Enhancing Water Resilience in Haiti: Understanding Extreme Events and Urbanization Effects.

Abstract

Water supply, management, and associated risks are constant concerns for the Haitian population. The country faces severe issues such as waterborne diseases, cyclones, floods, landslides, and droughts every year. These challenges are exacerbated by climate change, increasing the frequency and intensity of extreme weather events. In response, the Université d'État d'Haïti à Limonade (UEH-CHCL) and Auburn University have partnered to establish a water resilience unit in northeastern Haiti, aiming to improve water resource management, support climate change adaptation, and foster research and innovation. Addressing these issues in Haiti requires using novel data sources, as the ground-based monitoring networks that form the basis of water resources management in the United States and other developed countries are not available. This study explores the use of novel data sources to evaluate flood and drought scenarios in the Grande Rivière du Nord watershed, a critical region for agriculture and water supply, supporting thousands of people and essential for food security. We investigated hydrological trends and patterns to understand the impacts of climate change and urbanization on water resources. We analyzed two key hydrological variables: maximum annual precipitation and maximum instantaneous discharge. For precipitation data, we used the Global Precipitation Measurement Integrated Multi-Satellite Retrievals (GPM-IMERG) from Earthdata and for discharge, we used the GeoGLOWS Global Streamflow and Flood Forecasting System. Due to current travel restrictions in Haiti, which significantly impacted our ability to collect field data, we utilized gridded data products with suitable spatial and temporal resolution as a surrogate. An annual maxima series spanning approximately 25 years (2000–2025) was constructed for each variable. We fitted Gumbel and Lognormal distributions to the data using L-moment methods. We applied goodness-of-fit tests to identify the best fitting distribution, including root mean square error (RMSE), mean absolute error (MAE), and probability plot correlation coefficient (PPCC). Furthermore, we calculated exceedance probabilities, enabling the determination of return periods of 2, 10, 20, 50, and 100 years. Furthermore, hydrological modeling scenarios were developed using HEC-HMS to assess the impact of urbanization on streamflow patterns and flood risks. A baseline scenario with no impervious surface was compared to two post-urbanization scenarios: one based on impervious surface estimates from 2023 Environmental Systems Research Institute (ESRI) land cover data, and another representing a 20% increase in imperviousness. The model was calibrated using GeoGLOWS streamflow data corrected with the Simbi hydrological database. This research contributes to enhancing water resilience in Haiti and supports informed decision-making for improved water resource management and flood risk mitigation.