Estimation of soil organic carbon under different vegetation types on a hillslope of China's northern Loess Plateau using state-space approach

Soil organic carbon (SOC) plays a critical role in revegetation of semi-arid areas. The accurate estimation of SOC under various land-use types is fundamental to sustain ecosystem productivity. Thus, the dominant factors of the spatial distribution of SOC in shallow soil layers were determined at hillslope scale. The state-space modeling approach was used to quantify the relationship between SOC stock and land-use type, soil properties, topographic features, and fine root biomass (FRB) at 0-20 and 20-40 cm soil layers of a hillslope on the Loess Plateau. The best state-space models explained more than 96% of the variations in SOC stocks on the hillslope. The best multivariate state-space models including land-use type, FRB, soil pH, and total nitrogen were optimal for 0-20 and 20-40 cm soil layers. Land-use type was the dominant factor for identification of localized variation in SOC in the 0-40 cm soil layer. The results underscored the importance of land-use type in SOC variation on the hillslopes of the Loess Plateau. It also provided a useful insight into the accurate estimation of SOC using state-space modeling approach driven by other easily obtainable variables