Abstract
This study analyses Makurdi rainfall data from 1955 to 2015, for the purpose of detecting the rainfall variability, trends, and assessing its implications on flood occurrence. Data on daily rainfall and flood occurrences were collected from relevant government agencies. Rainfall trends were analyzed using Mann–Kendall, Spearman’s rank correlation and Student’s t tests, both giving almost similar results for the three methods. Rainfall variability was analyzed using standardized precipitation index (SPI). The result of the analysis indicated that 1955 and 1993 were the wettest years, while 1958 and 2003 were the driest. Also, September was the wettest, closely followed by August, while December was the driest. A negative (decreasing) statistically insignificant trend for the total annual rainfall data from 1955 to 2015, but positive (increasing) statistically insignificant trend for the extreme annual rainfall was observed. The findings revealed that extreme rainfall events corresponded to flood occurrence. Thus, the annual rainfall total is not necessarily an important cause of increased flood frequency, but the fraction of it that occurs as extreme rainfall. It is recommended that effective mitigation measures should be put in place for likely floods that may arise owing to the anticipated extreme rainfall magnitudes in the coming years in Makurdi.

Key words: Rainfall, trend, variability, extreme rainfall, flood, mitigation