Figures & data
Figure 1. Geographical location of the study area and spatial distribution of the main deserts in ACA.
![Figure 1. Geographical location of the study area and spatial distribution of the main deserts in ACA.](/cms/asset/f2440bb2-b363-4916-9bef-102795ea6397/tjde_a_2202421_f0001_oc.jpg)
Figure 3. Thematic maps of SDS source effective predictor variables and RGB image. (a) DLSTR. (b) Vol_Wt_S. (c) LCT. (d) Wt_C. (e) NDVI. (f) So_Sa. (g) Slope. (h) So_Wt. (i) Sur_R. (j) Wd_Sp. (k) Sn_C. (l) A_Temp. (m) T_Prec. (n) So_Temp. (o) RGB image.
![Figure 3. Thematic maps of SDS source effective predictor variables and RGB image. (a) DLSTR. (b) Vol_Wt_S. (c) LCT. (d) Wt_C. (e) NDVI. (f) So_Sa. (g) Slope. (h) So_Wt. (i) Sur_R. (j) Wd_Sp. (k) Sn_C. (l) A_Temp. (m) T_Prec. (n) So_Temp. (o) RGB image.](/cms/asset/0a09e158-7364-4edd-85c2-84c59b7ddf22/tjde_a_2202421_f0003_oc.jpg)
Table 1. Summary of the input variables in this study.
Table 2. Performance evaluation of four ML based models.
Figure 4. ROC curves for the four methods in SDS source susceptibility prediction based on nine model iterations.
![Figure 4. ROC curves for the four methods in SDS source susceptibility prediction based on nine model iterations.](/cms/asset/54494489-6c64-492d-b1cc-d6dae9bb3bd7/tjde_a_2202421_f0004_oc.jpg)
Figure 5. PR curves for the four methods in SDS source susceptibility prediction based on nine model iterations.
![Figure 5. PR curves for the four methods in SDS source susceptibility prediction based on nine model iterations.](/cms/asset/698fd2a1-733c-40b1-b48a-2cd894e7c307/tjde_a_2202421_f0005_oc.jpg)
Figure 6. SDS source susceptibility maps produced via the ML-based methods in spring. (a) 2008-4-29, (b) 2016-5-11, (c) 2018-4-18.
![Figure 6. SDS source susceptibility maps produced via the ML-based methods in spring. (a) 2008-4-29, (b) 2016-5-11, (c) 2018-4-18.](/cms/asset/4f65a7ad-4838-4fd1-9c84-9c613f10cecc/tjde_a_2202421_f0006_oc.jpg)
Figure 7. SDS source susceptibility maps produced using the ML-based methods in summer and autumn. (d) 2010-8-26, (e) 2015-6-10, (f) 2019-9-24.
![Figure 7. SDS source susceptibility maps produced using the ML-based methods in summer and autumn. (d) 2010-8-26, (e) 2015-6-10, (f) 2019-9-24.](/cms/asset/dd61d2a2-2ced-4267-969d-d30f63ecced9/tjde_a_2202421_f0007_oc.jpg)
Figure 8. SDS source susceptibility maps produced using the ML-based methods in winter. (g) 2007-3-1, (h) 2013-1-23, (i) 2019-1-2.
![Figure 8. SDS source susceptibility maps produced using the ML-based methods in winter. (g) 2007-3-1, (h) 2013-1-23, (i) 2019-1-2.](/cms/asset/4560c604-b193-48c7-814a-eaa287f767ab/tjde_a_2202421_f0008_oc.jpg)
Figure 9. Relative importance of the variables to the prediction model outputs based on the random forest.
![Figure 9. Relative importance of the variables to the prediction model outputs based on the random forest.](/cms/asset/2663d86f-241d-4b43-85cd-45081395f249/tjde_a_2202421_f0009_oc.jpg)