Abstract
Understanding the spatial pattern and driving factors of forest carbon density in mountainous terrain is of great importance for monitoring forest carbon in support of sustainable forest management for mitigating climate change. Methods:We collected the forest inventory data in 2015 in Shanxi Province, eastern Loess Plateau of China, to explore the spatial pattern and driving factors of biomass carbon density (BCD) for natural and planted coniferous forests using Anselin Local Moran's I, Local Getis-Ord G* and semivariogram analyses, and multi-group structural equation modeling, respectively.
The result of spatial pattern of BCDs for natural forests showed that the BCD was generally higher in the north but lower in the south of Shanxi. The spatial pattern for planted forests was substantially different from that for natural forests. The results of multi-group SEM suggested that elevation (or temperature as the alternative factor of elevation) and stand age were important driving factors of BCD for these two forest types. Compared with other factors, the effects of latitude and elevation on BCD showed much greater difference between these two forest types. The difference in indirect effect of latitude (mainly through affecting elevation and stand age) between natural and planted forests was to some extent a reflection of the difference between the spatial patterns of BCDs for natural and planted forests in Shanxi.
The natural coniferous forests had a higher biomass carbon density, a stronger spatial dependency of biomass carbon density relative to planted coniferous forests in Shanxi. Elevation was the most important driving factor, and the effect on biomass carbon density was stronger for natural than planted coniferous forests. Besides, latitude presented only indirect effect on it for the two forest types.
