Scale-Location Dependence Relationship between Soil Organic Matter and Environmental Factors by Anisotropy Analysis and Multiple Wavelet Coherence

Abstract

Soil organic matter (SOM) and environmental factors have been shown to have a scale-location dependence relationship. However, few studies have considered the anisotropy, and the scale-location dependence relationship may not be fully characterized. In this study, transects with dominant directions of SOM variability in the dryland farming regions of Songliao Plain, China were extracted by anisotropy analysis. The scale-location specific multivariate relationships between SOM and environmental factors along the two transects were examined using multiple wavelet coherence. Results indicated that the scale and location-specific variations in SOM and environmental factors were direction-specific. The major direction with the most significant SOM variations was 56° east by north, while the minor direction was perpendicular to the major direction. The strongest single factor for explaining SOM variations differed between two dominant directions, sand along the major direction (average wavelet coherence (AWC) = 0.57, percentage area of significant coherence (PASC) = 40.32% at all scales) and bulk density (BD) along the minor direction (AWC = 0.66, PASC = 50.16% at all scales). The combination of mean annual temperature (MAT) and BD was the best to explain SOM variations along the major direction (AWC = 0.78, PASC = 46.23% at all scales). A two-factor combination is adequate to explain SOM variability along the major direction, whereas a single factor is sufficient for the explanation along the minor direction. More factors did not evidently increase or even decrease the percentage of scale-location domains where SOM variations were significantly explained. This work has important implications for developing future sampling strategies and preparing detailed digital soil maps.