-cd-d

Out of range

1. a-ab-b

Classes

1. CP*

100.00

97.22

2. bc

2. DCP*

100.00

94.44

3. c-cd-d

3. Other Species

97.22

4. Out of range

4. All other areas

92.31

100.00

*CP: Corsican Pine DCP: Degraded Corsican Pine

Croat. j. for. eng. 35(2014)1

a: Youth and Density b: Pole and Mast c:Thin Wood d: Middle Woodland

105

B. Aricak et al.

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

Fig. 3a Classified satellite map, based on tree species

Fig. 3c Classified satellite map, based on stand age

Fig. 3b Classified satellite map, based on stand closure

Fig. 3d Fire risk potential map from the GeoEye satellite image of the Kastamonu Central Forest Directorate

Table 7 General classification accuracy by stand age Classes

Reference Classified Number of Producer’s User’s Total Total corrects Accuracy Accuracy

1. a-ab-b

75.00

90.00

2. bc

86.21

62.50

3. c-cd-d

85.37

87.50

4. Out of range

95.24

100.00

a: Youth and Density b: Pole and Mast

106

c: Thin Wood d: Middle Woodland

Fig. 3c illustrates the satellite image classified based on tree species stand age. The results of three separate classifications, by tree species, stand closure and stand age, were then converted to vector format and intersected; the newly created database allowed us to produce the fire risk potential map. Wise questioning was carried out through the created database. Table 8 illustrates the areas of the fire risk groups. Croat. j. for. eng. 35(2014)1

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

Table 8 Areas of fire risk groups distributed based on fuel material features Potential Fire Risk Groups

Area, ha

Area, %

586.5

2.41

821.3

3.45

1 514.6

6.23

2 371.2

9.75

1 046.9

4.30

526.4

2.16

2 474.4

10.14

3 803.9

15.61

3 867.3

15.90

7 307.8

30.05

TOTAL

24 320

100

100 different sampling points, obtained by taking 10 random sampling points from each group in the map of potential fire risk groups, were cross-checked through land surveying and observations of random sampling areas. The final accuracy matrix, obtained based on these data, is given in Table 9. According to Table 9, producers’ accuracy of very high fire risk groups (Group 1, 2, 3) is approximately 80%. Table 9 Error matrix produced according to the results of potential fire risk groups Potential Fire Risk Groups

–

5. Discussions and Conclusions Forest fires are one of the major causes of ecological disturbance and environmental concerns in forests of Turkey. The degree of forest fire risk is changeable from year to year and it is different for each kind of Croat. j. for. eng. 35(2014)1

B. Aricak et al.

stand structure. In this particular study, the fire risk potential map was produced by GeoEye satellite image according to fuel material features, such as tree species, stand closure and stand age gathered in Kastamonu central forest directorate (Fig. 3d). The distribution of potential fire risk in the map created using the satellite data acquired in September, 2011 in the region of Kastamonu central forest directorate is as follows: Þ 12.09% very high fire risk (Table 1, Group 1, 2, 3), Þ 16.21% medium risk (Table 1, Group 4, 5, 6), Þ 25.75% low risk (Table 1, Group 7, 8), Þ 45.90% fire risk free (Table 1, Group 9, 10). The fire risk potential map will change from year to year depending on the forest growing features. For this reason, the satellite images were taken periodically for creating the fire risk potential map. Using high resolution spatial satellite imagery to determine the fire potential according to fuel material features in forest lands is both time and cost efficient and provides highly accurate, up-to-date information. This is especially true for mountainous forests, where the road network is less extensive than in the other parts of the country. In such forests, due to ecological and economical constraints, forest fire probability is much higher. The implementation of the proposed method will increase the possibility of suppression of forest fires. By this method, according to the established fuel material features, forest fires could be monitored through fire surveillance systems, after determining the regions with increased fire risk potential. Especially in Mediterranean and Aegean regions, where the forest fire risk is rather high in most of the year, the acquisition of periodic satellite imagery will help the decision makers to effectively monitor the ever changing dynamic structure of forests.

Acknowledgements This particular study, code name 111O033, was fully supported by TUBITAK (The Scientific Technological Research Council of Turkey) »1002 Swift Support Program«.

6. References Adab, H., Kanniah, K. D., Solaimani, K., 2013: Modelling Forest Fire Risk in the Northeast of Iran Using Remote Sensing and GIS Techniques. Natural Hazards 65(3): 1723–1743. Alemdag, I. S., 1986: Estimating Oven Dry Mass of Trembling Aspen and White Birch Using Measurements from Aerial Photographs. Canadian Journal of Forest Research 16(1): 163–165.

107

B. Aricak et al.

Determining a Fire Potential Map Based on Stand Age, Stand Closure and Tree Species ... (101–108)

Bilgili, E., Kucuk, O., 2002: Orman Yangınlarının Uzaktan Algılama Teknigi ile Belirlenmesi [Determination of Forest Fires with Remote Sensing Technique]. IV.GAP Mühendislik Cong. Proc., 2002:1631–1637. Bilgili, E., 1998: Yangın Amenajmanı Planlamalarında Yanıcı Madde Amenajmanının Rolu (The Role of Fuel Material Managament in Fire Management Planning). Orman Yangınları Politikası ve Planlaması Eğitim Kursu, Ankara. Bilgili, E., Saglam, B., Baskent, E. Z., 2001: Yangın Amenajmanı Planlamalarında Yangın Tehlike Oranları ve Coğrafi Bilgi Sistemleri. KSU Fen ve Mühendislik Dergisi [Kahramanmaras Sutcu Imam University Journal of Natural Sciences 4(2): 288–97. Conard, S. G., Hartzell, T., Hilburuner, M. W., Zimmerman, G. T., 2001: Changing Fuel Management Strategies. Challenge of Meeting New Information and Analysis Needs 10: 267–275. Congaltion, R. G., 2001: Accuracy Assessment and Validation of Remotely Sensed and Other Spatial Information. International Journal Wildland Fire 10: 321–328. Eraslan, İ., 1982: Istanbul University, Orman Fakültesi Orman Amenajmanı Ders Notu [Istanbul University Department of Forest Engineering, Forest Management Textbook], Ankara. Eva, H., Lambin, F. E., 2000: Fires and Land-cover Changes in the Tropics: A Remote Sensing Analysis at the Landscape Scale. Journal of Biogeography 27: 765–776. Kourtz, P. H., 1984: Decision-making for Centralized Forest Fire Management. Forestry Chronicle 60: 320–327. Kucuk, O., 2004: Yanıcı Madde Özellikleri ve Yangın Davranisina Bagli Yangın Potansiyelinin Belirlenmesi ve Haritalanması [Fuel Material Properties and Determination Fire Risk Potential Related with Fire Behavior], (PhD Thesis). KTU Fen Bilimleri Enstitüsü [Karadeniz Technical University, Graduate School of Natural and Applied Sciences]. Kucuk, O., 2005: Yangın Potansiyelinin Belirlenmesinde Yanıcı Madde Haritalarının Önemi [Importence of Fuel Maps on Determination of Fire Potential]. SDU Orman Fakültesi Dergisi [Suleyman Demirel University Journal of Natural Sciences], A1, ISSN:1302-7085, pp: 104–116.

Loveland, T. R., 2001: Toward a National Mapping Strategy: Lessens From Selected Mapping Programs. International Journal of Wildland Fire 10: 289–299. Musaoglu, N., 2000: Multitemporal Forest Change Anlysis A Case Study: Sariyer Region. International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Oladi, D., 1996: Developing a Framework a Methodology for Plantation Assessment Using Remotely-Sensed Data, (PhD Thesis). University of New Brunswick, Faculty of Forestry and Environmental Management, Canada. Ottomar, R. D., Vihnanek R. E., 1998: Stereo Photo Series for Quantifying Natural Fuels. Vol. II: Black Spruce and White Spruce Types in Alaska, PMS 831 (National Wildfire Coordinating Group, National Interagency Fire Center: Boise, ID) 65. Robertson, F. C., 1971: Terminology of Forest Science, Technology. Pratice and Product, D.C., 349, Washington. Rothermel, R. C., 1983: How to Predict the Spread and Intensity of Forest and Range Fires. US, Department of Agriculture Forest Service, Gen. Tech. Rep. INT-143. Schaaf, M. D., 1996: Development of Fire Emission Trade off Model (FEMT) and Aplication to the Grande Ronde River Basin, Oregon. Final Report; USDA Forest Service, Pacific Northwest Region Contract 53-82FT-03-2. Available From: U.S. Department of Agriculture, Forest Service, Pacific Northwest Region, 333 SW First Avenne, Portland, OR 97208, USA. Stinson, K. J., Wright, H. A., 1969: Temperature and Headfires in the Southern Mixed Prairie of Texas. J. Range. Mgmt. 22:169–174. Sunar, F., Musaoğlu N., 1998: Merging Multiresolution Spot P and Landsat TM Data: The Effects and Advantages. International Journal of Remote Sensing 19(2): 219–224. Whelan, R. J., 1995: The Ecology of Fires. Cambrige Univ. Press, Cambridge, MA. Yomralioglu, T., 2002: Cografi Bilgi Sistemleri Temel Kavramlar ve Uygulamalar [Geographic Information Systems Basic Concepts and Applications]. Karadeniz Technical University, ISBN: 975-97369-0-X, Trabzon.

Authors' addresses:

Received: November 20, 2013 Accepted: December 30, 2013

108

Assist. Prof. Burak Aricak, PhD.* e-mail: baricak@kastamonu.edu.tr Assoc. Prof. Omer Kucuk, PhD. e-mail: omerkucuk@kastamonu.edu.tr Assist. Prof. Korhan Enez, PhD. e-mail: korhanenez@kastamonu.edu.tr Kastamonu University Faculty of Forestry 37200 Kastamonu TURKEY *Corresponding author Croat. j. for. eng. 35(2014)1

ISSN 1845-5719

9 771845 571000