Rainfall timing: variation, characteristics, coherence, and interrelationships in Nigeria

This study sought to improve general understanding on rainfall timing in Nigeria. Observed monthly rainfall data for 38 meteorological stations, spanning 30years was utilized, and the dates of onset and cessation of rainfall were estimated using Walter's method. The interrelationships among the variables and the dependency of the amount of rainfall were calculated using Pearson's moment correlation coefficient and multiple regression, respectively. The variance of the Standardized Anomaly Index (SAI) was used as a measure of spatial coherence. The study revealed divergent characteristics between the onset and the cessation of rainfall; progressing in opposite direction, with a very high latitudinal variation in each. Rainfall onset has a relatively high interannual variability, with an average CV of 21%, compared to 3.9% for cessation. Onset also has a larger spatial range of 137days, compared to 82days for cessation. Additionally, rainfall cessation was found to be more spatially coherent than the onset, but both revealed a weak spatial coherence. The relative spatial incoherence and high interannual variability of onset contrive to make it much less predictable than cessation. Interestingly, rainfall amount shows greater association with, and dependency on, onset, cessation, and rainy season length in stations of the extreme north, than in southern stations.