Detection of the Evolution Process of Desertification in Gulang County Based on Long Series and Similar Time Images

Panpan Liu,Bing Guo,Rui Zhang
DOI: https://doi.org/10.3390/land13101652
IF: 3.905
2024-10-11
Land
Abstract:Previous studies are mostly conducted based on sparse time series and often ignore the dramatic changes in desertification during the year. Utilizing the Landsat and MODIS data sets from 2000 to 2020, this study applied the spatio-temporal fusion algorithm to obtain the images of the study area taken at similar times in August over the past 20 years. The optimal desertification remote sensing monitoring index of Gulang County was constructed based on the feature space model, and then the spatial and temporal evolution patterns and the driving mechanism of desertification in Gulang County were revealed by using a geographic detector. The research results were as follows: (1) The ESTARFM algorithm had better applicability in constructing long time series and similar time images with the correlation coefficient R2 = 0.83 between the results of the ESTARFM fusion model and the original image; (2) the SWCI-MSAVI feature space desertification monitoring index model based on point-to-point mode had the best applicability with an overall accuracy of 95.39% and a Kappa coefficient of 0.94; (3) from 2000 to 2020, the desertification showed an increasing trend, and the degree of desertification gradually intensified from south to north in Gulang County; (4) the dominant factors in various historical periods were different, which were mainly composed of precipitation, temperature and population density. The dominant interactive factors changed from alternating dominance of natural factors and human activity factors to the co-dominance of natural factors and human activity factors. The research results could provide decision-making support for precise prevention and control of desertification in arid–semi-arid regions.
environmental studies
What problem does this paper attempt to address?