CROJFE - Volume 29, Issue 1 by Studio Bernardic

Issue 1

2008

Editorial

Are We on the Right Path? Dear readers, It has been three years since the magazine Croatian Journal of Forest Engineering (the successor to the journal which had been coming out under the title Forest Mechanization for twenty-five years) was published for the first time. We have great honor and joy, but also a pleasant duty, to give you a good news. The magazine Croatian Journal of Forest Engineering (CROJFE) has, among other referent bibliographic databases, been indexed in the Web of Science database (whose constituent part is the Science Citation Index Expanded – SCI database). Thus, one of the primary goals of the Editorial Board has been fulfilled. Both issues of CROJFE published in 2007 are indexed. The Croatian Journal of Forest Engineering is published twice a year. So far, it has been indexed in the following bibliographic databases: CAB Abstracts, Compendex, GeoBase, Global Health, Paperchem, SCOPUS and VINITI. It should also be pointed out that the Croatian Journal of Forest Engineering is the only magazine in Europe which specializes in the field of forest engineering. Division 3 of Forest Operations Engineering and Management of the International Union of Forest Research Organizations (IUFRO) cite this journal as one of the reference magazines in their field of activity (http://www.iufro.org/science/ divisions/division-3/30000/publications/). Articles published in the journal are available free of charge at the journal’s internet site (http:// crojfe.sumfak.hr/). In addition to the journal website, its international recognition is also enhanced by the home portal »Hr~ak« (http://hrcak.srce.hr/) and the international portal »Directory of Open Access Journals« (http://www.doaj.org/).

A Glance Backwards This is the moment to remind ourselves of the year 2005, when the journal CROJFE was first launched by the newly-appointed Editorial Board. Tibor Pentek was appointed Editor-in-Chief, @eljko Toma{i} was at the position of Responsible Editor and Dubravko Horvat was Technical Editor, while Tomislav Por{insky, Marijan [u{njar and Igor Stanki} were elected members of the Technical Board. Croat. j. for. eng. 29(2008)1

The first Editorial Board consisted mainly of experts working in the Department of Forest Techniques and Technologies at the Faculty of Forestry of Zagreb University. It should be emphasized, however, that the quality of the Editorial Board’s activities would be unthinkable without the responsible editor @eljko Toma{i}, the then head of the Development Service in the company »Hrvatske [ume« Ltd, Zagreb. He successfully integrated the publishing aspect of the science on the one hand and forestry operative on the other by encouraging the latter to recognize the value of the published results and apply them in practice. Stanislav Sever, the Editorial Advisor, whose recommendations, suggestions and ideas have greatly contributed to the final version of CROJFE, was an indisputable authority in the field of publishing magazines in the field of biotechnical sciences in the Republic of Croatia. In 2005 the financial situation was not promising. The only guaranteed financial means were those provided by the Commission of Scientific Research Work of »Hrvatske [ume« Ltd, Zagreb. The Faculty of Forestry of Zagreb University was unable to financially support the publication of the journal CROJFE. At the same time, the Croatian Ministry of Science, Education and Sport ceased to subsidize the journal »Mehanizacija {umarstva« due to its irregular publication. Viewed from the present time, the circumstances of that period were not conducive to the attainment of the goals set by the Editorial Board; yet, these goals were reached sooner than we had dared to hope. As we all know, fortune favors the brave, the hardworking and the persevering; this, combined with the exceptionally dedicated work of all the members of the Editorial Board, must have contributed to the achievement of our goals. The six issues of the journal CROJFE published in the past three years contain over 35 articles written by eminent Croatian and foreign authors. All the articles have been subjected to scrutinizing reviews by leading experts from relevant fields, which additionally strengthened the journal’s constant quality. The review list is relatively broad and is daily expanded with new reviewers.

T. PENTEK and T. POR[INSKY

Are We on the Right Path? (1–2)

A Glance Forward

Acknowledgement

As seen in the journal’s impressum, the current membership of the Editorial Board differs slightly from that at the beginning of its publication. As of July 2008, owing to the generous support by the Ministry of Science, Education and Sport of the Republic of Croatia, yet another step forward will be made in the development of the journal by the appointment of a junior researcher to the post of editor. The first and foremost duties and commitments of the new editor will be to enhance the development of CROJFE. Our primary goal is to collect increasing numbers of high quality papers by leading home and foreign experts from European and oversees countries. We hope that they will recognize our journal as one of the key magazines in this segment of forestry within Europe. Timely possibilities of selection from the broad array of reviewed and accepted scientific papers will allow us to publish thematic issues and increase the size of a particular issue from the current 6–8 papers to the future 9–12 papers. We will make further attempts to improve the quality of the English language in CROJFE and contribute to its globalization. Our insistence on more exhaustive Croatian abstracts at the end of the articles is aimed at promoting the journal in the Croatian forestry practice and increasing the possibilities of using the results of applied research. We will try to modernize the journal’s web site by conducting regular updates and complementing it with the papers published in all the past issues of the journal »Mehanizacija {umarstva«. This will be our tribute to the 25-year-long history of the journal. By doing so, we will ensure the continuity and integrity of the journal’s electronic version. We are aware of the complexity and duration of this work, and of the fact that the project will collide with day-to-day work involved in the publishing of current CROJFE issues. However, we are even more aware of the truth that a journal which does not have, or does not appreciate its history, cannot have a bright future. In addition, we plan to introduce an information system serving a multiple purpose, including receiving papers from the authors, forwarding them for reviews and communicating with the reviewers, keeping the authors updated on the evaluation procedure status, publishing the papers in the journal’s digital form, etc.

The CROJFE’s Editors-in-Chief wish to thank all members of the journal’s Editorial Board, members of the Editorial Council and members of the International Editorial Council on their hard work, effort, time and engagement invested in the preparation of the issues. We also thank the Croatian language adviser Ms. Branka Tafra and the English language adviser Ms. Maja Zaj{ek for always finding time and patience to meticulously and at short notice proofread the papers accepted for publication. All authors and reviewers of the published papers, as well as all those whose papers are in the process of publication, also deserve our sincere thanks for their contribution to the journal. Cooperation with the co-publisher, the company »Hrvatske [ume« Ltd, Zagreb, without which the continuing presence of CROJFE would be unthinkable, is at a very satisfactory level. The company’s Management has always been sensitive to our needs and problems and has, within their possibilities and short- and long-term business policy, regularly been at our service. We express our warmest gratitude to the Management of the Faculty of Forestry presided by the Dean Jozo Franji} and to all staff of the Forestry Department at the Faculty of Forestry of Zagreb University, who have continuously recognized our problems and appreciated the strategic significance of the publishing activity within the overall activity of the Faculty of Forestry. Our Department also provided considerable financial assistance to the journal during the years 2007 and 2008. We sincerely hope that their financial help will continue in the years to come. We thank the Croatian Ministry of Science, Education and Sport for including the journal CROJFE in the system of subsidies to scientific-professional journals in 2006. We hope that their very important financial support, although not too great at the moment, will further increase in the future, since the journal’s rising quality necessarily requires bigger financial investments.

With kindest regards, Yours Editors-in-Chief Tibor Pentek and Tomislav Por{insky

Croat. j. for. eng. 29(2008)1

Uvodnik

Jesmo li na pravom putu? Po{tovane ~itateljice i ~itatelji! Posebna nam je ~ast, zadovoljstvo i nadasve ugodna obveza {to vas, nakon tri godine od izlaska iz tiska prvoga broja ~asopisa Croatian Journal of Forest Engineering (sljednika ~asopisa koji je punih dvadeset i pet godina izlazio pod naslovom Mehanizacija {umarstva), moèmo razveseliti lijepom vije{}u. Naime, Croatian Journal of Forest Engineering (CROJFE), me|u ostalim referentnim bibliografskim bazama podataka, indeksiran je u bazi Web of Science (~iji je sastavni dio baza Science Citation Index Expanded – SCI), {to je i bio jedan od po~etnih ciljeva uredni{tva. Indeksirana su oba broja ~asopisa objavljena u 2007. godini. Croatian Journal of Forest Engineering izlazi dva puta na godinu, a do sada je indeksiran u ovim bibliografskim bazama podataka: CAB Abstracts, Compendex, GeoBase, Global Health, Paperchem, SCOPUS, VINITI. Tako|er valja napomenuti da je Croatian Journal of Forest Engineering jedini ~asopis u Europi koji je specijaliziran za {umarsko inènjerstvo, a Razred 3 [umarsko inènjerstvo i upravljanje {umskim radovima Me|unarodne udruge {umarskih istraìva~kih organizacija (IUFRO) navodi ga kao jedan od referentnih ~asopisa za svoje podru~je djelovanja (http:// www.iufro.org/science/divisions/division-3/30000/ publications/). Radovi objavljeni u ~asopisu, bez naknade, dostupni su na internetskoj stranici ~asopisa (http:// crojfe.sumfak.hr/). Osim mre`ne stranice ~asopisa njegovu me|unarodnu vidljivost osiguravaju doma}i portal »Hr~ak« (http://hrcak.srce.hr/) i me|unarodni portal »Directory of Open Access Journals« (http://www.doaj.org/).

Pogled unatrag U ovom se trenutku treba prisjetiti samih po~etaka pokretanja na{ega ~asopisa. Godine 2005. imenovano je uredni{tvo u kojem su ~lanovi bili glavni urednik ~asopisa Tibor Pentek, odgovorni urednik @eljko Toma{i}, tehni~ki urednik Dubravko Horvat, a tehni~ko su uredni{tvo ~inili Tomislav Por{insky, Marijan [u{njar i Igor Stanki}. U`e uredni{tvo, u tom njegovu prvom sazivu, ~inili su ve}inom ~lanovi Zavoda za {umarske tehCroat. j. for. eng. 29(2008)1

nike i tehnologije [umarskoga fakulteta Sveu~ili{ta u Zagrebu. Me|utim, uspje{an rad uèga uredni{tva nezamisliv je bez odgovornoga urednika @eljka Toma{i}a, tada {efa Razvojne slu`be poduze}a »Hrvatske {ume« d.o.o. Zagreb, koji je bio poveznica izme|u izdava~ke sastavnice znanosti s jedne i {umarske operative koja mora prepoznati rezultate objavljenih radova s druge strane te u suglasju sa svojim mogu}nostima primijeniti ih u praksi. Savjetnik je uredni{tva, koji svojim prijedlozima, sugestijama i idejama uvelike pridonosi kona~noj ina~ici CROJFE-a, od po~etka bio Stanislav Sever, ~ovjek koji je zasigurno veliki autoritet u izdavanju ~asopisa iz podru~ja biotehni~kih znanosti u Republici Hrvatskoj. Financijska situacija te 2005. godine nije bila sjajna. Jedino su financijska sredstva koje je osiguralo Povjerenstvo za znanstvenoistraìva~ki rad »Hrvatskih {uma« d.o.o. Zagreb bila zajam~ena. [umarski fakultet Sveu~ili{ta u Zagrebu nije imao raspoloìvih financijskih sredstava kojima bi potpomogao izdavanje ~asopisa CROJFE, a dotada{nji je ~asopis Mehanizacija {umarstva, zbog neredovitosti izlaènja, ispao iz sustava potpore Ministarstva znanosti, obrazovanja i {porta Republike Hrvatske. Unato~ okolnostima koje nam, daju}i realnu ocjenu s odre|enim vremenskim odmakom, nisu i{le na ruku, postavljeni je cilj uredni{tva ~asopisa ostvaren prije nego {to smo se i sami nadali. Tomu je, vjerojatno, osim iznimno predanoga rada svih ~lanova uèga uredni{tva pridonijela i mala doza sre}e, koja, kao {to svi znamo, prati hrabre, odva`ne, vrijedne i uporne. U tri je godine, u {est brojeva CROJFE-a objavljeno vi{e od 35 radova uglednih doma}ih i stranih autora. Svi su radovi pro{li stroge me|unarodne recenzije vode}ih stru~njaka iz podru~ja kojim se sadràjno bave te su na taj na~in prili~no ujedna~ene, propisane vrsno}e. Recenzentski su popisi ove}i i svakim se danom dopunjuju novim imenima recenzenata.

Pogled unaprijed Dana{nji je sastav u`ega uredni{tva ~asopisa pone{to druga~iji nego na po~etku njegova izla`enja, {to je i vidljivo iz impresuma ~asopisa.

T. PENTEK i T. POR[INSKY

Od srpnja 2008. godine, a zahvaljuju}i potpori Ministarstva znanosti, obrazovanja i {porta Republike Hrvatske, u ~asopisu }e po~eti raditi znanstveni novak – mladi urednik ~asopisa, ~iji }e temeljni zadaci i obveze biti vezani uz CROJFE. To je jo{ jedan korak naprijed u razvoju ~asopisa. @elja nam je osigurati {to ve}i broj {to boljih radova doma}ih i stranih, europskih i izvaneuropskih, vode}ih stru~njaka u podru~ju {umarskoga inènjerstva koji bi na{ ~asopis trebali prepoznati kao jedan od vode}ih i klju~nih ~asopisa iz ovoga segmenta {umarstva u Europi. Temeljem pravodobne ve}e mogu}nosti izbora recenziranih i prihva}enih znanstvenih radova bili bismo u mogu}nosti tiskati tematske brojeve ~asopisa, a obujam pojedinoga broja ~asopisa sa sada{njih 6–8 radova podi}i na 9–12 radova. Nastojat }e se jo{ pobolj{ati engleski jezik u ~asopisu i time pridonijeti globalizaciji CROJFE-a. Ustrajat }emo na {irim hrvatskim saècima na kraju radova radi daljnjega njegovanja hrvatskoga nazivlja te radi promid`be ~asopisa u hrvatskoj {umarskoj praksi i pove}anja mogu}nosti primjene rezultata ipak, u prvom redu, primijenjenih istraìvanja. Mre`nu }emo stranicu ~asopisa osuvremeniti, nastojimo ju redovito aùrirati, a èlja nam je da ju dopunimo ~lancima objavljenima u svim dosada{njim brojevima ~asopisa Mehanizacija {umarstva. Rije~ je o 25-godi{njoj povijesti ~asopisa. Na taj }e se na~in osigurati kontinuitet i cjelovitost elektroni~ke verzije ~asopisa. Iako smo svjesni sloènosti i dugotrajnosti toga posla, a znamo kako }e ovaj predvi|eni projekt kolidirati sa svakodnevnim aktivnostima povezanima s aktualnim brojevima ~asopisa CROJFE, znana nam je i ~injenica kako ~asopis koji nema ili ne cijeni svoju pro{lost ne moè imati svjetlu budu}nost. Planira se i uvo|enje informacijskoga sustava za upravljanje radom ~asopisa, koji bi obavljao prihvat radova od autora, upu}ivanje radova na recenzijski postupak i komunikaciju s recenzentima, obavje{}ivanje autora o stanju evaluacijskoga postupka, objavi radova u digitalnoj ina~ici ~asopisa itd.

Jesmo li na pravom putu? (3–4)

Zahvala Kao glavni urednici CROJFE-a, na dosada{njem radu, trudu, vremenu i anga`manu zahvaljujemo ponajprije svim ~lanovima u`ega uredni{tva ~asopisa, ~lanovima uredni~koga vije}a i me|unarodnoga uredni~koga vije}a. Hvala lektorici hrvatskoga jezika Branki Tafri i lektorici engleskoga jezika Maji Zaj{ek koje su uvijek prona{le dovoljno vremena da postupak lektoriranja radova prihva}enih za objavu obave brzo i uspje{no. Svim autorima i recenzentima objavljenih radova i onih u postupku objave tako|er velika hvala za prinos dostignutomu ostvaraju. Suradnja sa suizdava~em ~asopisa, tvrtkom »Hrvatske {ume« d.o.o. Zagreb, bez kojega bi opstojnost CROJFE-a bila nezamisliva, na zavidnoj je razini jer uprava poduze}a ima sluha za na{e potrebe i probleme te nam u svakoj prilici, u skladu sa svojim mogu}nostima, kratkoro~nom i dugoro~nom poslovnom politikom tvrtke izlazi ususret. Velika hvala poslovodstvu [umarskoga fakulteta na ~elu s dekanom Jozom Franji}em te svim ~lanovima [umarskoga odsjeka [umarskoga fakulteta Sveu~ili{ta u Zagrebu koji su imali razumijevanja za na{e probleme te pravodobno znali procijeniti strate{ki zna~aj izdava~ke djelatnosti u okviru sveukupne aktivnosti [umarskoga odsjeka. Zna~ajna je financijska pomo} ~asopisu tijekom 2007. i 2008. godine stigla i iz na{e ku}e, ~emu se iskreno nadamo i idu}ih godina. Zahvaljujemo Ministarstvu znanosti, obrazovanja i {porta Republike Hrvatske koje je CROJFE uvrstilo u sustav potpore znanstveno-stru~nih ~asopisa od 2006. godine. Nadamo se da }e ova nevelika, ali svakako zna~ajna financijska podr{ka u budu}nosti biti jo{ zna~ajnija jer daljnje pobolj{anje vrsno}e ~asopisa nu`no tra`i i ve}a financijska ulaganja. S po{tovanjem i uz srda~an pozdrav va{i glavni urednici ~asopisa, Tibor Pentek i Tomislav Por{insky

Croat. j. for. eng. 29(2008)1

Original scientific paper – Izvorni znanstveni rad

Modelling Stand Damages and Comparison of Two Harvesting Methods Bo{tjan Ko{ir Abstract – Nacrtak This paper deals with the problem of tree damage in a remaining stand. Two models were used for assessment of stand damage over the entire production period. Damage accumulates on the trees and in the stand, which is why the total share of damage tends towards the limit 100% if the number of thinnings increases. Parameters used in the models were established partly by previous field measurements and partly by simulations measurements. Motor-manual and cut-to-length technologies were analysed and compared according to the total number of damaged trees and the structure of trees according to the number of injuries. It was found that motor-manual technology causes more damage to trees and results in worse tree structure, meaning more than one injury. Many unanswered questions provide plenty of opportunities for further research. Keywords: stand damage, harvesting, technology, motor-manual, cut-to-length

1. Introduction – Uvod Slovenian forestry has a long tradition of sustainable, multifunctional and co-natural forest management. The doctrine of forest management in Slovenia has been abundantly discussed and presented to the international community (Mlin{ek 1977, Mlin{ek 1994, Diaci and McConnell 1996). Uniform and group shelterwood systems are predominant. It is obvious that such doctrine demands a specific approach and exerts a considerable influence on all aspects of practical forestry (Matthews 1999). Forest harvesting has developed and it is adapted today to the specifics of silvicultural philosophy and to restrictions which originate from different sources. Clear-cuttings have been forbidden for many decades. Frequent thinnings (one to two cuttings in the same stand per decade – state forests), difficult terrain and large tree dimensions are the reason why motor-manual cutting and tractor or cable skidding are still predominant today. Final cuttings are more an exception than a rule. Social changes in human resources, changes in energy prices, new possibilities in information techology and severe competition on the timber market are putting into question the economics and future of recent technologies, in which motor-manual cutting Croat. j. for. eng. 29(2008)1

and skidding with various tractors prevail (M-M technology). The main reasons why cut-to-length (CTL technology) technology (harvester and forwarder) took off after 2000 were: demand for cost reduction and problems with recruitment of new professional workers for traditional forest work (Ko{ir 2004). These reasons have been also crucial in the first debates in which forest enterprises tried to find a common language with the Slovenian Forestry Service. Compatibility of new CTL technology with the existing forest management doctrine raised a hot debate in which damage to existing trees and damage to forest soil were the most controversial issues. In this context damage to the remaining stand has been understood as injured trees with visible scars on stem, butt, roots or branches (scar area >10 cm2), broken branches in canopy and bent trees. In case of tree damage (MM technology) we already have reliable data from abroad and from field observations (Krivec 1975, Ivanek 1976, Eriksson 1981, Butora and Schwager 1986, Leinss 1991, Ko{ir 1998a, Sabo 1999 and other sources not mentioned here) as well as results gained from models (Ko{ir and Cedilnik 1996, Ko{ir 1996, Ko{ir 2000). At the time when we published quite unfavourable results concerning tree damage research nobody took too much notice when traditional technologies were in question, while

B. KO[IR

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

a negative attitude towards the new CTL technology introduction appeared following the first cases of damage. The performance of the general model of tree damage gave good results in the past, which were later confirmed by field observations. Therefore we decided to use the same model for comparing the damage to trees caused by traditional motor-manual and by the new CTL technology. We hope that in this manner we can present an unbiased assessment of this sensitive aspect of new technology introduction.

2. Research methods – Metode istra`ivanja Tree damage models Tree damage after forest operations decreases the value of forest yields in future, stability of stands, and it should be of utmost concern to every good forest owner (Spinelli 1999). Frequent thinnings also mean more chances for increased share of injured trees after finishing the work (Ko{ir 2001). The model used in this comparison was developed for shelterwood forest management where cutting intensity was defined as the proportion between marked trees and all trees in the stand. The stand damage intensity was defined as the proportion between damaged trees after operations and all remaining trees. We assumed that: Þ each tree has the same probability of being chosen for felling at a certain time, Þ each tree has the same probability of being damaged at the time of the last logging, Þ a tree which is damaged more than once in the same thinning is counted as one injury, and Þ a tree is marked as injured regardless of the severity of injury. The share of damaged trees remaining in each thinning during the production period depends on the probability of the tree being chosen for cutting or probability of the tree having been damaged up to now. Obviously some trees will be damaged once, twice or more times at the end of the rotation period. The basic model development and evaluation (Ko{ir and Cedilnik 1996) gives more criteria than required for this analysis (such as increment of injuries). For our purposes we used the following formula, the so-called »rule of tree damage accumulation«: n

Dn = 1 − (1 − D0 ) ∏ (1 − d i ) i=1

where: n number of thinnings in rotation period,

(1)

D0 Di di Dn

share of tree damage before forest operations started (beginning), share of tree damage in the stand after i thinning, share of tree damage in i thinning. obviously tends to the limit = 1 if speaking of known di > 0 values.

The share of damaged trees is not directly dependant upon the intensity of thinning (equation 1), but as it has already been proved (Ko{ir 2000), this dependence enters the model indirectly, because the intensity of cutting (ei) and skid trail density (G) impact the share of damaged trees (di) after i thinning. Instead of intensity of cutting (ei), concentration of wood cut per ha can be used (Vi). We may write: di = f(Vi , G) where: Vi intensity of cutting in i thinning (m3/ha), G density of skid trails (m/ha). From the second model (where some results of field studies were used) the following formulas were used in this article: d vi ⋅ N v + d si ⋅ ( N si − N v ) (2) di = N si where: share of tree damage after i thinning along dvi skid trails after i thinning, share of tree damage after i thinning bedsi tween skid trails after i thinning, Nv number of trees along skid trails, Nsi number of trees in the whole stand after i thinning. and: d vi = 1 − e ( Vi /G )

(3)

d si = a + b 1 ⋅ Vi − b 2 ⋅ G

(4)

where: e basis of natural logarithm, a, b1, b2 equation parameters. For distinction between technologies the values from Table 1 were used.

Tree damage simulation The above formulas give us a good estimation of total tree damage in the remaining stand after i thinnings, but we were also interested in the frequency of injuries to the trees at the end of the rotation period. The question was: how many injuries can we expect to the trees and how many removals? Croat. j. for. eng. 29(2008)1

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

B. KOŠIR

Table 1 Values of parameters in model computation Tablica 1. Vrijednosti parametara u modelu Parameter Parametar Number of trees in the whole stand after i thinning* Broj stabala u sastojini nakon i-te prorede* Intensity of cutting in i thinning (m3/ha)* Sje~na gusto}a i-te prorede (m3/ha)* Density of skid trails (m/ha) Gusto}a traktorskih vlaka (m/ha) Number of trees along the skid trails** Broj stabala uzdu` traktorske vlake** Equation parameters Parametri jednad`be

MM Nsi Vi

CTL

Taken from yield tables for spruce in Alpine region, strong thinning (Halaj et al. 1987) Preuzeto iz prirasno-prihodnih tablica, obi~na smreka u alpskom podru~ju, jake prorede (Halaj i dr. 1987)

200

500

= 0.05 Ns

= 0.15 Ns

a b1 b2

0.19 0.0005 0.00014

*Rotation period of 160 years has been analysed; 10 years thinning interval has been used *Analizirana je duljina ophodnje od 160 godina uz 10-godi{nji interval prorje|ivanja **In the original model (Ko{ir 2000) a different formula was used **U izvornom modelu (Ko{ir 2000) upotrijebljena je druga~ija jednad`ba

After i thinnings we can expect a certain number of trees to have di injuries, but this number will be small, as we must expect that during the thinning period the majority of injured trees have been already removed. To answer the question above we built a simulation matrix with Ns = N0 rows and i = n columns. Trees were randomly chosen for cut and marked as injured in the range of Vi and di for each technology separately. The final share of tree damage was the same as the share calculated using the basic model. Apart from numeric values of the simulated events we can also make a graphical picture of the situation after each i thinning. The main advantage of the simulation is that we can analyse the frequency of injury distribution after each thinning. This provides the possibility to calculate the value of money lost due to decay. The graphical part of the simulation is also valuable for student training.

3. Results of research– Rezultati istra`ivanja

Computed di do not differ significantly from field studies. For MM technology di are comparable with the averages for this combination of skid trail density and thinning density from the field measurement (Ko{ir 2000). Another example (Ko{ir and Robek 2000) shows that in an artificially established Scotch pine stand, about 50 years old, 8% of injuries were found in one cutting-unit and 12% of injuries in the second cutting-unit after the application of CTL technology (harvester Timberjack 1270 and forwarder Timberjack 1410). Fig. 1 shows the results of calculations of Di for MM and CTL technologies (formula 1). Early thinnings in the model start at the age of 30 years and continue up to 160 years when the rotation period ends and regeneration cuttings begin. The total number of damaged trees in the stand continuously grows and reaches around 90% at the end of the rotation period.

Table 2 Shares of injured trees after thinning with MM and CTL technology* Tablica 2. Udio o{te}enih stabala nakon prorede pri MM i CTL sustavu pridobivanja drva*

Results of the general model

di MM

di CTL

30 (i = 1)

Year – Godina

0.1701

0.1277

The results are partly shown in Table 2 (formulas 2, 3 and 4). The reason for this calculation was to get the whole picture of expected differences between the two technologies. First we calculated di as an input variable for formula 1 and for the simulation program.

60 (i = 4)

0.1882

0.1461

Croat. j. for. eng. 29(2008)1

90 (i = 7)

0.1896

0.1475

120 (i = 10)

0.1868

0.1447

150 (i = 13)

0.1841

0.1419

*30 – year age class (10 year interval has been used in the model) *30-godi{nji dobni razredi, s 10-godi{njim intervalima kori{teni su u modelu

B. KO[IR

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

Fig. 1 Di for MM and CTL technologies through the rotation period and difference between technologies Slika 1. Di za MM i CTL sustav pridobivanja drva u ophodnji i razlika me|u sustavima pridobivanja drva This is a rather shocking result, and the question is whether it is possible to prove such predictions by a field observation. The answer is positive if we take into account the fact that many smaller injuries disappear during the decades of tree growth. In practice it is thus possible to expect some smaller values of damaged trees (Robek and Medved 1997) than shown in Table 3, where the results of the field measurement are shown (Ko{ir 1998a). In old stands 64–70% of damage have been recorded. The difference between model predictions and the actual situation can be explained with some facts that are not

included in the model, but have an impact on field observations, such as the disappearing of visible injuries due to the healing of stem scars and the disapearing of wounded branches over a long period of time, insufficient knowledge of past technologies and the rate of damages they had induced, as well as insufficient accurate evidence about past cuttings. It can also be concluded from Figure 1 that the difference between MM and CTL technology is not constant during the whole tree age, but that it reaches its maximum value (0.10) at the age of 70–80 years and from then on decreases toward the limit = 0.

Table 3 Structure of tree damage in remaining spruce stands (databank of tree damages – Ko{ir 1998a) Tablica 3. Struktura o{te}enja stabala u preostaloj sastojini (baza podataka o{te}enja stabala – Ko{ir 1998a) Spruce stands on Alpine plateau Smrekove sastojine na alpskoj visoravni Polewood (40–60 years) Stupovlje (40–60 godina) Old and younger mature stand (80–100 years) Starije i mla|e zrele sastojine (80–100 godina) Mature stand (100–140 years) Zrele sastojine (100–140 godina) Stand in regeneration (> 140 years) Sastojine u pomla|ivanju (>140 godina) Old mature stand (> 140 years) Prezrele sastojine (> 140 godina)

Undamaged trees Neozlije|ena stabla

Old injuries Stare ozljede

Old and new injuries Stare i nove ozljede

New injuries Nove ozljede

0.58

0.31

0.05

0.06

0.33

0.47

0.16

0.05

0.32

0.50

0.12

0.06

0.21

0.64

0.13

0.03

0.15

0.70

0.15

0.00

Croat. j. for. eng. 29(2008)1

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

Motor-manual felling and extraction with tractors show very bad results in early thinnings in comparison with CTL technology.

Results of simulations An overall picture of tree damage behaviour during the rotation period should be close to the general

B. KOŠIR

model, though we can expect some minor deviations due to the stochastic process. The advantage of this approach is that we can obtain the structure of injuries according to the time of appearance and the number of trees, which have been injured several times (Table 4). We may also analyse the injuries of the removed part of the stand. These results enable

Table 4 Number of trees in stands in 30-year age classes Tablica 4. Broj stabala u sastojini za 30-godi{nje dobne razrede Year Godina

Undamaged trees Neo{te}ena stabla

30 60 90 120 150

3680 693 211 96 39

Year Godina

Undamaged trees Neo{te}ena stabla

30 60 90 120 150

3731 822 304 134 70

1 747 582 295 163 92

1 570 507 354 213 130

MM Number of each injured trees in different thinning period Broj ozlje|ivanja svakoga pojedinoga stabla tijekom razli~itih proreda 2 3 4 5 6 – – – – – 193 30 – – – 213 97 22 – – 155 120 49 8 1 123 109 58 24 8 CTL Number of each injured trees in different thinning period Broj ozlje|ivanja svakoga pojedinoga stabla tijekom razli~itih proreda 2 3 4 5 6 – – – – – 123 5 1 – – 157 48 3 – – 161 62 24 3 – 138 81 40 8 1

7 – – – – 1

7 – – – – –

Fig. 2 Differences in number of injured trees between MM and CTL technology Slika 2. Razlike u broju o{te}enih stabala izme|u sustavâ MM i CTL za pridobivanje drva Croat. j. for. eng. 29(2008)1

B. KO[IR

Modelling Stand Damages and Comparison of Two Harvesting Methods (5â&#x20AC;&#x201C;14)

us to make a better assessment of the lost value than ever made so far (Ko{ir 1998b). Nevertheless, this is not within the scope of this article. Differences in the number of injured trees after the whole rotation period are shown in Fig. 2. With CTL technology we can expect a greater number of undamaged trees. Later in the stand development

we can also expect a greater number of trees that have been injured once or twice on one hand, and a smaller number of trees that have been damaged more times on the other hand, than is the case when using MM technology. The whole structure of relative frequencies according to age is given in Fig. 3 and 4. First we notice

Fig. 3 Changes in the structure of injured trees for MM technology Slika 3. Promjene u strukturi o{te}enih stabala pri MM sustavu pridobivanja drva

Fig. 4 Changes in the structure of injured trees for CTL technology Slika 4. Promjene u strukturi o{te}enih stabala pri sustavu CTL za pridobivanje drva 10

Croat. j. for. eng. 29(2008)1

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

similar flows of undamaged stand curves, which are slowly closing towards the limit of 100% damaged trees. The share of tree injured once appears after the first thinning and at 70 (MM) and 90 (CTL) years it reaches its maximum, and thereafter slowly decreases as more and more trees with one injury receive more injuries or are removed from the stand. Trees with two injuries appear one thinning later and show a similar trend. The maximum is at 100 (MM) or 150 (CTL) years, and later the share of such trees slowly decreases. Other curves in both figures show similar logic flow. Comparison of both figures shows that the CTL curves have maximums at later stages of the stand, because there is less damage at each thinning. This can be better understood if we consider Figure 2 once more, where the situation after the end of the observed period is shown. The MM system shows a strong tendency of injury distribution towards more injuries per tree. This also means that the maximum of distribution is at three instead of two injuries per tree, as shown in the distribution of CTL technology. A compared distribution of injuries in earlier development stages shows the same relations between MM and CTL technologies, but with a higher number of undamaged trees (0 injuries) and a smaller number of trees injured more than once.

4. Discussion and conclusions – Rasprava i zaklju~ci Models of tree damage in the remaining stand are a valuable tool for analysing the consequences of forest operations in the time to come. It is of minor importance that the technologies are changing in the direction of less brute force, and what is even more important, machine operators, supervisors and forest owners are more aware than before of the damaging potential of modern technology. Education and training of machine operators and all others who deal with forest management is therefore a permanent issue. With respect to the described facts, the change of technology means that the same relations will appear on a different level, hopefully better for the forest. In this paper we compared two technologies which seem to compete at present in different conditions, where stand composition and terrain characteristics play a major role. An older technology, where motor-manual work prevails, has been studied much closer and for a longer period of time. Cut-to-length technology is well known, with a long tradition in the Nordic and many other countries, but the results of damage studies (Frödig 1992, Frohm 1993) or models (Siren 1999, Dvoøak 2005) cannot be used directly in Slovenian conditions, as stand composition, stand density, terrain characterCroat. j. for. eng. 29(2008)1

B. KOŠIR

istics and other circumstances of forest operations are different. This does not mean that the validation of the model is questionable, but there is a doubt whether the input variables are reliable and representative for a certain technology. There is also a challenging question of whether historical data can support our findings. This paper showed that under the given presumptions CTL technology causes less damage to the forest. In the long run that means a slower increase of damaged trees in the stand and better distribution of accumulated damages in the stand. This would improve the stability of stands (less damage by bark beetles, for example) and increase the yield value. Other advantages and disadvantages of CTL technology have not been discussed in the paper. There are still numerous questions connected to CTL technology and stand damages. Stand density influences considerably the share of injuries in the stand. If the reports of tree damage from northern countries (<0.05) are compared with those from Central Europe, a large discrepancy is obvious. Stand density is already included into the model, but not directly. In this respect the model should be upgraded by additional choices, based on actual field measurements. During recent observations and measurements (not yet published), according to the slightly modified Frödig method (1992), we noticed that tree damages are greater when the machine operator is working in a dark environment (with lights engaged). There is also no reliable research of extended work time (Nicholls et al. 2004) and work quality where work quality also includes tree damage. Answers to these and other questions will also improve the results of comparisons described above.

5. References – Literatura Butora, A., Schwager, G., 1986: Holzernteschaden in Durforstungbestanden. Berichte, 288, Birmensdorf, Eidgenossische Anstalt fur das forstliche Verzuchswezen, 51 p. Diaci, J., McConnell, S. 1996: Close-to-nature forestry and ecosystem management. Zb. gozd. lesar. 49: 105–127. Dvoøak, J. 2005: Variability of Tree Damage with Respect to Felling – Technological Factors that can be Changed in Short Term. Proceedings: FORMEC 2005, Ljubljana, 139–146. Eriksson, L. 1981: Strip roads and Damages Caused by Machines when Thinning Stands. The Swedish University of Agricultural Sciences, Dep. of Operational Efficiency, Garpenberg 1992, Rep. No. 193: 1–44. Frödig, A., 1992:. Thinning damage – A study of 403 stands in Sweden in 1988. The Swedish University of Agricultural Sciences, Dep. of Operational Efficiency, Garpenberg 1992, Rep. No. 193: 1–45.

B. KO[IR

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

Frohm, S., 1993: Efficient and Safe Thinning. In: Efficient, Sustainable and Ecologically Sound Forestry, Skogforsk Report 5: 43–49. Halaj, J., Grék, J., Pánek, F., Petrá{, R., Øehák, J., 1987: Rastové tabul’ky hlavnýh drevín ÈSSR. Príroda, Bratislava, p. 362. Harstela, P., 1995: Environmental impacts of wood harvesting in Nordic countries. Environmental impacts of Forestry and Forest Industry. EFI Proc. 3: 37–44. Ivanek, F., 1976: Vrednotenje po{kodb pri spravilu lesa v gozdovih na Pohorju. IGLG, Strokovna in znanstvena dela 51, Ljubljana, 142–147. Ko{ir, B., Cedilnik, A., 1996: The model of number increasing of tree damages at thinnings. Zb. gozd. lesar. 48: 135–151. Ko{ir, B., 1996: How to manage thinning with low damages of standing trees – experience from the model. Proceedings »Planning and implementing forest operations to achive sustainable forests« 19th Annual Meeting of COFE & IUFRO SG S3.04–00, July 29–August 1, 1996, Marquette, Michigan USA, 82–91. Ko{ir, B., 1998a: Damage to mountain spruce stands due to harvesting. Conference proceedings »Gorski gozd«, Ljubljana: Biotechnical Faculty, Department of Forestry and Renewable Forest Resources, p. 95–107. Ko{ir, B., 1998b: Critical evaluation of frequent thinnings from the aspect of energy consumption and damage in the stands. Zb. gozd. lesar. 56: 55–71. Ko{ir, B, Robek, R., 2000: Characteristics of the stand and soil damage in cut-to-lenght thinning on the @ekanc working site (SW Slovenia). Zb. gozd. lesar. 62: 87–115. Ko{ir, B., 2000: Primerjava rezultatov modela po{kodb drevja v sestoju zaradi pridobivanja lesa in terenskih opazovanj. – Research Reports, University of Ljubljana, Biotechnical Fac., Dep. of Forestry and Forest Resources, 62, p. 135–151. Ko{ir, B., 2001: Frequent thinning – impact on stand quality. In: Thinnings: a valuable forest management tool., Montreal, Quebec, Canada, September 2001, Canadian forest service, 2003. Ko{ir, B., 2004: Factors affecting technological changes. Gozd. vestn. 62(1): 3–11.

Krivec, A., 1975: Racionalizacija delovnih procesov v se~nji in izdelavi ter spravilu lesa glede na delovne razmere in po{kodbe. Research Reports, University of Ljubljana, Biotechnical Fac., Dep. of Forestry, 13, 2, Ljubljana, p. 145–193. Leinss, C., 1991: Unterzuchungen zur Frage der nutzungstechnischen Folgen nach Fall-und Ruckeshaden bei Fichte (Picea abies /L./ Karst.). Mitteilungen der Forstlichen Versuchs- und Forschungsanstalt Baden-Wurttemberg, Freiburg im Breisgau, Heft 157, 172 S. Matthews, J. D., 1999: Sylvicultural systems. Oxford Univ. Press, Oxford, p. 98–137. Mlin{ek, D., 1977: Übertragbarkeit und die Bedeutung des Prinzipes der Nachhaltigkeit und der Theorie der Waldpflege für die Naturgerechte Bewirtschaftung von erneuer-baren Naturgutern. Die Waldpflege in der Mehrweckforstwirtschaft. Österreichischer Agrarverlag, Wien, p. 45–57. Mlin{ek, D., 1994: Was ist naturnahe Waldwirtschaft?. In: HATZFELDT, Hermann Graf (Ed.). Ökologische waldwirtschaft : Grundlagen – Aspekte – Beispiele, (Alternative Konzepte, 88). Heidelberg: C. F. Müller, p. 67–76. Nicholls, A., Bren, L., Humphreys, N., 2004: Harvester Productivity and Operator Fatique: Working Extended Hours. Int. J. of For. Eng. 15(2): 57–65. Robek, R., Medved, M., 1997: Po{kodbe drevja zaradi izvajanja gozdarskih del po podatkih popisov propadanja gozdov Sloveniji. Research Reports, University of Ljubljana, Biotechnical Fac., Dep. of Forestry and Forest Resources, 52, p. 119-136. Sabo, A., 2000: Damaging Trees at Timber Skidding by the Skidder LKT 81 in Selection Forests of Different Openness in the Region of Gorski Kotar. Mehanizacija {umarstva 25(1–2): 9–27. Siren, M., 1999: One-Grip Harvesting Operations, Silvicultural Results and Possibilities to predict Tree Damage. In: Proc. IUFRO 3.09.00 Harvesting and Economic of Thinnings, Ennis, Ireland, p. 152–167. Spinelli, R., 1999: The Environmental Impact of Thinning: More Good than Bad? In: Proc. IUFRO 3.09.00 Harvesting and Economic of Thinnings, Ennis, Ireland, p. 136–143.

Sa`etak

Modeliranje o{te}enja sastojine i usporedba dvaju sustava za pridobivanje drva Gospodarenje se {umama u Sloveniji zasniva na prirodnoj odr`ivosti i multifunkcionalnosti. Prevladava grupimi~no gospodarenje raznodobnim sastojinama, a ~iste su sje~e ve} desetlje}ima zabranjene. Zbog ~estih

Croat. j. for. eng. 29(2008)1

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

B. KOŠIR

prorje|ivanja (u dràvnim {umama – 1 do 2 sijeka u istoj sastojini tijekom 10 godina), te{kih terenskih prilika i velikih dimenzija stabala naj~e{}e je kori{ten ru~no-strojni sustav za pridobivanje drva – MM (radnik s motornom pilom te zglobni traktor za privla~enje drva). Oplodne su sje~e vi{e iznimka nego pravilo. Zbog brojnih dru{tvenih promjena, promjena u cijenama energenata te razvoja informacijskih tehnologija, ali i zbog sve ve}e konkurencije na trì{tu drva dovedene su u pitanje ekonomi~nost i budu}nost primjene takva na~ina pridobivanja drva. Glavni razlog postupnoga uvo|enja potpuno mehaniziranoga sustava za pridobivanje kratkoga drva – CTL (sje~a i izradba harvesterom te izvoènje forvarderom) nakon 2000. godine bila je èlja za smanjenjem tro{kova, ali i nedostatak kvalificirane radne snage za ru~no-strojni rad motornom pilom. Uvo|enje je sustava harvester – forvarder (CTL) dovelo do brojnih rasprava zbog straha od mogu}ih o{te}enja preostalih stabala nakon sje~e (mogu}e ozljede na deblu, panju, korijenskom sustavu i granama, savijena stabla) i {umskoga tla te je od po~etka bio prisutan negativan stav i suzdrànost prema tomu »novomu« na~inu pridobivanja drva. Unato~ nepovoljnim rezultatima prethodnih istraìvanja o o{te}enju preostalih stabala nakon sje~e motornom pilom uobi~ajeni na~in pridobivanja drva (MM) nije bio predmetom rasprava zbog svoje tradicionalnosti. U radu je istraìvan problem o{te}enja preostalih stabala u sastojini nakon sje~e pri primjeni dvaju sustava za pridobivanje drva, razli~itih s obzirom na razinu mehaniziranosti sje~e i izradbe drva. Kori{teni su osnovni model procjene i simulacijski model za procjenu o{te}enja tijekom cijele ophodnje. Uspore|ena su oba sustava s obzirom na ukupan broj o{te}enih stabala, ali i struktura o{te}enih stabala ovisno o ponavljanju o{te}enja svakoga stabla pri izvo|enju {umskih radova u razli~itim razdobljima (prorede). Upotrebom osnovnoga modela za procjenu o{te}enja stabala dobiveni su dobri rezultati u pro{losti ({to je i potvr|eno terenskim mjerenjima), pa je tako i ovdje kori{ten isti model za usporedbu o{te}enja ovisno o primijenjenim sustavima za pridobivanje drva. Kori{teni osnovni model razvijen je za metodu oplodnih sje~a, gdje je sje~na gusto}a odre|ena odnosom dozna~enih stabala i svih stabala u sastojini. Intenzitet o{te}enih stabala jest odnos izme|u o{te}enih stabala po zavr{etku sje~e i svih preostalih stabala u sastojini. Postavljene pretpostavke su: Þ jednaka je vjerojatnost da svako stablo bude posje~eno u odre|enom vremenu Þ jednaka je vjerojatnost da svako stablo bude o{te}eno za vrijeme posljednje sje~e Þ svako stablo koje je o{te}eno vi{e od jedanput u istom sijeku broji se kao jedno o{te}enje Þ stablo je ozna~eno kao o{te}eno bez obzira na teìnu ozljede. Osnovni model razvoja i procjene primjenjuje vi{e kriterija nego {to je bilo potrebno za ovu analizu, stoga je primijenjeno tzv. »pravilo akumulacije o{te}enih stabala«. Udio o{te}enih stabala nije neposredno ovisan o intenzitetu prorje|ivanja, me|utim ta se ovisnost posredno javlja kroz intenzitet sje~e (ei), gusto}u vlaka (G) i udio o{te}enih stabala (di) nakon i-te prorede. Primijenjene su jednad`be omogu}ile dobru procjenu ukupne o{te}enosti stabala u sastojini nakon i-toga prorje|ivanja te je zbog èlje da se sazna u~estalost ozljeda na stablima na kraju ophodnje postavljeno pitanje: koliko se ozljeda moè o~ekivati na stablima i deblima? Za odgovor na to pitanje napravljena je simulacijska matrica sa Ns = N0 redova i i = n stupaca. Ukupni udio o{te}enih stabala bio je jednak kao i kod upotrebe osnovnoga modela. Slika 1 prikazuje usporedbu udjela o{te}enih stabala nakon i-te prorede za dva na~ina pridobivanja drva tijekom cijele ophodnje. Ukupni broj o{te}enih stabala stalno raste do vrijednosti od 90 % na kraju ophodnje. Iako su to zapanjuju}i rezultati, postavlja se pitanje da li se navedeno moè dokazati i na terenu. Naravno, treba uzeti u obzir da brojna o{te}enja s vremenom zarastaju – time i nestaju, zatim nepoznavanje prije kori{tenih sustava za pridobivanje drva, a s tim i koli~ine nastalih o{te}enja, ali i nedovoljno to~ni podatci o prija{njim sje~ama. Vidljivo je i da razlika izme|u kori{tenih sustava za pridobivanje drva nije stalna za vrijeme cijeloga ìvotnoga vijeka stabla, ali da doseè najve}u vrijednost (0,10) u 70-oj, odnosno 80-oj godini starosti te nakon toga opada (0). U tablici 4 prikazani su rezultati simulacije. Prednost je ove metode procjene u tome {to se mogu dobiti podatci o o{te}enjima ovisno o vremenu nastajanja te broj stabala koja su bila o{te}ena vi{e od jedanput. Mogu}a je analiza o{te}enja na deblu (nakon sje~e), pa je tako mogu}e odrediti i gubitak na vrijednosti zbog nastalih o{te}enja, ali to ipak nije bio cilj ovoga istraìvanja. Na slici 2 prikazane su razlike u broju o{te}enih stabala nakon cijele ophodnje. Upotrebom se sustava CTL za pridobivanje drva o~ekuje manje o{te}enih stabala. Slike 3 i 4 prikazuju strukturu relativne u~estalosti javljanja o{te}enja ovisno o godinama. Udio jednom o{te}enih stabala najve}i je u 70-oj godini starosti za ru~no-strojni na~in pridobivanja drva (MM), te u 90-oj godini za sustav CTL za pridobivanje drva, nakon ~ega udio opada dijelom zbog pove}anja broja ozljeda na stablu, a dijelom zbog same sje~e tih stabala. Udio dva puta o{te}enih stabala najve}i je za ru~no-strojni sustav za

Croat. j. for. eng. 29(2008)1

B. KO[IR

Modelling Stand Damages and Comparison of Two Harvesting Methods (5–14)

pridobivanje drva u 100-oj godini, a za sustav CTL u 150-oj godini starosti. Ostale krivulje pokazuju sli~an tijek. Usporedbom tih krivulja jasno je da sustav CTL za pridobivanje drva doseè vrhunac u kasnijim godinama ophodnje jer je manje o{te}enja za vrijeme svakoga prorje|ivanja. Istraìvanje je pokazalo da sustav CTL za pridobivanje drva uzrokuje manje o{te}enja u {umi, sporije pove}anje broja o{te}enih stabala tijekom vremena, ali i bolju distribuciju nakupljenih o{te}enja u sastojini, {to pove}ava stabilnost sastojine, ali i pove}ava vrijednost drva. Broj stabala po jedinici povr{ine utje~e na udio ozljeda pri primjeni sustava CTL te ako se usporede podatci o o{te}enjima stabala iz skandinavskih zemalja (<0,05) s podatcima iz srednje Europe, dolazi do odre|enih odstupanja pa je stoga potrebno unaprje|enje postoje}ega modela pomo}u stvarnih terenskih mjerenja. Tijekom nedavnih istraìvanja i mjerenja (koja jo{ nisu objavljena) primjenom modificirane Frödigove metode (1992) primije}eno je da su o{te}enja na stablima ve}a ako operater radi u tamnom okruènju (s upaljenim svjetlima). Tako|er ne postoje pouzdana istraìvanja o produljenom radnom vremenu i samoj kakvo}i rada koja uklju~uje i koli~inu nastalih o{te}enja na stablima. Odgovori na ta i druga pitanja pojasnit }e ovdje dobivene rezultate. Klju~ne rije~i: o{te}enja sastojine, sje~a, tehnologije, ru~no-strojna sje~a, strojna sje~a

Author’s address – Autorova adresa:

Received (Primljeno): April 7, 2008 Accepted (Prihva}eno): June 10, 2008

Prof. Bo{tjan Ko{ir, Ph.D. e-mail: bostjan.kosir@bf.uni-lj.si University of Ljubljana, Biotechnical Faculty Department of Forestry and Forest Resources Ve~na pot 83 1000 Ljubljana SLOVENIA Croat. j. for. eng. 29(2008)1

Orginal scientific paper – Izvorni znanstveni rad

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks and Possibility of Accurate Estimate Hideo Sakai, Tomas Nordfjell, Kjell Suadicani, Bruce Talbot, Ebbe Bøllehuus Abstract – Nacrtak An 8-WD forwarder loaded with 9,520 kg of timber, and fitted with low or high tire pressures, or tires rounded with tracks, was repeatedly driven on soil for 1, 8 and 24 passes to investigate mechanistic influences on soil compaction. Soil compaction occurred during the early passes, and heavy compaction occurred after 8 passes. High pressure tires caused heavy compaction in the deeper soil layer zones. The compacted zone for a loaded forwarder with tracks was shallow in depth and had the lowest degree of compaction. Linear regression between contact pressure and average depth of ruts after 24 passes was derived. An increase in contact pressure of 100 kPa caused a decrease of 5.7% in soil porosity at 10–15 cm depth after 24 passes. Maximum increment of cone index of 85 kPa, which occurred at depths of 14 to 28 cm, meant a decrease of 1% soil porosity between depths of 10-15 cm. Additionally, ruts of 10 cm in depth decreased porosity by 7%. Tracks kept original porosity with lowest compaction and should therefore be useful for preventing soil compaction. Key words: soil compaction, tracks, depth of ruts, porosity

1. Introduction – Uvod Harvesting and regeneration management must be solved scientifically and technologically so as to ensure sustainable forestry. Soil compaction caused by harvesting operations can affect future regeneration and growth of trees. Therefore, different kinds of tracks, machine size and load, and frequency of passes of forwarder transport should be decided in advance to minimize compaction for specific soil types and operational conditions. In this study, we intended to clarify differences of soil compaction among different kinds of tire pressures and tracks using the technique of cone-penetrometer. A cone-penetrometer has been used to measure soil compaction because it has quick application and can be used for interrelation of different soil conditions (Kumakura et al. 1993). The relationships between the increase of cone index and the decrease of soil porosity after harvesting by a forwarder/tractor have been shown to estimate soil compaction (Matangaran et al. 1999, 2006). Many results of soil compaction have been reported previously Croat. j. for. eng. 29(2008)1

(Kozlowski 1999, [u{njar et al. 2006), and they must be comparable by a standardized method such as cone index to provide an easy and accurate technique.

2. Methods and experimental site – Metode i mjesto istra`ivanja Experimental site was selected at a recently harvested forest in the Grib Forest in North Zealand, Denmark. The site was nearly flat, but small changes of inclination were inevitable especially by the existence of roots and stumps (Fig. 1). Clear-cutting of planted Norway spruce (Picea abies (L.) Karst.) was done just before the experiment, and timber was extracted using a forwarder system. The surface was disturbed at random by low frequency of forestry vehicle passes, and slash was left in the form of leaves and branches. Soil type at the experimental site was humus layer of 5 to 20 cm in depth, sandy A-layer of 15–30 cm thick, yellow-brownish soil of B-layer of 10–15 cm thick, and below this there was was gray clay C-layer. Soil moisture was about 60% at the surface,

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

Fig. 1 Experimental site and the experimented forwarder Rottne Rapid 8WD with tracks Slika 1. Mjesto istra`ivanja i istra`ivani forvarder Rottne Rapid 8WD s polugusjenicama 35% in the inner part, and 20% in the deeper part. It was estimated and characterized as dry condition. The forwarder used for experiment was Rottne Rapid 8-WD (Fig. 1). According to the catalogue, the unloaded total weight was 13,270 kg, the front part weigh was 7,980 kg and that of the rear part was 5,290 kg, and the loading capacity was 12,000 kg. The length, height, overall width, and width between outer side of the tires was 8,810 mm, 3,600 mm, 2,850 mm, and 2,650 mm, respectively. Tires were Trelleborg

Twin 421 Mark II 600/55–26.5 16PR Forestry Steel Belt. Tracks were made of steel for the experiment by Eco Baltic manufacturing via Olofsfors AB, whose weight of one series for a pair of tires was 850 kg. The experiment treatments were a forwarder with rubber wheels of low pressure of 120 kPa, high pressure of 350 kPa, and 500 kPa rounded with tracks, respectively. This corresponded to six treatments designed as unloaded and loaded. The load was set at 9,520 kg of logs as normal condition. In case of only wheels, two sets of tracks of 1,700 kg were loaded to reduce the difference of weight with tracks. The forwarder speed was 2.8 km per hour. Straight and nearly flat 18 courses of 20 m length were prepared before the movement of the forwarder, and three courses were randomly allotted for one treatment. Contact area of high pressure tires on the soil at the experimental site was 60´45 cm (rear) and 55´32 cm (front), and that of low pressure tires was 70´55 cm (rear) and 50´50 cm (front). Net contact area of one series of tracks was approximately 8,033 cm2 on hard ground, and 8,940 cm2 on soft ground, which was measured from footprint of track shoes, respectively. Contact pressure of each treatment was estimated from these contact areas (Table 1). After 1, 8 and 24 passes of a running forwarder, cone index was measured at the center of each of the ruts (Fig. 2). Five samples were randomly selected to

Table 1 Contact pressure of forwarder on soil per axles Tablica 1. Dodirni tlak forvardera na tlo po osovinama Treatment Djelovanje Low pressure tires – unloaded Gume s niskim tlakom (neoptere}en forvarder) Low pressure tires – loaded Gume s niskim tlakom (optere}en forvarder) High pressure tires – unloaded Gume s visokim tlakom (neoptere}en forvarder) High pressure tires – loaded Gume s visokim tlakom (optere}en forvarder) Tracks – unloaded Polugusjenice (neoptere}en forvarder) Tracks – loaded Polugusjenice (optere}en forvarder)

Front axle Prednja osovina 105 Pa

Rear axle Stra`nja osovina 105 Pa

0.78

0.45

0.88

0.99

1.11

0.63

1.25

1.41

0.07

0.05

0.07

0.11

Fig. 2 Ruts from loaded high pressure tires after 23 passes Slika 2. Kolotrazi nakon 23 prolaska optere}enih gumama s visokim tlakom Croat. j. for. eng. 29(2008)1

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15â&#x20AC;&#x201C;27)

the right and left of each rut center, respectively. Cone index at the center line between right and left ruts was measured at 5 points for comparison to the control. It was assumed that the cone index between ruts was the same as that outside the courses because the inner distance between ruts was 150 cm, which had enough distance available for identifying and locating unaffected soil. Also, 20 points outside the courses were measured for the control. Depth of ruts from the original surface of 10 points along a course was also measured to the right and left of each rut after 24 passes. Most woody debris and hu-

H. SAKAI et al.

mus layer on the surface of ruts disappeared after 24 passes. A cone-penetrometer used for the measurement of soil compaction corresponded to the American Society of Agricultural Engineering standard, according to the manufacturer Findlay, Irvine Ltd. The cone diameter was 12.83 mm. Cone resistance force (kgf) of various depths from 3.5 to 52.5 cm at an inerval of 3.5 cm was measured, but almost all measurements were disturbed by stones or roots. Cone index (100 kPa) was converted to the cone resistance (kgf) after multiplying each indexed value by 0.762.

Fig. 3 Average cone index before experiment Slika 3. Prosje~ni konusni indeks tla prije istra`ivanja Croat. j. for. eng. 29(2008)1

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

Soil samples were extracted after 24 passes of each treatment and the control outside the course by using soil sampler of 10-4 m3 at the surface and at the layer of 10–15 cm at the depth below the humus, which represented the most compacted zone. Three samples were taken for each treatment at each depth interval. Soil porosity was analyzed according to JIS test method (Japanese Society of Soil Mechanics and Foundation Engineering 1969).

3. Soil cone index of the experimental site – Konusni indeks tla na mjestu istra`ivanja There were numerous stones and roots with stumps in the underground. Data of cone index varied with increasing underground depth. Although the cone index outside the course from the surface to a depth of 10.5 cm was lower than the average one for 18 courses before applying treatments, the result of F-test among 18 courses and outside the course showed no significant differences (p > 0.05). The average cone index also showed similar values (Fig. 3). There were no significant differences among the averages of the six treatments and outside the course after F-test (p > 0.05). At 3.5 to 14 cm in depth, soil layer had thick humus and the cone index was stable, and from 14 to 31.5 cm the soil had sandy layer with a cone index that ranged over 500 kPa. From about 31.5 cm in depth, the layer became yellow-brownish clay, corresponding to a wider index range. A regression line for the total value was based on 110 measurements that had 18 courses and 20 points outside the course. This was therefore considered to be representative value of the cone index for the experimental site before soil compaction by forwarder, and thus could be used as the control for pre-treatment.

4. Resarch results – Rezultati istra`ivanja 4.1 Soil compaction – Zbijanje tla It was observed that the tires pushed surface soils downward even on a slightly downward sloping surface, and on these surfaces increased compression of soils was recorded. In the case of loaded treatments, increases of cone index were apparent with increased number of passes (Fig. 4). The range of lines was small at the zone between the surface around 14 cm in depth. The difference increased remarkably from the depth of more than 14 cm with the number of passes. The depth between about 14 and 21 cm was the most compacted zone. Under the depth of nearly 35 cm, data did not show any distinct trends. Cone indices at 14 to 21 cm in depth by

loaded low and high pressure tires, and fitted tracks after 24 passes were 2210–2390 kPa, 2350–2670 kPa, and 2100 kPa, respectively. Fitted tracks had the lowest compressed soil, whereas high pressure tires had the greatest. For the case of unloaded low pressure tires, cone index increased with increasing number of passes, and the maximum cone index after 24 passes was between 2250–2350 kPa, which was the same for the loaded treatment, whereas lines of unloaded high pressure tires and tracks were lower than those of the loaded treatments. However, depth of compacted zone for unloaded low pressure tires was about 3.5 cm shallower than the loaded treatment. For unloaded high pressure tires, although no passes had higher values, cone index increased with the number of passes. The increased cone index after 24 passes was 2250 kPa at 17.5 cm in depth, which was nearly the same as that by unloaded low pressure tires, but depth of compacted zone was around 3.5 cm deeper than unloaded low pressure tires. The lines after unloaded tracks were the lowest among all treatments, although the cone index after passes was lower than that of no passes because of low compaction. Although there were minus values probably due to the almost negligible difference compared to the average value, the average increment of cone index from the decided value of control with no passes was evident after the repetition of the passes (Fig. 5). The strong compaction occurred at the earliest stage of passes of loaded high pressure tires. Low pressure tires and tracks with and without load showed increased cone index at the surface, and at deeper layers, increase of cone index became smaller for both unloaded and loaded tracks. In the case of unloaded high pressure, the increase of cone index became also smaller with increasing depth of soil. Loaded low pressure tires and tracks showed high values of compaction from eighth passes onwards. Similarly, with unloaded low pressure tires and tracks compaction occurred after eight passes. In the range from the surface to 28 cm in depth, the maximum increment of the cone index was evident from the value of the control and its depth (Table 2). Differences of the increment of cone index between loaded and unloaded low pressure tires were small, whereas the increment of cone index of loaded high pressure tires and tracks were larger than that for unloaded condition. For unloaded treatments, the maximum increment of low and high pressure tires occurred at 3.5 to 24.5 cm in depth, whereas for tracks it occurred at 3.5 cm. On soil treated with loaded low pressure tires, high pressure tires, and tracks, depth of the maximum increment Croat. j. for. eng. 29(2008)1

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15â&#x20AC;&#x201C;27)

H. SAKAI et al.

Fig. 4 The average cone index per forwarder passes Slika 4. Prosje~ni konusni indeks po prolascima forvardera Croat. j. for. eng. 29(2008)1

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15â&#x20AC;&#x201C;27)

Fig. 5 The increment of cone index from none passes Slika 5. Pove}anje konusnoga indeksa u odnosu na neizga`eno tlo 20

Croat. j. for. eng. 29(2008)1

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

Table 2 Maximum increment of the cone index (105 Pa) and its depth after treatments Tablica 2. Najve}e pove}anje konusnoga indeksa (105 Pa) i dubina mjerenja nakon prolazaka forvardera Passes Prolasci 1 2 4 8 16 24

Low pressure tires unloaded Gume s niskim tlakom (neopter. forvarder) 0.9 (10.5 cm) 2.9 (17.5 cm) 3.3 (17.5 cm) 4.0 (3.5 cm) 3.7 (17.5 cm) 4.7 (14.0 cm)

Low pressure tires High pressure tires loaded unloaded Gume s niskim tlakom Gume s visokim tlakom (optere}en forvarder) (neopter. forvarder) 2.3 (3.5 cm) 0.8 (7.0 cm) 1.9 (3.5 cm) 1.3 (3.5 cm) 2.8 (3.5 cm) 1.1 (7.0 cm) 5.2 (28.0 cm) 3.1 (10.5 cm) 2.3 (14.0 cm) 1.1 (10.5 cm) 3.3 (17.5 cm) 3.2 (17.5 cm)

was found to be 28, 28, and 14 cm, respectively. Loaded low and high pressure tires had compaction at deeper soil layers, and the maximum increment of cone index reached 524 kPa after eight passes and 732 kPa after 24 passes, respectively. The compaction of loaded tracks was smaller and occurred at shallower depths compared to the other treatments. The maximum increment of cone index was derived (Fig. 6). Soil compaction occurred in the early passes (Shishiuchi and Adachi 1982). About 200 to 400 kPa of cone index increment occurred in the first pass and those of loaded low and high pressure tires reached at 317 and 532 kPa after four passes, respectively. The severe compaction was caused by loaded high pressure tires. The line of loaded tracks existed

High pressure tires loaded Gume s visokim tlakom (optere}en forvarder) 3.7 (3.5 cm) 5.4 (3.5 cm) 5.3 (10.5 cm) 5.8 (3.5 cm) 6.2 (24.5 cm) 7.3 (17.5 cm)

Tracks – unloaded Polugusjenice (neopter. forvarder)

Tracks – loaded Polugusjenice (optere}en forvarder)

–2.3 (3.5 cm) –1.8 (3.5 cm) –1.3 (3.5 cm) –0.2 (3.5 cm) –0.7 (3.5 cm) –0.4 (3.5 cm)

1.6 (3.5 cm) 1.6 (3.5 cm) 1.3 (3.5 cm) 1.9 (14.0 cm) 2.0 (14.0 cm) 3.0 (14.0 cm)

between those of unloaded low and high pressure tires. There were differences of compacted soil depth and the maximum increment of cone index between four and more than eight passes for all treatments. The relationship between contact pressure and maximum increment of cone index after 24 passes showed a tendency that increased contact pressure resulted in increasing compaction (Fig. 7).

4.2 Depth of ruts – Dubina kolotraga The depth of ruts was shallow for the low frequency passes. Once ruts were formed, repetition of passes made the ruts deeper. The observations endorsed the differences on the increase of ruts between four passes and more than eight passes on the

Fig. 6 Maximum increment of cone index Slika 6. Najve}e pove}anje konusnoga indeksa Croat. j. for. eng. 29(2008)1

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

Fig. 7 Maximum increment of cone index vs. contact pressure after 24 passes Slika 7. Ovisnost najve}ega pove}anja konusnoga indeksa o dodirnom tlaku na tlo nakon 24 prolaska maximum increment of cone index. When the forwarder moved over a stump or on a small downward slope, sudden increases of load pressure caused increased compression followed by shearing forces that accelerated the forming of ruts. A tendency was recorded that the deeper ruts were more compacted at some depths of soil (Fig. 8), but significant correlation coefficient between the average depth of ruts after 24 passes and maximum increment of cone index was not obtained. The treatments of high pressure tires unloaded and loaded, and loaded low pressure tires had deep ruts. Depth of ruts by unloaded low pressure tires and tracks was shallow and the same at about 4.5 cm, but the compaction by low pressure tires was much higher than that caused by tracks. For the loaded low pressure tires and tracks, maximum increment of cone index of tracks was lower than that of low pressure tires, and the depth of ruts by tracks was about 2 cm shallower than that by low pressure tires. Although thus the physical mechanism of compaction and forming of ruts was different between tires and tracks, linear relationship between contact pressure and average depth of ruts after 24 passes was obtained (p < 0.01) (Fig. 9).

Fig. 8 Maximum increment of cone index vs. depth of ruts after 24 passes Slika 8. Ovisnost najve}ega pove}anja konusnoga indeksa o dubini kolotraga nakon 24 prolaska

4.3Physical soil properties – Fizikalne zna~ajke tla The average specific gravity of the soil surface was 2.37 g/cm3 (Table 3), and there were no significant differences among treatments by F-test (p > 0.05). Inner part of soil at the depth of 10–15 cm was 2.48 g/cm3, and there were no significant differences among treatments (p > 0.05). Between the average 2.37 g/cm3 and 2.48 g/cm3, a significant difference was found by t-test (0.01 < p < 0.05). The inner part of soil of loaded high pressure tires had the most compacted soil, and the decrease of liquid and air phases from no passes after 24 passes were both about 5% (Table 3). The correlation coefficient (r) between soil bulk density and porosity of 24 samples was 0.987 (p < 0.05). A derived line showed theoretically the point of soil porosity of 100% for zero of bulk density and the point of the average specific gravity of 2.43 of all samples (Fig. 10). Porosity of the inner soil was considered as the most compacted zone, and smaller than that of the surface. There were no significant differences among porosities of the four treatments Croat. j. for. eng. 29(2008)1

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

H. SAKAI et al.

Fig. 9 Depth of ruts vs. contact pressure after 24 passes Slika 9. Ovisnost dubine kolotraga o dodirnom tlaku na tlo nakon 24 prolaska

Fig. 10 Soil porosity vs. bulk density Slika 10. Ovisnost poroznosti o prirodnoj gusto}i tla

at the surface soil by F-test (p > 0.05). However, a significant difference was recorded between loaded low and high pressure tires at the surface soil (p < 0.05). At the inner soil, there were no significant differences between loaded high and low pressure tires, and among loaded low pressure tires, tracks and no passes by t-tests (p > 0.05). But there was a

significant difference between loaded high pressure tires and tracks (p < 0.01). Loaded high pressure tires had the highest value of compaction for the inner soil, and porosity decreased by nearly 10% from the treatment with no passes. Tracks sustained soil porosity better than other treatments, as the clear relationship between contact pressure and soil porosity

Table 3 Physical properties of soil samples after treatments with loaded forwarder Tablica 3. Fizikalna svojstva uzoraka tla nakon prolazaka optere}enog forvardera Part Dio tla

Treatment Djelovanje

No passes – Bez prolaska Low pressure tires – Gume s niskim tlakom High pressure tires – Gume s visokim tlakom Tracks – Polugusjenice No passes – Bez prolaska Low pressure tires – Gume s niskim tlakom Inner Unutra{njost High pressure tires – Gume s visokim tlakom Tracks – Polugusjenice Surface Povr{ina

Croat. j. for. eng. 29(2008)1

Spec. gravity Bulk density Solid phase Liquid phase Air phase Spec. gusto}a tla Prir. gusto}a tla ^vrsta faza Teku}a faza Zra~na faza g/cm3 % 2.50 102.8 41.7 43.8 14.5 2.28 67.4 31.4 64.4 4.0 2.36 94.3 40.5 50.8 8.7 2.33 106.2 45.6 45.6 8.8 2.48 118.6 48.2 35.0 16.8 2.46 127.9 52.5 35.4 12.1 2.52 146.0 58.1 29.7 12.2 2.48 127.3 51.5 40.9 7.6

Porosity Poroznost 58.3 68.6 59.5 54.4 51.8 47.5 41.9 48.5

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

Fig. 11 Soil porosity in the inner part vs. contact pressure after 24 passes Slika 11. Ovisnost poroznosti tla na dubini 10–15 cm o dodirnom tlaku nakon 24 prolaska

Fig. 12 Soil porosity vs. maximum increment of cone index after 24 passes Slika 12. Ovisnost poroznosti tla o najve}em pove}anju konusnoga indeksa nakon 24 prolaska

in the inner part was nearly linear (0.01 < p <0.05) (Fig. 11). The increment of 100 kPa in the contact pressure caused the decrease of 5.7% porosity at 10–15 cm in depth. Although the most compacted depth was different among treatments, relationship between the maximum increment of cone index, which occurred at 14 cm to 28 cm in depth, and porosity was nearly linear (p < 0.05) (Fig. 12). The increase of cone index of 85 kPa equaled a decrease of soil porosity of 1%.

roots. The range of fluctuation was small in the zone between the surface and about 14 cm in depth because of a rapid increase of cone index in the thick

4.4 Estimation of soil compaction – Procjena zbijanja tla The formation of ruts differed between tires and tracks, and it was related to the contact pressure of tire/track surfaces to the soil matrix. The ruts of 10 cm in depth showed a tendency of decrease of 7% in soil porosity at 10–15 cm in depth (p < 0.05) (Fig. 13). It was recognized that the ruts caused by loaded tracks were the shallowest and any decrease of porosity was the smallest after 24 passes.

5. Conclusions with discussion – Zaklju~ci s raspravom The underground soil conditions varied greatly especially in deeper soil layers because the samples for cone index decreased by disturbance of rocks or

Fig. 13 Soil porosity vs. depth of ruts Slika 13. Ovisnost poroznosti tla o dubini kolotraga Croat. j. for. eng. 29(2008)1

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

homogeneous humus layers. The heavy compaction occurred in the sandy zone from 17.5 to 24.5 cm in depth on the hard layer of clay occurred after eight passes. Additionally, increment of cone index of tracks loaded and unloaded was smaller than those of tires. Soil compaction occurred in the early passes. Loaded treatments caused about 200 to 400 kPa of increment of cone index in the first pass. The results show that this increment of cone index of 200 kPa equaled a decrease of 2.6% of soil porosity. As the physical mechanism of compaction by surface-to-surface movement and forming ruts was different between tires and tracks, most compacted depth of soil was different among treatments. Rut depths were not significantly affected by tire pressures but they increased significantly with the number of two and five machine passes, and soil density increased significantly with increasing number of forwarder passages (Eliasson 2005). The regression line of contact pressure and the average depth of ruts were obtained, as well as contact pressure and maximum increment of cone index after 24 passes. Furthermore, regressed line of contact pressure to depth of ruts should be used to predict the depth of ruts from large forwarders. The significant difference between loaded low and high pressure tires at the surface part might be related to the difference of the increment of cone index. Loaded low pressure tires after 24 passes showed the increment of cone index at the surface soil of 150 kPa, and the maximum increment of cone index, which was 333 kPa, occurring between14 to 17.5 cm. On the contrary, loaded high pressure tires after 24 passes showed the increment of cone index at the surface part at 500 kPa, and the maximum increment of cone index of 732 kPa occurred at the deeper part of 17.5 cm compared to low pressure tires. Results for loaded high pressure tires that had the greatest maximum increment of cone index in the deeper soil zones suggested stronger physical forces influencing the soil than other treatments. Tracks could reduce rut depth by up to 40% and cone index by about 10% compared to wide and soft tires in spite of the increased mass of tracks by 10–12% (Bygdén et al. 2003). In our experiment, maximum increment of cone index of loaded low pressure tires and tracks was small after the first few passes, and both values could have been similar if the influence of added weight of tracks with wheels was eliminated. However over eight passes, the compaction by loaded low pressure tires increased deeper in the soil layers. Maximum increment of cone index of loaded tracks was lower than that of loaded low pressure tires, and the depth of ruts by loaded tracks Croat. j. for. eng. 29(2008)1

H. SAKAI et al.

was smaller than that by loaded low pressure tires by about 20% after 24 passes. It was recognized that high pressure tires were not practical, and that tracks were the best alternative. Advanced logging machines have been shown to increase soil bulk density and decrease both the total soil porosity and the water-holding capacity of the soil (Kamaruzaman 1991). According to our results, the ruts of 10 cm in depth meant the decrease of soil porosity of 7% in the inner part. It will, therefore, be possible to predict soil compaction only by the depth of ruts. The compacted zone caused by tracks was shallow and near the surface is showed low compaction. Tracks therefore sustained the original pre-treatment porosity of the soil. This is advantageous for the growth of plants because porosity is highly correlated to the growth and vigor of plants when nourishment is sufficient, and because undersurface soil is left undisturbed and roots are active within 10 cm in depth for absorbing nutrients and water (Shishiuchi and Adachi 1982). Compaction in the deeper zone means prolonged influence on the soils. However, according to the above results, tracks will be excellent and useful for preventing soil compaction for soft and sensitive ground. It is necessary to choose the direction of forwarding to avoid repetitions of the same course resulting in heavy soil compaction that indirectly avoids erosion, and to utilize strip roads. It was observed that the surface of compacted ruts was wet, and the water accumulated as a result of decreased porosity after compaction. Once soil was compacted, it would take several years to recover its non-disturbed physical soil properties, especially organic matter (Shishiuchi and Satomura 1995). Site preparations such as scarifying after harvesting operations will be needed before plantation or for the success of natural regeneration. The results can, therefore, be used for determining and indentifying the most appropriate harvesting systems, frequency of passes, route location, and sustainable forestry. In the future, recovery term of soil porosity, and quantity of lost nutrition due to soil compaction should be further clarified.

Acknowledgement – Zahvala We would like to express our thanks for the well skilled operator of the experimented forwarder, teacher, Esben Kruse Jensen, Danish Forestry College (University of Copenhagen), and Professor Henrik Have, The Royal Veterinary and Agricultural University, who prepared a cone-penetrometer for the experiment.

H. SAKAI et al.

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

6. References – Literatura Bygdén, G., Eliasson, L., Wästerlund, I., 2003: Rut depth, soil compaction and rolling resistance when using bogie tracks. Journal of Terramechanics 40(3): 179–190. Eliasson, L., 2005: Effects of forwarder tire pressure on rut formation and soil compaction. Silva Fennica 39: 549–557. Japanese Society of Soil Mechanics and Foundation Engineering, 1969: How to Test Soil Mechanics. Japanese Society of Soil Mechanics and Foundation Engineering, Tokyo, 1–675. Kamaruzaman, J., 1991: Effect of tracked and rubber-tired logging machines on soil physical properties of the Berkelah Forest Reserve, Malaysia. Pertanika 14(3): 265–276.

lection logged over area at Tokyo University Forest in Hokkaido. Journal of the Japanese Forest Society 21: 79–82. Matangaran, J. R., Iwaoka, M., Sakai, H., Kobayashi, H., 1999: Soil compaction by a processor and a forwarder on a thinning site. Journal of the Japanese Forest Engineering Society 14: 209–212. Shishiuchi, M., Adachi, K., 1982: Influence of tractor logging on soil surface condition (I) Effect of soil compaction from tractors on the growth of planted Japanese larch seedlings. Journal of the Japanese Forest Society 64: 136–142.

Kozlowski, T. T., 1999: Soil compaction and growth of woody plants. Scandinavian Journal of Forest Research 14: 596–619.

Shishiuchi, M., Satomura, Y., 1995: Recovery of soil physical properties and growth of planted seedlings in about 10 years following soil disturbance by tractor logging. Journal of the Japanese Forest Engineering Association 10: 139–144.

Kumakura, Y., Sato, T., Sakai, H., 1993: Relationships between soil types and their hardnesses. Journal of the Japanese Forest Society 75: 235–239.

[usnjar, M., Horvat, D., [eselj, J., 2006: Soil compaction in timber skidding in winter conditions. Croat. j. for. eng. 27(1): 3–15.

Matangaran, J. R., Aruga, K., Sakurai, R., Iwaoka, M., Sakai, H., 2006: The recovery of soil compaction in the se-

Sa`etak

Zbijanje {umskoga tla razli~itim tipovima guma i polugusjenica te mogu}nost dovoljno precizne procjene Cilj je ovoga rada utvrditi razlike u zbijanju tla pri vi{ekratnim prolascima forvardera. 8-kota~ni forvarder je tijekom istraìvanja bio opremljen gumama s niskim tlakom punjenja, gumama s visokim tlakom punjenja te polugusjenicama (slika 1). Na osnovi dimenzija guma i polugusjenica te raspodjele optere}enja po osovinama forvardera izra~unati su dodirni tlakovi kota~a osovine na tlo (tablica 1). Pri kretanju optere}enoga i neoptere}noga forvardera na sje~ini biljeìli su se prolasci na ispitnim izvoznim pravcima (slika 2). Nakon 1, 8. i 24. prolaska mjereno je zbijanje tla konusnim penetrometrom i dubina kolotraga te uzimani uzorci tla u nenaru{enom stanju na dubini tla 10–15 cm radi laboratorijskoga odre|ivanja poroznosti tla, prirodne i specifi~ne gusto}e tla. Zbijanje je tla odre|eno temeljem razlike konusnoga indeksa tla na izvoznom smjeru nakon prolazaka forvardera te konusnoga indeksa neizgaènoga tla (slike 3 do 7). Optere}eni je forvarder u prvom prolasku uzrokovao pove}anje konusnoga indeksa tla od 200 do 400 kPa ovisno o primjeni razli~itih guma ili polugusjenica. Najve}e je pove}anje zbijanja tla uo~eno nakon osmoga prolaska forvardera. Primjena je guma s visokim tlakom uzrokovala ve}e zbijanje tla u dubljim slojevima. Pove}anje konusnoga indeksa tla te dubina zbijanja tla po prolascima forvardera bila je najmanja pri primjeni polugusjenica. U odnosu na primjenu guma s niskim tlakom dubina je kolotraga za 20 % manja pri primjeni polugusjenica. Na dubinu kolotraga utje~e ponajprije broj prolazaka forvardera, te je odre|ena ovisnost dubine kolotraga o pove}anju konusnoga indeksa tla i o dodirnom tlaku kota~a na tlo (slika 8 i 9). Kretanje forvardera opremljenoga gumama s visokim tlakom uzrokovalo je najve}e smanjenje poroznosti tla – za 10 %, dok su najmanje promjene u poroznosti tla uo~ene pri kretanju forvardera s polugusjenicama (slika 11 i 12). Pove}anje dodirnoga tlaka kota~a forvardera na tlo za 100 kPa uzrokuje smanjenje poroznosti tla za 5,7 % na dubini 10–15 cm.

Croat. j. for. eng. 29(2008)1

Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks ... (15–27)

H. SAKAI et al.

Nastanak se kolotraga razlikuje s obzirom na primjenu razli~itih tipova guma i polugusjenica. Ustanovljena je ovisnost poroznosti tla o dubini kolotraga te je uo~eno da se kod kolotraga dubine 10 cm poroznost tla smanjuje za 7 % (slika 13). Iz navedenoga se zaklju~uje da je jednostavnim mjerenjem dubine kolotraga mogu}e procijeniti stupanj zbijanja tla. Na osnovi tih rezultata zaklju~uje se da polugusjenice treba primjenjivati na slabonosivim tlima radi spre~avanja o{te}ivanja {umskoga tla. Klju~ne rije~i: zbijanje tla, polugusjenice, dubina kolotraga, poroznost tla

Authors’ addresses – Adresa autorâ: Prof. Hideo Sakai, PhD. e-mail: sakaih@fr.a.u-tokyo.ac.jp The University of Tokyo Graduate School of Agricultural and Life Sciences Department of Forest Sciences Bunkyo-ku Yayoi 113-8657 Tokyo JAPAN Prof. Tomas Nordfjell, PhD. e-mail: tomas.nordfjell@srh.slu.se Swedish University of Agricultural Sciences Dept. of Forest Resource Management SE-901 83 Umeå SWEDEN

Received (Primljeno): September 29, 2007 Accepted (Prihva}eno): May 12, 2008 Croat. j. for. eng. 29(2008)1

Kjell Suadicani, PhD. e-mail: kjs@life.ku.dk Bruce Talbot, PhD. e-mail: beta@life.ku.dk Ebbe Bøllehuus e-mail: ebbe-boellehuus@os.dk University of Copenhagen Faculty of Life Sciences Hørsholm Kongevej 11 2970 Hørsholm DENMARK

Original scientific paper – Izvorni znanstveni rad

Probability of Occurrence of Soil Disturbances during Timber Harvesting Janusz Sowa, Dariusz Kulak Abstract – Nacrtak This paper deals with the probability of occurrence of soil disturbance during timber harvesting operations. Studies were carried out in southern Poland in 26 stands differing in terrain, skidding method, stand age, amount of timber harvested, forest site type, stocking, and soil type. In each stand the occurrence of the following forms of disturbances was measured: soil surface disturbance, soil cover tearing off and soil compaction. Logistic regression models describing the probability of occurrence of respective disturbance forms were constructed. It was found that the probability of occurrence of soil surface disturbance depends on forest site type and stocking. Soil cover tearing off depends mainly on harvesting density, while soil compaction depends both on the harvesting system and on harvesting density. Keywords: timber harvesting, soil disturbance, probability, modeling

1. Introduction – Uvod Each case of human interference in forest environment, especially harvesting timber crops, is associated with disturbances in forest ecosystems. These disturbances may be described in many different ways but their best qualitative impression may be obtained by estimating their size and the chance of their occurrence. The size of disturbances in forest environment caused by timber harvesting, as well as factors affecting their value, are relatively well recognized. Most often the size of disturbance is associated with the configuration of terrain (Suwa³a and Rzadkowski 2001), stand age (Habert 2003, Rzadkowski 1999), the harvesting system used (Ko{ir and Robek 2000, Suwa³a 1995), and the skidding method (Messingerova 1997, Porter 1997). On the other hand, the knowledge concerning the probability of occurrence of disturbances is fragmentary (Sowa 1997). We also do not know what factors decide that in a given cutting area we may expect a definite probability of occurrence of disturbances. Studies concerning this problem are complicated by the fact that many features of each cutting area may affect the chance of occurrence of disturbances. Each logging operation is conducted in a defined terrain that includes stands of a given age (thinnings, final cuttings), and involves the use of machines working Croat. j. for. eng. 29(2008)1

with a given harvesting system. Hence it is difficult to determine to what extent each of these characteristics makes the occurrence of environmental disturbances probable. In order to fully explain this problem, it would be necessary to carry out a great number of experimental replications, in which all factors are fixed with the exception of a variable factor. In practice, it is impossible to carry out such a study. However, with a larger number of replications, and using statistical methods, it is possible to obtain a certain idea about the effect of individual characteristics of the timber harvesting process on the probability of occurrence of disturbances. Based on previous studies (Sowa 2002) it was assumed that the occurrence of disturbance in the forest environment during timber harvesting operations is not an unforeseeable random event. Therefore it is possible to develop a mathematical model describing this process. The aim of this study was to determine the probability of occurrence of soil disturbance during timber harvesting operations. The scope of the study was limited to selected forest stands in southern Poland. Attempts were made to determine characteristics of the stands and features of the timber harvesting process connected with the probability of occurrence of disturbances.

J. SOWA and D. KULAK

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29â&#x20AC;&#x201C;39)

Table 1 Characteristics of research stands Tablica 1. Zna~ajke istra`ivanih sastojina

Croat. j. for. eng. 29(2008)1

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29â&#x20AC;&#x201C;39)

J. SOWA and D. KULAK

Table 1 Characteristics of research stands (continuation) Tablica 1. Zna~ajke istra`ivanih sastojina (nastavak)

Croat. j. for. eng. 29(2008)1

J. SOWA and D. KULAK

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

2. Material and methods – Materijal i metode The research was conducted in southern Poland in mountainous (Jeleœnia Forest District and Forest Experimental Station in Krynica), upland (Gromnik Forest District), and lowland (Krzeszowice Forest District) conditions. Sample plots were established in stands where the following categories of timber utilization were conducted: early thinnings, late thinnings, and final cuttings. The investigated harvesting systems were as follows: tree-length, half-tree-length, cut-to-length system. Skidding was performed by horses, farm tractors and skidders. Chain saws were used for tree felling, delimbing, and bucking. Three stands were chosen for each category of timber harvesting. Such an arrangement was repeated in three forms of terrain configuration (the exception are sample plots in mature stands in lowlands, represented by 2 stands). Thus, the experiment was carried out in 26 different stands in total. The description of their basic characteristics is presented in Table 1. Before the beginning of harvesting, on a 2-hectare working area in each stand, 16 circular sample plots of 1-are, set up on a network of rectangles (25 ´ 50 m),

were established. The stands where the present research was located were hardly accessible. Except division roads – 1st class trails – the stands had no lower class trails. The arrangement of sample plots in relation to 1st class trails is presented in Fig. 1. The centers of sample plots were marked with PVC poles and the plot number was painted on the nearest tree. Taking into account studies concerning soil disturbances by other authors (Dyrness 1965, Porter 1997, Sowa 2002, Wästerlund 1990, Wästerlund 1992) the following classification was assumed: Þ soil surface disturbance, i.e. translocation of litter and humus and exposure of mineral soil without its disturbance, most often caused by uncontrolled shifting or turning of the timber load; Þ soil cover tearing off, i.e. exposure of deeper layers and disturbing the mineral soil, most often in the form of furrows made by larger ends of skidded timber; Þ soil compaction, i.e. distinct marks of soil compaction made during skidding operation by a vehicle or a timber load. Directly after felling, the circular sample plots were found and assessment of soil disturbances was performed. If even one soil disturbance of the above-mentioned types was recorded, then – in the course of calculating the probability of occurrence of damage in the analysed stand – a given circular sample plot was classified as disturbed, regardless of the size of this disturbance. Each single disturbance of the soil cover, recorded in field, was ascribed only to one of the above-mentioned forms of soil disturbances. Disturbances of various types were often recorded on a single sample plot – such a sample plot was taken into consideration in the calculation of the probability of all forms of soil disturbances found on it. The data obtained during this study were analyzed statistically using the computer program STATISTICA 6.0 PL (StatSoft, Inc. 2004). All tests were performed at the significance level a = 0.05.

3. Results and discussion – Rezultati s raspravom

Fig. 1 Study design for single plot Slika 1. Postavljanje pokusnih ploha 32

The probability of occurrence of a random event, and soil disturbance is such an event, is one of its characteristics. To calculate this probability, individual circular sample plots were coded either (1) when a disturbance occurred, or (0) when there was no disturbance. For each working area (stand), the probability of occurrence of soil disturbances was computed by determining the percentage of circular sample Croat. j. for. eng. 29(2008)1

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

Table 2 The probability of occurrence of soil disturbances in research stands Tablica 2. Vjerojatnost pojave o{te}enja tla u istra`ivanim sastojinama Probability of occurrence of respective soil disturbances, % Vjerojatnost pojave odre|enoga o{te}enja tla, % Stand number Soil surface Soil cover Broj sastojine disturbance tearing off Soil compaction Povr{insko Premje{tanje Zbijanje tla o{te}enje tla slojeva tla 1 64 14 57 2 33 28 57 3 6 40 26 4 12 43 56 5 90 50 90 6 12 6 25 7 42 71 28 8 18 18 43 9 37 56 18 10 0 25 18 11 0 63 0 12 0 42 68 13 0 31 6 14 15 61 7 15 15 15 0 16 4 46 39 17 0 61 23 18 0 55 22 19 42 42 78 20 8 8 33 21 0 37 50 22 6 25 31 23 12 75 62 24 11 66 77 25 6 12 93 26 12 43 18 Total – Ukupno 14 37 37

plots with disturbances (code 1). The results are shown in Table 2. Analysis of the data in Table 2 showed that the occurrence of soil surface disturbances was statistically least probable. In total, in the entire experiment, surface disturbances were found in about 15% of circular sample plots. The tearing off and compacting of soil was equally probable – about 40%. The results of research by other authors, concerning the probability of occurrence of soil damage in the Croat. j. for. eng. 29(2008)1

J. SOWA and D. KULAK

course of timber harvesting indicate a large differentiation of this index depending on stand conditions in which felling is performed and on the technologies applied. For example, research done in British Columbia shows that the probability of occurrence of deep disturbances and compacting in the soil of the forests of New Zealand is relatively low – depending on the sampling method, the chance of their occurrence was described as 4–7% and 35–50%, respectively. In European conditions (France), during final cutting, much higher probability of soil disturbance was recorded, depending on the skidding means. In skidding by skidder, the probability of disturbance was 87% and by forwarder it raged from 65 to 86% (Gondard et al. 2003). Research on the probability of soil disturbance during harvesting has been conducted in Poland, too. Gil et al. (1987) reported that the probability of occurrence of soil disturbances in a mountain stand designated for final cutting (group cutting method) using mechanical skidding was 59–69%, while that for soil compaction was 19– 44%. The results obtained have confirmed the opinion of Giefing (1990) that the chance of occurrence of disturbance caused by timber harvesting increases as the stand age increases. It was found that the probability of occurrence of soil disturbances was on the average 68% during final cuttings, 65% during late thinnings, and 52% during early thinnings. In order to assess the usefulness of individual characteristics of stands, types of operations, and conditions occurring in circular sample plots for the estimation of the risk of soil disturbance occurrence, the correlation between these features and the occurrence of disturbance (zero-one codes) was calculated and expressed by Spearman’s coefficient of correlation R. It was assumed that the following characteristics (factors) may affect the probability of occurrence of soil disturbances: Þ A) configuration of terrain (mountains, uplands, lowlands); Þ B) category of timber utilization (final cutting, late thinning, early thinning); Þ C) harvesting system (tree-length, half-treelength, cut-to-length); Þ D) skidding means (horse, farm tractor, skidder); Þ E) harvesting density of circular sample plot (m3/are); Þ F) terrain slope in a circular sample plot (°); Þ G) harvesting density of entire cutting area (m3/ha); Þ H) forest site type;

J. SOWA and D. KULAK

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

Þ I) soil type and subtype (typical brown soil, acid brown soil, typical podzolic soil, podzolic gley soil); Þ J) soil texture (heavy loam, medium loam, sandy loam, coarse sand, compact sand, loose sand); Þ K) stocking; Þ L) crown closure (full, moderate, open); Þ M) stand age (years); Þ N) growing stock (m3/ha).

It may be observed that the correlation between the probability of occurrence of soil disturbances and the selected characteristics (factors) was not strong. The maximum value of the correlation coefficient R was 0.26, and this correlation was considered to be weak (Stanisz 1998). Thus, the correlation did not adequately explain what characteristics affected the probability of occurrence of soil disturbance. The individual characteristics are known, but perhaps their integration into one model would permit to determine which one of them, and to what degree, affects the occurrence of disturbance. The probability of occurrence of soil disturbance ranged from 0 to 1,

The results are presented in Table 3.

Table 3 Strength and significance of correlation between the probability of occurrence of analyzed forms of soil disturbances and the selected characteristics of stands Tablica 3. Jakost veze i zna~ajnost korelacije izme|u vjerojatnosti pojave promatranih oblika o{te}enja tla i odabranih zna~ajki sastojina Correlated characteristics Korelirane zna~ajke

Soil surface disturbance Povr{insko o{te}enje tla

Soil cover tearing off Premje{tanje slojeva tla

Soil compaction Zbijanje tla

All disturbance forms Svi oblici o{te}enja tla

Terrain configuration Konfiguracija terena

0.24

0.00

0.02

0.65

0.05

0.29

0.02

0.74

Timber utilisation category Vrsta prihoda

0.04

0.46

0.02

0.62

0.02

0.62

0.01

0.82

Harvesting system Metoda izradbe drva

0.03

0.49

0.08

0.24

0.00

0.02

0.61

Skidding means Sredstvo privla~enja

0.23

0.00

0.03

0.55

0.09

0.07

0.05

0.33

Harvesting density of circular plot Sje~na gusto}a pokusne plohe

0.11

0.02

0.13

0.01

0.22

0.00

0.27

0.00

Terrain slope Nagib terena

0.06

0.21

0.06

0.23

0.15

0.00

0.09

0.08

Harvesting density of cutting area Sje~na gusto}a sje~ine

0.13

0.01

0.14

0.00

0.10

0.04

0.21

0.00

Forest site type Tip sastojine

0.26

0.00

0.03

0.56

0.04

0.43

0.02

0.61

Soil type and subtype Tip i podtip tla

0.25

0.00

0.01

0.85

0.03

0.51

0.05

0.29

Soil texture Tekstura tla

0.23

0.00

0.05

0.26

0.10

0.03

0.04

0.41

Stocking Obrast

0.19

0.00

0.03

0.60

0.13

0.01

0.07

0.12

Crown closure Sklop sastojine

0.00

0.99

0.04

0.36

0.02

0.73

0.03

0.59

Stand age Dob sastojine

0.06

0.19

0.06

0.19

0.06

0.22

0.04

0.42

Growing stock Drvna zaliha

0.20

0.00

0.97

0.07

0.14

0.13

0.01

Croat. j. for. eng. 29(2008)1

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

J. SOWA and D. KULAK

Fig. 2 Regression model of the probability of occurrence of soil surface disturbance depending on forest type Slika 2. Regresijski model vjerojatnosti pojave povr{inskoga o{te}enja tla ovisno o tipu sastojine and in individual circular sample plots the soil disturbances were dichotomic, they either occurred or not. Therefore, it was assumed that by using a model of logistic regression it would be possible to describe most fully the effect of individual factors on the probability of soil disturbances. Models were elaborated separately for each soil disturbance form, and the consecutive factors were introduced to the model beginning with those characterized by the highest values of Spearman’s coefficient of correlation R. Currently the effect of characteristics being added on quality of the model were analyzed. The use of this method of model development prevented their basing on random variables mutually correlated with one another e.g. stand age and the category of timber harvesting. At the same time this permitted to choose the variable which would contribute more to the model. According to the assumption made, attempts were made to build the model determining the relationship between the forest site type and the probability of occurrence of soil surface disturbances. For this purpose a quasi-Newton method of estimation was used (Stanisz 2000). The model obtained and codes of forest site types are shown in Fig. 2. The chi-squared test showed statistical significance of the model (c2 = 35.02, p = 0.00). The percentage of circular sample plots for which, by using the model, it was possible to predict correctly the occurrence of damage was computed, assuming that all circular sample plots with the model probability of occurrence of soil damage smaller or equal to 0.5 will Croat. j. for. eng. 29(2008)1

be treated as 0 (a lack of damage) and those with the probability greater than 0.5 as 1 (damage occurred). The parameter calculated – and called the odds ratio – replaces the coefficient of the correlation R2, which is a measure of adjusting the type model that does not occur in logistic regression. The model presented above had simultaneously three parameters: terrain conditions – forests of mountain, upland, and lowland sites. The probability of occurrence of surface disturbances increased in the same order. It was verified whether basing of a logistic model on terrain conditions only (mountains, uplands, and lowlands) would give the same results as in the case of using forest site types. However, the model obtained, although a significant one, was characterized by only 48% efficiency in estimating the probability of occurrence of soil surface disturbance. According to the assumed model development method, the following random variables were successively introduced into the model (Fig. 2): soil type and subtype, terrain configuration, skidding means, soil texture, and growing stock. However, these operations did not significantly improve the model, i.e. the test c2 did not show differences between new models and the model shown in Fig. 2. Only the use of stand stocking resulted in model improvement (c2 = 314.06, p = 0.00) (Fig. 3). This model was statistically significant (c2 = 52.76, p = 0.00). The model was very efficient because it correctly estimated the probability of occurrence of surface disturbances in 87%. The addition of new vari-

J. SOWA and D. KULAK

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29â&#x20AC;&#x201C;39)

Fig. 3 Regression model of probability of occurrence of soil surface disturbance depending on forest type and stocking Slika 3. Regresijski model vjerojatnosti pojave povr{inskoga o{te}enja tla ovisno o tipu i obrastu sastojine

rence of soil disturbance in mountain and upland stands. In the case of lowland stands its effect was insignificant. The results of this study showed that the probability of occurrence of soil surface disturbances in nearly 90% depended on stand factors on which manâ&#x20AC;&#x2122;s activities had no direct influence. This, however, does not mean that the forest manager and the firm conducting timber harvesting are free of responsibility for the condition of the forest because the level of disturbances is greatly affected by their size. And this, as shown, is also connected with the harvesting system selected. The data in Table 3 show that the harvesting density had the greatest effect on the occurrence of soil cover tearing off. However, the model using this random variable turned out to be statistically insignificant (c2 = 5.75, p = 0.09). On the other hand, the attempt to subject the probability of occurrence of soil cover tearing off to harvesting density in a circular sample plot permitted to obtain a significant model (c2 = 7.75, p = 0.01) (Fig. 4). The model estimated correctly the probability of occurrence of soil tearing off in 62% of cases (odds ratio). The graph (Fig. 4) shows that the performance of harvesting operations, even those not too intensive, resulted in over 30% probability of occurrence of disturbances in the form of soil cover tearing off. However, the extrapolation of values of the logistic function for greater volumes of harvested timber

Fig. 4 Regression model of the probability of occurrence of soil cover tearing off depending on harvesting density Slika 4. Regresijski model vjerojatnosti pojave premje{tanja slojeva tla ovisno o sje~noj gusto}i ables, i.e. growing stock and harvesting density in a circular sample plot, did not result in any statistically significant difference. When analyzing the model in Fig. 3, it may be noticed that the factor of stocking had the greatest effect on the probability of occur-

Fig. 5 Regression model of the probability of occurrence of soil compaction depending on harvesting system Slika 5. Regresijski model vjerojatnosti pojave zbijanja tla ovisno o metodi izradbe drva Croat. j. for. eng. 29(2008)1

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

showed that a 100% certainty of the occurrence of soil cover tearing off is attained not sooner than when intensity of harvesting is 15 m3/ha, and this is unattainable in practice. The attempt to explain the occurrence of soil compaction caused by the harvesting system used resulted in obtaining a logistic model shown in Fig. 5. This model was statistically significant (c2 = 289.62, p = 0.04). It permitted to determine correctly 62% of the probability of occurrence of soil compaction. Its analysis showed that soil compaction most often occurred during timber harvesting in the tree-length system. The occurrence of this kind of soil disturbance was less probable in the half-tree-length system, and the least probable in the cut-to-length system. However, the differences between individual systems in this respect were not large (up to 10%). Development of the model by addition of harvesting density in a circular sample plot (Fig. 6) resulted in a significant (c2 = 399.24, p = 0.00) increase in the efficiency of forecasting the occurrence of soil compaction to the level of 68%. This model was significantly different from the model with only free term (c2 = 17.24, p = 0.00). The graph presenting the logistic function (Fig. 6) shows that in spite of a significant effect of the harvesting system on the probability of occurrence of soil compaction, the harvesting density had a greater effect on the occurrence of this form of disturbance.

J. SOWA and D. KULAK

4. Conclusions – Zaklju~ci On the basis of research results, the following conclusions should be pointed out: Þ The probability of occurrence of soil disturbance during timber harvesting operations is a highly foreseeable characteristic. Þ Mathematical models developed during this study permit to determine the probability of occurrence of a definite form of soil disturbance as well as to determine characteristics of stands and operations affecting the value of this probability. Þ The probability of occurrence of soil surface disturbances significantly depends on the forest site type and stand stocking. The latter factor is of greater importance, and it is inversely proportional to the value of the probability. Þ Soil surface tearing off may be deduced in over 62% from the harvesting density. Þ The harvesting system and harvesting density significantly affect the probability of occurrence of soil compaction. Þ The probability of occurrence of soil compaction depends on the length of skidded timber. The highest one was found when the tree-length system was used. Under the half-tree-length system the probability was smaller, and under the cut-to-length system its value was the smallest. Þ The probability of occurrence of any of the analyzed forms of soil disturbance is directly proportional to the harvesting density.

5. References – Literatura Dyrness, C. T., 1965: Soil surface condition following tractor and high-lead logging in the Oregon Cascades. Jour. of For. 63: 272–275. Giefing, D. F., 1990: Wp³yw pozyskania i transportu drewna w ciêciach przedrêbnych na szkody i jakoœ} techniczn¹ pozostaj¹cych drzewostanów sosnowych. In: Reakcja ekosystemów leœnych i ich elementów sk³adowych na antropopresjê, Wydawnictwo SGGW – AR, Warszawa, Poland, p 77–85. Gil, W., Sosnowski, J., Stanibu³a, S., 1987: Szkody wyrz¹dzane przy zrywce drewna przez ci¹gniki Tree Farmer C-5D oraz Ursus C-328. Zeszyty Naukowe AR w Krakowie, 215, p 107–119.

Fig. 6 Regression model of the probability of occurrence of soil compaction depending on harvesting system and harvesting density Slika 6. Regresijski model vjerojatnosti pojave zbijanja tla ovisno o metodi izradbe drva i sje~noj gusto}i Croat. j. for. eng. 29(2008)1

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: 165–174.

J. SOWA and D. KULAK

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

Haberle, A., 2003: Aufarbeitungskonzepte für Steillagen. AFZ Wald. 4: 164–166. Hope, G., 2007: Changes in soil properties, tree growth, and nutrition over a period of 10 years after stump removal and scarification on moderately coarse soils in interior British Columbia, Forest Ecology and Management 242: 625–635. Ko{ir, B., Robek, R., 2000: Characteristics of the stand and soil damage in cut-to-lenght thinning on the @ekanc working site (SW Slovenia). Zb. gozd. lesar. 62: 87–115. McMahon, S., 1995: Accuracy of two ground disturbance methods for assessing site disturbance. Journal of Forest Engineering 6(2): 27–34. Messingerova, V., 1997: Skody na ostavajucom poraste a podnom povrchu po sustredovani dreva vo flysovej oblasti. Acta Fac. For. 39: 205–216. Porter, B., 1997: Techniczne, ekonomiczne i przyrodnicze aspekty zrywki drewna w sosnowych drzewostanach przedrêbnych. Wyd. Fundacja Rozwój SGGW, p 79. Warszawa. Rzadkowski, S., 1999: Pozyskiwanie drewna metod¹ sortymentow¹ w górskich drzewostanach œwierkowych. In: Tendencje i problemy mechanizacji prac leœnych w warunkach leœnictwa wielofunkcyjnego. AR Poznañ, p 132–136. Sowa, J. M., 1997: Podstawy metodyczne modelu szacowania szkód pozyskaniowych w œrodowisku leœnym. In: Materia³y i dokumenty Kongresu leœników Polskich. Warszawa. Tom 2: 130–135.

Sowa, J. M., 2002: Empiryczna weryfikacja teoretycznego modelu szacowania szkód œrodowiskowych od pozyskania drewna w wybranych jednostkach gospodarczych Lasów Pañstwowych. Sprawozdanie z badañ – projekt badawczy P06H 00216/KBN, p 127. Kraków. StatSoft, Inc., 2004: STATISTICA (data analysis software system), version 6. www.statsoft.com. Stanisz, A., 1998: Przystêpny kurs statystyki w oparciu o program STATISTICA PL na przyk³adach z medycyny. StatSoft Polska, p 362, Kraków. Stanisz, A., 2000: Przystêpny kurs statystyki w oparciu o program STATISTICA PL na przyk³adach z medycyny. Tom II. StatSoft Polska, p 407, Kraków. Suwa³a, M., Rzadkowski, S., 2001: Wydajnoœ} pracy, koszty i uszkodzenia drzew przy pozyskiwaniu drewna w trzebie¿ach drzewostanów górskich. Prace IBL, seria A. 911: 85–111. Suwa³a, M., 1995: Wp³yw wybranych metod i œrodków pozyskiwania drewna na uszkodzenia nadziemnych czêœci drzew oraz powierzchniowych warstw gleby w póŸnych trzebie¿ach drzewostanów sosnowych. Prace IBL, seria A. 786: 82, Warszawa. Wästerlund, I., 1990: Soil strength in forestry measurend with a new kind of test ring. Proc. 10th Intern. Conference of the ISTVS. Kobe, Japan, p 73–82. Wästerlund, I., 1992: Extent and causes of site damage due to forestry traffic. Cand. Jour. For. Res. 7: 135–142.

Sa`etak

Vjerojatnost pojave o{te}enja tla pri pridobivanju drva Cilj je rada odre|ivanje vjerojatnosti pojave o{te}enja tla pri pridobivanju drva. Istraìvanje je provedeno u 26 {umskih sastojina u planinskom, brdskom i nizinskom podru~ju. [umske su sastojine razvrstane po vrsti prihoda na rane prorede, kasne prorede i dovr{ni sijek. Na svakom podru~ju izabrane su po tri sastojine iste vrste prihoda, osim u nizinskom podru~ju gdje je dovr{ni sijek zastupljen samo u dvije sastojine. Stabla su se sjekla i izra|ivala motornim pilama uz primjenu deblovne, poludeblovne i sortimentne metode izradbe. Drvo se privla~ilo konjima, poljoprivrednim traktorima ili skiderima. Zna~ajke su istraìvanih {umskih sastojina te primijenjenih metoda izradbe drva i privla~enja drva prikazane u tablici 1. Prije izvo|enja radova u svakoj je sastojini na povr{ini od 2 ha postavljeno 16 pokusnih ploha kru`noga oblika povr{ine 1 ar u prostornom rasporedu 25 x 50 m (slika 1). Nakon sje~e i izrade stabala te privla~enja drva na pokusnim su plohama zabiljeèna o{te}enja tla koja su razvrstana u 3 oblika: Þ povr{insko o{te}enje tla – skidanje sloja humusa i listinca zbog pomicanja i okretanja tovara Þ premje{tanje slojeva tla – o{te}enje tla u obliku brazdi zbog guranja tla debljim krajem drvnih sortimenta pri privitlavanju Þ zbijenost tla – izrazito o{te}enje tla pri privla~enju drva zbog prolaska vozila i tovara. Pokusne su plohe razvrstane po vrsti o{te}enja tla. Vrlo je ~esto na istoj plohi zabiljeèno vi{e oblika o{te}enja tla te se ta pokusna ploha razmatrala u prora~unu vjerojatnosti svih zabiljeènih oblika o{te}enja tla.

Croat. j. for. eng. 29(2008)1

Probability of Occurrence of Soil Disturbances during Timber Harvesting (29–39)

J. SOWA and D. KULAK

Podaci su mjerenja statisti~ki obra|eni u ra~unalnom programu STATISTICA 6.0 PL. U prora~unu vjerojatnosti pokusna ploha s pojavom o{te}enja tla ozna~ena je s 1, a pokusna ploha bez o{te}enja tla s 0. Za svaku {umsku sastojinu odre|ena je vjerojatnost pojave odre|enoga oblika o{te}enja tla temeljem postotnoga odnosa o{te}enja tla u pokusnim plohama (tablica 2). Pretpostavka je da ove zna~ajke mogu utjecati na pojavu o{te}enja tla: konfigurcija terena, vrsta prihoda, metoda izradbe drva, sredstvo privla~enja drva, obujam posje~enoga drva na pokusnoj plohi, nagib terena na pokusnoj plohi, sje~na gusto}a, tip sastojine, vrsta tla, tekstura tla, obrast, dob sastojine i drvna zaliha. Rezultati korelacija izme|u navedenih zna~ajki i pojave o{te}enja tla pokazuju slabu povezanost podataka (tablica 3). U daljnjoj analizi podataka kori{tena je logisti~ka regresija s obzirom na to da se pojavnost o{te}enja tla na pokusnim plohama vrednovala s 0 (bez o{te}enja) i 1 (s o{te}enjem). Logisti~kom regresijom odre|eni su modeli za svaki oblik o{te}enja tla te c2 testom utvrdila statisti~ka zna~ajnost modela. Na slici 2 prikazan je model vjerojatnosti pojave o{te}enja tla u ovisnosti o tipu sastojine te su opisani kodovi tipova sastojine. Prikazani model prikazuje da je vjerojatnost pojave o{te}enja najmanja u planinskim uvjetima, a najve}a u nizinskim uvjetima. Uvr{tavanjem ostalih varijabli u model nije dobiveno statisti~ki zna~ajno pobolj{anje modela. Obrast je sastojine jedina nezavisna varijabla koja je utjecala na pobolj{anje modela. Obrast sastojine ima najve}i utjecaj na pove}anje vjerojatnosti pojave o{te}enja tla u planinskim i brdskim uvjetima, dok je njegov utjecaj nezna~ajan u nizinskim uvjetima (slika 3). Rezultati istra`ivanja pokazuju da vjerojatnost pojave o{te}enja tla zna~ajno ovisi o stani{nim uvjetima, ponajprije o tipu sastojine i obrastu. Statisti~ki zna~ajnim pokazao se model procjene vjerojatnosti pojave premje{tanja slojeva i tla u ovisnosti o obujmu posje~enoga drva na pokusnoj plohi (slika 4). Model u 62 % slu~ajeva ispravno procjenjuje vjerojatnost pojave premje{tanja slojeva i tla. Logisti~ki model procjene pojave zbijanja tla izra|en je u ovisnosti o metodi izradbe drva (slika 5) te je uo~eno da je pri primjeni deblovne metode naj~e{}a pojava zbijanja tla. Vjerojatnost je pojave zbijanja tla najmanja pri primjeni sortimentne metode. No, razlike u vjerojatnosti pojave zbijanja tla izme|u tri promatrane metode izradbe drva nisu velike (do 10 %). Logisti~ki model prikazan na slici 6 pokazuje da unato~ zna~ajnu utjecaju primjene odre|ene metode izradbe drva najve}i utjecaj na vjerojatnost pojave zbijanja tla ima obujam posje~enoga drva. Op}enito, obujam posje~enoga drva ima najve}u utjecaj na sve oblike o{te}enja tla. Klju~ne rije~i: pridobivanje drva, o{te}enje tla, vjerojatnost, modeliranje

Authors’ address – Adresa autorâ:

Received (Primljeno): January 11, 2007 Accepted (Prihva}eno): May 12, 2008 Croat. j. for. eng. 29(2008)1

Prof. Janusz Michal Sowa, PhD. e-mail: rlsowa@cyf-kr.edu.pl Dariusz Kulak, PhD. e-mail: rlkulak@cyf-kr.edu.pl Agricultural University of Cracow Faculty of Forestry Department of Forest and Wood Utilization Al. 29 Listopada 46 31–425 Kraków POLAND

Orginal scientific paper – Izvorni znanstveni rad

Morphological Analysis of Forest Tractor Assemblies Marijan [u{njar, Dubravko Horvat, Andrija Kristi}, Zdravko Pandur Abstract – Nacrtak The results of this paper present the morphological analysis of nine different types of tractor assemblies used in forestry practice in timber forwarding from thinning operations of lowland forests. Among these tractor assemblies 4 types are older, equipped mechanical cranes. The remaining 5 tractor assemblies, manufactured more recently, are equipped with hydraulic cranes, and two of them are additionally equipped with double-drum winches. According to the research results, older tractor assemblies have very favourable morphological characteristics, according to which they are classified as ecologically acceptable vehicles for the timber forwarding from thinning operations in lowland forests. However, work technology requires from the vehicle to reach each timber assortment processed in the stand, which also means higher possibility of damage to forest soil of poor bearing strength in lowland areas. Technical design of more recent types of tractor assemblies ensures environmental suitability in timber forwarding from thinning operations. The installation of an articulated joint on the semi-trailer shaft ensures small turning diameter and hence lower risk of causing damage to standing trees. The adequate breadth of tractor assemblies is achieved by use of narrower tyres on rear tractor wheels. Installation of the forest winch on the tractor assembly causes lower soil tracking and lower damage to soil and standing trees. They also exert imaginary pressure on the soil between 2 kPa and 4 kPa and hence they are more adequate for the work on forest soils of poor bearing strength than forwarders, as the lowest imaginary pressure exerted by forwarders is around 4 kPa. Keywords: tractor assembly, morphological analysis, timber forwarding, ecological suitability, thinnings

1. Introduction – Uvod Timber harvesting from thinning operations in lowland forests of the east part of Croatia is very significant due to considerable quantities of wood – around 50% of the total annual allowable cut, ecological sensitivity and economic efficiency. Naturally regenerated forests in the east part of Croatia are considered the most valuable forests in the country, but also ecologically the most sensitive. These forests developed on deep pseudoglay type of soils characterised by poor load-carrying capacity and high water content. These forests are operated and maintained on the principle of sustainable management, and hence problems arise in carrying out wood production operations. Due to the above mentioned characteristics of forest soil, timber is mostly forwarded so as to minimise the damages to soil. During winter, the extraction of timber from the Croat. j. for. eng. 29(2008)1

main felling is carried out by forwarders. However, forwarders are not suitable for extracting timber from thinning operations during vegetation periods. Due to their large mass (also including load mass), they exert large contact pressure on the soil, which is of poor bearing strength in that period, so that serious damage is caused. Based on the experience of forestry experts, and along with a generally accepted standpoint on the need of timber forwarding from lowland forests since the beginning of the mechanisation of timber extraction operations, tractor assemblies have been used in thinning operations of the said forests. Tractor assembly means an adapted farming tractor with forest semi-trailer and installed crane. The advantage of tractor assemblies in timber forwarding from thinning operations of lowland forests lies in the vehicle mass and lower contact pressure on forest soil,

M. [U[NJAR et al.

Morphological Analysis of Forest Tractor Assemblies (41–51)

which results in lower damage to forest soil and the remaining standing trees and young trees. The tractor assembly is also less costing than forwarder, which affects the reduction of cost per product unit. The application of forest machines in thinning operations in Croatia started in early 70s of the last century. After the use of the first assembly in 1968, the production and use of the domestic tractor assembly started in 1972. The first tractor assembly of that kind was the so-called Pionir (Fig. 1) which was equipped with mechanical crane and mechanically driven winch. In mid eighties the number of tractor assemblies Pionir reached its maximum of 70 pieces. After 25 years this assembly is still in use, primarily because of its extremely simple design and consequently also a relatively low purchase price, as well as low operating and maintenance costs (Beuk et al. 2007). Ecological issues demanded winching of processed assortments (logs) from the stand (felling site) to one of the parallel strip roads (laid out at a 75 m distance between them). Due to technical deficiencies of the Pionir assembly, it was difficult to carry out such operations, so that they used to enter into the stand until the stump, by which both ecological and economic features were disturbed. Further improvements were made by providing better forest opening with a 37.5 m distance between parallel strip roads. On the other hand these tractor assemblies are technically and technologically out-of-date and hence they cannot meet many modern criteria related to ergonomic and safety requirements (driver’s cabin and seat, need for frequent rising to the loading area, etc.). Since the 90s, different designs of tractor assemblies have been used. Mechanical cranes on the tractor assembly have been replaced with hydraulic cranes, which enabled lifting of heavy timber assortments

and an ergonomically more favourable hydraulic steering, later replaced with electro-hydraulic steering. Based on practice in using tractor assemblies Horvat et al. (2004A) have made the following recommendations related to the basic technical characteristics of the tractor assembly: semi-trailer loading capacity – 6 t, tractor power – approximately 60 kW, hydraulic crane of net lifting capacity – more than 40 kNm, the same tractor and semi-trailer track width – less than 1.7 m, total length of the assembly – up to 9 m, clearance – larger than 300 mm, double-drum winch with tractive force of more than 50 kN and reduction of the turning diameter with the use of the articulated joint shaft or movable bogie wheels. The latest design of tractor assembly from 2004 is based on the above recommendations. The so-called new tractor assembly Formet consists of the following elements: farming tractor Steyr 8090 with narrow track, double-drum winch Igland 6002 Pronto TL, hydraulic crane Igland 43–65, semi-trailer Metalac S-6 with loading capacity of 6 t (Fig. 2). Nowadays farming tractors Belarus 920 and Belarus 952 are often used instead of the tractor Steyr 8090. The most significant technical and exploitation characteristics of the new tractor assembly Formet lie in its independent hydraulic crane of good lifting capacity, bigger cabin with turning seat, double-drum winch with good tractive force, higher semi-trailer loading capacity as well as the tractor of a modern design in terms of better technical and safety features. Additional equipping of tractor assemblies with forest winch enabled them to move exclusively on parallel trails, from which winching of timber assortments was carried out and then loading by crane,

Fig. 1 Tractor assembly Pionir Slika 1. Traktorski skup Pionir

Fig. 2 Tractor assembly Formet Slika 2. Traktorski skup Formet

Croat. j. for. eng. 29(2008)1

Morphological Analysis of Forest Tractor Assemblies (41–51)

without having to enter into the stand and approach each processed assortment at the distance of the crane reach. When the tractor assembly moves exclusively on forest roads and trails the possibility is reduced of causing damage to forest soil (soil compaction, formation of wheel rut) by passage of the loaded vehicle, especially in conditions of its poor bearing strength. However, a considerable part of felled trees is not in the reach of the crane and then winching is required. Wästerlund (1994) emphasises that winching within a stand, where thinning operations are carried out, may cause extremely serious damage to the remaining trees in the stand. During winching, relatively high resistances can occur on the soil, depending on winching coefficients. Horvat et al. (2005) determined that the winching coefficient depends on the weight, form and position of logs in the felling site, and that higher resistances were recorded in skidding with the larger end turned forward. When winching logs, as opposed to for example timber skidding, where one end of the load is elevated on the rear part of the skidder, the log is dragged on the soil in full weight. During such operation higher resistances occur, and the negative effect is enhanced by ploughing effect of the front end of the log. When the tractor assembly is engaged in thinning stands, lower ground tractive resistance of wood assortments from the felling site to the skid trail can be achieved by directional felling. In this way, the assortments would be directed with their thinner end toward the skid trail already in the phase of felling and the winching distance would also be reduced. Directional felling of trees has also some ecological advantages, because it causes less damage to forest soil, and economic advantages as higher productivity is achieved by the tractor assembly due to lower winching distances. The chosen technology of timber forwarding from thinning operations in lowland fellings, which is represented in the use of tractor assemblies equipped with semi-trailer, double-drum winch and hydraulic forest crane, is acceptable from both the theoretic/scientific and professional point of view, and it is in accordance with the ecological requirements on soil and stand protection.

2. Research methods – Metode istra`ivanja The aim of this paper is to carry out the morphological analysis of different types of forest tractor assemblies and, based on the obtained results, to indicate the specifics of their technical and technological characteristics with respect to forwarders, and ecoCroat. j. for. eng. 29(2008)1

M. [U[NJAR et al.

logical suitability of tractor assemblies for performing the operations of timber forwarding from thinnings in lowland forests. Morphological analysis is used for determining the current state, features and patterns, as well as possible development trend of forestry machines. Based on the selected geometric, mass and other values, dependencies are expressed and assessment is made of the machine choice. The results of the performed analyses are used by forestry experts in choosing new machines, selection of the machine for the most favourable use in different work conditions, for defining framework parameters in designing new machines within known families (groups). This method was introduced by Bekker in 1956 primarily with the aim of assessing the advantages of vehicles moving off road. He stated that the ratio of the vehicle geometric indicators, and especially the so-called factor of the vehicle load-bearing surface, determine the vehicles manoeuvrability on soft soils (Horvat et al. 2004B). Horvat and Kristi} (1999) outlay the first morphological analysis of thinning tractor assemblies (tractor with semi-trailer and hydraulic crane) as the starting point in seeking the optimum solution for lowland forests. Morphological analysis was carried out so that length, breadth, track width, height, clearance, loading capacity, maximum crane reach and maximum lifting moment were observed with respect to semi-trailer mass. The research was carried out by measurement of the basic dimension and mass characteristics on 9 different types of tractor assemblies. The method of morphological analysis was used for the determination of dependences between the selected characteristics of tractor assemblies: Þ mass of tractor assembly, tractor, forest semitrailer with hydraulic crane, Þ length, breadth and height of tractor assembly, Þ turning diameter of tractor assembly, Þ imaginary pressure on soil. The measurement of mass of tractor, tractor semitrailer and the whole tractor assembly was carried out with four scales of the Swedish manufacturer TELUB. All scales are connected to a measuring amplifier HBM Spider 8, which is directly connected to a laptop. The results of measurement were read from each scale by use of the software programme Catman 4.0. In order to measure the mass accurately, all wheels of the tractor assembly have to be positioned horizontally. This is why the four scales were first placed under the tractor wheels, and the semi-trailer wheels on wooden supports of the scales’ height. Then the scales were placed under the semi-trailer wheels,

M. [U[NJAR et al.

Fig. 3 Measuring of tractor mass Slika 3. Mjerenje mase traktora

Fig. 4 Measuring of semi-trailer mass Slika 4. Mjerenje mase poluprikolice and wooden supports under the tractor wheels. The sum of all readings represents the total mass of the unloaded tractor assembly. The sum of masses on the scales under the semi-trailer wheels represents the semi-trailer mass (Fig. 4). The forest semi-trailer is connected to the tractor by a shaft and hence part of the semi-trailer weight is transferred to the rear wheels of the tractor. In order to measure accurately the weight of the tractor alone, it is necessary to lift the semi-trailer shaft with a manual hydraulic crane (Fig.3). During shaft lifting, the scales readings change up to the moment when the semi-trailer stops to exert load on the rear part of the tractor and only then the scales readings show a stable mass value, which is also the tractor mass. The measurement tape was used for determining the dimensions of the tractor assembly in accordance with the following definitions:

Morphological Analysis of Forest Tractor Assemblies (41–51)

Þ length – horizontal distance from the vertical surface touching the most remote point of the front part of the vehicle (front part of the tractor) to the vertical surface touching the most remote point of the rear part of the vehicle (rear part of the semi-trailer). Þ breadth – horizontal distance between two vertical surfaces parallel with the longitudinal vehicle axis touching the most remote points on both sides of the axes (in measuring the maximum breadth of the tractor assembly it was the distance between outer edges of the semi-trailer wheels or outer edges of wheels of the rear tractor axis), Þ height – vertical distance between the horizontal soil surface and horizontal surface touching the highest point of the vehicle (the highest point of the hydraulic crane in transport position). The measurement of the diameter of the turning circle was carried out in accordance with the standard ISO 789-3:1996 (Agricultural tractors – Test procedures – Part 3: Turning and clearance diameters). The standard defines the diameter of the turning circle as the diameter of the smallest circle made by the outer peripheral parts of unloaded vehicle in turning without braking. The measurement of the smallest turning circle was made in turning of each tractor assembly to the left and right. Each tractor assembly started this turning drive on a plane surface until the smallest turning circle was achieved (Fig. 5). The smallest turning circle is achieved at the moment when due to turning, the shaft comes quite close to the tyre of the tractor rear wheel (Fig. 6). Some types of tractor assemblies (more recent ones) have an articulated

Fig. 5 Tractor assembly in achieving the smallest turning circle Slika 5. Traktorski skup pri postizanju najmanjega kruga okretanja Croat. j. for. eng. 29(2008)1

Morphological Analysis of Forest Tractor Assemblies (41–51)

Fig. 6 Contact between semi-trailer shaft and tractor rear wheel tyre in achieving the smallest turning circle Slika 6. Dodir ruda poluprikolice i gume stra`njega kota~a traktora pri postizanju najmanjega kruga okretanja

M. [U[NJAR et al.

Fig. 8 Schematic presentation of the measured values Slika 8. Shema mjerenih vrijednosti

2ABC 2( A B + A C + B 2 C2 ) − ( A 4 + B 4 + C4 ) 2

In assessing the morphological features of the vehicles travelling off road, Bekker (1956) introduced the term imaginary pressure (pim), defining it as the ratio between the vehicle weight (G) and surface of the imaginary rectangle (Aim) defined by the vehicle length (L) and breadth (B). pim =

Fig. 7 Marking of the smallest turning circle Slika 7. Ozna~ivanje najmanjega kruga okretanja joint shaft so that before measuring the smallest turning circle, the articulated joint shaft was positioned at the highest deviation angle from the longitudinal axis of the tractor assembly. During circling of the tractor assembly in achieving the smallest turning circle, several places that define the outer parts of the tractor assembly were marked on the ground and namely: tractor front wheels moving on the outer edge of the circle or the edge of the front tractor bumper (Fig. 7). Fig. 8 shows schematically the marked places on the turning circle of the tractor assembly as well as the marked measurement distances (A, B and C). The diameter of the smallest turning circle was calculated by the following equation: Croat. j. for. eng. 29(2008)1

m⋅ g B⋅L

It can be clearly seen from this definition that this is a theoretic feature suitable for comparison of vehicles and that it has not a physical meaning. Based on the research of the actual pressure of the vehicle on the wheel’s contact surface, the same author concluded that there is a strong correlation between this parameter and imaginary pressure. The imaginary pressure was calculated based on the measured values of the total mass and the basic dimensions for each tractor assembly. The regression analysis was used for the research of possible stochastic dependencies between satisfactorily correlated variables. The exponential regression software programme REG.EXE, primarily developed for the research of the wheel’s slip curves (Hitrec and Horvat 1987), was used for the regression of dependencies of data pairs, which showed by the increase of the independent variable that the data of the dependent variable have an asymptotic trend of increase. The above said software programme shows the correlation strength of the selected re-

M. [U[NJAR et al.

Morphological Analysis of Forest Tractor Assemblies (41–51)

Table 1 Measured masses and basic dimensions of tractor assemblies Tablica 1. Izmjerena masa i osnovne dimenzije traktorskih skupova

Tractor assembly Traktorski skup

1 2 3 4 5 6 7 8 9

Pionir Tigar 42 Pionir IMT 541 Pionir Tigar 49 DV Pionir IMT 549 FMV Steyr 964 Formet Belarus 920 Formet Belarus 952 Vinkum Belarus 952 Vinkum Steyr 9094

Tractor assembly Traktorski skup m kg 3573 3870 4136 4186 6478 8758 8774 8811 8824

Mass Masa Tractor Traktor mt kg 2754 3059 3301 3328 4798 6418 6576 6188 6469

Semi-trailer Poluprikolica mp kg 819 811 835 858 1680 2340 2198 2623 2355

gression models by use of three parameters: r – correlation coefficient, R – correlation index and R > – testing of correlation index based on number of observations.

3. Research results – Rezultati istra`ivanja In this paper tractor assembly is considered a special forest vehicle, although it is composed of a farming tractor as the driving means, forest semitrailer, crane and forest winch if required. The results of this paper show the morphological analysis of nine different types of tractor assemblies used in forestry practice in timber forwarding from thinning operations in lowland forests. Among these tractor assemblies 4 types were older, equipped mechanical cranes – tractor assemblies Pionir with different driving means – tractor. The remaining 5 tractor assemblies were equipped with hydraulic cranes, and two of them were additionally equipped with double-drum winches (tractor assemblies Formet). The following tractor assemblies were researched along with the description of individual components: Þ 1. Pionir – tractor Tigar 42, semi-trailer Metal S-5 of loading capacity of 5t, mechanical crane, Þ 2. Pionir – tractor IMT 541, semi-trailer Metal S-5 of loading capacity of 5t, mechanical crane, Þ 3. Pionir – tractor Tigar 49 DV, semi-trailer Metal S-5 of loading capacity of 5t, mechanical crane,

Dimension of tractor assembly Dimenzije traktorskoga skupa Length Breadth Height Duljina [irina Visina L B H mm mm mm 7890 1880 3685 8380 1820 3415 8230 1850 3450 8320 1800 3081 9300 1940 2844 10480 2060 2995 10420 1930 2780 10230 2130 3190 10310 2200 3790

Turning diameter Promjer kruga okretanja D m 11.8 14.3 12.8 13.6 14.3 13.7 12.0 14.2 12.5

Þ 4. Pionir – tractor IMT 549, semi-trailer Metal S-5 of loading capacity of 5t, mechanical crane, Þ 5. FMV – tractor Steyr 964, semi-trailer Metalac S-5 of loading capacity of 5t, hydraulic crane FMV470 (reach – 6 m, gross lifting capacity – 47 kNm), Þ 6. Formet – tractor Belarus 920, semi-trailer Metalac S-6 of loading capacity of 6t, hydraulic crane Igland 43-65 (reach – 6.5 m, gross lifting capacity – 52 kNm), winch Igland Pronto 6002, Þ 7. Formet – tractor Belarus 952, semi-trailer Metalac S-6 of loading capacity of 6t, hydraulic crane Igland 43-65 (reach – 6.5 m, gross lifting capacity – 52 kNm), winch Igland Pronto 6002, Þ 8. Vinkum – tractor Belarus 952, semi-trailer Graditelj S-8 of loading capacity of 8t, hydraulic crane Cranab 40-55 (reach – 5.5 m, gross lifting capacity – 56 kNm), Þ 9. Vinkum – tractor Steyr 9094, semi-trailer Metal S-8 of loading capacity of 8t, hydraulic crane HDM 340 (reach – 4.8 m, gross lifting capacity – 60 kNm). Table 1 shows the measured masses of tractor assemblies, tractors and semi-trailers and basic dimensions and turning diameters in accordance with the methods described above. Based on the presented results, the differentiation of older types of tractor assemblies Pionir can be seen. They have lower masses because they are not equipped with up-to-date cranes, which have a Croat. j. for. eng. 29(2008)1

Morphological Analysis of Forest Tractor Assemblies (41â&#x20AC;&#x201C;51)

higher mass. Semi-trailers with Pionir tractor assemblies are also less heavy as they have a lower loading capacity (5 tons). They also differ in size as they have lower length and breadth, but at the same time they are higher than more recent tractor assemblies. The reason of higher height of Pionir tractor assemblies lies in the design of the mechanical crane, which is in fact a fix console so that the winch rope passes through a pulley. More recent tractor assemblies have hydraulic cranes that can be assembled in transport position. Only the tractor assembly Vinkum with the driving means tractor Steyr 9094 is separated due to the redesign of arms of the hydraulic crane HDM 340, because of which the crane cannot be assembled into a proper transport position. In 1956 Bekker outlays the opinion that the object (vehicle) moving in a media have the tendency of acquiring the form that causes the lowest moving resistance (Horvat et al. 2004B). If the object (vehicle) is presented in the form of a prism, then the ratios H/L (height/length) and B/L (breadth/length) can show significant volume values that are characteristic of each vehicle family. The ratios H/L and B/L are called form indexes and they are used as the initial information on the researched vehicle and its classification in an already known vehicle family. It is characteristic of tractors used for timber extraction (adapted farming tractors, skidders with winch, forwarders), that they are usually in the area under the direction H = B. Form indexes were calculated from the basic dimensions of the researched tractor assemblies and Figure 9 shows their dependence. This Figure also shows the ratio between form indexes of forwarder family according to the research carried out by Horvat et al. (2004B). The survey of form indexes of tractor assemblies and forwarders was given with the aim of comparing the dimensional characteristics of two types of driving means for timber forwarding. It can be seen that the values of form indexes of all tractor assemblies and forwarders covered by the analysis are under the direction H = B, or in other words in the area where height prevails over breadth. It can be observed that form indexes of tractor assemblies have lower values of form index B/L (lower than 0.25) than the forwarder and that they are completely separated from the area of forwarders. Lower values of form index of tractor assemblies can be explained by their higher length compared to breadth. Lower breadth of tractor assemblies makes them suitable for performing the operations of timber forwarding in thinning stands, as an easier vehicle passage between standing trees in the stand is provided. Croat. j. for. eng. 29(2008)1

M. [U[NJAR et al.

Fig. 9 Dependence of form index B/L on form index H/L Slika 9. Ovisnost indeksa oblika B/L o indeksu oblika H/L Higher length implies a larger turning circle of the tractor assembly and hence its poorer manoeuvrability. However the design of the tractor assembly, where the semi-trailer is connected to the tractor by a shaft, enables a large turning angle of the shaft (approximately 60Â°) from the tractor longitudinal axis, thus providing a lower turning circle and better manoeuvrability of the vehicle. With more recent types of semi-trailers, an additional articulated joint on the shaft enables turning of the semi-trailer longitudinal axis from the tractor longitudinal axis at an angle of 90Â°. Based on the analysis of the form index, it can be concluded that tractor assemblies may be treated as a separate family of vehicles for timber forwarding, which is ecologically acceptable for thinning operations owing to its dimensional characteristics. In further research the measured values were correlated and then subjected to regression analyses. In this procedure, the mass of tractor assemblies was chosen as the independent variable. Figure 10 shows the dependence of length on mass of the tractor assembly, tractor and semi-trailer. High correlation coefficients and correlation indexes were achieved, which implies a very strong correlation between the independent and dependent variable. The increase of mass causes the increase of length of the tractor, semi-trailer and tractor assembly. The fastest growth can be seen of the regression curve of length of the tractor assembly with the increase of semi-trailer mass. Although semi-trailers have the

M. [U[NJAR et al.

Morphological Analysis of Forest Tractor Assemblies (41â&#x20AC;&#x201C;51)

lowest mass, their mass has the highest effect on the increase of length of the tractor assembly. Tractor assemblies are divided into two groups. Older types of tractor assemblies Pionir are characterised by lower masses and lengths. Their position is very close to tractor masses, which can be explained by their light semi-trailers of lower loading capacity and installation of mechanical crane (fix console with pulleys and rope drum), which is supported and leaned on the tractor. More recent types of tractor assemblies have a hydraulic crane installed on semi-trailer of higher mass and load bearing capacity (6 and 8 tons). These tractor assemblies have the highest masses with respect to length. The tractor assembly FMV with the tractor Steyr 964 is singled out in the middle of the regression line due to semi-trailer of lower loading capacity (5 tons). Dependence of breadth on mass of the tractor assembly, tractor and semi-trailer is shown in Fig. 11. The same trend of the listed data and regression lines can be observed. Tractor assemblies are divided into older and more recent types. Older types of tractor assemblies are characterised by the breadth lower than 2 m due to the use of narrower tractors and semi-trailers of lower loading capacity. With more recent types, two tractor assemblies (Formet with tractor Belarus 952 and FMV with tractor Steyr 964) have the breadth lower than 2 m. The tractors used as the driving means in these assemblies had a narrow track width (narrower tyres), which affected their smaller breadth. As semi-trailers in these as-

semblies had an additional articulated joint on the shaft and a small breadth, these assemblies were selected as the most suitable for timber forwarding in thinning operations due to exceptionally good manoeuvrability in the stand. Figure 12 shows the dependence of the smallest turning diameter on the mass of the tractor assembly. From an ecological point of view, smaller turning circle means better manoeuvrability of the vehicle in the stand and consequently lower soil tracking and damage to soil and trees in the stand. Tractor assemblies have low turning diameters â&#x20AC;&#x201C; from 11.8 m to 14.3 m. The smallest turning diameter has been recorded with an old type of tractor assembly Pionir with the driving tractor Tigar 42, and it also has the smallest length of all researched assemblies. The turning circle diameter increases with the increase of mass of the tractor assembly, which is related to a strong dependence of length on mass of the tractor assembly. However, under the regression line two more recent types of tractor assemblies of higher mass can be clearly seen. The said assemblies have an articulated joint on the semi-trailer shaft, which enables them to make a very small turning diameter regardless of higher mass and total length. Dependence of imaginary pressure on the mass of the tractor assembly is shown in Fig. 13. The regression line indicates that the imaginary pressure of the tractor assembly on the ground (forest soil) in-

Fig. 10 Dependence of length on mass Slika 10. Ovisnost duljine o masi

Fig. 11 Dependence of breadth on mass Slika 11. Ovisnost {irine o masi

Croat. j. for. eng. 29(2008)1

Morphological Analysis of Forest Tractor Assemblies (41â&#x20AC;&#x201C;51)

M. [U[NJAR et al.

Fig. 12 Dependence of turning circle diameter on mass of tractor assembly Slika 12. Ovisnost promjera kruga okretanja o masi traktorskoga skupa

Fig. 13 Dependence of imaginary pressure on mass of tractor assembly Slika 13. Ovisnost imaginarnoga tlaka o masi traktorskoga skupa

creases proportionally with the increase of mass. The values of the imaginary pressure of the tractor assemblies Pionir range between 1.2 kPa and 1.6 kPa. More recent tractor assemblies have imaginary pressures of 2.2 kPa to 3 kPa. Lower imaginary pressure of the vehicle on the soil means better ecological acceptability, i.e. lower risk of causing damage to forest soil (soil compaction) during the vehicle movement.

mental suitability in timber forwarding from thinning operations. The problem of higher total length does not affect the vehicle movability as the installation of the articulated joint on semi-trailer shaft ensures small turning diameter and hence reduced risk of causing damage to the standing trees. The breadth of new tractor assemblies does not differ considerably from Pionir assemblies despite the use of tractors of higher mass and semi-trailers of higher loading capacity. Favourable breadth of tractor assemblies is achieved by use of narrower tyres on rear wheels of the tractor. Better forest accessibility is also provided by the installation of hydraulic cranes whose arms can be assembled during travel. The installation of the forest winch on the tractor assembly also contributes to ecological suitability of the vehicle. The tractor assembly with winch does not have to move on the stand so as to reach each processed assortment in the reach of the crane. In this way soil tracking is reduced as well as the risk of causing damage to the soil and standing trees. On the other hand, higher mass of more recent tractor assemblies results in higher imaginary pressure on the soil. However, by comparing imaginary pressures of tractor assemblies and forwarders, it can be concluded that tractor assemblies are more environmentally friendly in timber forwarding under conditions of poor bearing strength of the soil. According to the research carried out by Horvat et al. (2004B) the lowest possible imaginary pressures of

4. Conclusions â&#x20AC;&#x201C; Zaklju~ci According to the research results Pionir tractor assemblies have very favourable morphological characteristics (low masses and dimensions, small turning diameter, low imaginary pressure on the soil) by which they are classified as ecologically acceptable vehicles for timber forwarding from thinning operations in lowland forests. However, work technology with Pionir tractor assemblies requires from the vehicle to move on the stand so as to approach each processed assortment, which implies higher soil tracking and possibility of causing damage to forest soils of poor bearing strength in lowland areas. Furthermore, out-of-date technical and design solutions (type of crane, older types of tractors) are limiting factors in extending their use primarily due to unfavourable ergonomic and safety requirements. More recent types of tractor assemblies are characterised by technical designs that ensure environCroat. j. for. eng. 29(2008)1

M. [U[NJAR et al.

Morphological Analysis of Forest Tractor Assemblies (41–51)

6-wheel and 8-wheel forwarders have the value around 4 kPa. The highest imaginary pressure of the researched tractor assemblies is 3 kPa. Consequently, more recent types of tractor assemblies, which exert the imaginary pressure on the soil ranging between 2 kPa and 3 kPa are better from the point of view of soil protection and ecological acceptability for the operations performed on forest soils of poor bearing strength. To conclude, tractor assemblies are considered ecologically more favourable forest vehicles under conditions of poor bearing strength of the soil as well as in thinning operations in lowland forests.

5. References – Literatura Bekker, M. G., 1956: Theory of land locomotion. The University of Michigan Press, p. 1–499. Beuk, D., Toma{i}, @., Horvat, D., 2007: Status and development of forest harvesting Mechanisation in Croatian state forestry. Croat. j. for. eng. 28(1): 63–82.

Operational planning and control; work study, 3.06.00 Forest operations under mountainous conditions, 3.07.00 Ergonomics, Opatija 27. 9. – 1. 10. 1999, Faculty of Forestry of Zagreb University, 1999, 99 – 100. Horvat, D., [u{njar, M., Toma{i}, @., 2004A: New technical and technological solutions in thinning operations of lowland forests. Proceedings of the International scientific conference »Forest Engineering: New Techniques, Technologies and the Environment«, IUFRO, The Ukraine Forestry Academy of Sciences (LANU), The Ukrainian Mountain Forestry Research Institute (UkrNDIGirlis), The State Forestry Management Association »Lvivlis«, The National Nature Park »Hutsulshchyna«, October 2004, Lviv, Ukraine. Poster. Horvat, D., Por{insky, T., Krpan, A., Pentek, T., [u{njar, M., 2004B: Suitability Evaluation of Timberjack 1710B Forwarder based on morphological analysis. Strojarstvo 46(4–6): 149–160. Horvat, D., Spinelli, R., [u{njar M., 2005: Resistance coefficients on ground-based winching of timber. Croat. j. for. eng. 26(1): 3–11.

Hitrec, V., Horvat, D., 1987: Jedna metoda odre|enja regresijskog modela na primjeru krivulje klizanja kota~a. Meh. {umar. 12(11–12): 177–181.

ISO 789-3, 1996: (Agricultural tractors – Test procedures – Part 3: Turning and clearance diameters).

Horvat, D., Kristi}, A., 1999: Research of some morphological features of thinning tractor assemblies with semi-trailer. Abstracts »Emerging harvesting issues in technology transition at the end of century« IUFRO Division 3, RGs: 3.04.00

Wästerlund, I., 1994: Forest response to soil disturbance due to machine traffic. Interactive seminar and workshop »Soil, tree, machines interaction«, Feldafing, Germany, p. 1–23.

Sa`etak

Morfolo{ka ra{~lamba {umskih traktorskih skupova Rezultati ovoga rada prikazuju morfolo{ku ra{~lambu devet razli~itih tipova traktorskih skupova koji se koriste u {umarskoj praksi pri izvoènju drva iz proreda nizinskih {uma. Pri tome su zastupljena ~etiri tipa traktorskih skupova starije proizvodnje Pionir, opremljenih mehani~kim dizalicama. Preostalih pet novijih traktorskih skupova bilo je opremljeno hidrauli~nim dizalicama, a od toga su dva jo{ dodatno opremljena dvobubanjskim {umskim vitlima. Prema rezultatima istraìvanja traktorski skupovi Pionir imaju vrlo povoljne morfolo{ke zna~ajke (malu masu i dimenzije, mali promjer kruga okretanja, mali imaginarni tlak na tlo) koje ih svrstavaju u ekolo{ki prihvatljiva vozila za izvoènje drva iz proreda nizinskih {uma. No, tehnologija rada s traktorskim skupovima Pionir zahtijeva kretanje vozila po sastojini do svakoga izra|enoga sortimenta, {to zna~i i ve}e gaènje tla i mogu}nost o{te}ivanja slabonosivih {umskih tala u nizinskom podru~ju. Tako|er, zastarjela tehni~ka i konstrukcijska rje{enja ograni~avaju njihovu uporabu zbog nepovoljnih ergonomskih i sigurnosnih zahtjeva. Noviji tipovi traktorskih skupova isti~u se tehni~kim rje{enjima koja im osiguravaju okoli{nu pogodnost pri izvoènju drva iz proreda. Problem ve}e ukupne duljine ne utje~e na kretnost vozila jer se ugradnjom zgloba na rudu poluprikolice osigurava mali promjer kruga

Croat. j. for. eng. 29(2008)1

Morphological Analysis of Forest Tractor Assemblies (41–51)

M. [U[NJAR et al.

okretanja te time manja opasnost od o{te}ivanja dube}ih stabala. Povoljna se {irina traktorskih skupova postiè kori{tenjem uìh guma na stra`njim kota~ima traktora. Ugradnjom se {umskoga vitla na traktorski skup smanjuje gaènje sastojine, o{te}ivanje tla i dube}ih stabala. Noviji tipovi traktorskih skupova ostvaruju imaginarni tlak na tlo izme|u 2 kPa i 4 kPa te imaju ve}u prednost za rad na {umskim tlima slabe nosivosti nego forvarderi koji ostvaruju najmanje imaginarne tlakove oko 4 kPa. Zaklju~no, traktorske skupove smatramo ekolo{ki povoljnim {umskim vozilima u uvjetima slabe nosivosti tla te rada u proredama nizinskih {uma. Klju~ne rije~i: traktorski skup, morfolo{ka ra{~lamba, ekolo{ka pogodnost, izvoènje drva, prorede

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

Received (Primljeno): January 29, 2008 Accepted (Prihva}eno): May 12, 2008 Croat. j. for. eng. 29(2008)1

Andrija Kristi}, MSc. e-mail: andrija.kristic@hrsume.hr »Croatian Forests« Ltd. Forest Adminstration Vinkovci Trg bana Josipa [ok~evi}a 20 HR – 32100 Vinkovci CROATIA

Original scientific paper – Izvorni znanstveni rad

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest Igor Poto~nik, Tibor Pentek, Dragutin Pi~man, Ivica Papa, Anton Poje Abstract – Nacrtak This paper deals with the required width of forest road formation. Forest roads are necessary for emergency forest management. Besides positive effects, forest roads also bring negative effects. One of them is the loss of forest area due to their construction in the forest environment. The estimation of the possible loss of forest area was the main objective of this research. The clearance of a forest road cross-section affects both the forest and the road. The crowns of edging trees that fill in the clearance have a number of positive effects. The intent of this study was to investigate the filling in process of the clearance of a forest road cross-section in beech stands of south-east Slovenia. At the age of 25–35, forest roads undergo the most intensive differentiation in the filling in process of the clearance. The distance between crowns is 6 m in a 15-year old forest road, 0.74 m in a 35-year old road, and 0.24 m in a 50-year old road. Therefore it may be expected that the distance will be reduced to an average roadway width (3.63 m) in 18 years, and in the next 10 years to 50% of the roadway width. The expected number of profiles with the distance between crowns larger or smaller than 2 m and distance between crowns larger than 2 m at the age of 15 is 90 times higher than in case of older roads. On the other hand, the distance between crowns larger than 2 m with roads of the same age is expected to be 1.5 times higher for every additional meter of distance between stems. It can be concluded that the knowledge of the filling in process of the clearance of a forest road cross-section can be of great help in planning forest road maintenance. Keywords: forest road, clearance of a forest road cross-section, filling in, forest road maintenance

1. Introduction and research problem – Uvod i problematika istra`ivanja In the seventies and eighties of the 20th century intensive construction of both primary and secondary forest communication system was carried out in Slovenia. Recently, however, there has been a considerable reduction of construction of new forest communications. Current problems of forest construction in Slovenia concern primarily maintenance of the existing network of forest roads. In addition to bedrock, particularly traffic and atmospheric factors affect the condition of forest roads. If we restrict ourselves to atmospheric factors, meteoric water constitutes the main problem. Striking directly against the roadway, rain drops damage the upper road layer of a forest road by washing out minute bonding agents, and thus they make the road susceptible to destrucCroat. j. for. eng. 29(2008)1

tive power of water. Appropriate granulation of material, suitable inclination of the levelling line and suitable cross inclination, along with properly installed drainage facilities allow roadway drainage and thus they help preserve the forest road. Trees and other vegetation, which fill in the clearance of a forest road cross-section play a special role. Branches and leaves intercept rain drops, preventing them from striking directly against the roadway. They also absorb part of meteoric water, mitigate the effects of direct solar radiation and prevent excessive drying up of the roadway. In this way permanent humidity and coherence of the upper layer are ensured, which means less damage to the roadway and lower maintenance costs. The effectiveness of a green protective shelter above a forest road depends on a number of factors.

I. POTO^NIK et al.

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

Barna (2001) concluded that twigs of beech trees in each third of the crown responded to cutting by means of enhanced length increment always in the 2nd vegetation period after cutting. The width of a forest road clearance and tree speies are probably the most important factors (Poto~nik 2003). Rock base and terrain slope were analysed as some of the factors that impact the road formation width. Comparison of the road formation width and basic rocks shows that limestone is the most favourable and tonalite the least favourable. It was concluded that road formation width could increase up to 80% more on steeper terrain (comparing to gentle terrain slope) and 20% on less solid rock base regardless the terrain slope. It varies between 5.4 m (solid rock base, gentle slope) and 11.4 m (soft rock base, steep slope). Conifers do not fill in the gaps at all or only to a smaller extent, while deciduous trees are more aggressive. Webster and Lorimer (2004) report that minimum opening sizes and time required for successful canopy recruitment of tree saplings are not well known because the gap capture process is slow and difficult to monitor. They investigated canopy recruitment in harvest openings created by group selection in hemlock-hardwood forests to determine if yellow birch (Betula alleghaniensis Britt.), a species with intermediate shade tolerance, could successfully reach the canopy via small openings. On the other hand Brisson (2001) paid attention to tree crown plasticity. A tree neighbouring a new gap must show a certain degree of morphological plasticity in its lateral growth to take advantage of the available space, expanding branches preferentially on the side of the gap. The morphological plasticity of sugar maple (Acer saccharum Marsh.) was evaluated by measuring crown asymmetry with respect to four several neighbourhood contexts. Isolated trees had the most symmetrical crown, while all trees at the edge of a field had the largest part of their crown growing away from the forest. Seiler and McBee (1992) developed a simple, rapid technique for estimating the projected tree crown area on the ground. The technique utilizes photographs of tree crowns taken from two directions with an object of known scale present in each photograph. The photographs are then displayed onto a dot grid and the projected (two dimensional) crown surface area in square feet is estimated. The intent of the study was to investigate the filling in of the clearance of a forest road cross-section in relation to road age. It was based on the a priori assumption: the older a road the more filled in the clearance of its cross-section. Such information may be of help in planning maintenance measures of forest roads.

2. Research area and research methods – Podru~je i metode istra`ivanja Study sites were selected according to similar ecological conditions (site, forest community, developmental stage, bedrock, terrain inclination, inclination of levelling line). Thus three forest roads of different ages were selected: 15 years, 35 years, and 50 years, respectively. They are located in the forest management unit Soteska (district unit of Novo Mesto Forest Service in the south-east of Slovenia). Bedrock consists of Cretaceous limestone, the forest community is Enneaphyllo-Fagetum. The selected roads are situated in forests dominated by beech. The stands are uniform and mature, the growing stock being 305–360 m3/ha with an annual increment of about 10 m3/ha. The average terrain inclination is around 35%. According to Fig. 1 fields the following data were taken: distance between crown of beech trees (Fagus sylvatica L.) as a vertical projection of their edges (Cd), roadway width (Cw), road body width (Rw) and distance between the first stems from each side of the road (Sd). Due to field work, accuracy of 10 cm was implemented. A flat segment was selected from each forest road (age classes) and 30 cross-sections were sampled for

Fig. 1 Cross-section data Slika 1. Podatci o popre~nim profilima Croat. j. for. eng. 29(2008)1

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53â&#x20AC;&#x201C;62)

measuring roadway width, road body width, distance between stems, and distance between crowns. The other elements of cross-sections were considered less important for the purposes of this pilot study.

3. Results of research â&#x20AC;&#x201C; Rezultati istra`ivanja Average values of individual parameters were determined for each sample group of cross-sections

I. POTO^NIK et al.

separately (Fig. 2). Roadway width slightly decreases with age, which may be attributed to differences in construction technology, rather than to differences in technical elements. The oldest road was built manually and thus the differences in relation to machine construction are evident. Road body is the widest in the most recently built road and the narrowest in the oldest road. Different construction techniques are again responsible for the differences. Distance between trunks, too, decreases with road age. But the decrease also results from the filling of

Fig. 2 Differences in parameters of the cross-section in relation to forest road age Slika 2. Razlika pojedinih sastavnica popre~noga profila {umskih cesta razli~ite starosti Croat. j. for. eng. 29(2008)1

I. POTO^NIK et al.

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

Table 1 Average values and significance of differences between means controlled by age of forest road Tablica 1. Statisti~ka analiza utjecaja starosti {umske ceste na pojedini parametar popre~noga profila Road age Starost ceste

years – godine

Roadway width Širina planuma

Road body width Širina tijela ceste

Distance between stems Udaljenost izme|u debala stabala

Distance between crowns Udaljenost izme|u kro{anja stabala

Mean ± St. Dev. Arit. sred. ± stand. dev. m

4.01 ± 0.30

3.55 ± 0.21

3.34 ± 0.31

Total – Ukupno

3.63 ± 0.39

10.65 ± 1.43

8.53 ± 1.29

7.10 ± 0.85

Total – Ukupno

8.76 ± 1.90

15.39 ± 2.65

12.39 ± 2,67

9.45 ± 2.42

Total – Ukupno

12.41 ± 3,53

5.97 ± 2.63

0.74 ± 1.46

0.24 ± 0.48

Total – Ukupno

2.32 ± 3.13

the forest road edge with tree species as well as from different construction techniques. Comparison of the distance between crowns is very interesting. Regardless of the fact that more recent roads are wider (roadway and road body), the differences in the distance between crowns are highly characteristic of each age group. The extreme example is the most recently built road with an average distance between crowns of nearly 6 m. There is not much difference in the distance between crowns between the 35-year-old (0.74 m) and 50-year-old road (0.24 m). In the author’s view, there are hardly any gaps between crowns after 25 years. It should be underlined, however, that the clearance of a forest road cross-section was filled in here by the beech, which has a highly adaptable crown. Conifers would have filled in the clearance to a much lesser extent or not at all. Measuring selected factors by stratums (three different groups by the road age) turned out as a

Mean difference between years Srednja razlika izme|u starosti m 0.46 0.21

m/year – m/god. 0.02 0.01

Significance of differences between means – p Zna~ajnost razlike izme|u srednjih vrijednosti – p (15–50) 0.000

(15–35) 0.000 (35–50) 0.003

2.12 1.43

0.11 0.10

(15–50) 0.000

(15–35) 0.000 (35–50) 0.000

3.00 2.94

0.15 0.20

(15–50) 0.000

(15–35) 0.000 (35–50) 0.000

5.23 0.50

0.26 0.03

(15–50) 0.000

(15–35) 0.000 (35–50) 0.080

good solution. Average values are significantly different among the age groups and they get reduced by the age, respectively (Table 1). Differences are not only significant within age groups but also in majority of paired comparisons between age groups. Therefore differences are significant in all factors between the age group of 15 and 35 years and greater when comparing groups of 35 and 50 years. It may be concluded that changes of factors investigated are more intensive in the first two or three decades than later. Two results should be pointed out: relative change of distance between stems is larger in older groups (0.20 m/year) than younger groups (0.15 m/year). Assuming that the forest road was of the same width, and the width of road formation and roadway was enlarged due to machinery construction in time, a larger growing space would be available at the road edge without a crown shelter. On the other hand distances between crowns in older groups were not different (p = 0.080) Croat. j. for. eng. 29(2008)1

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53â&#x20AC;&#x201C;62)

I. POTO^NIK et al.

Fig. 3 Frequency distributions of distances between crowns in respect to road age Slika 3. U~estalost raspodjele udaljenosti izme|u kro{anja stabala u ovisnosti o starosti {umske ceste

Fig. 4 Dependence of distance between tree crowns and road age Slika 4. Ovisnost razmaka izme|u kro{anja stabala o starosti {umske ceste

which means that older groups could be merged and analyzed as a single group. When comparing frequency distribution of distance between crowns, differences are small as well (Fig. 3). Croat. j. for. eng. 29(2008)1

Fig. 3 shows that frequency distribution of distance between crowns is changing by age from left asymmetrical distribution through normal distribution to right asymmetrical distribution (J distribution) 50 years after the road construction. The share of cross

I. POTO^NIK et al.

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

Table 2 Results of logistics regression Tablica 2. Rezultati logisti~ke regresije Estimated parameter Procijenjeni parametar B

Standard Error Standardna pogreška

4.498

Road age under 15 years Starost cesta do 15 godina Distance between stems Udaljenost izme|u debala Constant Stalnica

Wald test

Degree of freedom Stupnjevi slobode df

0.932

23.285

0.393

0.177

–7.474

2.451

95.0% confidence interval for eB 95 % interval pouzdanosti eB

Odds ratio Omjer šanse eB

Lower limit Donja granica

Upper limit Gornja granica

0.000

89.835

14.455

558.316

4.925

0.026

1.481

1.047

2.096

9.297

0.002

0.001

Table 3 Pearson correlations coefficients between parameters Tablica 3. Pearsonov koeficijent korelacije izme|u istra`ivanih parametara

Roadway width Širina planuma ceste Road body width Širina tijela ceste Distance between stems Udaljenost izme|u debala stabala

Roadway width Širina planuma ceste

Road body width Širina tijela ceste

Distance between stems Udaljenost izme|u debala stabala

0.764(**)

0.562(**)

0.764(**)

0.602(**)

0.562(**)

0.602(**)

** Correlation is significant at the 0.01 level (2-tailed) – Korelacija je statisti~ki zna~ajna uz razinu zna~ajnosti 0,01 (dvosmjeran test)

sections with connected crowns from both side of the forest road is also growing by age: 0% after 15 years to 60% after 35 and 50 years after the construction. It can be concluded that tree crowns begin to fill up the clearance of the cross section utmost 15 to 20 years after construction, when reducing the distance between tree crowns is the most intensive (Fig. 4). With no regard to limited validation besides high regression coefficient (R2 = 0.604; p = 0.000) due to lack of data, the result is still reliable enough to estimate reducing of tree crowns distance by age. Therefore it may be expected that the distance will be reduced to an average width of roadway (3.63 m) in 18 years, and in the next 10 years to 50% of the roadway width. It has to be pointed out that two facts were taken into consideration when the function was established. The inverse function fits the data the best (R2 = 0.686; p = 0.000), but it is only good for the age over 25 years. Using inverse function and reducing the age the clearance in the cross-section (distance between crowns) theoretically means unlimited enlargement of clearance which is unrealistic. Therefore we used 10 m distance between crowns as a first distance, which is 2 m more than the width of a road formation at 40% terrain slope (Poto~nik 2003).

Further in the study we tested if any other factor, besides the age, impacts the distance between tree crowns. Binary multivariate logistic regression was used for calculating the model. Distances between crowns were split into 2 groups (below 2 m – 58 profiles and above 2 m – 32 profiles) to enable all necessary data in both groups. Forest road age was divided into 2 groups (up to 15 years and more than 15 years) due to insignificant differences. Only one additional factor was added because the number of profiles with distance between crowns over 2 m was relatively small. Theoretically, at least 10 records per factor were to be included. The model shows that besides the age only distance between stems impacts the clearance of the cross-section. Table 2 shows that odds ratio (number of profiles with distance between crowns less or more than 2 m) for distance between crowns larger than 2 m by the age of 15 years is 90 times higher than in the case of older roads. On the other hand, odds ratio for distance between crowns larger than 2 m at the same age are 1.5 times higher for every additional meter of distance between stems. We could not include other two technical elements of the cross section (roadway width and road Croat. j. for. eng. 29(2008)1

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

body width) because their impact was already comprised with parameters included in the model. All parameters are significantly correlated (Table 3) which brings the conclusion that one parameter also takes the explanatory role for other parameters. Thus, when excluding the age of the road its contribution is taken over by the width of the roadway. Naturally, the variance explained by the model, is therefore lower.

4. Discussion – Rasprava This pilot study confirmed the assumption that distance between stems and tree crowns as well as road body width and roadway width are decreasing by age, which means the older the road the more it is incorporated in the natural environment. The result is expected and logical in discussing distance between stems and crowns, which is not the case with studied technical parameters of the forest road. Theoretically, road body width and roadway width should remain unchanged in time if regular maintenance were provided. Using different road construction technologies (manual and machine work) is a possible answer to explain the reduction of road body width and roadway width in time. The fact is that machine construction was introduced in late 50’s and early 60’s of the 20th century which is around the older age classes of forest roads in the present study. Manually constructed roads (older age class) are narrower in terms of roadway compared to those constructed by machines due to hard working conditions. Hence, it may be concluded that the efficiency of overgrowing the clearance of the cross-section depends on time (road age), width of the forest belt removed due to construction and road body width. Some differences and trends of overgrowing the clearance of the cross-section may also be explained by tree species and their age. The study dealt with beech but we may expect different results with conifers and other broadleaves, too. Younger trees, regardless of tree species, are able to develop tree crowns faster and consequently overgrow the clearance faster. Kmet' and Ditmarová (2001) warn that pollution could cause negative processes (especially chlorophyll fluorescence parameters) and state that a deteriorated physiological state of young beech trees could be observed at more polluted sites even at the stage of a latent (hidden) damage. Under specific ecological conditions after construction, the clearance of a forest road cross-section is quite rapidly filled in with crowns of edging trees, particularly in beech stands. The filling in of the Croat. j. for. eng. 29(2008)1

I. POTO^NIK et al.

clearance has the following positive aesthetic, ecological, technical and commercial effects: Þ the forest road becomes part of the forest, Þ damage caused by construction work is rectified, Þ the stand affected by construction work is more stable, Þ green shelter protects the forest road against direct solar radiation and against harmful effects of precipitation. From the point of view of the most efficient maintenance of forest road it can be concluded that: Þ at the age of forest road over 30 years we may expect up to 50% lower maintenance costs which is directly caused by the impact of precipitation on the road surface, Þ the belt of the forest, which has to be removed due to construction, should be as wide as minimally necessary to enable other trees to overgrow the clearance above the cross-section, Þ tree removal, especially on the fill slope, should be minimal because of increased stability of forest road and shorter time necessary to connect tree crowns above the forest road (Poto~nik et al. 2005), Þ at the edge of the road formation broadleaves should be promoted trying not to develop asymmetrical crowns (lower mechanical stability) performing adequate nursery measures, Þ all measures should be thoroughly implemented in the areas of protected nature (Poto~nik 2006).

5. References – Literatura Barna M., 2001: The impact of shelterwood cutting on twig growth in understoried beech trees (Fagus sylvatica L.). Ekológia Bratislava 20(2): 191–199. Brisson J., 2001: Neighborhood competition and crown asymmetry in Acer saccharum. Can. J. For. Res. 31(12): 2151–2159. Go³¹b J., 2004: Influence of forest slope road on water storage in adjoining soils. Electronic Journal of Polish Agricultural Universities, Forestry 7(2) () Kmet' J., Ditmarová ¼., 2001: Bioindication of the beech (Fagus sylvatica L.) photosynthetic apparatus function conditions under stress circumstances. Ekológia Bratislava 20(2): 200–208. Pentek, T., Neve~erel, H., Pi~man, D., Por{insky, T., 2007: Forest road network in the Republic of Croatia – Status and perspectives. Croat. j. for. eng. 28(1): 93–106.

I. POTO^NIK et al.

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

Poto~nik, I., Yoshioka, T., Miyamoto, Y., Igarashi, H., Sakai, H., 2005: Maintenance of forest road network by natural forest management in Tokyo University Forest in Hokkaido. Croat. j. for. eng. 26(2): 71–78. Poto~nik, I., 2003: Forest road formation width as an indicator of human impact on forest environment. Ekológia 22(3): 298–304, Bratislava. Poto~nik, I., 2006: Road Traffic in Protected Forest Areas – Case Study in Triglav National Park, Slovenia. Croat. j. for. eng. 27(2): 115–121.

Seiler J. R., McBee, N. P., 1992: A rapid technique for the evaluation of mature tree crown growth. Journal of Arboriculture 18(6): 325–328. Simon S. Using SPSS to Develop a Logistic Regression Model. (http://www.childrens-mercy.org/stats/training/ hand04.asp) Webster, C. R., Lorimer C. G., 2004: Minimum opening sizes for canopy recruitment of midtolerant tree species; a retrospective approach. Ecological Applications 15(4): 1245–1262.

Robek, R., Klun, J., 2007: Recent developments in forest traffic way construction in Slovenia. Croat. j. for. eng. 28(1): 83–91.

Sa`etak

Popunjavanje svijetloga otvora popre~noga profila {umske ceste u bukovim sastojinama Sedamdesetih i osamdesetih godina 20. stolje}a u Sloveniji se intenzivno gradila mreà primarnih i sekundarnih {umskih prometnica. Posljednjih je godina zna~ajno smanjen obujam radova na izgradnji {umskih prometnica, a teì{te je stavljeno na odràvanje postoje}ih sastavnica primarnoga {umskoga transportnoga sustava. Najve}a o{te}enja na {umskim cestama uzrokuju oborine. [tete i tro{kovi odràvanja zna~ajno se mogu smanjiti ugradnjom kvalitetnoga kamenoga materijala propisane granulacije u gornji stroj {umske ceste, pogodnim uzdu`nim nagibom nivelete, odgovaraju}im popre~nim nagibom te izvedbom objekata povr{inske i podzemne odvodnje gdje god je, na trasi {umske ceste, to potrebno. Stabla i ostala vegetacija koja popunjava svijetli otvor iznad {umske ceste ima velik utjecaj na zadràvanje {umske ceste u dobrom stanju. Grane i li{}e sprje~avaju izravan dolazak oborinske vode na {umsku cestu (smanjuju njezinu udarnu snagu), a dio vode i apsorbiraju (manja je koli~ina vode koja do|e do {umske ceste). Zatvoreni sklop stabala zasjenjuje {umsku cestu i na taj na~in {titi gornji stroj od nagloga isu{ivanja. Time je osigurana stalna vla`nost, ~vrsto}a i povezanost kolni~ke konstrukcije, {to tako|er ide u prilog niìm tro{kovima odràvanja {umske ceste. Cilj je ovoga rada istraìti popunjavanje svijetloga otvora iznad {umske ceste u bukovim sastojinama u ovisnosti o starosti {umske ceste. Pretpostavljeno je da je na starijim {umskim cestama popunjeniji (uì) svijetli otvor. Potvrda navedene hipoteze istraìvanja moè pomo}i pri planiranju vrste, intenziteta i u~estalosti radova na odràvanju {umskih cesta. Istraìvanje je provedeno u GJ Soteska, U[P Novo mesto. Rije~ je o bukovoj {umi (Enneaphyllo-Fagetum) na vapnena~koj geolo{koj podlozi. Drvna zaliha iznosi od 305 do 360 m3/ha s godi{njim prirastom od 10 m3/ha. Prosje~ni je nagib terena oko 35 %. Kao objekti istraìvanja odabrane su tri {umske ceste razli~ite starosti (15 godina, 35 godina i 50 godina) smje{tene u vrlo sli~nim sastojinskim, stani{nim, terenskim i tehni~kim uvjetima (geolo{ka podloga, pedolo{ka podloga, {umska zajednica, razvojni stadij {ume, prosje~ni nagib terena, popre~ni nagib terena i uzdu`ni nagib nivelete {umske ceste). Prikupljanje je podataka na terenu, u 30 odabranih popre~nih profila svake {umske ceste, obuhva}alo izmjeru: udaljenosti izme|u kro{anja stabala (Cd), {irine planuma {umske ceste (Cw), {irine tijela {umske ceste (Rw) i udaljenosti izme|u debala stabala (Sd) sukladno prikazu na slici 1. Izmjera se obavljala s to~no{}u od 10 cm. Za svaki je od mjerenih parametara izra~unata prosje~na vrijednost po svakom objektu istraìvanja (slika 2). [irina planuma i {irina tijela {umske ceste lagano pada sa staro{}u {umske ceste, {to moèmo objasniti prije razli~itim metodama gradnje (najstarija je {umska cesta gra|ena ru~no) nego razli~itim tehni~kim zna~ajkama. Razmak se izme|u debala stabala smanjuje sa staro{}u {umske ceste, za {to je tako|er zasluàn na~in gradnje {umskih cesta u razli~itim razdobljima. Usporedba udaljenosti izme|u kro{anja stabala daje najzanimljivije

Croat. j. for. eng. 29(2008)1

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

I. POTO^NIK et al.

rezultate. Na {umskoj je cesti starosti 15 godina razmak izme|u kro{anja stabala 5,97 m, dok razlika izme|u udaljenosti kro{anja stabala na {umskoj cesti starosti 35 godina (0,74) i na onoj starosti 50 godina (0,24) nije velika. Iako je bukva vrsta ~ija se kro{nja brzo i lako prilagodi novonastaloj situaciji u {umi (svijetli otvor {umske ceste), 25 godina nakon izgradnje {umske ceste smanjenje udaljenosti kro{anja stabala gotovo se i ne primje}uje. U tablici 1 prikazana je statisti~ka analiza pojedinoga istraìvanoga parametra popre~noga presjeka {umske ceste za svaku od triju {umskih cesta razli~ite starosti unutar grupe i izme|u grupa. Slika 3 prikazuje promjenu frekvencije udaljenosti izme|u kro{anja stabala kroz godine starosti {umske ceste od lijeve asimetri~ne distribucije preko normalne distribucije do desne asimetri~ne distribucije (J distribucija) 50 godina nakon izgradnje {umske ceste. Udio izmjerenih popre~nih profila sa sastavljenim kro{njama bukovih stabala u odnosu na ukupno izmjereni broj popre~nih profila (30) u svakoj starosnoj grupi iznosio je 0 % za {umsku cestu starosti 15 godina te 60 % za {umske ceste starosti 35 i 50 godina. Zatvaranje sklopa iznad izgra|ene {umske ceste u bukovim sastojinama najintenzivnije je 15 do 20 godina nakon izgradnje {umske ceste (slika 4). Razmak izme|u kro{anja stabala trebao bi se smanjiti na prosje~nu {irinu planuma {umske ceste (3,63 m) nakon 18 godina, a u idu}ih 10 godina udaljenost }e izme|u kro{anja stabala iznositi 50 % prosje~ne {irine planuma {umske ceste (tablica 4). U nastavku istraìvanja ispitali smo da li bilo koji drugi ~imbenik, osim starosti {umske ceste, utje~e na razmak izme|u kro{anja stabala. Pri modeliranju smo se koristili binarnom multivarijantnom logisti~kom regresijom. Izmjerenu smo udaljenost izme|u kro{anja za 90 popre~nih profila {umske ceste podijelili u dvije grupe: udaljenost do 2 m (u ovu grupu je u{lo 58 popre~nih profila) i udaljenost preko 2 m (32 profila). Starost {umskih cesta podijeljena je tako|er u dvije grupe: do 15 godina starosti i preko 15 godina starosti (u drugoj su grupi objedinjene dvije {umske ceste, ona starosti 35 i ona starosti 50 godina, {to je bilo mogu}e u~initi zbog statisti~ki nezna~ajnih razlika). Modeliranje je rezultiralo tvrdnjom kako uz starost {umske ceste na {irinu svijetloga otvora iznad {umske ceste utje~e samo razmak izme|u debala rubnih stabala. U tablici 2 prikazan je o~ekivani razmak ve}i od 2 m izme|u kro{anja stabala na {umskoj cesti starosti do 15 godina. O~ekivani razmak ve}i od 2 m izme|u kro{anja stabala na popre~nim profilima {umske ceste starosti do 15 godina 90 je puta ve}i nego na popre~nim profilima starije {umske ceste. S druge je strane o~ekivani razmak izme|u kro{anja stabala ve}i od 2 m kod {umskih cesta iste starosti 1,5 puta ve}i za svaki dodatni metar udaljenosti izme|u debala rubnih stabala. Pilot-projekt potvrdio je pretpostavku da udaljenost izme|u debala, udaljenost izme|u kro{anja te {irina tijela i {irina planuma {umske ceste padaju s godinama starosti {umske ceste. To zna~i da su starije {umske ceste bolje uklopljene u svoje okruènje. Rezultat je o~ekivan u pogledu udaljenosti izme|u debala i kro{anja rubnih stabala, dok bi, teoretski, {irina tijela i {irina planuma {umske ceste trebali ostati nepromijenjeni tijekom vremena uz dobro odràvanje. Strojna izgradnja {umskih cesta u Sloveniji uvedena je kasnih pedesetih i ranih {ezdesetih godina pro{loga stolje}a, dakle nakon izgradnje najstarije {umske ceste istraìvane u ovom radu. Zbog te{kih radnih uvjeta pri ru~noj se gradnji {umskih cesta izvodio uì planum {umske ceste (a time i uè tijelo {umske ceste) u usporedbi sa strojnim na~inom izvedbe. Moè se zaklju~iti da intenzitet zatvaranja sklopa iznad {umske ceste ovisi o vremenu izgradnje (starosti {umske ceste), {irini pojasa {ume koji je posje~en zbog izgradnje {umske ceste i {irini tijela {umske ceste. Neke razlike u popunjavanju svijetloga otvora mogu se objasniti i vrstom drve}a te godinama starosti sastojine. Ovdje je istraìvana bukva, ali se bitno razli~iti rezultati mogu o~ekivati u istraìvanju ~etinja~a te drugih vrsta lista~a. Mla|a su stabla, bez obzira na vrstu, sposobnija razviti svoju kro{nju brè te posljedi~no brè popuniti svijetli otvor od starijih stabala. U istraìvanim sastojinskim i stani{nim uvjetima svijetli se otvor iznad {umske ceste uistinu brzo zatvarao kro{njama rubnih bukovih stabala. Zatvaranje sklopa kro{anja iznad {umske ceste ima ove pozitivne estetske, ekolo{ke, tehni~ke i komercijalne u~inke: Þ Þ Þ Þ

{umska cesta postaje neraskidiva, estetski oblikovana, cjelina sa {umom posljedice i {tete za {umski ekosustav nastale izgradnjom {umske ceste umanjene su i ispravljene sastojina i stani{te poreme}eno izgradnjom {umske ceste postaju puno stabilniji zeleni zastor kro{anja bukovih stabala {titi {umsku cestu od izravnoga sun~anoga zra~enja i od {tetnih posljedica oborina.

Croat. j. for. eng. 29(2008)1

I. POTO^NIK et al.

Filling in the Clearance of a Forest Road Cross-Section in Beech Forest (53–62)

S gledi{ta provo|enja {to u~inkovitijega odràvanja {umskih cesta mogu se donijeti ovi zaklju~ci: Þ sa staro{}u {umske ceste preko 30 godina moè se o~ekivati smanjenje tro{kova odràvanja za 50 %, {to je uzrokovano sprje~avanjem izravnoga pridolaska oborina na povr{inu {umske ceste Þ {umski pojas koji je potrebno posje}i pri izgradnji {umske ceste treba biti {to je mogu}e uì kako bi rubna stabla {to prije zatvorila sklop iznad {umske ceste Þ sje~a stabala pri izgradnji {umske ceste (pripremni radovi), posebno na nasipnoj strani mje{ovitoga profila (zasjeka), trebala bi biti {to manja radi pove}anja stabilnosti {umske ceste i skra}ivanja vremena potrebnoga za sklapanje kro{anja stabala iznad {umske ceste (Poto~nik i dr. 2005) Þ rubna bjelogori~na stabla uz {umske ceste trebala bi imati simetri~nu kro{nju (jer asimetri~na kro{nja smanjuje mehani~ku stabilnost stabla), {to se postiè odgovaraju}om njegom stabala Þ sve bi navedene mjere u potpunosti trebalo primijeniti u za{ti}enim podru~jima prirode (Poto~nik 2006). Klju~ne rije~i: {umska cesta, svijetli otvor {umske ceste, popunjavanje svijetloga otvora, odràvanje {umskih cesta

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

Received (Primljeno): April 08, 2008 Accepted (Prihva}eno): June 10, 2008

Asst. Prof. Tibor Pentek, PhD. e-mail: pentek@sumfak.hr Asst. Prof. Dragutin Pi~man, PhD. e-mail: picman@sumfak.hr Ivica Papa, BSc. e-mail: papa@sumfak.hr Forestry Faculty of Zagreb University Department of Forest Engineering Sveto{imunska 25 HR–10 000 Zagreb CROATIA Croat. j. for. eng. 29(2008)1

Original scientific paper – Izvorni znanstveni rad

Planning and Assessment of Alternative Forest Road and Skidding Networks Akbar Najafi, Hoshang Sobhani, Arastou Saeed, Majid Makhdom, Mohammadreza Marvi Mohajer Abstract – Nacrtak Road construction and harvesting operation have always been the two most expensive activities in forestry. The aim of this paper is to define a method to evaluate the forest road network variants. For this purpose,, the forest was sampled with the use of a systematic grid of 150 m by 200 m spacing by an inventory group. In each grid point the terrain conditions and stand data collected by the inventory group were analyzed using GIS. The forest was evaluated and maps of »Forest Potential for Road Construction« (FPRC) and »Forest Capacity for Harvesting« (FCH) were prepared, and then the first network was designed and costs were calculated. The skidding costs of each cross point were calculated for 3 types of extraction machines (two skidders and one farm tractor) and considering the results some better networks were designed. Finally a network was accepted in which not only the environmental impacts were decreased, but the costs of road network and skidding were minimized, so that the most suitable place for each skidder was determined and presented on a map using Linear Programming (LP). Keywords: forest road density, road network analysis, linear programming, road construction cost, skidding cost, Iranian forests

1. Introduction and research problems – Uvod i problematika istra`ivanja For decades forest engineers have tried to define methods for minimizing the sum of road costs (RC) and log extraction costs (LEC) because road construction and harvesting have always been the two most expensive activities in forestry. The planning of a road network to access timber harvest site is a difficult and time–consuming task (Murray 1998). Further more, the viability and profitability of operational forest management plans are very much influenced by road building and maintenance cost as well as the road network structure (Kirby et al. 1986). Recent research has discussed the need for tools and techniques to assist in the development of road network systems in forest management, particularly within the context of accessing timber harvest site (Heinimann et al. 2003, Dean 1997, Newnham 1995, Lui and Sessions 1993). The basic problem is to build a road network in a forested region that provides access to identified harvesting site and minimizes overall road building (Kirby et al. 1986). Croat. j. for. eng. 29(2008)1

Computer and Linear Programming are the two means that help scientists to solve the problem and many researchers developed a variety of methods to determine the optimum forest road density. To reach this goal some heuristic methods have been developed (Akay and Sessions 2005, Tan 1999, Dahlin and Fredriksson 1995). Optimization models for forest tactical planning problems are notoriously difficult to solve, hence various heuristic strategies and specialized solution procedures have been developed (Richards and Gunn 2000) and so some researchers have used a combination of a heuristic method and LP for optimum network development (Tan 1999, Dean 1997, Newnham 1995). Some researchers have used DEM or systematic grids or both of them to identify feasible road segments. Liu and Sessions (1993) have evaluated the fixed and variable costs and the optimum set of road segments to be used, while others have enlarged the link pattern to 24 links, 8 to the nearest and 16 to the second nearest neighbour node. Branch evaluation, a heuristic method developed by Dean (1997), is an automated method of develop-

A. NAJAFI et al.

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

ing a road network designed to access any number of potential harvesting sites while minimizing the overall cost of the network. Both models by Liu and Sessions (1993) and Dean (1997) have the ability to solve multiple target access problems with designated targets and predefined cost matrix or link variables. In some cases, however, especially in developing skid trails or temporary access roads for timber harvest, the ending location of each road branch is not always fixed. Anderson and Nelson (2004) have developed a computer algorithm to generate road networks under a variety of assumptions related to road design standards. This method does not create an optimized road network, but rather mimics the procedure a professional might use when projecting roads by hand. Chung et al. (2001) described a methodology for optimizing cable logging layout using a digital terrain model (DTM). The methodology formulates a cable logging layout as a network problem. Each grid cell containing timber volume to be harvested is identified as an individual entry node of the network. To calculate the network efficiency and total cost (road network and skidding cost), Pentek et al. (2005) considered an existing network. Then they measured the distance between the centre of each compartment and the network, and then they calculated the costs and applied a mathematical function to determine buffer and dead zone of the road network.

2. Aim of research – Cilj istra`ivanja The aim of this study was to define a method to evaluate forest road network and related skid trails efficiency from the cost point of view. We hypothesized that: Þ forest road networks could be evaluated by GIS and a dense grid field observation, Þ skidding cost is reduced using Linear Programming.

Field trips and soil studies showed that the southern parts with a surface of 1000 hectares contained shallow soil and hence decision was made to designate these parts as conservation areas. There were two alternatives for choosing the network entry points, east and north. The forest network could be connected to a public road in East or to another district network in North. The logs were decked on the road banks and there was no place constructed for landing, so the landing construction cost would be zero. Ground skidding is the dominant method of harvesting and accounts for approximately 60% of the log volume in these mountainous uneven aged hardwood forests. The following extraction machines are used for primary transport: Timberjack 450C, HSM 904 and Zetor.

4. Materials and Methods – Materijal i metode 4.1 Designing of Forest Evaluation System – Dizajniranje sustava procjene With the use of a systematic grid of 150 m by 200 m spacing, the forest was sampled by an inventory group (Fig. 1). In each grid point the terrain conditions and stand data collected by the inventory group was analyzed using GIS (Fig. 2). The result was named »Evaluation Map« that showed land use of each land unit. The »Map of unstable area« was created and improved through numerous field trips to avoid road construction in these areas. The Map of unstable area and the Evaluation Map were overlaid and this

3. Study site – Podru~je istra`ivanja The study site is located on 51°15’ E and 36°30’ N in the North of Iran. The study area was covered with hardwood of 14 forest types in approximately 2000 hectares. There were a large number of constraints (steep terrain, unstable area, environmental constraints, machines, etc.) for forest road network designing and construction in the area. There was a permanent stream flooding through the district and the banks of the stream were identified as landslides based on the studies done by geologists through field trips. And so, the construction costs and especially the maintenance costs increased noticeably in this area.

Fig. 1 Inventory grid Slika 1. Plan pokusa Croat. j. for. eng. 29(2008)1

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

A. NAJAFI et al.

Fig. 2 GIS layers used in the designed forest evaluation system Slika 2. Sastavnice GIS-a primijenjenoga u procjeni dizajniranoga sustava procjene resulted in the »Forest Potential for Road Construction Map« (FPRC). To overcome the above shortcoming, the map of »volume per hectare« was created and overlaid on the FPRC map and this resulted in the »Forest Capacity for Harvesting Map« (FCH) – Fig. 3. The map was used as a guide in network designing.

4.2 Road Network Designing and Evaluation – Planiranje mre`e {umskih cesta i procjena kakvo}e Evaluation is necessary to compute road costs (RC) and log extraction costs (LEC) and these data are analyzed so as to find the network which contains Croat. j. for. eng. 29(2008)1

the minimum sum of RC and LEC. The procedure of calculating RC and LEC will be demonstrated below. 4.2.1 Road costs (RC) – Tro{kovi povezani sa {umskim cestama Calculating the benefits of forest roads, where a number of factors must be considered, is a complicated operation (Dahlin and Fredriksson 1995) However, RC generally consists of three elements: Þ depreciation costs, Þ maintenance costs, Þ lost productive area costs. In Iran roads are constructed by constructors and experience shows that they estimate the road

A. NAJAFI et al.

Planning and Assessment of Alternative Forest Road and Skidding Networks (63â&#x20AC;&#x201C;73)

Fig. 3 Forest Capacity for Harvesting Map (FCH)* Slika 3. Potencijal istra`ivanoga podru~ja za pridobivanje drva*

Fig. 4 A schematic map of forest road, skid trails and cells Slika 4. Shematski prikaz obilje`be {umskih cesta, traktorskih putova i vlaka te nositelja informacija 66

Croat. j. for. eng. 29(2008)1

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

construction and maintenance costs according to the average conditions of the forest and hence depreciation and maintenance costs per unit are unchanged within a specific planning area. The stock growth of the surface allocated to the road construction is called the growth of lost productive area. This method calculates the growth value of lost productive area (m3/ha) in each cell, where the road segment may be located. Road costs are calculated as follows: RC = annual depreciation + annual maintenance + value of growth of lost productive area. RC is converted to $/m3 for comparison with skidding costs and so RC per m3 = RC / total annual allowable cut. 4.2.2 Log Extraction Costs (LEC) – Tro{kovi privla~enja drva Centre of all cells were accessible by skid trail and more cells were connected to a skid trail. There are three costs involved in the calculation of LEC: Þ landing costs, Þ skid trail construction costs, Þ skidding costs. The maximum allowable gradient in every access spur is 25%, which is limited based on erosion hazards. If skid trails are constructed on slopes of more than 25%, the costs are estimated as construction costs, and if skid trails are cleared by skidder, the costs are considered as skidding cost. The Division of Forest Techniques of the Austrian Federal Forest Research Centre of the Federal Ministry of Agriculture Forestry Environment and Water Management (FBVA 2000) used to determine the skidding costs, depending on the annual utilization. Time study of HSM 904 has been done and production per hour has been calculated by some authors because the related data were not available. Evaluation of skidders shows that the skidding time is a function of many independent variables such as total volume of logs (m3) in each travel, skid trail gradient, skidding distance, etc. These variables are estimated in each cross point of the grid, using

A. NAJAFI et al.

GIS and inventory data and the related skidding costs in each cell and trail are calculated. For example, Fig. 5 shows ten cells (E4 … E13) which are connected by skid trail with a slope gradient of more than 25%, so an access spur should be constructed. The skidding distance between E4 and the road was measured on the access spur. The number and volume of logs in each cycle are also estimated using the inventory data. All independent variables that affect the skidders’ productivity are explicitly calculated by use of skidding distance, as explained above. Then the skidding time in each cell is calculated and converted to skidding costs based on the cubic meter considering the skidders’ hour productivity cost. For instance the mathematical model of productivity for Timberjack 450C calculated by time study was: T = 4.61 + (0.073 ⋅ V) + (1.152 ⋅ N) + (0.0162 ⋅ D) (1) where: T time of travel to extract all harvested timbers from cell E4, min V total volume of logs in each travel, m3 N number of log in each travel D skidding distance, m If the calculated skidding time for total harvested timber in E4 is 114.07 minutes and its hour cost is $36.37, the total skidding cost will be $69.14 and if the annual allowable cut in E4 is 7.7 m3 per cell, the skidding cost based on cubic meters will be $8.98. As mentioned above the access spur should be constructed between E4 and E10. The distance between E9 and E10 to be constructed is 80 meters through which the logs of 6 cells are carried and if the construction costs of the access spur per kilometre is $341, this cost in E9 will be $4.55. The total distance between E4 and E9 is 790 m but only 200 m of it should be constructed, so the share of E4 considering the above mentioned method will be 49.33 m and the access spur construction cost in E4 is $16.82 The construction cost of the access spur based on cubic meter is $2.18. The landing construction cost is zero, and the log extraction cost (LEC) will be $11.16. The sum of LEC of all cells divided by the sum of annual allowable cut will result in the LEC based on cubic meter. 4.2.3 Machines Allocation – Planiranje rada {umskih traktora

Fig. 5 Access spur Slika 5. Pristupni pravac (traktorski put) Croat. j. for. eng. 29(2008)1

Although RC and LEC are calculated in each cell and can be compared in a Forest district, dividing a

A. NAJAFI et al.

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

The first constraint shows that the total allowable cut is equal to the sum of allowable cuts of all cells. The second constraint emphasizes that total skidding time of each machine cannot be more than total allowed time. Total time is determined by the forest manager with regard to their facilities. Tji is estimated according to PMH of the machines and some other independent variables such as distance to road, trail gradient, etc. in each cell of the trail. Now LEC and RC can be summed and the sum can be compared with other districts’ costs because they are both calculated based on cubic meter. When each network is designed, these elements are also calculated and compared. The costs of each road segment and skid trail and their effects on the total costs were evaluated separately and the results were used to design the next network. In this way the best network was designed with the minimum sum of LEC and RC.

district based on 3 ha cells does not seem to be practical because a very complicated procedure is needed to manage them. To solve the problem, the costs are calculated and compared in skid trails. Productive machine hours (PMH) of various skidders in an identical skid trail were not the same because of the difference among the engine powers of skidders and, as the time allocated to each skidder by the company were limited, we were not able to allocate a skidder an unlimited working time in an area, that is we had to apply a combination of skidders in the district. We used LP to show how much log each skidder in each skid trail should extract to minimize the total skidding costs. The model could be described as: Min∑ ∑ Cij Xij

(2)

i ∈ I j∈ J

∑∑X i ∈ I j∈ J

∑∑t j∈ J i ∈ I

= ∑ Vi ; ∀ ij

(3)

Xij ≤ ∑ Tj ; ∀ ij

(4)

i ∈I

j∈ J

5. Research results and discussion – Rezultati istra`ivanja i rasprava 5.1 Road network »A« – Mre`a {umskih cesta »A«

where: I J Vi Cij

A major segment of the road network »A«, designed by the Forest Company, is located in unstable area so the forest damage, erosion and annual maintenance costs would certainly be much higher than the average. The segment is close to a permanent stream and the erosion caused by unstable road will have negative impact on the stream ecosystem. The network can neither be acceptable from the economical point of view, because of high annual maintenance costs. To evaluate the road network, the annual allowable cut was determined according to FCH map (low volume and sensitive areas were eliminated). Table 1 shows the results.

number of trails, number of skidders, total annual allowed cut on the trail (i), m3 skidding costs on the trail (i) extracted by skidder (j), $/m3 Xij volume of logs extracted by skidder (j) on the trail (i), m3 tji log extraction time of skidder (j) on the trail (i), min/m3 Tj time allocated for skidder (j) by the ∑ j∈ J company, hour

Table 1 Road costs (RC), Log extraction costs (LEC) and Total costs for different alternatives of forest road networks Tablica 1. Tro{kovi povezani sa {umskim cestama, tro{kovi privla~enja drva te ukupni tro{kovi pridobivanja drva za razli~ite ina~ice dizajniranih mre`a {umskih cesta Road network variant Ina~ica mre`e {umskih cesta

Log Extraction Costs – Tro{kovi privla~enja drva Zetor

Timberjack 450C

HSM 904

LP Solution Ina~ica dobivena linearnim programiranjem

Road Costs Tro{kovi povezani sa {umskim cestama

Total Costs Ukupni tro{kovi pridobivanja drva

– – – 2.94

3.87 6.42 4.74 3.25

8.32 7.70 6.74 6.19

$/m3 A B C D

3.83 1.90 2.00 2.95

2.60 4.12 2.67 3.44

4.01 6.98 4.74 5.55

Croat. j. for. eng. 29(2008)1

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

A. NAJAFI et al.

Fig. 6 Designed alternatives of forest road network Slika 6. Dizajnirane ina~ice mre`e {umskih cesta 5.2 Road network »B« – Mre`a {umskih cesta »B« Road network »B« was the first road network designed in this research (Fig. 6). Our objective was to avoid unstable areas and to access high stock land based on quality and quantity investigations. Evaluations showed that total costs were decreased but it could not be a candidate for the best road network. There was not any road segment on the north part of the forest, so it could not be acceptable regarding other goals of forest management such as forest protection. On the other hand log extraction from the north part required the design and construction of many skid trails, which had to cut the stream in many places and could be harmful for the fish.

5.3 Road network »C« – Mre`a {umskih cesta »C« The stream and the unstable area were located in the middle of the district and divided it into two Croat. j. for. eng. 29(2008)1

parts and constructing a road to connect the two parts of the district was impossible so, two segments were added to network »B«. The end part of the south segment was eliminated to prevent RC increment and a new road network called network »C« was created. The new segments were connected to the road network of the northern neighbouring district (Fig. 6). Results (Table 1) showed that total costs were decreased.

5.4 Road network »D« – Mre`a {umskih cesta »D« The eastern segment of the north road and a small ending part of the south road were eliminated (Fig. 6) and the changed network (i.e. »D«) was evaluated. The result showed that total RC and LEC in the new road network were lower than in others. This network was also unacceptable regarding environmental and forest management aspects.

A. NAJAFI et al.

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

Fig. 7 Range of Road costs (RC), Log Extraction Costs (LEC) and total costs for different alternatives of forest road networks Slika 7. Prikaz tro{kova povezanih sa {umskim cestama, tro{kova privla~enja drva te ukupnih tro{kova pridobivanja drva za razli~ite ina~ice dizajniranih mre`a {umskih cesta

Fig. 8 Extraction machine operations area Slika 8. Podru~je rada pojedinoga stroja pri privla~enju drva

The next step was to design new road networks by adding some feasible segments or omitting some other segments to determine the lowest total costs

(results not shown). These issues were considered in the network »D« after their evaluations (Fig. 7) and it was determined to be the best road network. Croat. j. for. eng. 29(2008)1

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

Figure 7 shows that factors such as road density or road network efficiency cannot determine the optimum road network density in the mountainous uneven age forests and factors such as stock distribution, slope and unstable area are more important and effective and they can be measured by the method developed by this research. This method can evaluate the road networks easily, precisely and in detail. Many LEC lines (Fig. 7) can be drawn for a single road density and the gradient of LEC depends not only on road density but also on road distribution. As the inventory groups in Iran gathered most of the above data, they incurred no additional costs. Doing the same study in other forests with different conditions can be beneficial for a better evaluation of this study. Regarding the limitation of skidders power, allowable cut, budget and skidder availability LP was used to increase the work productivity and to decrease the skidding costs and also to determine the skidders spatial allocation. Then a valuable and applicable map was created which showed where and how long each skidder must work (Fig. 8). In the current research not only the road network costs and linkage costs are considered but also the skidding and distribution costs. We developed a method to design the network and its related skid trails manually and – using LP and GIS – have investigated transportation (road network and skid trail) efficiency and total costs. We used a grid point which was overlaid on the forest and our judgment was based on the real data in details. As in this method the forest has been divided into 3 hectare mosaics, data analysis was done in a more explicit manner and also it became possible to calculate the cost of each skid trail or each segment separately. Next skid trail networks were systematically designed and not randomly, so that each network was more complete than the previous network and also closer to the objective of the research. In this way the procedure to reach the best skid trail network has been enhanced, providing the possibility to this method to be applied in vast areas as it saves a lot of time in designing networks. Furthermore data analysis and machine allocation became possible in each skid trail. Machine allocation in each skid trail leads to an optimum combination of skidders. This method can also be used in forests where there are already some built transportation networks (road and skid trail network). This method analysis enables us to add or eliminate some parts of skid trails or roads.

Croat. j. for. eng. 29(2008)1

A. NAJAFI et al.

6. References – Literatura Akay, A. B., Sessions, J., 2005: Applying the Decision Support System, TRACER, to Forest Road Design. Western Journal of Applied Forestry 20(3): 184–191. Anderson, A. E., Nelson, J., 2004: Projecting Vector-based Road Networks with a Shortest Path Algorithm. Canadian Journal of Forest Research 34(7): 1444–1457. Chung, W., Session, J., 2001: Transportation Planning Under Multiple Objectives. The International Mountain Logging and 11th Pacific Northwest Skyline Symposium 2001. Dean, J. D., 1997: Finding Optimal Routes for Networks of Harvest Site Access Roads Using GIS-based Techniques. Canadian Journal of Forest Research 27: 11–22. Dahlin, B., Fredriksson, J., 1995: Computer-Assisted Forest Road Planning. (A Proposed Interactive Model with Special Emphasis on Private Forest Land). Journal of Forest Engineering 6(2): 35–39. Dean, J. D., 1997: Finding Optimal Routes for Networks of Harvest Site Access Roads Using GIS-based Techniques. Can. J. For. Res. 27: 11–22. Heinimann, H. R., Stückelberger, J., Chung, W., 2003: Improving Automatic Grid Cell Based Road Route Location Procedures. Proceedings of Austro2003 – High Tech Forest Operations for Mountainous Terrain, October 5–9, 2003, Schlaegl, Austria, University of Natural Resources and Applied Life Sciences Viena, CD-ROM. Kirby, M., Hager, W., Wong, W., 1986: Simultaneous Planning of Woodland Management and Transportation Alternatives. TIMS Stud. Mnage. Sci. 21: 371–387. Liu, K., Sessions, J., 1993: Preliminary Planning of Road System Using Digital Terrain Models. Journal of Forest Engineering 4(1): 27–32. Murray, T. A., 1998: Rout Planning for Harvest Site Access. Canadian Journal of Forest Research 28(7): 1084–1087. Neve~erel, H., Pentek, T., Pi~man, D., Stanki}, I., 2007: Traffic load of forest roads as a criterion for their categorisation – GIS analysis. Croat. j. for. eng. 28(1): 27–38. Newnham, R. M., 1995: A Tool for Designing Forest Road Network. Journal of Forest Engineering 6(2): 17–26. Pentek,T., Pi~man, D., Poto~nik, I., Dvor{~ak, P., Neve~erel, H., 2005: Analysis of an existing forest road network. Croat. j. for. eng. 26(1): 39–50. Richards, E. W., Gunn, E. A, 2000: A Model and Tabu Search Method to Optimize Stand Harvest and Road Construction Schedules. Forest Science 46(2): 188–203. Tan, J., 1999: Locating Forest Roads by a Spatial and Heuristic Procedure Using Microcomputers. Journal of Forest Engineering 10(2): 91–100.

A. NAJAFI et al.

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

Sa`etak

Planiranje i procjena kakvo}e razli~itih ina~ica mreè {umskih prometnica Tro{kovi povezani sa {umskim cestama (RC) te tro{kovi pridobivanja drva (LEC) oduvijek su, poglavito zbog izgradnje i odràvanja mreè {umskih cesta te privla~enja drva, imali zna~ajan udio u ukupnoj strukturi tro{kova radova u {umarstvu. Planiranje dobre mreè {umskih cesta radi omogu}avanja pristupa {umskomu radili{tu (sje~ini) i smanjivanja srednje udaljenosti privla~enja drva zahtjevan je i vremenski dugotrajan posao (Murray 1998). Odrìvost i isplativost cjelokupnoga operativnoga {umarskoga planiranja i gospodarenja {umom pod sna`nim je utjecajem tro{kova izgradnje i odràvanja mreè {umskih cesta te ovisi o koli~ini i kategoriji {umskih cesta (Kirby i dr. 1986). Najnovija istraìvanja raspravljaju o potrebi primjene novih alata i metoda koji }e pomo}i unaprje|enju postoje}e mreè {umskih cesta radi boljega i u~inkovitijega obavljanja {umskih radova uz poseban naglasak na radove pridobivanja drva (Dean 1997, Newnham 1995, Liu i Sessions 1993). Temeljni je problem kako uz {to manje ukupne cestovne tro{kove izgraditi dovoljno dobru mreù {umskih cesta koja }e osigurati pristup svim, u fazi planiranja determiniranim kao klju~nim, {umskim radili{tima. Cilj je ovoga istraìvanja razviti metodu pomo}u koje }e biti mogu}e optimizirati mreù {umskih prometnica ({umskih cesta i traktorskih putova) s tro{kovnoga gledi{ta. Istraìvanje je temeljeno na dvjema pretpostavkama (hipotezama): Þ optimizacija se mreè {umskih prometnica moè posti}i modeliranjem cestovnih tro{kova i tro{kova pridobivanja drva Þ najbolji se raspored strojeva za privla~enje drva na istraìvanom podru~ju moè odrediti pomo}u linearnoga programiranja (LP). Istraìvanje je provedeno na povr{ini od oko 2000 ha u 14 razli~itih tipova {uma tvrdih lista~a. Terenski su uvjeti, s gledi{ta planiranja, projektiranja i izgradnje najbolje mogu}e mreè {umskih prometnica, bili vrlo zahtjevni. Nagibi su terena vrlo veliki, tlo je nestabilno (klizi{ta) i slabo nosivo, vodotok koji prolazi istraìvanim podru~jem je poplavio, a na samim je obalama potoka zemlji{te klizilo (utvr|eno geolo{kim studijama). Zbog navedenoga tro{kovi su izgradnje i odràvanja {umskih prometnica na istraìvanom podru~ju vrlo visoki. U ovim se prebornim planinskim {umama drvo naj~e{}e privla~i po tlu (60 % drva privu~e se na taj na~in), a pri tome su kori{tena dva zgobna traktora: Timberjack 450C i HSM 904 te jedan prilago|eni poljoprivredni traktor Zetor. Na terenu je postavljena mreà pravilnih ~etverokutnih nositelja informacija dimenzija 150 x 200 m (slika 1). Izra|en je GIS istraìvanoga podru~ja ~ije su sastavnice vidljive na slici 2. Dizajnirane su dvije temeljne karte: karta pogodnosti (potencijala) pojedinoga nositelja informacija za prolazak {umske ceste (FPRC) i karta pogodnosti (potencijala) pojedinoga nositelja informacija za radove na pridobivanju drva (FCH – slika 3) koja prikazuje sje~nu gusto}u u m3/ha. Planiranje mreè {umskih cesta i procjena kakvo}e pojedine ina~ice (izra|ene su ~etiri ina~ice: A, B, C i D) temelji se na minimaliziranju ukupnih tro{kova povezanih sa {umskim cestama (RC) i tro{kova pridobivanja drva (LEC). Cestovni se tro{kovi dijele u tri skupine: tro{ak amortizacije, tro{ak odràvanja i tro{ak gubitka proizvodne {umske povr{ine. Jedini~ni tro{ak povezan sa {umskim cestama izra~unava se pomo}u formule 1. Tro{kovi se pridobivanja drva sastoje od: tro{ka izgradnje pomo}nih stovari{ta, tro{ka izgradnje traktorskih putova i tro{ka privla~enja drva. Pri izra~unu tro{kova privla~enja drva za razli~ite strojeve koji se za navedene radove koriste preuzete su kalkulacije Austrijskoga dràvnoga {umarskoga istraìva~koga centra pri Ministarstvu poljoprivrede, {umarstva i vodnoga gospodarstva Austrije (FBVA 2000). Zbroj svih tro{kova privla~enja drva (po svim nositeljima informacija) podijeljen je ukupnim godi{njim etatom i tako je dobiven jedini~ni tro{ak privla~enja drva po kubnom metru. Kako organizacija rada, ponajprije planiranje strojeva za privla~enje drva, nije prakti~na (zbog kompliciranoga upravljanja) na tako malim povr{inama, godi{nji je etat razdijeljen na traktorske putove (drvo koje gravitira pojedinomu traktorskomu putu) te je za svaki stroj izra~unat tro{ak privla~enja i prona|eno najbolje (najjeftinije) rje{enje.

Croat. j. for. eng. 29(2008)1

Planning and Assessment of Alternative Forest Road and Skidding Networks (63–73)

A. NAJAFI et al.

Dizajniranjem vi{e ina~ica mreè {umskih cesta (slika 6), s tim da su se svakom novom ina~icom nastojali ukloniti prethodno uo~eni nedostaci (~ime se iz ina~ice u ina~icu postizao vi{i stupanj kakvo}e mreè {umskih cesta) ostvarilo se optimalno rje{enje s ina~icom D mreè {umskih cesta (slika 7 i tablica 1). Zbog raznolikosti terena, koli~ine i strukture etata te raspoloìvosti skidera pomo}u linearnoga programiranja (LP) nastojao se pove}ati u~inak (PMH) i ujedno smanjiti tro{kovi privla~enja. Kreirana je i karta kojom je utvr|en prostorni raspored i podru~je rada pojedinoga stroja, odnosno kombinacije strojeva (slika 8). Planiranje {umskih prometnica ({umskih cesta i traktorskih vlaka) opisanom metodom uz primjenu GIS-a dobro je rje{enje pri otvaranju neotvorenih ili slabo otvorenih {umskih podru~ja gdje je, radi postizanja optimalne otvorenosti, potrebno izgraditi ve}u koli~inu {umskih prometnica. Metoda je tako|er primjenjiva u onim {umskim podru~jima u kojima ve} postoji mreà {umske prometne infrastrukture, ali njezin prostorni razmje{taj nije zadovoljavaju}i pa odre|ene sastavnice mreè treba ukloniti, a nove izgraditi. Linearnim se programiranjem osigurava rasporedom strojeva za privla~enje drva bolja organizacija posla i u~inkovitost strojeva. Klju~ne rije~i: gusto}a mreè {umskih cesta, ra{~lamba mreè {umskih prometnica, linearno programiranje, cestovni, tro{kovi pridobivanja drva, iranske {ume

Authors’ addresses – Adresa autorâ: Asst. Prof. Akbar Najafi, PhD. e-mail: a.najafi@modares.ac.ir Tarbiat Modares University Faculty of Natural Resources & Marine sciences P.O.Box 46414–356 Noor, Mazandaran Province IRAN

Received (Primljeno): February 12, 2007 Accepted (Prihva}eno): April 8, 2008 Croat. j. for. eng. 29(2008)1

Assoc. Prof. Hooshang Sobhani, PhD. Asst. Prof. Arastou Saeed, PhD. Prof. Majid Makhdom, PhD. Prof. Mohammadreza MarviMohajer, PhD. Tehran University College of Natural Resources Department of Forestry P.O.Box 31585–4314 Karaj IRAN

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

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre Tibor Pentek, Hrvoje Neve~erel, Tomislav Por{insky, Dragutin Pi~man, Kruno Lepoglavec, Igor Poto~nik Abstract â&#x20AC;&#x201C; Nacrtak Nowadays, under conditions in which forest areas are being opened with increasing number of forest roads due to great need for forest resources, the need for a well-kept system of forest roads becomes a necessary prerequisite in the process of intensification of forest ecosystems management. Cadastre of primary forest traffic infrastructure on the territory of the Republic of Croatia has already been established, and now the need for secondary forest road cadastre arises. Development of secondary forest road cadastre methodology will make a great and important step in inventarisation of the entire secondary forest traffic infrastructure. Although this is a comprehensive and very demanding job, it will result in numerous benefits to forest profession. Once established, a comprehensive forest traffic infrastructure cadastre will be able to easily and relatively quickly supplement newly built forest communications. Based on numerous analyses of primary and secondary forest traffic infrastructure, forest experts can enhance the quality of forest communications spatial configuration and direct the process of further primary and secondary forest opening up with the final goal of establishing the optimal primary and secondary forest network from technical-technological, economic, environmental-ecological and sociological-aesthetic point of view. Key words: cadastre of forest traffic infrastructure, secondary forest roads, secondary relative openness, GIS

1. Introduction â&#x20AC;&#x201C; Uvod Under the conditions of contemporary intensive forest resources management, there is an increased need for forest communications, both primary and secondary ones. The construction of a larger number of primary forest roads contributes to a decrease of the mean skidding distance, which results in smaller costs of timber harvesting. On the other hand, technology selection for timber harvesting (forwarders, skidders, tractors) has its effect on the increased number of newly built skid roads. The newly arisen situation, where an increased number of forest communications is met, requires taking of an inventory of forest traffic infrastructure and establishment of a complete forest communications cadastre. Establishment of the forest traffic infrastructure cadastre enabled precise and detailed insight in the existing traffic resources of the forest area (Pentek et al. 2007), analysis of the existing conditions in the Croat. j. for. eng. 29(2008)1

primary and secondary openness of forests, observng of possible deficiencies and shortcomings (Pentek et al. 2005a), quality primary and secondary opening of unopened and insufficiently opened areas (Pi~man et al. 2006a), planning (Pentek et al. 2005b) and cost-control in the construction and maintenance of forest roads (Pentek et al. 2007), planning and cost-control in the construction and repair of skid roads (Pentek 2002), preparing of harvesting plan and similar. If there is no such a cadastre or if it is incomplete (Pentek et al. 2003), then it is necessary to carry out field inventory of the primary (all forest roads and public roads that can be used during forest management works) and secondary forest traffic infrastructure (skid roads and skid trails). According to Pi~man et al. (2006b) the preparing of secondary forest communications cadastre has not taken hold in the practice yet, as opposed to the primary forest traffic infrastructure cadastre. How-

T. PENTEK et al.

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

ever, the necessity of starting such a process, under the unquestionable profits to be achieved upon completion of the job, is only a question of days. During the construction of new forest communications, the existing cadastre should be systematically updated in order to dispose with the actual conditions of the forest traffic infrastructure.

2. Research issues – Problematika istra`ivanja Secondary forest roads ([iki} et al. 1989) are constructed facilities, which are occasionally used for performing the tasks provided by the Management Plan. They are, primarily, intended for skidding or forwarding operations. Secondary forest traffic infrastructure may be divided into skid roads and skid trails. Skid roads are constructed facilities, in which earth-moving works are present, whereby the upper layer is missing. Skid trails are temporary constructed facilities, obtained by intersection through the forest and by successive tractor passing over the same track. After they have completed their function, the forest takes them over again. The term »skid road« dates back to the 19th century, when it referred to a corduroy road made of logs, used to skid or drag logs through woods and bog (Turner 1986). The existence of a secondary forest road network is a precondition for timber transport in hilly, highland, upland and mountain forests. Under extremely heavy terrain conditions, secondary forest roads most often have to be constructed (skid roads), and more rarely they arise by multiple tractor passing over the same track (skid trails). Without well prepared skid roads, operating on such terrains would be almost impossible for the ground moving machines used for timber extraction (cable skidders). The density of the secondary forest road network and their arrangement depend on many factors (Neve~erel et al. 2007): Þ Technical means used in timber harvesting, Þ Configuration, Þ Terrain rockiness, Þ Number of trees on a specific surface, Þ Dimensions of trees, Þ Position of the main truck forest road, to which timber is extracted. Rebula (1983) comes to the conclusion that the secondary forest communications openness, i.e. the necessary density of skid roads and trails amounts to 250–300 m/ha in young forest stands, while in the

older forest stands, where the tree distance is larger, it amounts to 100–180 m/ha. For the area of selective forests in Gorski Kotar, an optimal density of secondary forest traffic infrastructure of 150 m/ha (Zdjelar 1990) was established. Depending on their horizontal (position) and vertical (altitude) route development, the following skid roads can be distinguished: Þ Lowlands skid roads, Þ Valley skid roads, Þ Hillside, and Þ Ridge skid roads. When selecting the skid road route, efforts are made to use the gravitational force in the direction of timber extraction. In downhill travel, the tractor is far less burdened, faster movement is possible, and skidding of a larger load quantity is possible. The result is the reduction of timber transport costs. Uphill skidding of timber is not justified, if the possibility for downhill skidding of timber exists. A skid road adjusts itself to terrain configuration to a higher extent than a primary forest road, and the width of the driving surface on a skid road is less than on the truck forest road. Consequently, it is understandable that during the construction of skid roads, the scope of the earth-moving works is less than during the construction of truck forest roads. According to Jeli~i} (1983) 10 to 30 times less financial funds are necessary for the construction of the same length of a skid road under similar terrain conditions than for the construction of a primary forest road. The first step that should be made, before the beginning of the opening up procedure for a certain forest area with secondary forest roads, is the analysis of the existing secondary forest road network. In order to perform the analysis procedure, the cadastre of secondary forest traffic infrastructure should be prepared in digital format. According to Pi~man et al. (2006b), the secondary forest traffic infrastructure is categorized in the following manner: Þ Skid roads/trails of the I order – all skid roads/trails diverging from the public or primary forest roads, Þ Skid roads/trails of the II order – all skid roads/trails diverging from skid roads/trails of the I order, Þ Skid roads/trails of the III order – all skid roads/trails diverging from skid roads/trails of the II order, Croat. j. for. eng. 29(2008)1

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

Þ Skid roads/trails of the IV order – all skid roads/trails diverging from skid roads/trails of the III order.

The data obtained were processed by the program package GPS Pathfinder Office 2.80., and then drawn in on the previously prepared basis. 3.2.3

3. Goals and Research Methods – Cilj i metode istra`ivanja 3.1 Goal of research – Cilj istra`ivanja The goals of this research are defined by the following phases of work: Þ Establishment of GIS for the research area, Þ Preparing the cadastre of secondary forest roads, Þ Defining the criteria for categorization of secondary forest traffic infrastructure, Þ Analysis of secondary relative openness.

3.2 Method of research – Metode istra`ivanja 3.2.1

Establishment of GIS for the research area – Uspostava GIS-a istra`ivanoga podru~ja

For the management unit of Veprina~ke {ume, Forestry Office Opatija, a geographic information system (GIS) was established in such a manner that layered maps to the scale of 1: 5000 were scanned, whereby data processing and digitalization of the management division (borders of management unit, compartment and subcompartment) was performed. Then the physical units of a subcompartment, as the smallest accounting forestry units, were joined with the data from the Management Plan. The processing was performed by ArcGIS program package. 3.2.2

Preparing the cadastre of secondary forest roads – Formiranje katastra sekundarnih {umskih prometnica

The whole job of preparing the secondary forest road cadastre is divided in two basic parts: the first one includes terrain measurement for the required data, and the second part includes computer data processing (data entry, transformation, preparation for drawing on forest-management maps in digital format). During the survey of secondary forest roads, we have used a contemporary method of work – GPS device, Trimble, GeoExplorer 3 with the use of an internal antenna, whereby the recording interval was 5 seconds. The surveying was performed by the so-called snap-back method, in which the surveying is performed by walking in both directions, which enables a better fitting of data in the space. Croat. j. for. eng. 29(2008)1

T. PENTEK et al.

Defining the criteria for categorization – Utvr|ivanje kriterija za kategorizaciju

The prerequisite for defining the categorization criteria was a way of separating certain secondary forest roads. For each secondary forest roads, a specific sequence of opening up was provided for each area. 3.2.4

Analysis of secondary relative openness (method of bordered surfaces) – Analiza sekundarne relativne otvorenosti (metoda ome|enih povr{ina)

During evaluation of and commenting on the secondary relative openness, the modified evaluation system of primary relative openness will be used (Pentek 2002). The modified evaluation system of secondary relative openness looks as follows: Þ Up to 60 % – insufficient openness (1), Þ From 60 to 70 % – poor openness (2), Þ From 70 to 80 % – barely good openness (3), Þ From 80 to 90 % – very good openness (4) and Þ Over 90 % – excellent openness (5)

4. Area of research – Podru~je istra`ivanja Compartments of the management unit Veprina~ke {ume, situated between 45°20’ and 45°24’ of the northern geographic latitude, and 14°11’ and 14°16’ of the eastern geographic longitude, have been selected as the area of research (Fig. 1). The total sur-

Fig. 1 Management unit Veprina~ke {ume Slika 1. Gospodarska jedinica Veprina~ke {ume 77

T. PENTEK et al.

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

face of the management unit is 1,950.87 ha, of which 1,899.23 ha is overgrown, 43.36 ha productive, not overgrown, 3.12 ha unproductive, not overgrown, and 5.16 ha unfruitful. During felling and processing, half-tree-length method is used. Timber is extracted by skidders with cable winches. This method of primary timber transportation requires a well-developed secondary openness. Due to the terrain configuration and a developed orography, skid roads have to be constructed. The scaling of roundwood is performed at the felling site. The basic characteristics of forest opening-up and timber harvesting are steep and orographically diverse mountainous terrain, richness of the karst relief phenomena, shallow soils, rocky bedrock and heavy construction material categories. The average slope inclination of the terrain ranges between 5 and 30°. The mentioned features indicate the need for a good primary and secondary openness. The annual cut (8,100 m3/year) is of a very good quality, and the main economic species is beech. The primary forest openness amounts to 16.78 m/ha, and the secondary forest openness is 101.94 m/ha.

5. Results of research – Rezultati istra`ivanja 5.1 Methodology for development of secondary forest road cadastre – Metodologija uspostave katastra sekundarnih {umskih prometnica Determination of cadastre numbers for the secondary forest roads is performed according to a unified system, while observing the management unit as a whole. The marking of the secondary forest roads (Fig. 2 and Fig. 3) is performed by combining the management unit code, capital letters and Arabic numerals (they represent the significance and the sequence of the traffic route).

Fig. 2 Cadastre code of secondary forest roads Slika 2. Katastarska oznaka sekundarnih {umskih prometnica The first marking is an Arabic numeral (three digits) and represents the number of the management unit, given to it in the management division of the forest/management area of Croatia. The second marking is an Arabic numeral and represents the number of the compartment within the management unit. The third marking is a capital letter, by which the manner of construction of the secondary forest road is defined: C – skid road, D – skid trail and an Arabic numeral (1–4), which represents the marking of the sequence of opening-up. The fourth marking (two digits) defines the sequence of the secondary road route in the management unit. The sequence of the secondary road is determined separately for the manner of construction, and for every secondary road category. The secondary forest traffic infrastructure is categorized observing the principle of the opening sequence, i.e. separation of lower order skid roads/ tracks from the higher order skid roads/tracks.

Table 1 Analyses results of secondary relative openness in the selected subcompartments Tablica 1. Rezultati analize sekundarne relativne otvorenosti u odabranim odsjecima Rope length Duljina u`eta

Corrected rope length Korigirana duljina u`eta m

Open surface Otvorena povr{ina

Relative openness Relativna otvorenost

40 50 60

Evaluation of relative openness Ocjena relativne otvorenosti %

78.38

64.11

95.89

78.42

107.82

88.18

115.24

94.25

Croat. j. for. eng. 29(2008)1

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

T. PENTEK et al.

Fig. 3 Cadastre of secondary forest roads – scheme Slika 3. Katastar sekundarnih {umskih prometnica – shema 5.2 Analysis of secondary relative openness – Ra{~lamba sekundarne relativne otvorenosti During the analysis of the secondary relative openness, four different values of approach to the forest area were selected – 30, 40, 50 and 60 meters, which represent the length of the skidder winch rope. For the purpose of this research, inclinations of skid roads were established, as well as inclinations perpendicular to skid roads, which represent the direction of the rope extraction and winching. Terrain configuration with surface obstacles and spatial arrangement of trees in the forest stand were also considered. Going around the horizontal obstacles, and the difference in inclination in relation to the place of winching shortens the length of the winch rope by 20%, whereby the distances of the approach to the forest area are also changed. Taking into consideration the totally open area (Fig. 4) of single variants in relation to the total area, the values of average secondary relative openness were obtained (Table 1).

6. Discussion – Rasprava In Croatia, timber harvesting is performed on the ground, by movable machinery, while cable-yarders Croat. j. for. eng. 29(2008)1

are very rare, and helicopters and other forms of aerial transport have not gained any broader use. Because of such an orientation of the forestry profession, good primary openness is necessary, as well as an especially good secondary openness. Skidders are the most frequent means of work in timber harvesting, which may be explained by the conditions of work and costs. The obtained results are quality information, suggesting which length of winch rope should be chosen, and how many new skid roads should be constructed, so as to achieve the maximum openness. In the selected area, the length of the winch rope of 40 m was selected because of terrain configuration, horizontal obstacles, and humanization of forest work. For the mentioned variant, relative openness should be increased by 11.58% for an excellent openness (21.58% for a maximum openness), which would amount to 23.03 km of newly constructed skid roads on the level of the management unit. The calculation is based on the optimal planning of new skid roads, where the length of newly constructed skid roads would definitely increase because of inevitable multiple opened areas. The end result would certainly be within the limits of optimal density of secondary forest traffic infrastructure, as established by Rebula (1983) and Zdjelar (1990).

T. PENTEK et al.

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75â&#x20AC;&#x201C;83)

Fig. 4 Analyses of secondary relative openness in the selected subcompartments Slika 4. Ra{~lamba sekundarne relativne otvorenosti u odabranim odsjecima 80

Croat. j. for. eng. 29(2008)1

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

7. Conclusions – Zaklju~ci Terrain measurement of the secondary forest roads by a GPS receiver, with the use of the snap-back method of surveying, represents a sufficiently quick and exact method for the establishment of the secondary forest traffic infrastructure cadastre (for drawing in of secondary forest roads in digital maps up to 1:5,000). The methodology for preparing the secondary forest road cadastre is a safeguard for a unified system of coding of the components at the level of the whole Republic of Croatia. The presented pilot project for the formation of the secondary forest road cadastre, in the m.u. of Veprina~ke {ume, is a guidepost for forestry experts as to the manner and methods to be used in preparing the secondary forest road cadastre at the level of the public company of »Hrvatske {ume« d.o.o., Zagreb, as well as of privately owned forests. The benefits of the cadastre are numerous because insight in the actual resources is gained, a quality analysis of the existing secondary forest traffic infrastructure is made possible, and guidelines are set defining further opening-up, as necessary. The cadastre also serves as the basis for the selection of contractors, who dispose of adequate machinery and equipment. The variant selection (40 m) with a smaller length of winch rope contributes to a significant humanization of forest work because in timber harvesting steel ropes are used, and they are heavy to manipulate under severe terrain conditions.

8. References – Literatura Akay, A. B., Sessions, J., 2005: Applying the Decision Support System, TRACER, to Forest Road Design. Western Journal of Applied Forestry 20(3):184–191. Anderson, A. E., Nelson, J., 2004: Projecting Vector-based Road Networks with a Shortest Path Algorithm. Canadian Journal of Forest Research 34(7): 1444–1457. Dahlin, B., Fredriksson, J., 1995: Computer-Assisted Forest Road Planning (A Proposed Interactive Model with Special Emphasis on Private Forest Land). Journal of Forest Engineering 6(2): 35–39. Dean, J. D., 1997: Finding Optimal Routes for Networks of Harvest Site Access Roads Using GIS-based Techniques. Canadian Journal of Forest Research 27: 11–22. Heinimann, H. R., Stückelberger, J., Chung, W., 2003: Improving Automatic Grid Cell Based Road Route Location Procedures. Proceedings of Austro2003 – High Tech Forest Operations for Mountainous Terrain, October 5–9, 2003, Schlaegl, Austria, University of Natural Resources and Applied Life Sciences Viena, CD–ROM. Croat. j. for. eng. 29(2008)1

T. PENTEK et al.

Jeli~i}, V., 1983: [umske ceste i putevi. SIZ odgoja i usmjerenog obrazovanja {umarstva i drvne industrije SRH, Zagreb, 1–193. FAO, 1998: A Manual for the planning, design and construction of forest roads in steep terrain., p. 1–188. Häyrinen, T., 1998: Forest road planning and landscaping. Proceedings of the »Seminar on Environmentally sound forest roads and wood transport«, Sinaia (Romania), June 17–22, 1996, p. 50–61. Heinimann, H. R., 1998: Opening-up planning taking into account environmental and social integrity. Proceedings of the »Seminar on Environmentally sound forest roads and wood transport«, Sinaia (Romania), June 17–22, 1996, p. 62–72. Murray, T. A., 1998: Rout Planning for Harvest Site Access. Canadian Journal of Forest Research 28(7): 1084–1087. Neve~erel, H., Pentek, T., Pi~man, D., Stanki}, I., 2007: Traffic load of forest roads as a criterion for their categorisation – GIS analysis. Croat. j. for. eng. 28(1): 27–38. Pentek, T., 2002: The computer models for forest road network optimisation with regard to the dominant influential factors. Dissertation, Forestry Faculty of Zagreb University, 1 – 271. Pentek, T., Pi~man, D., Krpan, A. P. B, Por{insky, T., 2003: Inventory of primary and secondary forest communications by the use of GPS in Croatian mountainous forest. Proceedings of Austro2003 – High Tech Forest Operations for Mountainous Terrain, October 5–9, 2003, Schlaegl, Austria, University of Natural Resources and Applied Life Sciences Viena, CD-ROM, 1–12. Pentek,T., Pi~man, D., Poto~nik, I., Dvor{~ak, P., Neve~erel, H., 2005a: Analysis of an existing forest road network. Croat. j. for. eng. 26(1): 39–50. Pentek, T., Pi~man, D., Neve~erel, H., 2005b: Planning of forest roads – current status, identifying problems and trends of future activities. Nova meh. {umar. 26: 55–63. Pentek, T., Neve~erel, H., Por{insky, T., Horvat, D., [u{njar, M., Ze~i}, @., 2007: Quality planning of forest road network – precondition of building and maintenance cost rationalisation. Proceedings of Austro 2007 – FORMEC’07 »Meeting the Needs of Tomorrows« Forests: New Developments in Forest Engineering«, October 7–11, 2007, Wien-Heiligenkreuz, Austria, University of Natural Resources and Applied Life Sciences Viena, CD–ROM. Pi~man, D., Pentek, T., Neve~erel, H., 2006a: Forest opening by forest roads – choosing the potential locations of the future forest road routes. Glasnik za {umske pokuse, Special issue 5: 617–633. Pi~man, D., Pentek, T., Neve~erel, H., 2006b: Forest road cadastre – the present condition, the working methodology and obtained uses. Glasnik za {umske pokuse, Special issue 5: 635–646. Poto~nik, I., Pentek, T., Pi~man, D., 2005: Traffic characteristics on forest roads due to forest management. Croat. j. for. eng. 26(1):51–57.

T. PENTEK et al.

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

Rebula, E., 1983: Optimalna gusto}a traktorskih vlaka. Meh. {umar. 8(3–4): 317–321.

agement in the Mountains, Osiach (Austria), 18–24 June 2001, p. 1–5.

Richards, E. W., Gunn, E. A., 2000: A Model and Tabu Search Method to Optimize Stand Harvest and Road Construction Schedules. Forest Science 46(2): 188–203.

Shiba, M., 1992: Optimization of road layout in opening of forest. Proceedings of IUFRO workshop »Computer supported planning of roads and harvesting«, Feldafing, Germany, p. 1–12.

Sessions, J., 1992: Using network analysis for roads and harvest planning. Proceedings of IUFRO workshop »Computer supported planning of roads and harvesting«, Feldafing, Germany, p. 36–41. Sessions, J., Chung, W., Heinimann, H. R., 2001: New algorithms for solving large transportation planning problems, paper on Workshop on New Trends in Wood Harvesting with Cable Systems for Sustainable Forest Man-

[iki}, D., et al., 1989: Tehni~ki uvjeti za gospodarske ceste. Znanstveni savjet za promet JAZU, Zagreb, p. 1–40. Turner, W., 1986: A Clash Over Aid Effort on the First »Skid Row«, The New York Times, December 2, 1986, p. A20. Zdjelar, M., 1990: The methods of building skidding tracks in view of productiveness, work economy, tree damage and worker’s exertion. Meh. {umar. 15(1–2): 3–26.

Sa`etak

Metodologija izrade katastra sekundarne {umske prometne infrastrukture Saètak Pri dana{njem intenzivnom gospodarenju {umama pove}ana je potreba za {umskim prometnicama, i primarnima i sekundarnima. Izgradnja ve}ega broja primarnih {umskih prometnica pridonosi smanjenju srednje udaljenosti privla~enja, {to pak smanjuje tro{kove pridobivanja drva. S druge strane odabir na~ina pridobivanja drva (forvarderi, skideri, traktori) utje~e na pove}anje broja izgra|enih traktorskih putova. Novonastala situacija, u kojoj se susre}e pove}ani broj {umskih prometnica, zahtijeva inventarizaciju {umske prometne infrastrukture te uspostavu potpunoga katastra {umskih prometnica. Ako taj katastar ne postoji ili je nepotpun (Pentek i dr. 2003), tada je potrebno obaviti inventarizaciju primarne (sve {umske ceste i one javne ceste koje se mogu koristiti pri radovima u {umarstvu) i sekundarne {umske prometne infrastrukture (traktorski putovi i traktorske vlake) na terenu. Ciljevi su ovoga istraìvanja definirani sljede}im fazama rada: uspostava GIS-a istraìvanoga podru~ja, stvaranje katastra sekundarnih {umskih prometnica, definiranje kriterija za kategorizaciju sekundarnih {umskih prometnica, analiza sekundarne relativne otvorenosti. Katastarski se brojevi sekundarnih {umskih prometnica (slika 2) utvr|uju po jedinstvenom sustavu promatraju}i gospodarsku jedinicu kao cjelinu. Sekundarne se {umske prometnice ozna~uju kombinacijom {ifre gospodarske jedinice, velikih slova i arapskih brojeva (predstavljaju zna~enje i redoslijed prometnice). Prva je oznaka arapski broj (tri znamenke) i predstavlja broj gospodarske jedinice koji ona ima u gospodarskoj podjeli {umskogospodarskoga podru~ja Hrvatske. Druga je oznaka arapski broj i predstavlja broj odjela unutar gospodarske jedinice. Tre}a je oznaka veliko slovo kojim je definiran na~in izgradnje sekundarne {umske prometnice: C – traktorski put, D – traktorska vlaka i arapski broj (1 – 4) koji predstavlja oznaku redoslijeda odvajanja. êtvrta oznaka (dvije znamenke) odre|uje redoslijed prometnice u gospodarskoj jedinici. Redoslijed prometnice odre|uje se posebno za na~in gradnje te za svaku kategoriju. Sekundarnu {umsku prometnu infrastrukturu kategoriziramo po{tuju}i na~elo redoslijeda otvaranja, odnosno odvajanja traktorskih putova/vlaka nièga reda od traktorskih putova/vlaka vi{ega reda. Pri analizi sekundarne relativne otvorenosti odabrane su ~etiri razli~ite vrijednosti pristupa povr{ini – 30, 40, 50 i 60 metara, koje predstavljaju duljine uèta vitla zglobnoga traktora. Za potrebe ovoga istraìvanja utvr|eni su nagibi traktorskih putova te nagibi okomito na traktorske putove koji predstavljaju smjer izvla~enja uèta i privitlavanja. Konfiguracija terena s povr{inskim preprekama te prostorni razmje{taj stabala u sastojini tako|er su uzeti u obzir. Zaobilaènje horizontalnih prepreka te razlika u nagibu u odnosu na mjesto privitlavanja skra}uje duljinu uèta vitla za 20 %, ~ime se mijenjaju i udaljenosti pristupa {umskoj povr{ini. Uzev{i u obzir ukupno otvorenu povr{inu (slika 4) pojedine ina~ice u odnosu na ukupnu povr{inu, dobivene su srednje sekundarne relativne otvorenosti (tablica 1). U Hrvatskoj se pridobivanje drva obavlja po zemlji kretnim strojevima, dok su {umske ì~are vrlo rijetke, a helikopteri i drugi oblici zra~noga transporta nisu u{li u {iru primjenu. Zbog takve orijentacije {umarske struke

Croat. j. for. eng. 29(2008)1

Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre (75–83)

T. PENTEK et al.

potrebna je dobra primarna otvorenost {umskim prometnicama te osobito dobra sekundarna otvorenost. Zglobni su traktori naj~e{}e sredstvo rada pri pridobivanju drva, {to se moè objasniti uvjetima rada i tro{kovima. Dobiveni su rezultati kvalitetan podatak koji nam sugerira koju duljinu uèta vitla odabrati te koliko je potrebno novih traktorskih putova izgraditi da bi se postigla maksimalna otvorenost. Na odabranom je podru~ju, zbog konfiguracije terena, horizontalnih prepreka te humanizacije rada odabrana duljina uèta vitla od 40 m. Za navedenu je varijantu do odli~ne otvorenosti potrebno za 11,58 % pove}ati relativnu otvorenost, {to bi na razini gospodarske jedinice iznosilo 23,03 km novoizgra|enih traktorskih putova. Izra~un je baziran na optimalnom planiranju novih traktorskih putova, pri ~emu bi se definitivno pove}ala duljina novoizgra|enih traktorskih putova zbog neizbje`nih vi{estruko otvorenih povr{ina. Krajnji bi rezultat zasigurno bio u granicama optimalne gusto}e sekundarnih {umskih prometnica do koje su u svojim istraìvanjima do{li Rebula (1983) i Zdjelar (1990). Temeljem dobivenih rezultata istraìvanja mogu se donijeti ovi zaklju~ci: Þ Metodologija izrade katastra sekundarnih {umskih prometnica jamstvo je ujedna~enoga sustava {ifriranja sastavnica na razini ~itave Republike Hrvatske. Þ Predstavljeni pilot-projekt stvaranja katastra sekundarnih {umskih prometnica u GJ Veprina~ke {ume putokaz je {umarskim stru~njacima na koji na~in te kojim metodama treba pristupiti izradi katastra sekundarnih {umskih prometnica na razini javnoga poduze}a »Hrvatske {ume«, d.o.o. Zagreb, a zatim i u {umama privatnih {umovlasnika. Þ Koristi su od katastra {umske prometne infrastrukture brojne jer se dobiva uvid u stvarne resurse, a omogu}ena je i analiza kakvo}e postoje}e sekundarne {umske prometne infrastrukture te po potrebi definiranje smjernica daljnjega otvaranja. Þ Katastar tako|er sluì kao podloga pri odabiru izvoditelja radova koji imaju odgovaraju}e strojeve i opremu. Þ Izbor varijante (40 m) s manjom duljinom uèta vitla pridonosi znatnoj humanizaciji rada jer se u pridobivanju drva upotrebljava ~eli~na uàd kojom je u zahtjevnim terenskim uvjetima te{ko manipulirati. Klju~ne rije~i: katastar {umskih prometnica, sekundarne {umske prometnice, sekundarna relativna otvorenost, GIS

Authors’ addresses – Adresa autorâ: Asst. Prof. Tibor Pentek, PhD. e-mail: pentek@sumfak.hr Hrvoje Neve~erel, BSc. e-mail: hnevecerel@sumfak.hr Asst. Prof. Tomislav Por{insky, PhD. e-mail: porsinsky@sumfak.hr Asst. Prof. Dragutin Pi~man, PhD. e-mail: picman@sumfak.hr Kruno Lepoglavec, BSc. e-mail: lepoglavec@sumfak.hr Forestry Faculty of Zagreb University Department of Forest Engineering Sveto{imunska 25 HR–10 000 Zagreb CROATIA

Received (Primljeno): March 10, 2008 Accepted (Prihva}eno): May 12, 2008 Croat. j. for. eng. 29(2008)1

Assoc. Prof. Igor Poto~nik, PhD. e-mail: igor.potocnik@bf.uni-lj.si University of Ljubljana, Biotechnical Faculty Department of Forestry and Forest Resources Ve~na pot 83 SI–1000 Ljubljana

Preliminary note â&#x20AC;&#x201C; Prethodno priop}enje

Application of RFID (Radio Frequency Identification) in the Timber Supply Chain Sven Korten, Christian Kaul Abstract â&#x20AC;&#x201C; Nacrtak This paper deals with the use of Radio Frequency Identification (RFID) in the timber supply chain. A distinction is made between the use in motor-manual and highly mechanised timber supply chains. A technical specifications profile is elaborated for both types. As a result it can be summarised that the use of RFID in the timber supply chain is technically feasible. In motor-manual harvesting, it was possible to mark the logs with RFID tags manually. Several tag types could be read out at all stations between the forest and timber industry. A RFID tag-fixing device for harvester heads was developed for the use in highly mechanised timber harvesting. Thus it was possible to attach RFID tags automatically to the logs. Automatic reading of tags during loading processes for skidding and transportation could also be realised. The tag-fixing device and the antennas that were developed are preliminary models. For the practical use of RFID technology in the timber supply chain, further research is necessary for the development tag design and fixing and reading methods. Keywords: timber supply chain, logistics, RFID, tag-fixing device, log identification, technical requirements

1. Introduction â&#x20AC;&#x201C; Uvod In Germany a large number of mainly small and medium sized enterprises take part in the process of timber supply. Consequently, the timber supply chain from the forest to the timber industry is characterised by several fractions between these individual process participants, which results in a high disintegration of material and information flows. However, it is impossible to supervise or control timber logistic processes without knowing how much timber is where at a particular time. Long delivery times and related quality losses become inevitable. Attempts to compensate for information deficits by multiple measurements and inventory of round wood at individual points of the supply chain are costly and often produce errors when entering, assessing and passing on data. As the timber industry often lacks precise knowledge of the actual status of timber supply it holds large stocks of round wood as an intermediary buffer. This buffer ensures the current production but it also creates important costs. For supplies from smallscale private forests, in particular, accounting by maCroat. j. for. eng. 29(2008)1

nufacturer is very expensive because separate storing of timber by individual owner is often required. In Germany, the supervision and control of product and information flows across businesses in the sense of Supply Chain Management (SCM) has recently been considered to have the highest potential for improving the entire supply chain from the forest to the timber industry (Bodelschwingh 2005, Friemel 2005, Holzmann et al. 2006). SCM aims at developing material, information and value flows across interfaces in the best time and at optimum cost. Automatic identification systems can be used in this context to clearly identify the products in the supply chain and, thus, make an essential contribution to the supervision and even control of the product flows. This produces a clear picture as to which products or quantities of products are at a specific place at a particular time. Thus rotation cycles in the timber supply chain can be shortened and quality losses can be reduced. In addition, an overview of the flow of products allows the clients to keep smaller stocks, which results in financial advantages due to reduced capital commitment. Clear identification

S. KORTEN and C. KAUL

of the forest owner at the saw mill allows, besides the accurate accounting of mixed loads from small private forest owners, a proof of origin in line with the Chain of Custody. At present, the logistics sector is paying increasing attention to Radio frequency Identification (RFID) technology, which enters into competition with established systems like barcoding. RFID is less susceptible to dirt and, in addition, it is suited for simultaneous recording of several units (bulk reading). This is what makes its application in forestry particularly interesting. Dykstra et al. (2003) discussed the strengths and weaknesses of various identification technologies for wood tracking. RFID was described as one of the most promising technologies of the future. The LINESET EU research project examined, among other things, the potential use of RFID technology in the timber supply chain, and first tests were carried out (Uusijärvi 2003). Another research project at the FVA Freiburg focussed on the use of RFID tags in motor-manual timber harvesting. Emphasis was placed on the development of a special fixing method to allow an automatic recovery of the tags at the saw mill (Holzmann et al. 2006). The forest consulting enterprise Cambium Forstbetriebe developed a log tracking system for improving internal logistic processes based on the identification of timber logs by means of RFID tags (Friemel 2005). So far it has not been examined whether and under which conditions the application of RFID technology is technically feasible in the entire logistic process. Only a few tags have been tested. That is why there are no concrete specifications for using RFID in the timber supply chain. The following explanations attempt to close this gap.

2. Main characteristics of RFID technology – Glavne zna~ajke RFID-a RFID-technology makes possible an exchange of data via radio between a data carrier (transponder; tag) and a data reader. A transponder is basically composed of a microchip for storing data and a coil used as antenna. If the transponder is in the electromagnetic field of a reader, data can be exchanged between the transponder and the reader. This alternation of transmitting and responding has led to the creation of the name transponder composed of TRANSmitter and resPONDER (Franke und Dangelmaier 2006, Finkenzeller 2002). With regard to energy supply, there is a distinction made between active and passive systems. Active transponders are equipped with an independent energy supply by an integrated battery. This battery provides the energy needed for

Application of RFID in the Timber Supply Chain (85–94)

data storing, data transmission and for other integrated elements such as sensors. Passive transponders use non-volatile data memories which do not need energy for storing data. For the transfer of data they receive the necessary energy without contact by means of an electromagnetic field (inductive coupling) established by the antenna of the reading device. The radio waves are used both for the energy supply and data transmission. Passive transponders on which globally unique identification numbers are stored are often used for controlling logistic processes. By means of these numbers it is possible to clearly identify marked objects. This is mainly done in LF range (Low Frequency / e.g. 125 kHz transponder) with a reading range from a few centimetres to one metre, and in HF range (High Frequency / 13.56 MHz transponder) with a reading range of up to 1.5 metres. In the UHF range (Ultra High Frequency / e.g. 868 MHz transponder) the reading range may reach several metres. Important characteristics for the practical application such as reading range, bulk reading capacity (simultaneous reading of several tags) or sensitivity to liquids and metals are determined decisively by the working frequency of the transponder. HF and UHF transponders allow bulk reading. So far LF transponders have not often been used for bulk reading despite technical feasibility because of the low reading range. Since data transmission is done via electromagnetic waves, liquids and metals have a negative influence on the reading field and, hence also on the reading range of the system which increases with higher working frequency. Consequently, LF transponders are relatively resistant to metals and liquids. Whereas the sensitivity of HF transponders to liquids is low, they can be strongly influenced by a metallic environment. UHF transponder tags are usually strongly influenced by metals and liquids. The full reading range can only be achieved by an optimal, parallel orientation of the tag antenna and the reader antenna. Deviating from the optimal orientation may significantly reduce the possible reading range. In principle, transponder tags can be integrated into almost every case design type. However, depending on the work frequency there are minimum requirements for the dimensions and geometry of the transponder antenna. Round die-casting cases, the so called coins or discs, are frequently used for integrating transponders. Often transponder tag labels (»smart labels«) are made of paper or plastic films. Transponder cards are of a special design, where tags are incorporated into several layers of PVC films. The integration of transponders into glass capsules Croat. j. for. eng. 29(2008)1

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

S. KORTEN and C. KAUL

Fig. 1 Tested RFID tags Slika 1. Testirane RFID plo~ice

allows a particularly slender design. Such glass capsules, when incorporated into plastic nails, for example, can be driven into wood. In comprehensive tests, potentially suitable LF and HF transponder tags of different design types were examined for their durability and weather resistance. Furthermore, some design types were submitted to load tests (Korten and Schneider 2006). Due to their high sensitivity to liquids and metals, transponder tags of the UHF range were eliminated beforehand. Finally, the selected tag types (Fig. 1) were used in several practical trials in both the motor-manual and highly mechanised timber supply chain.

3. RFID in the Motor-Manual Timber Supply Chain â&#x20AC;&#x201C; RFID u ru~no-strojnom lancu dobave drva For the use in the motor-manual timber supply chain it is a prerequisite that transponder tags can be fixed manually. After each work step the tag number can be recorded manually by means of a reader and a MDR (mobile data registration) device with subsequent transmission via GSM to a trial database. The experiment took place in an old stand of Norway spruce and was carried out with transponder Croat. j. for. eng. 29(2008)1

tag type (Fig. 1). Immediately after cutting and processing long lengths, 30 tags of each type were fixed on the front side of the thick log ends. Another 30 pieces of the nail transponder tags were driven through the bark laterally at a distance of approx. 2 cm to the cutting surface. The tags were fixed depending on the tag type always one tag per one log. The Signumat plates were fixed with a special hammer, the cards were fastened with a stapling machine, adhesive labels were also glued and stapled in order to prevent loss. The coins were screwed and the nail transponders were driven in. The experiment was carried out in December and January with strong frost. Fixing the nail transponders turned out to be the biggest problem. About 12% of the nail tags to be fixed on the front side, and 67% of the nail tags to be fixed laterally were destroyed. This was due to the fact that the penetration resistance for fixing across the fibre was often too high. Previous drilling or punching of holes for the nail tags could be of help. All the other tags were operational after fixing. Once the tags were fixed, timber data were registered in the stand for each stem individually by means of the MDR device TimbaTec Recon with the software programme MobileForst. Subsequently, the number registered on the tag was read and linked with the timber data. The MDR device had a reading

S. KORTEN and C. KAUL

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

Fig. 2 Reading-antenna Slika 2. Antena za o~itanje range of a few centimetres in the tested frequency ranges (LF and HF). After cable-skidding the logs to the road, the transponder tags were read manually. As no timber data were entered there, and only the tag numbers were registered, the reading was clearly faster than the first reading in the stand. It turned out that two of the nails fixed laterally were destroyed during the extraction process. All other tags were still in place and operational. However, depending on the tag type and the situation of the stems on the pile, reading was partly difficult when the pile was not flush. For log transportation with a whole tree trailer, the tags of a respective load were read manually after loading in the forest. It was easy to register the tags showing the direction of motion, because the stems were generally stacked flush and could be reached easily from the truck. On the contrary, it was often difficult to read tags looking backwards, because shorter stems or stems lying on top could hardly be reached with the MDR device. If two short stems are to be loaded one behind the other, atten-

tion must be paid to the fact that the marked stem end looks towards the outside and can be reached from the front or from the back. Except for one adhesive label, all transponder tags could be read after loading. No mechanical destruction was visible on the defective label. The reason for the defect was not clear. After transportation to the saw mill and subsequent stacking, the tags were read again by means of the MDR device. No other tags were destroyed during transportation and unloading. At the saw mill the tag numbers were read after debarking and before chop cutting (Fig. 2). As the debarking knives reacted to pressure and were not controlled optically, 75% of the transponder tags fixed on the front were removed and/or destroyed during debarking. The laterally fixed nail tags did not survive the debarking process either. The rest of the functional tags were read by means of LF and HF antenna systems on the log-pusher. The log-pusher aligns the stems on the cross conveyor by exerting pressure on the thick-end of the stem in a way that the chop cutting saw cuts the stem by approx. 5 cm. The tags were read during the pushing process and, Croat. j. for. eng. 29(2008)1

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

therefore the necessary reading distance was only a few cm. Recording of the intact tags with the reading antenna on the log-pusher was done without any problems except for the nail tags fixed on the front side (Fig. 1). They could not be read because of insufficient field-strength, which was due to the difference in size between the transponder antenna and the reading antenna.

4. RFID in the Highly Mechanised Timber Supply Chain â&#x20AC;&#x201C; RFID u visoko mehaniziranom lancu dobave drva In the highly mechanised sorting of short lengths with harvesters, the tags should be fixed automatically during processing. It would be neither very reasonable nor productive to interrupt this highly productive harvesting process for fixing tags manually. In the same way, the tags must be read automatically after hauling and transportation because manual recording would slow down the work considerably. Therefore, the subject of this study was automatic fixing of tags by the harvester, forwarder (automatic reading) and short timber truck. The tag-fixing device developed for the harvester was designed in such a way that its incorporation

S. KORTEN and C. KAUL

into the saw box of the harvester-head Ponsse H73 was possible. For study purposes, the device was mounted in a separate saw box (Fig. 3). The additional saw box was necessary because it was not possible to permanently make tests on a complete head. The device is based on a pneumatic stapler. Staples of 25 mm leg length are used. A cartridge with 160 flexible ISO-cards (Fig. 1) is located behind the stapler head. The ISO-cards were used for the experiment because of the high reading range and because of the possibility to store them in a magazine and to automatically place them for fixing. The experiment took place in an old stand of Norway spruce. In principle the device was working satisfactorily. During processing it was possible to mark several segments subsequently with a tagging card. Thereby the technical feasibility was proven in principle. During the tests, however, two weak points of the tag-fixing device were discovered. On several occasions the card supply did not work when cards from the cartridge got jammed in the guiding grooves. Furthermore, the device sometimes did not return completely into the saw box. This was due to the fact that the guiding bolt for stabilising the rotational motion in the guide rail produced a stronger friction resistance in the practical test than in the laboratory trial.

Fig. 3 Harvester head with RFID tagging-device Slika 3. Sje~na glava s ure|ajem za postavljanje plo~ica za RFID Croat. j. for. eng. 29(2008)1

S. KORTEN and C. KAUL

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

Fig. 4 Reading-antenna behind the front grid of the forwarder Slika 4. Antena za o~itanje iza prednje re{etke forvardera The pneumatic cylinder for the rotational motion was not strong enough for this purpose. A HF antenna was developed and then used for the automatic reading of tags on the forwarder, on the short timber truck and later on in the saw mill. The antenna is composed of two one-turn copper coils with a diameter of 60 cm each. Each coil produces an individual reading field generating in total two partly overlapping reading fields. On the forwarder the antenna was fixed behind the grid in front of the stakes (Fig. 4) and on the side at the stakes. The antenna was fixed as high up as possible behind the front grid, so that all sections could pass the reading field during the loading process. In analogy to the tag-fixing device developed for harvester heads the flexible ISO-cards were in use. When reading the tags it turned out that the front-grid metal reduced the reading range considerably. Whereas a reading range of about one meter could be reached without metal, this range was reduced to 50 cm due to the grid with the tag- and reading antennas in parallel, i.e. with fixing the tag cards to the sections on the front. However, this reading range still allowed good reading of the tags fixed on the front side. Laterally fixed tags produced an unfavourable mutual position of the antennas and, thus, the reading range partly sank to under

20 cm, which made automatic reading of all tags during loading almost impossible. Lateral fixing of the antenna at the stakes had the advantage that no interference was produced by the metal of the stakes. Here again the tags were passed through the reading field of the antenna, which was possible without any problem during loading of the forwarder in the stand. With optimum orientation of the transponder tags (front side fixing) the reading range was more than 100 cm, with lateral fixing it was close to 50 cm. This range was sufficient for reading the tags during loading of the forwarder. In later practical application two reading antennas would be necessary, one on each side of the stakes. Whereas reading consumed no additional time as it could be integrated into the normal work process easily, there were problems with manipulating timber by crane. The reading antenna blocked the free space between two stakes and even overlapped it, which made smooth crane movements impossible. In analogy to the lateral attachment on the forwarder, the antennas were also fixed to the stakes of a short timber truck (Fig. 5). The tags were read during loading in the forest. The findings were the same as in the forwarder experiment. For recording the transponder tag numbers in the context of factory measurement, the reading antenna Croat. j. for. eng. 29(2008)1

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

S. KORTEN and C. KAUL

Fig. 5 Reading antenna at the stakes of the short-wood truck Slika 5. Antena za o~itanje na ru~icama kamiona

Fig. 6 Reading antenna in front of the debarker Slika 6. Antena za o~itavanje ispred linije za otkoravanje Croat. j. for. eng. 29(2008)1

S. KORTEN and C. KAUL

was mounted in the saw mill. As the reading of the tags fixed on the sides had to be done before debarking, the antenna was mounted directly in front of the debarker at a height of 100 cm above the longitudinal conveyor (Fig. 6). The tag cards in use could be read without problems regardless of their orientation. With the corresponding reading adjustment, even laterally fixed tags lying directly on the metal surface of the cross conveyor and, thus, passing the antenna at maximum distance (100 cm) could be read. Cards that were later experimentally fixed on the front side could also be read reliably with the antenna equipment installed. The tag-fixing device and the antenna system are only utility models which have not been designed for practical application and, hence, do not meet the requirements of long-term practical use. They are rather meant for gathering first experience and serve as a basis for further industrial development work.

5. Requirements for RFID in the Timber Supply Chain â&#x20AC;&#x201C; Zahtjevi RFID-a u lancu dobave drva For both the motor-manual and highly mechanised timber supply chains, the following requirements can be deducted regarding work frequency, with respect to the reading range, design and location of tags to be met by the RFID systems (Kaul 2007). In the motor-manual timber supply chain it is of advantage to fix the transponder tag on the front side of the stem, because there is less mechanical load during cable logging than on the sides. After skidding, manual reading with a MDR device can be done by a forest worker on site. Here it is of advantage to have the tags fixed to the front side because they can be found more easily in stacked timber. In the motor-manual timber supply chain there is no need for a high reading range because the tags are read manually. This allows the use of both LF and HF transponder tags. A high reading range and bulk reading at the stack and on the long-timber truck would facilitate and speed up the work process. However, when reading the tag numbers on the cutting area and in the saw mill, individual stems must be recorded specifically. Therefore, bulk reading is probably counterproductive in those cases. The tag design used should be robust and weather resistant. Apart from that there are no further special design requirements. However, the fixing method and the necessary tools depend on the tag design. Regarding dimension and weight, every tag-fixing tool should be as compact and light as possible. In order to be handy it should be possible to

Application of RFID in the Timber Supply Chain (85â&#x20AC;&#x201C;94)

wear it on the working belt. Thus, it can be reasonably expected for a forester to carry an extra stapling machine or light hammer. In the highly mechanised timber supply chain the tags should be attached without any considerable delay and without interruption of the working cycle, because even minor losses in productivity would show clearly in the cost statement due to the high share of fixed costs. Therefore, the tags should be fixed by the harvester during tree processing. This implies that a tag-fixing device must be incorporated directly into the harvester head. Then the tag numbers can be linked to the relevant data of the section in the harvesterâ&#x20AC;&#x2122;s records. For lateral attachment the tags can be fixed during cross-cutting, which does not need extra time. Furthermore, it is possible to mark the log section already processed. Front side attachment of the tags would highly facilitate the subsequent logistic processes because of the more favourable orientation of the tag antenna. However, it can only be done after cross-cutting, which necessarily needs more time. In addition, log marking is done before processing. The problem is that it is often not clear before processing and measuring if there is another log section to be cut. Marking on the front side of the section already processed is technically not possible, because the log section falls down after cross-cutting. Reading of the transponder tags should take place only after successful tag attachment in order to ensure the operability of the tags after their attachment. In the process of short-timber extraction on the cutting area the logs sections are loaded by crane onto the forwarder and not hauled by cable over the forest ground. Thus, the lateral mechanical pressure is limited to the contacts between the log sections and the contact with the grapple during the loading processes. Therefore, a lateral attachment is feasible in principle. However, fixing the tags on the front side would be more suitable because, on one hand the tags would be better protected and, on the other hand, they would be better oriented towards the reading antennas, which would clearly increase the reading range. This is also the case for loading on a short-timber truck. For automatic reading during forwarder or truck loading, reading ranges of more than 0.5 m are obligatory and more than 1 m would be desirable. In any case, bulk reading is necessary as several log sections are transported simultaneously by the loading crane. Therefore, LF tags are not suitable in the highly mechanised timber supply chain. The HF tags used met the requirements. Magazine feeding is a precondition for automatic tag fixing and requires a magazine suitable tag deCroat. j. for. eng. 29(2008)1

Application of RFID in the Timber Supply Chain (85–94)

sign. Every re-loading of magazines means an interruption of the working process, which should happen as seldom as possible. Therefore, a high feeding capacity of the magazine is desirable although the small space available in the harvester head is a limiting factor. At the saw mill reading of both the tags fixed laterally and on the front side is technically feasible. The tags fixed laterally must be read in any case before the debarking process. In case of unsatisfactory debarking, however, stems are again put in front of the debarker, which changes the order of the stems that are not marked any more after debarking. Thus, the tag information can no longer be linked clearly with those of the later factory measurements. Consequently, it would be preferable to read the tags after debarking but before chop cutting. To realize this, the tags must go through the debarking process without being damaged. This is only possible with central fixing on the front side and debarking devices that are controlled optically. In view of different stem diameters, laterally fixed tags require reading ranges of up to one meter. If the position of the antenna is always adjusted to the respective stem diameter (e.g. antenna curtain) lower reading ranges are also possible. In principle, LF and HF tags can be used here. These results confirm that RFID technology is suitable for use in the timber supply chain. The tagfixing device and the antennas that were developed are preliminary models. Further research is necessary for the development of tag design as well as for fixing and reading methods.

S. KORTEN and C. KAUL

the Technische Universität München in cooperation with the Department of Logistics of the Faculty of Mechanical Engineering at the Technische Universität Dortmund. The authors wish to thank for the financial support and good cooperation.

6. References – Literatura Bodelschwingh, E. von, 2005: Analyse der Rundholzlogistik in der deutschen Forst- und Holzwirtschaft – Ansätze für ein übergreifendes Supply Chain Management. Dissertation am Lehrstuhl für Forstliche Arbeitswissenschaft und Angewandte Informatik der TU München, 214 S. Dykstra, D., Kuru, G., Taylor, R., Nussbaum, R., Magrath, W., Story, J., 2003: Technologies for Wood Tracking. EASES Discussion Paper. 68 S. Finkenzeller, K., 2002: RFID-Handbuch. Carl Hanser Verlag, Wien. 446 S. Franke, W., Dangelmaier, W., 2006: RFID-Leitfaden für die Logistik. Gabler-Verlag, Wiesbaden. 299 S. Friemel, G., 2005: Forstunternehmer nutzt Vorzüge von Transpondern. Holz-Zentralblatt 131(60): 788. Holzmann, M., Verhoff, S., Sauter, U., 2006: Der Freiburger Transponderzyklus. AFZ – Der Wald 61(13): 716–721. Kaul, C., 2007: Technische Anforderungen für einen Einsatz der RFID-Technologie in der Holzerntekette. Diplomarbeit, Lehrstuhl für Forstliche Arbeitswissenschaft und Angewandte Informatik, TU München. 77 S.

Acknowledgement – Zahvala

Korten, S., Schneider, J., 2006:. Reorganisation der Informations- und Warenflussprozesse in der Holzerntekette mit Hilfe der Transpondertechnologie. Schlussbericht AiFProjekt Nr. 14186, Lehrstuhl für Forstliche Arbeitswissenschaft und Angewandte Informatik der TU München; Fachgebiet Logistik der Universität Dortmund. 161 S.

The research project was funded by the Federal Ministry of Economics and Technology. It has been carried out at the Department of Forest Work Science and Applied Computer Science of

Uusijärvi, R., 2003: Linking raw material characteristics with industrial needs for environmentally sustainable and efficient transformation processes (LINESET). Final Report QLRT-1999-01467, Trätek. 196 S.

Sa`etak

Primjena RFID-a (radijske frekvencijske identifikacije) u lancu dobave drva U radu se prikazuju mogu}nosti primjene radijske frekvencijske identifikacije (RFID) u lancu dobave drva. U procesu dobave drva od {ume do drvne industrije u Njema~koj sudjeluje velik broj malih i srednje velikih {umarskih tvrtki, {to ima za posljedicu visok stupanj dezintegracije toka materijala i informacija. Ipak, nadzor i/ili kontrola dobave nemogu}a je bez poznavanja koliko se drva nalazi gdje u odre|eno vrijeme. Dugo vrijeme isporuke i povezani gubici u kakvo}i pritom postaju neizbje`ni. Poku{aji da se nedostatak informacija nadoknadi vi{estrukim izmjerama i inventurom obloga drva na odre|enim to~kama u lancu dobave su skupi i ~esto stvaraju pogre{ke pri procjeni, uno{enju i preno{enju podataka. Unapre|enje cjelokupnoga procesa mogu}e je uspostavom sustava za upravljanje lancem dobave drva. U tom se smislu u posljednje vrijeme pove}ana pa`nja pridaje mogu}nosti

Croat. j. for. eng. 29(2008)1

S. KORTEN and C. KAUL

Application of RFID in the Timber Supply Chain (85–94)

primjene radijske frekvencijske identifikacije u kontekstu jasnoga prepoznavanja proizvoda u lancu dobave. Time bi se ostvario zna~ajan doprinos nadzoru i kontroli toka proizvodnje i informacija te pruìla jasna slika o vrsti i koli~ini proizvoda na odre|enom mjestu u odre|eno vrijeme. Radijska frekvencijska identifikacija zasniva se na razmjeni podataka elektromagnetskim valovima izme|u nosa~a podataka (primopredajnik, plo~ica) i ~ita~a podataka. Primopredajnik se u osnovi sastoji od mikro~ipa za pohranu podataka i zavojnice koja sluì kao antena. Ako se primopredajnik nalazi u elektromagnetskom polju ~ita~a, tada je mogu}a razmjena podataka izme|u primopredajnika i ~ita~a. U radu je ispitana mogu}nost primjene RFID-a u ru~no-strojnom te visoko mehaniziranom lancu dobave drva. U oba slu~aja istraìvanjima su obuhva}ene sve karakteristike va`ne za prakti~nu primjenu RFID-a. Testirane su razli~ite vrste primopredajnih plo~ica prema tipu, obliku, na~inu pri~vr{}ivanja na trupce, podru~ju rada (LF – niske frekvencije, HF – visoke frekvencije). Promatrani su domet o~itanja, osjetljivost na teku}ine i metale, kapacitet o~itanja obujma (istodobno o~itanje vi{e plo~ica), mjesto postavljanja plo~ica na trupcima i gubitak plo~ica pri sje~i, izradi, privla~enju i transportu. Za primjenu u ru~no-strojnom lancu dobave drva preduvjet je ru~no postavljanje primopredajnih plo~ica. U sastojini su zabiljeèni podaci svakoga trupca posebno i povezani s brojem plo~ice. Nakon svake faze rada pojedina je plo~ica ru~no evidentirana pomo}u ~ita~a i mobilnoga ure|aja za registraciju podataka te naknadno putem GSM-a odaslana u probne baze podataka. Nekoliko vrsta primopredajnih plo~ica bilo je mogu}e o~itati na svim dijelovima puta od sastojine do drvne industrije. Pri visoko mehaniziranom postupku pridobivanja drva pomo}u harvestera plo~ice se moraju automatski postaviti prilikom sje~e i izrade. Ru~no postavljanje plo~ica neproduktivno je i nerazumno ometanje i prekidanje u visokoproduktivnom procesu pridobivanja drva. Na isti na~in plo~ice je potrebno o~itavati automatski nakon izvoènja i transporta trupaca. Stoga je u istraìvanjima promatrano automatsko postavljanje plo~ica pri radu harvestera te automatsko o~itanje plo~ica pri utovaru drva na forvarder, na {umski kamion te pri istovaru u pilani. Na harvestersku je glavu ugra|en poseban ure|aj za postavljanje primopredajnih plo~ica ~ime je omogu}eno njihovo automatsko postavljanje. Plo~ice se tako|er automatski o~itavaju pri utovaru trupaca za izvoènje iz sastojine ili prijevoz do pilane. Rezultati su istraìvanja pokazali da je primjena RFID-a u lancu dobave drva tehni~ki mogu}a. Ure|aj za postavljanje plo~ica i antene razvijene za istraìvanja su po~etni modeli. Za prakti~nu primjenu RFID-a potrebna su daljnja istraìvanja i unapre|enja posebno s obzirom na oblik i vrstu primopredajnih plo~ica te na~in njihova postavljanja i o~itavanja. Klju~ne rije~i: lanac dobave drva, RFID, ure|aj za postavljanje plo~ica, identifikacija trupaca, tehni~ki zahtjevi

Authors’ address – Adresa autorâ:

Received (Primljeno): April 29, 2008 Accepted (Prihva}eno): June 10, 2008

Sven Korten, PhD. e-mail: korten@wzw.tum.de Christian Kaul, MSc. e-mail: kaul@wzw.tum.de Chair of Forest Work Science and Applied Computer Science Technische Universität München Am Hochanger 13 D-85354 Freising GERMANY Croat. j. for. eng. 29(2008)1

Preliminary note – Prethodno priop}enje

Energy Analysis of Pellets Made of Wood Residues Stjepan Risovi}, Igor \uki}, Kre{imir Vu~kovi} Abstract – Nacrtak The use of high technology in the manufacture of pellets and their use in modern furnace chambers of boiler plants for remote heating and heating of family houses, are the reason why pellets as the energy source find an ever increasing application. The biggest problem and obstacle for increasing their use is the starting investment in the boiler and pellet tank, which is considerably higher than the heating oil installation. In order to obtain the best possible indicators on economic feasibility of manufacturing wood pellets, a comprehensive research has been carried out in a wood processing plant. Pellets are manufactured from beech and oak residues. Beech accounted for about 90%. In order to get a complete picture of the consumption of electrical power per unit of manufactured pellets and of required power, the operating and idle power was measured at measurement sites during operations of individual electromotors. The measuring results were stored directly into computer and analysed later by a software package LabVIEW. The machines’ consumption of electrical power in the observed period was derived from the measuring results by numerical integration, based on which the average electrical power was then determined. Regarding the results of measuring carried out on four groups of electromotors, the highest consumption of electrical power has been measured during pressing, and then on electromotor group for the preparation of chips for pelleting. The highest consumption of electrical power has been recorded for the manufacture of one ton of pellets in pressing 138.98 kWh/t and transport system 74.25 kWh/t, which makes a total of 213.23 kWh/t or almost 95% compared to total consumption of electrical power. The measured electrical power for the manufacture of one ton of pellets in pressing hardwood residues, (in this research – beech), is by 100% higher than the electrical power referred to in literature, which relates to softwood. Consequently, the difference should be made in each study between the consumption of electrical power in pressing softwood and hardwood. Lower consumption of electrical power has been recorded with chippers with sharpened knives amounting to 24.3% compared to unsharpened knives. Keywords: forest biomass, pellets, energy

1. Introduction – Uvod The original ideas on secondary energy source, by which liquid fossil fuel could be replaced by solid renewable energy source, ranged in a wide area of use of wood, historically the most represented biomass as means of heating – all the way from split logs and cut cordwood to wood dust. Only by completely turning wood into dust, fuel was made that reached the level of automation, which was obtained in using Croat. j. for. eng. 29(2008)1

fossil energy sources, and hence also full comfort in handling and use of liquid and gaseous fuels. However, the price of wood dust was high due to high energy consumption required for transforming the primary energy source into the secondary energy source, as well as due to serious requests for removing the risk of explosion of the mixture of wood particles of specific size (smaller than 0.3 mm), moisture (lower than 12%) and minimum quantity in a specific air volume in which fuel was transferred by air convey-

S. RISOVI] et al.

ors into burning systems. Equalisation of sizes went through split logs, splitters and chips to the beginning of its compression into a uniform shape of the same dimensions – production of wood briquettes. Mechanical handling systems were determined by their size, and the price enabled their use for special purposes. The development went on to the idea of production of the so-called mini briquettes – pellets. Generally, the process of pelleting is the compression of a material – animal food, metal for the production of bullets, and similar, in the shape of small balls or rollers. Thus, wood pellet was produced by densification, mechanical compression of sawdust, shavings or grindings, with or without binders. Energy consumption was reduced with respect to dust, and the risk of developing explosion mixture was reduced or completely removed. Three options were possible during transport from manufacturer to user – transport by tank trucks, transfer to the tank in the air flow and fully automated burning and combustion control. Pellets can also be produced from wood bark, paper, different kind of house waste, agricultural waste and wood residues. Most European countries, including Croatia, are not interested in pelleting from other materials, apart from wood. In future, pellet production may be expected from forest residues, fast-growing energy wood, urban plantations residues, orchards, etc. The choice of material and possible binder for pellet production is a significant issue in defining the sale strategy in the market, as pellets have to remain an energy source that causes no environmental pollution, either by emission of smoke toxicants or by solid residue – ashes. Anyway, compared to fossil fuels, pellets are ecologically clean fuels that cause no greenhouse effect because they are CO2 neutral, and they are renewable meaning that plants use CO2 from the air in the process of photosynthesis releasing oxygen. In Croatia the establishment of a biomass market is at the very beginning in the broadest sense. In the domestic market there are only sporadic cases of limited marketing of forest biomass, so that pellets will surely take a significant place in the market of ecological fuels when this proves to be economically feasible. In most European countries the market of wood forest biomass has already been established and represents a significant source of income of wood industry, forestry and even agriculture. The social significance should also be emphasised of utilisation of local forests wood residues for raising the living standard of rural areas, by provid-

Energy Analysis of Pellets Made of Wood Residues (95–108)

ing heating and hot water to houses and apartments, thereby providing additional employment. These are all some starting points that have been enhancing the development of production and use of pellets as a suitable energy source for providing heating energy in highly developed countries of Mid-Europe, Scandinavia and North America during the last twenty years.

2. Use of biomass in the countries of the European Union and in the Republic of Croatia – Uporaba biomase u zemljama Europske unije i u Republici Hrvatskoj In the past period, high price of oil, with the tendency of further increase, as well as frequent disruptions in the supply of natural gas, have considerably affected the increased interest for the use of forest biomass. New and efficient technologies are more and more available in the market, and they are equally suitable for individual users and industrial plants. Primary production of biomass energy (apart from forest biomass here also including agricultural biomass) increased by 5.6% on the area of the European Union during 2004 (Table 1). The share of wood biomass in total consumption of primary energy in 2004 was 3.2%, which represents a slight increase with respect to 3.0% in 2003. The use of wood biomass increased considerably for the production of electric power and to be specific by 23.5% with respect to 2003, and in 2004 a total of 34.6 TWh was produced. The main reason of the said increase lies in considerable development of biomass cogeneration in several European countries with significant areas covered by forests (Sweden, Finland, Austria). The current development of use of wood biomass is very uneven. Many countries are just beginning to use their potentials (Poland, Check Republic, Slovak Republic and Baltic States), while others, such as Finland and Sweden, have already developed highly advanced technologies and they are making considerable use of their natural potential. The White Book of 1997 does not set individual goals for wood biomass. It only outlines the total goal for biomass (also including biogas and biofuels) of 135 million toe. If these sub-sectors are exluded, according to EurObserv’ER the goal of 100 million toe remains for wood biomass in 2010. In order to reach this goal, additional efforts must be taken considering the fact that, at the current increasing trend, the production energy from wood biomass will only be 77.9 million toe in 2010. In the Republic of Croatia, forests cover almost half of the continental area, out of which 78% are Croat. j. for. eng. 29(2008)1

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Table 1 Primary production of wood mass energy on the area of the European Union in million of tons of oil equivalent – toe Tablica1. Primarna proizvodnja energije iz drvne biomase u Europskoj uniji u milijunima tona ekvivalentne nafte – toe Country Zemlja France Sweden Finland Germany Spain Poland Austria Portugal Latvia Great Britain Denmark Italy Check Republic Greece Hungary Netherlands Lithuania Slovenia Belgium Slovak Republic Estonia Ireland Luxemburg Cyprus Malta Total – EU25

Energy in 2003, toe Energija u 2003. toe 9.002 7.927 6.903 5.191 4.062 3.921 3.222 2.652 1.240 1.084 1.071 1.015 0.895 0.909 0.777 0.561 0.672 0.422 0.346 0.300 0.150 0.145 0.015 0.006 0.000 52.488

state forests, and 22% are privately owned forests. Total growing stock is approximately 400 million m³. The annual allowable cut is approximately 6.5 million m³ of wood raw material, of which 2.6 million m³ is energy wood. Based on total production, wood industry was also developed and it now has the capacity of processing the total production of industrial roundwood. The result of wood processing is approximately 1.7 million m³ of wood residues suitable for both energy use and pellet production. Further to the above, it can be seen that forestry and wood industry produce considerable quantities of raw material suitable for energy utilisation. Having in mind the increasing trend of pellet demand in the foreign market, by implementing some Croat. j. for. eng. 29(2008)1

Energy in 2004, toe Energija u 2004. toe 9.180 8.260 7.232 6.263 4.107 3.927 3.499 2.666 1.300 1.231 1.113 1.083 1.007 0.927 0.805 0.720 0.697 0.422 0.382 0.303 0.150 0.144 0.015 0.006 0.000 55.439

Change, % Promjena, % +2.0 +4.2 4.8 20.7 1.1 0.2 8.6 0.5 4.8 13.6 3.9 6.7 12.5 1.9 3.6 28.2 3.7 0.0 10.4 1.1 0.0 –0.6 0.0 0.0 0.0 +5.6

incentive measures further broadening of pellet production will surely be enhanced and it should bring an additional source of income to their producers.

2.1Possible yield of wood residues in wood sector – Mogu}i prinos drvnoga ostatka u drvnom sektoru In 2003 market research was carried out through a questionnaire in 92 companies on the area of the whole republic covering approximately 55–60% of the basic group. The basic group consisted of the total number of processed logs in Croatia in 1991, and they were estimated to approximately 1,652,207 m3. The coverage percentage is higher than 55% and it is sufficiently representative for the total results to

S. RISOVI] et al.

be broadened to the basic group. It should be noted that the number of persons interviewed in different counties were not equally covered due to time limits, so that the questionnaire are carried out on the expected sample. The coverage percentage by individual counties ranges between 20 and 90%, and this, however, is not an obstacle for expressing the results relevant for this project. The last available processing of data from the point of view of capacity and quantity of logs processing by forest offices refers to 1989, and hence it is not comparable with the questionnaire. However, this data may contribute the assessment of the questionnaire results relevance, as it is known that the primary processing of wood is increasing, as well as the production of logs. It is, however, well known that significant ownership and programme transformations have been carried out, that large wood processors have almost disappeared and a considerable number of small wood processors have appeared, and this only in the primary phase of processing, and that this process has not necessarily occurred only in the area of raw materials but also in more distant areas. Due to such changes, research had to be carried out so as to establish the actual state and level of the current wood processing, the basis for understanding the quantity and quality of wood waste, which was the aim and objective of this paper. The quantity of wood residues in industrial processing depends on the species of raw material, way of processing, intended assortment, but also on the quality of the input raw material. The results of the questionnaire show that in all production phases wood residue is on average around 45.7%, while in a higher phase of wood processing this percentage can be as much as 65% with respect to the input raw material. When only speaking of primary and final wood processing, the percentage of wood residue is around 31%. These are theoretical notions and practical experience of which: Þ approximately 230 thousand m3 of untrimmed timber of commercial quality I–IV, Þ approximately 920 thousand in finishing quality, with a part of sawmill waste, Þ 170 thousand m3 of bark (not included in the input quantity because it is measured and purchased without bark), Þ 136 thousand m3 of sawdust, Þ other wood residues – 299 thousand m3, and loss due to exceeded measurement of 115 thousand m3.

Energy Analysis of Pellets Made of Wood Residues (95–108)

From final processing of 920 thousand m3 of sawn timber the following can be obtained: Þ approximately 92 thousand m3 of sawn timber that is not processed, Þ approximately 400–490 thousand m3 of sawing elements, Þ approximately 99.5 thousand m3 of sawdust, Þ remaining useful material – 158–248 thousand m3.

3. Characteristics of pellets – Svojstva peleta In the central heating fuel market, pellets represent a new type of fuel. They appeared about twenty years ago in wood industry of the United States and in Europe about ten years ago.

Fig. 1 Illustration of pellets Slika 1. Prikaz peleta

The requirements of fuel quality are contained in the German standard DIN 51731, DINplus and Austrian standard ÖRNORM M 7135. The difference between the German and Austrian standard is that the Austrian standard allows the content of potato or corn concentration up to 2%, as auxiliary binder. Table 2 shows the characteristics of pellets in accordance with the above standards, and Fig. 1 shows the pellet appearance. The use of high technology in the manufacture of pellets and their use in modern furnace chambers of boiler plants for remote heating and heating of family houses, are the reason why pellets as the energy source find an ever increasing application. The biggest problem and obstacle for increasing their use is the starting investment in the boiler and pellet tank, which is considerably higher than the heating oil installation. Croat. j. for. eng. 29(2008)1

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Table 2 Survey of pellet characteristics pursuant to ÖNORM M 7135, DIN 51 731 and DINplus Tablica 2. Prikaz svojstava peleta prema ÖNORM M 7135, DIN 51 731 i DINplus Pellet diameter – Promjer peleta, mm Max. length – Najve}a duljina, mm Density – Gusto}a, kg/m3 Moisture – Mokrina, % Ash content – Sadràj pepela, % Heating power – Ogrjevnost, MJ/kg Sulphur content – Sadràj sumpora, % Chlorine content – Sadràj klora, %

ÖNORM M 7135 4–20 100 ³1000 £12.0 £0.5 ³18.0 £0.04 £0.02

4. Technology of pellet manufacturing – Na~in izradbe peleta The process of pellet manufacturing can be divided into the following steps: Þ a) Collecting wood residues at the central landing. The collected biomass is usually of different size and hence it has to be sorted and prepared for further process of pellet manufacturing. Þ b) Chipping/grinding of large biomass. The most favourable size of chipped biomass is up to 4 mm. This is obtained by chippers and grinders, and for the choice of electromotor power, wood chipping of 15 kW/t is recommended.

DIN 51731 4–10 50 1000–1400 £12.0 £1.5 17.5–19. £0.08 £0.03

DINplus 4–10 ³5 ´ d ³1120 £10.0 £0.5 ³18.0 £0.04 £0.02

cess in pellet manufacturing. By cooling, the pellets are stabilised, hardened and finally shaped. After compression, the moisture content of pellets is approximately 14%, and by cooling their moisture is reduced by approximately 6%. The energy of approximately 5 kWh/t is consumed for cooling the pellets. Þ f) Sieving. Wood dust generated in pellet manufacturing is separated by air flow. Þ g) Packaging and storage. Pellets are packaged automatically in bags of 15 to 20 kg, or in big-begs of the volume of 1 m3. As pellets are highly prone to moisture, it is highly important to provide adequate storage of finished pellets. Besides the above, pellets may also be stored in closed silos.

Þ c) Drying, depending on biomass moisture. The moisture of the material for manufacturing pellets must not exceed 17%. Drying can be carried out by all kinds of dryers, steam dryer (direct and indirect), hot air dryer or drum dryer. When applying artificial drying, the value of the consumed energy goes up to 15% of pellets price. Regarding the boiler unit as the heat generator for drying, 1 MW/t of water released from biomass is usually used. Þ d) Compression/pelleting. Biomass, previously wetted by saturated water steam, is pressed in high-pressure presses with ring moulds (Fig. 2a) or horizontal plain moulds (Fig. 2b), where pellets are shaped of 4 to 20 mm in diameter and up to 100 mm in length. The usually required power for manufacturing 1 000 kg of pellets is approximately 60 kW for soft wood. Þ e) Cooling. The temperature of pellets after compression is pretty high, about 90°C, so that cooling is an important part of the proCroat. j. for. eng. 29(2008)1

Fig. 2 Illustration of pellets manufactured Slika 2. Shematski prikaz izradbe peleta 99

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Fig. 3 Illustration of the press for manufacturing pellets Slika 3. Prikaz pre{e za izradbu peleta from beech and oak residues. Beech accounted for about 90%. Table 3 shows the survey of the annual raw material required for manufacturing and wood residues intended for the production of briquettes, pellets and production of heating energy. The data on average utilisation of the basic material were used for the calculation. The daily quantity of saw-

5. Materials and methods – Materijal i metode In order to obtain the best possible indicators on economic feasibility of manufacturing wood pellets, a comprehensive research has been carried out in a wood processing plant. Pellets are manufactured

Table 3 Structure of wood residues Tablica 3. Struktura drvnoga ostatka Raw material Sirovina Logs Trupci

Required raw material Potrebna sirovina m3/year

–

m3/god.

21,000

Moisture Mokrina

Residues Ostatak

35–40

12.00 20.00 40.00

Unedged sawn timber Samica

14,280

Elements of square sawn timber Elementi – ~etvrta~e

12,940

5–12

15.00 3.00

10,611

5–12

10.36

Square sawn timber ^etvrta~e Total residues Ukupni ostatak Total residues Ukupni ostatak

100

17–25 10.00

Residues – Ostatak Type of residues Vrsta ostatka Slab – Okorak Sawdust – Piljevina Side trimmings in bucking Okrajak pri krojenju Sawdust in bucking Piljevina pri krojenju Side trimmings – Okrajak Sawdust – Piljevina Sawdust, grindings, turning waste Piljevina, blanjevina, ostatak pri tokarenju

m3/year m3/god.

kg/year kg/god.

2,520 4,200

2,016,000 3,360,000

5,712

4,569,600

1,428

1,142,400

1,941 388

1,552,800 310,400

1,100

880,000 13,831,200 10,471,200

Croat. j. for. eng. 29(2008)1

Energy Analysis of Pellets Made of Wood Residues (95–108)

dust, shavings, side trimmings and grindings of approximately 35 m3/1 day was determined based on experience. Since logs are sawn in sawmills owned by the wood-processing plant, where the research was carried out, there is a difference between total residues generated in wood working and processing and total residues in the research DI plant. Sawdust generated in sawing logs whose annual quantity is 4,200 m3 or 3,360,000 kg remains in the place where it was generated, while in the plants of the above said factory slabs are chipped for the requirements of manufacture of pellets. In total structure of residues generated in processing, side trimmings, sawdust, grindings, shavings and particles separated in turning are produced from 5,800 m3/year of purchased square sawn timber, which together with 7,140 m3/year of their own processed square sawn timber produced by sawing 21,000 m3 of logs, makes a total of 12,940 m3 of square sawn timber. The utilisation of energy potential of wood residues can bring numerous benefits such as the increase of energy efficiency, use of domestic energy source, decrease of adverse environmental effects and use of wood residues that remain as waste.

S. RISOVI] et al.

After analysing the quantity and type of residues, heating energy requirements, as significant energy source in the manufacture of wood furniture, and upon understanding the need and demand for a high quality and environmentally clean fuel, two starting points were determined on disposal of wood residues: Þ 1. utilisation of wood biomass as fuel for obtaining heating energy for own needs, Þ 2. manufacture of briquettes and pellets for sale. Further to the above described, the following assessment was made: Þ 1. positive solution of economic and organisational problems of disposal of wood residues, Þ 2. in the long run an ecologically valuable contribution to the solution of the arisen problem and continuous environmental protection with the wish to leave it cleaner to future generations. Since the form of residues is not suitable for the manufacture of pellets, these residues are additionally processed before compression. Fig. 4 illustrates the preparation of residues for pellets as well as their manufacture.

Fig. 4 Illustration of manufacture of pellets and measurement sites for the determination of driving power and consumption of electrical power Slika 4. Shematski prikaz izradbe peleta i mjernih mjesta za utvr|ivanja pogonske snage i utro{ka elektri~ne energije Croat. j. for. eng. 29(2008)1

101

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Fig. 5 Schematic presentation of the measurement chain used for measuring total electrical power Slika 5. Shematski prikaz mjernoga lanca upotrijebljenoga za mjerenje ukupne djelatne snage Slabs and side trimmings of different size produced in sawing logs and processing of unedged sawn timber and elements are chipped in two chippers Mion & Mosole, type: TRH 240X650 4W. The following items are installed on each chipper: four electromotors; the main driving engine with the power of 75 kW, two auxiliary electromotors and electromotor for conveyor vibrations. The produced chips are air transported to the tank No. 1. From the above said silo, chips of inadequate size for pelleting are removed by an electromotor driven remover (M3) (Fig. 4) and by a snail transporter (M4) they are transported to the grinder (M5) by which the chips are reduced to particles smaller than 4 mm. Small chips are transported by a fan (M6) into tank No 2 and then they are ready for being compressed in one of two CPM Europe presses. In a high-pressure press, pellets are manufactured without any binder. Pellets are shaped in a conical roller-shaped matrix of the diameter of 6 mm. A part of chips may be air transported by a fan (M11) to the press for the production of briquettes. In order to get a complete picture of the consumption of electrical power per unit of manufactured pellets and of required power, the operating and idle power was measured at measurement sites during operations of individual electromotors. Total operating electrical power was measured by a measurement chain (Makar 1987) shown in Fig. 5. The measuring results were stored directly into computer and analysed later by a software package LabVIEW. The machines’ consumption of electrical power in the observed period was derived from the measuring results by numerical integration, based on which the average electrical power was then determined.

102

The measurement was carried out on four electromotor groups, all together making the line for the manufacture of pellets. Two measurements were carried out on chippers, and one on transport system and pellet press with a group of electromotors.

6. Results and discussion – Rezultati i diskusija Chipper 1 Mion & Mosole, type: TRH 240X650 4W – unsharpened knives Þ measurement on the chipper involves the following electromotors: main electromotor, power of 75 kW, cosj = 0.84, n = 1485 min–1 two auxiliary electromotors, power of 3.0 kW each, cosj = 0.75, n = 1410 min–1 electromotor for transport conveyor vibrations, P = 3 kW, cosj = 0.81 With the use of the appropriate measurement equipment, the measurements were carried out of the operating and idle power between 10:43 and 11:55, and during that time three tons of material were prepared for further processing. The calculation of the unit consumption of electrical power for chipping of 1 t of side trimmings and slabs, kWh/t: t = 4321 s; t' = 4321 – 70 = 4251 s = 1.18 h or 1h 10’ 51’’ where: t measurement time t'

measurement time without interruption Croat. j. for. eng. 29(2008)1

Energy Analysis of Pellets Made of Wood Residues (95–108)

S. RISOVI] et al.

Fig. 6 Measurement results of total electrical power on chipper M1 Slika 6. Rezultati mjerenja ukupne djelatne elektri~ne snage na ivera~u M1

Fig. 7 Measurement results of total electrical power on chipper M2 Slika 7. Rezultati mjerenja ukupne djelatne elektri~ne snage na ivera~u M2

The total measured power consumption in the observed period was 39.44 kWh, and power consumption per ton of processed material was: Et = E / 3 = 39.44 / 3 = 13.14 kWh/t The average time for processing 1 ton of raw maerial (without interruption) was 23 min and 37 s. Chipper 2 Mion & Mosole, type: TRH 240X650 4W – sharpened knives Croat. j. for. eng. 29(2008)1

Þ measurement on the chipper involves the following electromotors: main electromotor, power of 75 kW, cosj = 0.84, n = 1485 min–1 two auxiliary electromotors, power of 3.0 kW each, cosj = 0.75, n = 1410 min–1 electromotor for transport conveyor vibrations, P = 3 kW, cosj = 0.81

103

S. RISOVI] et al.

Three tons of slabs and side trimmings were processed: Þ measurement started at 12:11, and finished at 13:19, Þ interruption time is 801 s. The calculation of the unit consumption of electrical power for chipping of 1 t of side trimmings and slabs, kWh/t is: t = 3 947 s; t' = 4321 – 801 = 3 947 – 801 s = 3 146 s = 0.847 h The total measured power consumption in the observed period was 31.70 kWh, and power consumption per ton of processed material was: Et = E / 3 = 31.70 / 3 = 10.57 kWh/t The average time for processing 1 ton of raw material (without interruption) was 18 min and 59 s. Transport system Transport system involves the following consumers: Þ fan electromotor for silo suction S5, P = 30 kW, cosj = 0.86, n = 1465 min–1 Þ remover, P = 7.5 kW, cosj = 0.84, n = 1460 min–1 Þ electromotor of snail transporter, P = 2.20 kW, cosj = 0.78, n = 1410 min–1

Energy Analysis of Pellets Made of Wood Residues (95–108)

Þ hammer-mill, P = 75 kW, n = 2965 min–1, cosj = 0.92 Þ fan, P = 37 kW, n = 1470 min–1, cosj = 0.86 The total measurement time was t = 4688 s and during measurement there were no interruptions, and approximately 2 tons of chipped biomass was transported. The total measured power consumption in the observed period was 148.50 kWh, and power consumption per ton of processed material was: Et = E / 2 = 148.50 / 2 = 74.25 kWh/t The average time for processing 1 ton of raw material (without interruption) was 39 min and 4 s. Pellets press The fourth group measurement involves the following electromotors: Þ press with ring mould CPM Europe Þ main press electromotor, P = 160 kW, cosj = 0.85, n = 1 485 min–1 Þ auxiliary electromotor, P = 11 kW, cosj = 0.80, n = 1460 min–1 Þ remover 2, electromotor power P = 7.5 kW, cosj = 0.84, n = 1 460 min–1 Þ electromotor of snail transporter (from refrigrator to vertical elevator) P = 1.5 kW, cosj = 0.76, n = 1 390 min–1

Fig. 8 Measurement results of total electrical power on transport system Slika 8. Rezultati mjerenja ukupne djelatne elektri~ne snage na transportnom sustavu 104

Croat. j. for. eng. 29(2008)1

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Fig. 9 Measurement results of total operating electrical power on pellet press Slika 9. Rezultati mjerenja ukupne djelatne elektri~ne snage na pre{i za pelete Þ exhaust system for dust and hot air from refrigerator, P = 0.55 kW, cosj =0.72, n = 1 385 min–1 Þ the forth measurement also involves 5 inaccessible electromotors, whose total power is estimated to 10 kW Þ there were no interruptions during measurement The hourly press efficiency is 800 kg of pellets. The measurement was carried out in 3722 seconds or 1h 2min 2s or 1.034 h, during which time 830 kg of pellets were pressed. The total measured power consumption in the observed period was 115.36 kWh, and power consumption per ton of processed material was: Et = E / mp = 115.36 / 0.83 = 138.98 kWh/t

The average time required for pressing 1 ton of pellets (without interruption) was 75 min or 1.25 h. The measurement values were determined in the regular working regime for sufficient time to gather enough data on electrical power consumption. Table 4 shows the obtained measurement results of average power in the process of manufacturing pellets. Table 4 clearly shows the highest consumption of electrical power for manufacturing 1 ton of pellets, the press 138.98 kWh/t and transport system 74.25 kWh/t, totalling 213.23 kWh/t or almost 95% with respect to total consumption of electrical power. Table 4 also shows the difference in the consumption of electrical power in chipping with sharpened and unsharpened knives. If the starting point is the consumption of electrical power with sharpened knives, then the

Table 4 Survey of consumers of electrical power and measured operating electrical power Tablica 4. Prikaz potro{a~a elektri~ne energije i izmjerene djelatne elektri~ne snage Measurement Mjerenje

Machine Ure|aj

1. 2.

Chipper M1 – Ivera~ M1 Chipper M2 – Ivera~ M2 Transport system Transportni sustav Pellet press Pre{a za pelete

3. 4.

Croat. j. for. eng. 29(2008)1

Installed power Instalirana snaga kW 84.00 84.00

Measured power Izmjerena snaga kW 39.44 31.70

Unit consumption of electrical power for manufacturing 1 t of pellets Jedini~na potro{nja el. energije za proizvodnju 1 t peleta kWh/t 13.14 10.57

151.70

114.03

74.25

198.05

107.92

138.98

105

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Table 5 Results of moisture measurement Tablica 5. Rezultati mjerenja mokrine Sample Uzorak 1 2 3 4 5 6

Chips after wetting – Iverje nakon navla`ivanja Pellets after pressing – Peleti nakon pre{anja Pellets after cooling – Peleti nakon hla|enja

Mass of wet sample Masa vla`noga zraka kg 0.517 0.305 0.318 0.632 0.391 0.434 0.288 0.373 0.215

consumption is higher by 24.31% with unsharpened knives per 1 ton of chips. The Croatian standard HRN ISO 3130 was used for determining moisture of wood residues. Slabs and side trimmings made by chipper M1 are marked from 1 to 3, and those made by chipper M2 are presented in Table 5 as samples 4 to 6. The sample moisture is uniform and it ranged between 7.21% and 9.67%. Chips are wetted in order to achieve better quality of pellets and to make simpler the process of pressing. The measured moisture was 18.05%. After pressing, the pellets moisture was approximately 5%, which indicates that 13% of water contained in chips evaporated during pressing.

7. Conclusions – Zaklju~ci Generally biomass is increasingly significant because it is environmentally friendly and the product of combustion – ash is suitable as fertiliser, by which the »life« cycle of biomass is closed. The interest for biomass, in the form of pellets, is increasing and they have become a standard item in the range of commercial goods. Regarding the results of measuring carried out on four groups of electromotors, the highest consumption of electrical power has been measured during the fourth measurement, i.e. during pressing, and then on electromotor group for the preparation of chips for pelleting. The highest consumption of electrical power has been recorded for the manufacture of one ton of pellets in pressing 138.98 kWh/t and transport system 74.25 kWh/t, which makes a total of 213.23 kWh/t

106

Mass of dried sample Masa osu{enoga uzorka 0.475 0.283 0.295 0.583 0.359 0.392 0.236 0.354 0.203

Moisture Mokrina % 8.12 7.21 7.23 7.75 8.18 9.67 18.05 5.09 5.58

or almost 95% compared to total consumption of electrical power. The measured electrical power for the manufacture of one ton of pellets in pressing hardwood residues, (in this research – beech), is by 100% higher than the electrical power referred to in literature, which relates to softwood. Consequently, the difference should be made in each study between the consumption of electrical power in pressing softwood and hardwood. Lower consumption of electrical power has been recorded with chippers with sharpened knives amounting to 24,3% compared to unsharpened knives. Based on the above, it can be concluded that the investment into manufacture of pellets is justified. The pellet market has an increasing trend but it cannot be expected (at least not in near future) that the offer could exceed the demand.

8. References – Literatura Anon. 1999: HRN ISO 3130 Odre|ivanje sadr`aja vode za ispitivanje fizikalnih i mehani~kih svojstava. Hrvatski zavod za norme. Anon. 1996: DIN 51731 – Test of solid fuels – Compressed untreated wood Requirements and testing, Deutsches Institut für Normung. Anon. 1997: White Paper for a Community Strategy and Action Plan, Energy for the Future: Renewable Sources of Energy, EC. Anon. 1998: SS 187120. Biofuels and peat – Fuel pellets – Classification. Swedish Standards Institution. Anon. 2000: ÖNORM M 7135 – Presslinge aus naturbelassenem Holz oder naturbelassener Rinde – Pellets und Croat. j. for. eng. 29(2008)1

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Briketts – Anfoderungen und Prüfbestimmungen. Österreichisches Normunginstitut.

ence on Biomass for Energy And Industry, Sevilla, Spain, 5–9 June 2000, p. 601–604.

Mlakar, F., 1987: Op}a elektri~na mjerenja. Tehni~ka knjiga, Zagreb.

Risovi}, S., Domac, J., Pe{kan, M., Ostro{ki, S., 2000: Energijska i gospodarna ra{~lamba briquettes od drvnog ostatka. Energija i okoli{, Vol. II, Opatija, p. 59–65.

Risovi}, S., 2003: Briketi i pellets – novi energent na hrvatskom tr{i{tu, [umska biomasa, Akademija tehni~kih znanosti Hrvatske, p. 123–141 Risovi}, S., Domac, J., 2000; Briquetting of wood waste – an energy and economy system analyis, 1st World Confer-

Risovi}, S., Dundovi}, S., Slunjski, M., Sever, S., 2003: Razvrstavanje {umskih strojeva za pretvorbu pramarnog energenta u koji oblik sekundarnog goriva. [umska biomasa, Akademija tehni~kih znanosti hrvatske, p. 54–96.

Sa`etak

Energijska ra{~lamba peleta izra|enih od drvnoga ostatka Polazne prosudbe o sekundarnom energentu kojim bi se moglo zamijeniti fosilno teku}e gorivo krutim obnovljivim energentom kretale su se u {irokom rasponu kori{tenja drva, povijesno naj~e{}e upotrebljavane fitotvari kao ogrjeva – sve od cjepanica i sje~enica pa do drvnoga praha. Tek se potpunim usitnjavanjem drva u prah uspjelo stvoriti gorivo koje je doseglo razinu automatiziranosti postignutu pri uporabi fosilnih energenata, a time i svu lagodnost koju su pri rukovanju i kori{tenju osiguravala teku}a i plinovita goriva. No, cijena je drvnoga praha bila velika zbog utro{ka energije za pretvorbu primarnoga u sekundarni energent te mnogoga zahtjeva pri uklanjanju mokrine (ispod 12 %) i minimalne koli~ine u odre|enom obujmu zraka kojim se gorivo zra~nim konvejerima prenosi do loì{nih sustava. Ujedna~ivanje je dimenzija teklo preko cijepanoga drva, sje~ke i iverja do po~etka njegova ugu{}ivanja u jednoliki oblik istih dimenzija, do proizvodnje drvnih briketa. Razvoj je tekao dalje sve do ideje izradbe tzv. mini briketa – peleta. Peletiranje je zapravo ugu{}ivanje u obliku kuglice ili valj~i}a. Tako je i nastao drvni pelet zgu{}ivanjem, mehani~kim zbijanjem piljevine, blanjevine ili bru{evine s veznim sredstvom ili bez njega. Utro{ak je energije u odnosu na prah smanjen, a i opasnost je od nastanka eksplozivne smjese smanjena ili je uklonjena. Pri prijevozu od proizvo|a~a do korisnika mogao se primijeniti prijevoz cisternama, preno{enje u struji zraka do spremnika te potpuno automatiziranje loènja i nadzor izgaranja. Peleti se mogu proizvoditi i od kore drva, papira, razvrstanoga ku}noga sme}a, ostataka poljoprivrednih kultura te otpadnoga drva. Peletiranje iz drugih materijala osim drva za ve}inu europskih zemalja, pa tako i za Hrvatsku, nije zanimljivo. U budu}nosti se moè o~ekivati izradba peleta od {umskoga ostatka, od brzorastu}ega energijskoga drva, ostatka drve}a urbanih nasada, vo}njaka i sl. Izbor je materijala i moèbitnoga vezivnoga sredstva za izradbu peleta va`no pitanje pri osmi{ljavanju strategije prodaje na trì{tu jer peleti moraju ostati energent koji ne one~i{}uje okoli{. U svakom su slu~aju peleti u odnosu na fosilna goriva ekolo{ki ~isto gorivo koje ne uzrokuje stvaranje stakleni~kih plinova jer je CO2 neutralan i obnovljiv zato {to ga biljke koriste iz zraka u procesu fotosinteze osloba|aju}i kisik. U Hrvatskoj je stvaranje trì{ta biomase u naj{irem smislu tek u za~etku. Na doma}em trì{tu postoje osamljeni primjeri ograni~enoga trgovanja {umskom biomasom (fitotvari), tako da }e peleti zasigurno zauzeti zna~ajno mjesto na trì{tu ekolo{kih goriva u slu~ajevima kada je to i gospodarski opravdano. U ve}ini je europskih zemalja trì{te drvne {umske biomase ve} uspostavljeno i predstavlja zna~ajan izvor prihoda drvne industrije, {umarstva, pa i poljoprivrede. Udio drvne biomase u ukupnoj potro{nji primarne energije u 2004. iznosio je 3,2 %, {to je blagi porast u odnosu na 3,0 % u 2003. godini. Zna~ajno je porasla uporaba drvne biomase za proizvodnju elektri~ne energije, i to za 23,5 % u odnosu na 2003. te je u 2004. ukupno proizvedeno 34,6 TWh. Glavni je uzrok navedenoga porasta prili~an razvitak kogeneracije na biomasu u nekoliko europskih zemalja s velikim povr{inama pod {umama ([vedska, Finska, Austrija). Koli~ina drvnoga ostatka u industrijskoj preradbi drva ovisi o vrsti sirovine, na~inu preradbe, èljenom izlaznom asortimanu, ali i o kakvo}i ulazne sirovine. Rezultati ankete pokazuju da drvni ostatak u svim fazama proizvodnje u prosjeku iznosi oko 45,7 %, a s vi{om fazom preradbe drva taj postotak u odnosu na ulaznu sirovinu iznosi i 65 %. Croat. j. for. eng. 29(2008)1

107

S. RISOVI] et al.

Energy Analysis of Pellets Made of Wood Residues (95–108)

Ako se govori samo o primarnoj i doradnoj obradbi drva, postotak drvnoga ostatka iznosi oko 31 %. Radi dobivanja {to potpunijih pokazatelja o gospodarstvenoj probita~nosti proizvodnje drvnih peleta provedeno je obuhvatno istraìvanje u drvnoindustrijskom pogonu. Peleti se izra|uju od ostatka bukve i hrasta. Bukve je bilo oko 90 %. Radi dobivanja cjelovite slike utro{ka elektri~ne energije po jedinici proizvedenih peleta i zahtjevane snage na mjernim mjestima tijekom rada pojedinih elektromotora mjerena je djelatna i jalova snaga. Mjerni rezultati pohranjivani su izravno na ra~unalo i poslije analizirani programskim paketom LabVIEW. Iz rezultata mjerenja numeri~kom je integracijom dobivena potro{nja elektri~ne energije na strojevima u promatranom razdoblju, iz ~ega je odre|ivana prosje~na elektri~na snaga. [to se ti~e rezultata mjerenja obavljenih na ~etiri skupine elektromotora, najve}a potro{nja elektri~ne energije izmjerena je pri samom pre{anju, a potom na skupini elektromotora za pripremu iverja za peletiranje. Najve}a je potro{nja elektri~ne energije za izradbu jedne tone peleta pri pre{anju 138,98 kWh/t i transportnoga sustava 74,25 kWh/t, {to ukupno iznosi 213,23 kWh/t ili gotovo 95 % u odnosu na ukupno utro{enu elektri~nu energiju. Izmjerena je elektri~na energija za izradbu jedne tone peleta pri pre{anju od tvrdoga drva (u istraìvanom slu~aju bukva) za 100 % ve}a nego ona prikazana u literaturi koja se odnosila za meko drvo. Zbog navedenoga potrebno je u svakoj studiji razlikovati utro{ak energije pri pre{anju mekoga ili tvrdoga drva. Zabiljeèna je manja potro{nja energije kod ivera~a s nao{trenim noèvima u iznosu od 24,3 % u odnosu na nenao{trene noève. Klju~ne rije~i: {umska biomasa, peleti, energija

Authors’ addresses – Adresa autorâ: Assoc. Prof. Stjepan Risovi}, PhD. e-mail: risovic@sumfak.hr Igor \uki}, MSc. e-mail: dukic@sumfak.hr Forestry Faculty of Zagreb University Department of Process Tehniques Sveto{imunska 25 HR–10 000 Zagreb CROATIA

Received (Primljeno): March 13, 2008 Accepted (Prihva}eno): June 10, 2008

108

Kre{imir Vu~kovi}, BSc. e-mail: kresimir.vuckovic@fsb.hr University of Zagreb Faculty of Mechanical Engineering and Naval Architecture I. Lu~i}a 5 HR–10 000 Zagreb CROATIA Croat. j. for. eng. 29(2008)1

Editorial – Uvodnik TIBOR PENTEK, TOMISLAV POR[INSKY Are We on the Right Path?? ......................................................... 1 Jesmo li na pravom putu?

Orginal scientific papers – Izvorni znanstveni radovi BO[TJAN KO[IR Modelling Stand Damages and Comparison of Two Harvesting Methods ........................ 5 Modeliranje o{te}enja sastojine i usporedba dvaju sustava za pridobivanje drva HIDEO SAKAI, TOMAS NORDFJELL, KJELL SUADICANI, BRUCE TALBOT, EBBE BØLLEHUUs Soil Compaction on Forest Soils from Different Kinds of Tires and Tracks and Possibility of Accurate Estimate .................................................... 15 Zbijanje {umskoga tla razli~itim tipovima guma i polugusjenica te mogu}nost dovoljno precizne procjene JANUSZ SOWA, DARIUSZ KULAK Probability of Occurrence of Soil Disturbances during Timber Harvesting ...................... 29 Vjerojatnost pojave o{te}enja tla pri pridobivanju drva MARIJAN [U[NJAR, DUBRAVKO HORVAT, ANDRIJA KRISTI], ZDRAVKO PANDUR Morphological Analysis of Forest Tractor Assemblies ..................................... 41 Morfolo{ka ra{~lamba {umskih traktorskih skupova IGOR POTO^NIK, TIBOR PENTEK, DRAGUTIN PI^MAN, IVICA PAPA, ANTON POJE Filling in the Clearance of a Forest Road Cross-Section in Beech Forest ........................ 53 Popunjavanje svijetloga otvora popre~noga profila {umske ceste u bukovim sastojinama AKBAR NAJAFI, HOSHANG SOBHANI, ARASTOU SAEED, MAJID MAKHDOM, MOHAMMADREZA MARVI MOHAJER Planning and Assessment of Alternative Forest Road and Skidding Networks ................... 63 Planiranje i procjena kakvo}e razli~itih ina~ica mre`e {umskih prometnica TIBOR PENTEK, HRVOJE NEVE^EREL, TOMISLAV POR[INSKY, DRAGUTIN PI^MAN, KRUNO LEPOGLAVEC, IGOR POTO^NIK Methodology for Development of Secondary Forest Traffic Infrastructure Cadastre ............... 75 Metodologija izrade katastra sekundarne {umske prometne infrastrukture

Preliminary notes – Prethodna priop}enja SVEN KORTEN, CHRISTIAN KAUL Application of RFID (Radio Frequency Identification) in the Timber Supply Chain ............... 85 Primjena RFID-a (radijske frekvencijske identifikacije) u lancu dobave drva STJEPAN RISOVI], IGOR \UKI], KRE[IMIR VU^KOVI] Energy Analysis of Pellets Made of Wood Residues ...................................... 95 Energijska ra{~lamba peleta izra|enih od drvnoga ostatka ISSN 1845-5719

9 771845 571000