Abstract

Continuous remote-sensing monitoring of sand in desert areas and the exploration of the spatio–temporal evolution characteristics of soil–wind erosion has an important scientific value for desertification prevention and ecological restoration. In this study, the Kubuqi Desert was selected as the study area, and the Landsat series satellite remote sensing data, supplemented by satellite remote sensing data such as GE images, SPOT-5, ZY-3, GF-1/2/6, etc., integrated object-oriented, decision tree, and auxiliary human–computer interaction interpretation methods, developed the Kubuqi Desert area dataset from 1990 to 2020, and established a soil erosion intensity database of the past 30 years based on the soil–wind erosion correction equation. The results show that the application of the training samples obtained by a high-score collaborative ground sampling to land use/cover classification in desert areas can effectively improve the efficiency of remote-sensing mapping of sand changes and the accuracy of change information identification, and the overall accuracy of the classification results is 95%. In general, the sandy area of the Kubuqi Desert area has decreased year by year, during which the mobile sand in the hinterland of the desert has expanded in a scattered distribution. The overall soil–wind erosion intensity showed a downward trend, especially since 2000; the ecological improvement trend after the implementation of desertification control projects is obvious. Changes in the sand type contributed the most to the reduction of soil–wind erosion intensity (contribution 81.14%), ecological restoration played a key role in reducing the soil–wind erosion intensity (contribution 14.42%), and the increase of forest and grass vegetation covers and agricultural oases played a positive role in solidifying the soil- and wind-proof sand fixation. The pattern of sandy land changes in desert areas is closely related to the national ecological civilization construction policy and the impact of climate change.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call