Machine learning algorithm for estimating karst rocky desertification in a peak-cluster depression basin in southwest Guangxi, China

Abstract

Karst rocky desertification (KRD) has become one of the most serious ecological and environmental problems in karst areas. At present, mapping KRD with a high accuracy and on a large scale is still a difficult problem in the control of KRD. In this study, a random forest (RF) based on maximum information coefficient and correlation coefficient feature selection is proposed to predict KRD. Nine predictors stood out as feature factors to estimate KRD. Rock exposure was the most important predictor, followed by fractional vegetation cover for the prediction of KRD processes. The kappa and classification accuracy indexes were to evaluate the performance of the model. We recorded overall accuracy rate and kappa index values of 94.7% and 0.92 for the testing datasets respectively. The RF model was then used to predict the KRD in 2001, 2011, 2016, and 2020, and it was found that the KRD in the study area has exhibited a positive trend of improvement. Therefore, the use of multisource remote sensing data combined with the RF model can obtain better prediction results of KRD, thereby providing a new idea for large-scale estimation of the KRD in peak-cluster depression.

Zoning strategies for ecological restoration in the karst region of Guangdong province, China: a perspective from the “social-ecological system”

“…The primary task is to delineate comprehensive zoning. Previous land spatial zoning studies in karst areas have not comprehensively considered the spatial variations of desertification and human disturbances (Zhang et al, 2020). Some researchers have established ecological restoration zoning at the basin and grid scales, they have elucidated the inherent characteristics and relationships between various levels of zoning, but precise boundary delineation for implementation was not addressed (Wang and Zhao, 2022;.…”

Monitoring and classification of karst rocky desertification with Landsat 8 OLI images using spectral indices, multi-endmember spectral mixture analysis and support vector machine

“…This comparison revealed different results from the spectral indexes with a high accuracy rate, as in other studies in the DPM literature[7,45].Traditional methods (field studies) for monitoring karst rocky desertification mainly rely on field surveys, which require a quite deal of time and finance. Although using remote sensing images for monitoring karst rocky desertification is not limited by the topography, erosion or land cover, it also has disadvantages, including poor image quality (cloudiness), readily affected by human subjectivity, and strictly guaranteeing the accuracy[46]. In this study, field studies from traditional karst rock desertification methods were also carried out in order to detect the inadequacies that may occur as well as the low…”