Quantitative discrimination of the influences of climate change and human activity on rocky desertification based on a novel feature space model

Abstract

Under the stress of the global change, rocky desertification has become a serious environmental problem in the Karst Mountain area. At present, few remote sensing monitoring research works on rocky desertification based on feature space model have been conducted and reported. In this study, the Albedo-LST feature space remote sensing monitoring index based on point-point model has been proposed, and subsequently the spatio–temporal evolution pattern and driving mechanism of rocky desertification in Dafang district from 1986 to 2019 were analyzed. The results show that: (1) The point-point Albedo-LST feature space model of rocky desertification has good applicability with the overall accuracy of 90.79%; (2) From 1986 to 2019, the rocky desertification in Dafang district first showed an increasing trend (1986–2005) and then a decreasing trend (2005–2019); (3) The comprehensive evolution frequency of rocky desertification during 2001–2005 was the largest with 7.51% a−1, which was related to the implementation of the Grain for Green Project; (4) The single factor with the largest contribution rates to rocky desertification are land use type, landform, and temperature. The interactive factors with the largest explanatory power are temperature ∩ land use type and landform ∩ land use type. The research results can provide decision support for the prevention and control of rocky desertification in Southwest China.

Keywords

• albedo-lst feature space
• rocky desertification
• spatial–temporal evolution
• driving mechanism
• geodetector