TITLE:
Forecasting Catches of Lates stappersii in Lake Tanganyika (Burundi): Insights for Adaptive Fisheries Governance and Food Security
AUTHORS:
Vianney Ngabo Mulema, Wales Singini, Gabriel Balagizi Baguma, Odette Karerwa, Eunice Musau Ntambwe, Fiston Bokwala Bobitshe, Aristote Yofemo Bifinole, Alidor Kankonda Busanga
KEYWORDS:
ARIMA, Pelagic Fish Dynamics, Lake Tanganyika Fisheries, Adaptive Management, Food Security
JOURNAL NAME:
Agricultural Sciences,
Vol.17 No.11,
November
18,
2026
ABSTRACT: Fisheries in Lake Tanganyika are vital for food security in Burundi, yet Lates stappersii catches face pressures from overexploitation, climate variability, and weak governance. This study applies Autoregressive Integrated Moving Average (ARIMA) modelling to forecast catches, making the first application in this approach to Burundi’s pelagic fisheries. Annual catch data (2007-2025) were analysed using Box-Cox transformation, stationarity tests, and Box-Jenkins methodology. The ARIMA (0, 1, 1) model emerged as the most parsimonious specification, validated through residual diagnostics and comparative benchmarking against ETS and Random Walk models. Forecasts for 2026-2035 project a stable mean catch of ~232 tons annually, while the 80% and 95% prediction intervals widen substantially with the forecast horizon, reflecting increasing uncertainty in long-term projections. These results provide a quantitative basis for adaptive fisheries management and highlight the importance of precautionary strategies to safeguard food security and biodiversity in Burundi and the wider Great Lakes region.