Estimation and spatial pattern analysis of forest biomass in fenglin nature reserve based on geostatistics

The biomass of 120 randomly selected from 158 sample plots, were calculated based on forest inventory data in 1997, Fenglin National Nature Reserve, Northeastern China, through the biomass estimation model. In addition, the remaining 38 sample plots were used to validate the model with correlation coefficients between the predicted and real values. Furthermore, the spatial distribution of forest biomass for each cell of the study region was derived by the spatial interpolation with geostatistics of ArcGIS on the biomass of those plots, and the relationship between the spatial distribution pattern of biomass, forest structures (including 8 forest type and 5 age groups) and terrain factors (including elevation, aspect and slope) were analyzed. The results showed that the interpolation method is mostly available method to estimate forest biomass based on forest inventory point data, since the predicted and real biomass was mostly correlated (R² = 0.78). Spatially, the biomass varied obviously in this region, it ranged from 41. 6 t/ hm² to 376. 9 t/ hm², and at average the biomass was 171. 5 t/ hm². The total biomass of this region kept up to 3. 08 million ton for that time. Respect to forest biomass and forest structure, the biomass varied significantly with forest age, and increased with increasing forest age. The oldest forest yielded more biomass than younger forest; the biomass for the former was up to 174. 8 t/ hm² and for the latter reached 155. 1 t/ hm². In addition, the biomass varied with forest type, and decrease from coniferous forest (e. g. fir- spruce forest), conifer-deciduous mixed forest (e. g. mixed Korean pine forest) to deciduous forest. The biomass in coniferous forest, conifer-deciduous mixed forest and deciduous forest were 175. 0 t/ hm², 162. 5 t/ hm², and 156. 7 t/ hm², respectively. Respect to forest biomass and terrain factors, the biomass was significantly affected by the terrain factors. The sites at the increasing elevation yielded lower biomass, and at the gradient of elevation (from 250 m to 700 m) the biomass decreased from 175. 0 t/ hm² to 155. 1 t/ hm². The sites at the wet-cold aspect showed higher biomass than that on dry-warm aspect, and at the gradient it ranged from 174. 0 t/ hm² to 163. 5 t/ hm². However, the relationship between slope index and the biomass showed more complicated results, the biomass decreased from 176. 0 t/ hm² to 167. 0 t/ hm² while the slope increased from 0^oto 24^o, and then increased to 175. 0 t/ hm² while slope continue to increase. Our results would be a guide to forest manager to estimate the biomass based on the inventory data, and provide a way to mitigate forest type and age to control local biomass and carbon stocks.