Drought Analysis for Water Planning in Agriculture (A Case-study of Gourma Province, Eastern Burkina Faso)

Abstract

Drought is a naturally occurring event that human beings grapple with the question of how to mitigate its impacts. Certainly, Agriculture is one of the biggest casualties of drought events but the extent to which the sector is affected is imperceptible. This thesis report presents a method of quantifying the impact of drought on Agriculture and proposes a method of selecting mitigation strategies suited to Gourma province. Stakeholders in the agricultural sector have to make policy and technical decisions affecting their citizens based on inaccurate information and thus end up reacting to droughts while the damage is already done. Farmers get reduced yields at the end of the season and have nothing to eat or generate income due to inadequate planning at start of season. Based on the Effective Reconnaissance Drought index (eRDIst), this study characterises all the agro-meteorological drought events in Gourma province between 1979 and 2013. Potential yield (Yp) and water limited yield (Yw) simulations help determine the yield gap (Yg) and by regression modelling, the relationship between drought and yields is established through a mathematical model and is further used to estimate farmer’s losses. Multi Criteria Decision Making (MCDM) provides a way of selecting suitable Drought Mitigation Options (DMOs) from the available options. Analysis of eRDIst shows occurrence of three drought events spread out in nine years with different levels of severity, frequency and duration. The years 1983 and 2008 show low drought intensity while the event between 1999 and 2005 has the highest intensity. From the resulting mathematical models, both drought and extremely wet conditions affect the yields. Analysis using the mathematical models show that a unity change in the drought conditions only results in a reduction of yields by 775Kg/Ha translating to losses of 676$/Ha while in the case of both drought and extremely wet conditions, the mathematical model shows that yield reduction stands at 995 Kg/Ha which is equal to 760$/Ha. Under short term DMOs, the over exploitation of aquifers, restricted irrigation during droughts and reallocation of water were found to be suitable while for long term solutions, insurance was found to be the best performing. This study finally recommends Le Guardien de l’espoir insurance scheme as a safety net for the farmers. From this study, a more accurate method of demonstrating the impact of drought on Agriculture has been established. It is also key to note that more research needs to be carried out on why the relationship between drought and yields in medium textured soils is stronger than in fine textured soils. The suitability of insurance schemes for farmers managed by the farmers themselves as a tool for mitigating the drought impacts needs to be further researched on too.