Understanding the rapid increase in drought stress and its connections with climate desertification since the early 1990s over the Brazilian semi-arid region

The aridity-related drought studies have been carried out extensively in Brazilian semi-arid ecosystems, although there is no report on relating aridity with different drought indices over the recent decades. Due to drought impact on agriculture and natural ecosystems, it has attracted extensive attention in the academic community. In this study, the Standardized Precipitation Evapotranspiration Index (SPEI), the Soil Moisture and Ocean Salinity-derived Soil Water Deficit Index (SWDI), the Meteosat Spinning Enhanced Visible and InfraRed Imager (SEVIRI) radiance-derived solar and infrared estimates, and the SEVIRI-derived Normalized Difference Vegetation Index (NDVI) datasets were employed to investigate the spatial and temporal characteristics of the drought episodes and the vegetation dynamic responses. An approach was implemented for identifying drought using a combination of mathematical and statistical features derived from the SPEI. The results showed that: (1) The frequency, duration, and intensity of drought severity identified by the SPEI and SWDI from 1990 to 2022 showed different characteristics in the Brazilian semi-arid region from 1990 to 2022. The frequency of moderate, severe, and extreme drought ranged from 5% to 92%, the duration was mostly concentrated in 5–6 dry months. (2) The whole Brazilian semi-arid region exhibited an overall drying tendency. (3) The NDVI-derived vegetation in the Brazilian semi-arid ecosystems exhibited an overall decreased trend during 2004–2022, indicating a degrading trend of the vegetation cover. The region where degrading trend of NDVI was negatively correlated with SPEI accounted for approximately 13% of the region. (4) The frequency, and average duration of the flash drought were 21% concentrated in 6 pentads, during the region growing season. (5) The spatial and temporal trends with the significance test of the rainfall, air temperature, and SEVIRI radiance-based estimations suited well with those of the SPEI data. These research results were significant for the region to respond to and prevent drought through human-induced land degradation.