Droughts, which consist of natural and transient water deficits, endanger the water, food, and energy security of ecosystems. An effective drought preparedness strategy must focus on mitigating socioeconomic vulnerabilities to the phenomenon. In this context, the article aims to analyze the multi-annual characteristic of droughts in the Northeast region of Brazil. The study identifies each pixel under a spatial-temporal perspective of the phenomenon's severity, based on data from the Drought Monitor, providing a multi-annual and cumulative view of the extreme event. The approach aggregates a series of spatial-temporal drought severity data in a single product with clusters of monthly data referring to 5 years of outputs from the Drought Monitor. The results reveal that from July 2014 to June 2019, over 75% of the Northeast region of Brazil (NEB) area registers exceptional drought for, at least, 1 month along the 5 years. This result means impacts on water supply and crops that correspond respectively to the water dams collapsing volumes and widespread agricultural losses. Despite the natural recurrence and constancy of drought phenomena in the region, water management policies have often based the building of disproportionate and costly water infrastructure works. Cumulative multi-annual analysis of the outputs of drought monitoring can be an early warning system for water basins, regions, or municipalities, notably those who usually live with recurrent highly severe droughts, and remain neglected upon a remarkable vulnerability. This approach also supports proactive and preventive drought management instead of the usual emergency crisis management model.
Keywords: Cumulative approach; Drought categories; Northeast region Brazil; Pixel trajectory; Spatial analysis.