From Methodological Evaluation to Operational Implementation: Development of an Integrated Climate Service for Drought Monitoring and Forecasting in Cuba

Meteorological drought is one of the main climate-related hazards affecting water resources, agriculture, ecosystems and socioeconomic activities in Cuba. In response to the need for timely, spatially consistent and operational drought information, an integrated climate service is being developed within the SIN-SEQUÍA project, whose objective is to strengthen drought resilience under climate change conditions in Cuba. The system is conceived not only as a web platform, but as a complete methodological and technological workflow for drought monitoring, seasonal forecasting and dissemination. The development process integrates several complementary components. First, different statistical and hybrid approaches for seasonal precipitation prediction were evaluated, including candidate predictor selection, principal component regression schemes and the implementation of the Climate Predictability Tool/PyCPT framework. These studies supported the identification of a robust configuration for operational seasonal precipitation forecasting in Cuba. Second, the monitoring component builds on R_SVsequia, a Cuban-developed tool that uses gridded monthly precipitation and recent rain-gauge observations to calculate drought-related indicators, including precipitation, standardized precipitation indices at different time scales, drought status and hazard levels. These outputs are consistent with the products planned for the SIN-SEQUÍA platform, which include monitoring, forecasting, precipitation, SPI, status and hazard information. A key operational challenge is the manual preparation of the most recent precipitation information. To address this limitation, the integration of GPM/IMERG satellite precipitation estimates is proposed as an operational alternative for the latest month, after temporal aggregation, interpolation to the national 4 km grid and bias correction using the historical observational database. The final framework links national observations, satellite precipitation, seasonal prediction methods and a Django-based online platform to improve the timeliness, reproducibility and accessibility of drought information for decision-making in Cuba.