Modelling and forecasting Lake Malawi water level fluctuations using stochastic models

Abstract

Fluctuations of water level in Lake Malawi has become a big concern among stakeholders owing to its hydro-ecological and socio-economic implications. The study was aimed at modelling and forecasting patterns of Lake Malawi water levels in Malawi to provide a likely trend in the future. The study used Seasonal Autoregressive Integrated Moving Average (SARIMA) processes to select an appropriate stochastic model for forecasting the monthly data for water levels in Lake Malawi for the period 1986 to 2015. The appropriate model was chosen based on SARIMA (p, d, q) (P, D, Q)S process. The SARIMA (1, 1, 0) (2, 1, 1)12 model was selected to forecast the monthly data of water levels for Lake Malawi from August, 2015 to December, 2021. The plotted time series data showed that water level in Lake Malawi has been decreasing since 2010 to date, but not as much as was the case between 1995 and1997. The forecast of water level in Lake Malawi until 2021 showed a mean of 474.45masl ranging from 473.93 to 475.04masl with a confidence interval of 95% against registered mean of 473.40masl in 1997 and 475.48masl in 1989, which were the lowest and highest water levels on the lake respectively since 1986. The forecast implies that by the year 2021, water level in Lake Malawi will fall below the actual recorded mean by 0.15masl and 0.69masl from the maximum ever recorded. It is however unlikely to be lower than the level recorded in 1997.