Flood prediction of the Black Volta in a climate change scenario

The Lawra District of the Upper West Region (the driest region in Ghana) has experienced periodic and devastating flash flood resulting from high intensity short duration rainfall, a characteristic of semi-arid and arid regions. Statistical analysis showed the correlation between rainfall, temperature and flow rates and their variability and using different statistical approaches for determining which analysis best fit a particular parameter for forecasting. With increasing temperatures comes high rainfall from thunderstorm activities, therefore predicting future climate give an idea of future rainfall patterns and even the occurrence of the associated flash floods. Time Series, Trend Analysis, Moving Average, Weighted Moving Average, Exponential Smoothing, Percentage Growth and Seasonality, were used for analysis on 31 years of data. For 2011 the observed average values were 894.48mm, 1044.26m³/s and 28.8°C for rainfall, flow rates and temperature respectively. Trend and seasonality analysis best responded to the forecasting of the parameters with 902.52mm, 1056.26m³/s and 28.70°C for annual averages. Annual mean rainfall intensities above 1000mm can cause flood but those above 1100mm are more likely to cause severe damages. There is a significant change in temperature over long periods whereas changes for short periods are quite minimal. Nonetheless any change in temperature is a major concern testifying that flood events will be experienced more often than before.