Determining the influence factors of soil organic carbon stock in opencast coal-mine dumps based on complex network theory.

Abstract The soil organic carbon (SOC) stock is severely affected by the exploitation of opencast coal mines, and the relationships of factors influencing SOC stock are complex. The influence factors of SOC stock in reclaimed lands in opencast coal mines are unclear, and the existing models can not characterize the complex relationship of SOC stock sufficiently. This paper analyzed the influence factors of SOC stock using complex network theory in the Antaibao opencast coal-mine in Shanxi province of China. An investigation of the soil, topography and vegetation in 50 reclaimed plots was performed. Soil factors, i.e., rock content (RC), total nitrogen (TN), available phosphorus (AP), available potassium (AK), soil bulk density (BD), soil water content (SWC), SOC stock, electrical conductivity (EC), clay content, silt content, sand content and pH, vegetation factors, i.e., above-ground biomass (AGB), tree volume (TV), herb coverage (HC), canopy density (CD), and topography factors, i.e., slope, slope aspect (SA) and slope position (SP), were selected as the nodes, and the relationship among the various factors were considered as the edges to construct a complex network using Gephi. The network characteristics, including degree, betweenness and average shortest path so on, were calculated. SOC stock, SWC and BD played an important role in the complex network of SOC stock. The SOC stock was affected by three clusters: soil texture cluster, soil physicochemical property cluster, and vegetation-topography cluster. SOC stock network was not stable and was sensitive when some important nodes were disturbed. The complex network theory could be used to analyze the influence factors of SOC stock. This study provided a reference for selecting rational land reclamation measures to increase soil carbon stock. Highlights • Complex network was used to analyze SOC stock in reclaimed lands of mining area. • Influence factors of SOC stock were BD, SWC, TN, RC and clay content. • SOC stock network was divided into three clusters structure. • SOC stock network was sensitive to interference. [ABSTRACT FROM AUTHOR]