Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

The geographical location map of the study area (a) displays the elevation distribution and meteorological station locations in the Beijing-Tianjin-Hebei-Shandong-Henan region, while the land cover map (b) categorizes all land types into seven classes: forest, grassland, wetland, cropland, urban, barren, and water.

More »

Fig 1 Expand

Table 1.

Data sources and specifications of remote sensing variables used for drought monitoring.

More »

Table 1 Expand

Table 2.

Drought classification criteria based on SPEI values.

More »

Table 2 Expand

Table 3.

Definition, calculation formula, and source references of drought monitoring indices used in this study.

More »

Table 3 Expand

Fig 2.

Flowchart of the drought prediction model construction.

More »

Fig 2 Expand

Fig 3.

A comparison of annual total rainfall (mm) recorded by 30 ground-based observation stations with CHIRPS data, along with their linear regression curves, is presented.

More »

Fig 3 Expand

Fig 4.

A comparison of the monthly mean rainfall (mm) data with CHIRPS records.

The inset in the upper left corner shows boxplots of precipitation in spring, summer, autumn, and winter.

More »

Fig 4 Expand

Fig 5.

Maps illustrating the spatial distribution patterns of 5-year average rainfall in the Beijing-Tianjin-Hebei-Shandong-Henan Region from 2000 to 2020.

More »

Fig 5 Expand

Fig 6.

Scatter plots comparing drought predictions from an ensemble ML model with observed values across four spatial/temporal scales:

(a) SPEI-1, (b) SPEI-3, (c) SPEI-6, and (d) SPEI-12, respectively.

More »

Fig 6 Expand

Table 4.

Drought categorization consistency rate at each scale (n = 5016).

More »

Table 4 Expand

Fig 7.

Correlation heatmap (* indicates significance, ***, **, * represent P < 0.001, P < 0.01, P < 0.05 respectively), with histograms of frequency distributions for each variable displayed along the diagonal.

More »

Fig 7 Expand

Fig 8.

Bar and hive plots depicting variable importance based on SHAP values for the RF model’s prediction of SPEI-3 (SPEI-3 characterizes agricultural drought).

More »

Fig 8 Expand

Fig 9.

Time series curves of two temporal scales (SPEI-1, SPEI-6) and sites PDSI bar charts calculated from observational data (2017–2020).

More »

Fig 9 Expand

Fig 10.

Spatial Distribution of Drought Severity Classifications at Two Temporal Scales (SPEI-1 and SPEI-6) in the Beijing-Tianjin-Hebei-Shandong-Henan Region.

More »

Fig 10 Expand