Abstract

Drought is a complex climatic phenomenon with cascading effects on natural ecosystems and human systems, particularly in semi-arid regions such as Guanajuato, Mexico. Its impacts are multifaceted, affecting water availability, agricultural productivity, public health, and environmental stability. Prolonged periods of below-average precipitation can lead to water shortages for domestic, industrial, and agricultural use, while also increasing the risk of wildfires and degrading water quality. These conditions can trigger economic losses, especially in rural communities dependent on rain-fed agriculture and livestock. To effectively respond to these challenges, robust monitoring tools and data-driven decision-making frameworks are essential. This study presents an analytical and visualization framework based on the Standardized Precipitation Index (SPI) to assess drought conditions across Guanajuato from January 2000 to November 2023. The SPI enables multi-temporal analysis of precipitation anomalies, allowing for the identification of short- and long-term drought events. Using Python for data processing and interactive visualization, we restructured raw precipitation data from 29 meteorological stations, cleaned inconsistencies, and developed dynamic time series and geospatial heatmaps. The final dataset comprises 8,294 records organized into 286 monthly entries across 29 stations. Missing values, identified as NaN, zero, and placeholder values (-999.99), were imputed using inter-annual interpolation. Temporal trends and spatial patterns reveal recurring drought episodes, particularly in the northern and central zones of the state. The integration of interactive maps and time series visualizations provides stakeholders with intuitive tools for early drought detection and response planning. This work establishes a foundational model for drought monitoring in Mexico, emphasizing the importance of open data, reproducible methods, and user-friendly interfaces in climate resilience strategies.