Abstract
Desertification is one of the main environmental threats to the Brazilian Semi-Arid region, causing biodiversity loss, soil degradation, and a decline in agricultural productivity. Monitoring droughts through remote sensing is fundamental to understanding climate dynamics and supporting mitigation strategies. In this respect, the objective of this study was to analyze the temporal variability of drought in four recognized desertification nuclei in Northeast Brazil: Cabrobó, Gilbués, Irauçuba, and Seridó, from 2001 to 2024. The methodology consisted of evaluating drought scenarios using the Vegetation Condition Index (VCI), Temperature Condition Index (TCI), and Vegetation Health Index (VHI), derived from the MODIS sensor. The results indicated a strong seasonality, with more critical values in the second half of the year, when VCI, TCI, and VHI frequently fell below 30, characterizing moderate to severe drought. The core of the Seridó region presented the highest levels of vulnerability, with average VCI and VHI values below 40 and a strong correlation with precipitation (r > 0.60). The Mann-Kendall test did not indicate statistically significant monthly trends, suggesting stability in the historical drought pattern, although regression analysis revealed signs of worsening in Gilbués. Thus, the effectiveness of SR-derived indices in characterizing drought events and highlighting the combined influence of climatic, edaphic, and anthropogenic factors was confirmed. It is concluded that systematic monitoring using SR is essential to support adaptation and mitigation policies in the face of desertification in the Brazilian Semi-Arid region.