DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagoðeno pretraživanje po punom tekstu



ŠUMARSKI LIST 11-12/2018 str. 50     <-- 50 -->        PDF

increase disproportionately (Figure 5). These changes can be easily understood from the s.d. and S.E.E. of the foliage fuel distributions in the DBH proportions (Table 5).
DISCUSSION AND CONCLUSIONS
RASPRAVA I ZAKLJUÈCI
Regression models were developed to assess the foliage, branch, and active and total fuel biomass of 84 Turkish red pine. The regression models aimed to describe the relationship between fuel biomass components and tree properties (H, DBH, CW, CL). When the adjusted coefficient of determination (R2adj) values were evaluated, it was observed that fine branches had the highest explanation percentage with 90%, while total fuels had second highest with 89,8%, and thick branches had third highest with 89%. The lowest R2adj value was observed with fine branches, at 60.4%, followed by foliage, at 64.8%. Although the R2adj value was generally high in equations involving both CW and CL, and generally low in equations involving H, it was still understood that all tree properties could be used for determining crown fuel loads in Calabrian pines. Comparisons with other studies on crown fuel load in Calabrian pines indicate that H, DBH, CW, and CL are all important predictive variables for this tree (Küçük et al., 2008; Zianis et al., 2011).
In a study performed by Küçük et al. (2008) on Turkish red pine forests across northwestern Turkey, the mean oven-dried weight of total crown fuel biomass for trees and saplings together, and its properties was 2.47 kg/tree (n= 324, mean DBH= 15.49 cm., mean H= 2.27 m., mean CW=1.1 m., mean CL= 1.6, s.d.= 5557.2, S.E.E.= 319.781, R2adj= 0.944). The mean value of total crown fuel biomass for trees and saplings in our study was 14,098 kg/tree (Table 2). Another study performed by Küçük and Bilgili (2008) found results similar to our study. In the said study, the mean oven-dried weights of fuel biomass was 16.54 kg/tree (n= 35, mean DBH= 15.91 cm, mean H= 10.25 m., mean CW=3.74 m, mean CL= 5.75, R2adj= 0.799). In a study performed by Zianis et al. (2011) in the Turkish red pine forests of Greece’s island of Crete and Lesvos. The total fuel load results on island Create were closer to the findings of Küçük et al. (2008). In this study performed in Crete, the mean oven-dried weight of the total crown fuel biomass (T1+T2) was determined as 3.83 kg/tree (n= 12, mean DBH= 18 cm, mean H= 8 m). T1 is dry biomass of needles and twigs up to 0.63 cm in diameter and T2 is dry biomass of branch wood 0.64-2.5 cm in diameter. As these comparisons indicate, the fuel biomass of Calabrian pines can differ between regions. In this study, an increase in DBH did not inevitably result in an increase in foliage fuel biomass, while CR was generally found to scatter disproportionately. However, this does not mean that trees with high DBH values consistently have lower foliage fuel biomass than trees with low DBH values. It must be considered that such a generalization might lead to erroneous and inaccurate assessments. A study performed by Affleck et al. (2012) to characterize the crown profile and crown mass of conifer forests showed that the total crown fuel biomass distributed disproportionately from the relationship between CR and DBH. The same study concluded that in crown biomass studies, large conifer trees are generally present in smaller number, while their effect on overall biomass per unit area is disproportionate. In fact, the species included in the study of Affleck et al. (2012), which was conducted in the Interior Northwest of USA, are quite different from Calabrian pine. The similar results compared with our study regarding to same tree species appeared in the study of Zianis et al. (2011). The distribution of average total crown fuel biomass to the DBH sizes (7,3 – 30 cm) in their study were disproportionately and differentiated in each site. The variability between biomass equations are generally due to the increasing size between the independent variables (Zianis and Mencuccini, 2004).
Although fuel characterization and classification is a mathematical modelling (Alexander 2007), the differences in these models is generally due to the distinguishing features of individuals in nature, and the complex compositions that stem from structural and spatial distributions (Affleck et al., 2012; Fernandes, 2009). In addition, it is believed that the hazard, risk, and severity of forest fires are also associated with the ecological context, which includes components such as historical natural fire regimes, time, space, and process (Hardy 2005). For this reason, there is a need to simply and constantly renew and develop fuel classification approaches (Sandberg et al., 2001). Although fuel characterizations and classifications have great importance in fire behavior modelling, using them on their own is not sufficient for fire decision support systems. Especially in large administrative areas, fuel loads will not be unique due to the reasons mentioned above, and there is consequently a need for different fuel load standards rather than a single fuel load standard. In decision-making processes for fire