Professional paper
ProductionAspect of Silver Fir (Abies alba Mill.) Dieback on Example of a few Beech-fir Stands
Elvis Paladinić
orcid.org/0000-0003-4886-4476
; Hrvatski šumarski institut Jastrebarsko
Denis Štimac
orcid.org/0000-0001-6323-0581
; Hrvatske šume, UŠP Delnice
Hrvoje Marjanović
orcid.org/0000-0001-5701-7581
; Hrvatski šumarski institut Jastrebarsko
Ivan Balenović
orcid.org/0000-0001-7422-753X
; Hrvatski šumarski institut Jastrebarsko
Maša Zorana Ostrogović
orcid.org/0000-0001-9369-0685
; Hrvatski šumarski institut Jastrebarsko
Abstract
The research problem, raised by Croatian Forests Ltd., Forest Administration Delnice, is the frequent discrepancy in planned versus realised yields during selecting cutting in fir-beach forests. This discrepancy between planned and realised wood assortments structure and at the same time between revenues, is significant for fir-beach stands affected by fir trees dieback. The result is that performed sylvicultural activities in such disturbed stands, in order to get timber assortments, have questionable cost-effectiveness, and in some cases are fully unprofitable. The key point in process of calculating planned values according to timber assortments are so called “Assortment tables”, empirical models for estimation of timber assortments structure, which are definitely inappropriate for current climatic and disturbed stand conditions. Hence, the aim of this research is to study existing Assortment tables for fir, which are used by the Production department of Forest Administration branch Delnice, in order to satisfy criteria for sufficiently reliable planning in the production of timber assortments.
Material and methods. In order to ensure quality and uniformity of research from the be ginning, the detailed plan of field activities was designed and agreed for implementati on with Production department. On the research area in target Forest management un i ts, a few forest compartments were chosen for field data acquisition each year. In each of selected forest compartments it was required to collect data for around 100 marked fir tr ees, which were felled during selecting cutting. Diameter distribution of those hundred tr ees should represent fir diameter distribution of whole compartment. Each of those fir tr ees got measured diameter at breast height, total height, height of the first branch, and all bucked timber assortments. From those data we got volumes (per trees and per timber assortment). Ultimately, assortment structure data were collected from seven forest co mpartments. Each compartment has two dataset, one representing produced timber assortments (done by foresters) versus planned assortment structure (calculated using assortments tables).
Concerning research aim, two hypotheses (H1, H2) were constructed. H1 states: If planned and realised timber assortments production from forest compartment significantly affected by fir dieback is compared, than statistically significant differences between planned and realised values according to assortments classes will be confirmed. H2 states: If timber assortments production is compared according to three general health categories of fir stands, than statistically significant differences between same assortment classes will be confirmed. For testing H1, One-way ANOVA was used together with Test of Homogeneity of Variances. Analysed variables are veneer logs volume, sawlogs (1st, 2nd and 3rd class), pulp-wood and firewood, and volume of wood residues (recovered wood). Pulp-wood and firewood are in the same category, because only 14 pieces of firewood were measured in whole sample. Wood residue category refers to timber assortments thicker than 16 cm and left on forest floor because of its uselessness for technical purposes (damaged, decayed).
Statistical significance of differences between samples of planned and realised volume according to assortment classes was analysed. Testing homogeneity of variances between these two samples showed in most cases that variances of the same assortment classes’ volumes are not homogenous. Considering this findings, Independent-samples T test was used for analyzing H1. Analog to H1 testing, the same tests were used for H2 analyzing. Forest compartments which were compared represented two of three health categories of fir stands (stands of moderately damaged health condition were missing). Those two compared groups of data were unequal in size, one consisting of data acquired from 5 forest compartments, and another of data acquired from two forest compartments.
This research was partially extended with sampling of fallen fir trees to get a clue about insect and pest species which could be one of main factors causing fir dieback. So, in the 2008 woody samples collected from fallen trees in one forest compartment, were analysed in CFRI laboratory.
Research area. In wider context, research area represents fir-beach stands of Gorski Kotar region, and narrower area is determined by boundaries of Forest Administration Branch Delnice i.e. forest area affected by intensive fir trees dieback. The stands in focus represent main fir-beach forest community in Gorski Kotar, Abieti-Fagetum illyricum Ht. Research activities on field data acquisition were started in forest management unit “Brloško”. Data about forest compartments included in research are presented in Table 1.
Results of research and discussion. Firstly, the results of H1 are presented and discussed. Statistically significant differences between planned and realised volumes were confirmed for veneer assortment, for all forest compartments except one. The results were expected because five of seven compartments are considered as stands of significantly disturbed health condition, where possibility for production of high quality timber assortments from fir trees is minimal. Another fact considerably affects on differences between planned vs realised timber assortment structure is assortments tables, according to which each tree with dbh higher than lowest threshold for veneer class gets certain veneer volume from tree volume. Veneer volume participates in the whole sample of produced timber assortment volume with 0.59 %, and concerning forest compartments from 0.00 % to maximum 3.01 % (see Table 2).
T-test results for 1st class saw logs confirmed discrepancies between realised and planned quantities for all forest compartments except compartment no.15 in forest management unit “Brloško” (t value is 0.80). Volume of 1st class saw logs participates in the whole sample of produced timber assortment volume with 5.85 % to 35.95 %, concerning forest compartments. Different from previous, t-test for 2nd class saw logs confirmed discrepancies between realised and planned quantities for three forest compartments, two representing category of good health condition, and one category of worse health condition (table 3). Volume of 2nd class saw logs participates in the whole sample of produced timber assortment volume with 13.93 % to 31.28 %, concerning forest compartments. Testing volume of 3rd class saw logs confirmed discrepancies between realised and planned quantities for four forest compartments (two in category of good health condition, and two in category of worse health). This assortment class after pulp-wood and firewood has the highest share in whole sample volume of assortment classes. Volume of 3rd class saw logs participates in the whole sample of produced timber assortment volume with 16.29 % to 36.90 %, concerning forest compartments.
The most frequent timber assortment is pulp-wood and firewood category, leading by number of logs. This fact is in line with disturbed wood quality of standing fir trees which are in process of dieback or completely dead trees in the felling moment. Analysing discrepancies between realised and planned quantities for this assortment category, tests confirmed statistically significant difference for all forest compartments except one. Tested wood volume of the category participates in the whole sample of produced timber assortment volume with 6.38 % to 54.41 %.
Recovered wood (i.e. wood residue) assortment is problematic because of its technical inapplicability. This assortment category consisted of small number of pieces, and according to assortment tables each tree gets certain volume of the assortment (so called wood waste). This discrepancy is supported with T-test results confirming significant differences between realised and planned quantities for all forest compartments. Hence, further consideration of the recovered wood assortment was cancelled.
Regarding to obtained results, it can be concluded that H1 is accepted because statistically significant differences have been confirmed for majority of tested combinations. Testing H2 by ANOVA and T-test, for each assortment class were confirmed differences between assortment volume variances comparing two health categories of stands (table 4). Other words, the relationship between health categories of fir stands and realised assortment structure has been confirmed. Concerning this, in process of timber assortment production planning more attention on assortments tables should be done (carefully choosing and constructing new ones adapted for different health categories).
Conclusions. This project was aimed to give additional contribution in resolving discrepancy between planned and realised timber assortment production from fir-beech stands with different degree of fir trees dieback. The main reason for this discrepancy, especially in case of same health category stands, most probably lays on current assortments tables which were constructed in 90’is and updated a few times mostly on empirical basis. Another reason, but with considerably smaller influence on final production plan than assortment tables is inappropriate tariff selection.
Performed statistical tests confirmed significant differences between planned and realised timber assortment quantities in case of veneer. And such results are in strong correlation with worse health condition of the majority ofresearched forest compartments (except the ones in f.m.unit “Široka Draga”). Tests have been confirmed H1 on other timber assortments mostly for 1st and 3rd class of saw logs, pulp-wood and firewood category, and for 2nd class saw logs haven’t. Worse wood quality of fir trees from stands with significant fir dieback resulted in the statistically significant differences between planned and realised values of pulp-wood and firewood assortment volumes. Recovered wood category hasn’t been further analysed because of quite a small sample collected on the field.
Testing H2 on realised production data of the same timber assortment types between forest compartments belonging different health category, significant discrepancies in quantities of assortment volumes between two observed health categories have been confirmed.
Final recommendation or question to forest Management is “Would it be possible to adjust spatio-temporally the realisation of sanitary selecting cutting in fir-beach forests on yearly basis, in order to minimise timber production costs?”
Keywords
health status category of stand; silver fir; dieback; forest woody assortments; planned production; realised production
Hrčak ID:
72334
URI
Publication date:
14.10.2011.
Visits: 3.069 *