Revealing the scale-specific controls of soil organic matter at large scale in Northeast and North China Plain

Abstract

Soil organic matter (SOM) plays an important role in terrestrial ecosystem functioning and is closely related to soil fertility and soil health. Better understanding on the spatial distribution of SOM is vital to agriculture and C-cycle management. With the advancement of digital soil mapping framework and data mining technology, selection of environmental covariates become critical to identify the controls of SOM spatial distribution at different scales and were rarely discussed in previous studies. The objectives of this study were to separate the scale-specific variations in SOM and their dominant controls at those scales along two transects from Northeast and North China Plain. Spatial distribution of SOM was separated into seven scale components (six details, D1 through D6 and one approximation, A6) along each transect using the discrete wavelet transform. The largest variations in SOM were separated in A6 (>1280km) along the northeast transect (91.2% of the total) and D5 (320–640km) along the north transect (40.6% of the total). Unlike the northeast transect, considerable amount of variations was also separated in other scale components of the north transect. There were no significant correlations between the scale-components of SOM and terrain factors along both the transects. While a relatively stronger correlation was observed between SOM and climatic and vegetation factors along the northeast transect, no significant correlation was observed along the north transect at large scales. This may be due to the long-term cultivation in the Northern China Plain. Principal components 1 and 3 identified from the proximally sensed visible and near infrared (vis–NIR) spectra had strong correlation with SOM along the north transect while the principle component 2 was highly correlated with SOM along the northeast transect. The scale-specific controlling factors at different locations may help in selecting environmental factors in digital soil mapping at different scales for improving mapping accuracy.