Agriculture in Nigeria is mainly dependent on rainfall which is variable in nature. Therefore, the need to have a full knowledge of its pattern and trend is pertinent in order to achieve food sufficiency. This study examined the effect of precipitation effectiveness indices: Onset date, cessation date, length of rainy season and occurrence of pentad dry spells (5, 10 and ≤15) days on the yield of millet, sorghum, rice and maize in Sokoto State, between 1993 and 2008. Walter‟s 1967 method was used to derive the selected precipitation effectiveness indices while dry spell parameters, was derived in pentads. Crop yield data in (ton/ha) and the selected precipitation effectiveness indices were harmonised using log10 method. Trendline and linear trendline equations where fitted to show the direction of change. While simple correlation was used to analyse the relationship between crop yield and the selected precipitation indices. Furthermore, regression analysis was used to determine to what extent the selected precipitation effectiveness indices influence the yield of the selected crops. The findings revealed that the trend of these precipitation effectiveness indices are characterized by marked “noises” and variability.
Additionally, onset dates of rainfall are arriving earlier while cessation dates are arriving latter. Consequently the length of rainy season is increasing. Dry spells of 5 days is common while dry spell incidents of 10 days and ≤15 consecutive days is decreasing. Similarly rice and maize yield are increasing, due to early onset date of rains. The regression summary shows that the selected precipitation effectiveness indices account for only 52.1%, 50.4%, 52.1% and 68% of yield variation in maize, millet, rice and sorghum respectively. The study therefore, recommends government support for farmers to increase the cultivation of rice and maize in the study area, and the introduction of more crop varieties that are tolerant to dry spells of 5 consecutive days.