Rachel Rubanguka
Research Mentor(s): Andrew Gronewold
Mentor Department: School for Environment and Sustainability
Authors: Rachel Hoops, Andrew Gronewold
Session: Session 5 (2:00pm – 2:50pm)
Presentation Type: Oral
Abstract
Lake Malawi and the Shire River system are essential water resources for Malawi, supporting hydropower, irrigation, fisheries, water supply, transportation, tourism, wildlife conservation, and agriculture. However, extreme lake level changes, including those associated with a lake outflow blockage (1915–1937) and the record-high level of 1980, have made water management challenging. Maintaining a minimum outflow from the lake (170 m³/s) for hydropower is also crucial, yet existing models do not always provide accurate estimates of key hydrological components. So far, no study has applied the Large Lake Statistical Water Balance Model (L2SWBM) to Lake Malawi, leaving a gap in understanding its long-term water balance. L2SWBM is a Bayesian statistical framework designed to improve estimates of precipitation, evaporation, inflows, outflows, and water levels by integrating multiple datasets and resolving inconsistencies across different models. It has been successfully applied to the Laurentian Great Lakes, and a terminal lake in North America’s Great Basin, improving water balance calculations. Unlike traditional models, L2SWBM combines multiple data sources, producing more reliable estimates based on data quality. This study applies L2SWBM to Lake Malawi using water level records from 1954 to 2015, with a gap between 1986 and 1991. Using multiple datasets, the model refines estimates for precipitation, evaporation, inflow, outflow, and water level. Results show that the L2SWBM improves water balance estimates and reconciles differences in existing datasets. This study is the first to use L2SWBM for Lake Malawi, providing a valuable tool for hydrological assessments. The findings offer better estimates for water balance components, helping policymakers, hydropower planners, and water managers make informed decisions.
