American Journal of Water Resources, 2020, Vol. 8, No. 1, 31-37 Available online at http://pubs.sciepub.com/ajwr/8/1/4 Published by Science and Education Publishing DOI:10.12691/ajwr-8-1-4 Statistical Study of Dry Spells and Their Impact on Rainfed Corn in the Burkinabe Sahel Chaim Vivien Doto * , Dial Niang, Malicki Zorom, Hamma Yacouba International Institute for Water and Environmental Engineering, 01 P.O. BOX poisson@594, Ouagadougou, Burkina Faso *Corresponding author: vivien.doto@2ie-edu.org Received December 10, 2019; Revised January 12, 2020; Accepted February 03, 2020 Abstract This study aimed at characterizing dry spells and their potential impact in corn rainfed agriculture in the Burkinabe Sahel in order to help managers of rural environments to better plan the corn growing season in this area characterized by high climatic variability. This characterization was made possible through a typology of intra- dekadal dry spells and an inter-dekadal analysis of dry spells with the first-order Markov chains, allowed proposing a mitigation alternative of the effect of dry spells on corn growing. Results showed that climatic risk which arises for rainfed corn production is reduced if sowing was done in the third 10-day period (or dekad) of June. This is justified by the fact that the probability of having two consecutively dry dekads during the most sensitive periods to water stress (i.e the mid-season) was less than 20% while it was over 30% for two consecutively wet dekads. However, because of the increased potential impact of the dry spells longer than 7 days during the mid-season, the coupling of this sowing scenario to supplemental irrigation would be the robust alternative adaptation to climate variability in the Burkinabe Sahel. Nevertheless, subsequent studies can be carried out to quantify this potential impact of dry spells on corn under rainfed farming. Keywords: dry spells occurrence, rainfed farming, climatic risk, corn, Sahel, Burkina Faso Cite This Article: Chaim Vivien Doto, Dial Niang, Malicki Zorom, and Hamma Yacouba, “Statistical Study of Dry Spells and Their Impact on Rainfed Corn in the Burkinabe Sahel.” American Journal of Water Resources, vol. 8, no. 1 (2020): 31-37. doi: 10.12691/ajwr-8-1-4. 1. Introduction Rainfalls in semi-arid areas, particularly in the Sahelian zone are very erratic and characterized by a very high inter-annual variability. This inter-annual variability of rainfall in the Sahel is characterized by its persistence and its magnitude which date back to the years 1966-1968 where the rainfall deficit reached 50% [1]. In the Burkinabe Sahel a reduction of 15-30% of rainfall in the years 1970 and 1980 was reported by [2]. Some authors were able to show that this variability has caused the drying of the Sahelian zone [3,4,5] due to a descent of isohyets of 100 to 150 km southwards for the period 1945- 1969 compared to the period 1969-1990 and a persistent degradation of arable lands [6]. This situation makes vulnerable the Burkinabe Sahel farming activity which remains largely rainfed and of which alone further characterization of frequent, and recurrent dry spells during the short rainy season of three months would be one of the better alternatives to establish adaptation scenarios. This characterization is particularly important for growing corn, a high value cereal but little grown in the Burkinabe Sahel because of climatic risk that the crop faced [7]. Although corn is the third most grown cereal in Burkina Faso after sorghum (Sorghum bicolor) and millet (Pennisetum glaucum) [8], it better uses water than these cereals most preferred in the Sahel due to their higher ability to resist drought. This ability to withstand drought has already been demonstrated by [9,10,11] who have shown that this adaptation emanates from the calibration of the duration of the crop cycle on the dates of start and end of the rainy season. Thus, in view of these conditions, analysis of the rainfall must be done in terms of probability of occurrence for significantly reducing corn farming climatic risk. This is justified by the fact that agricultural option is a bet for agro-climatic conditions. It is therefore essential to establish a correspondence between the risk incurred by the farmer and a climatic risk model to facilitate its decision-making [12]. But the efficient forecasting of certain weather events is not easy. However, authors have already shown in the 1950s [13,14,15] and recently [16,17,18] that the statistical study is an excellent tool for identifying some aspects of the phenomenon. Among these tools, the Markov chain [19,20,21] is one of such tools that allows calculating the probability of occurrence of an event knowing that it occurred in a recent past. The first work on the use of Markov chains in rainfall are from Gabriel and Newman [22]. Thanks to a first-order Markov chain, these authors were able to describe the intermittence of daily rainfall in Tel Aviv. Thus, through this process the persistence and magnitude of drought in the Sahel can be better understood. It is in this way that this study was carried out and aimed at not only analyzing the statistical distribution