Rainfall Stability and Depth Dependent Soil Moisture Coupling in Southern Nigeria: A 45 Year CHIRPS and SMAP Assessment of Edo State

Robust characterization of rainfall variability and its coupling with soil moisture is central to climate adaptation, agricultural planning, and water resource management in data sparse tropical regions, yet subnational hydroclimatic assessments remain scarce across southern Nigeria. This study integrates a 45-year (1981 to 2025) monthly rainfall record from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) with an 11-year (April 2015 to December 2025) record of Soil Moisture Active Passive (SMAP) Level 4 surface and root zone soil moisture to characterize rainfall climatology, long term trends, and rainfall to soil moisture coupling across the 18 Local Government Areas (LGAs) of Edo State. Rainfall analyses cover the full 1981 to 2025 record, whereas rainfall to soil moisture coupling analyses cover only the overlapping April 2015 to December 2025 period (129 months). Trends were assessed using the Mann Kendall test, Sen's slope estimator, two autocorrelation robust modified Mann Kendall tests, and pixel level false discovery rate correction; coupling was quantified through standardized monthly anomaly analysis, Pearson correlation with effective sample size adjusted significance and false discovery rate control, seasonal stratification, and lagged response analysis, with the lag of maximum correlation tested against a permutation null, using CHIRPS aggregated to the SMAP grid by area weighted averaging. Mean annual rainfall averaged 1801 millimeters per year (coefficient of variation of 10.6 percent), with no statistically significant trend detected at either the statewide (Sen's slope −0.71 millimeters per year, 95 percent confidence interval −5.53 to 4.10, or −0.4 percent of the mean per decade) or pixel scale (731 valid of 733 CHIRPS pixels) under any test applied after false discovery rate correction. Surface soil moisture anomalies were more strongly coupled to rainfall (mean r of 0.39, significant in 96.5 percent of 230 cells after correction) than root zone anomalies (mean r of 0.20, significant in 34.3 percent of cells); the surface response was essentially immediate (0-month lag at every cell), whereas the root zone optimal lag ranged from 0 to 3 months and was weakly resolved, because mean correlations at lags 0 to 3 differed by less than 0.03. Surface coupling was strongest in the dry season and weakest during the peak monsoon, no seasonal root zone correlation was significant, and coupling varied among LGAs independently of mean annual rainfall. These findings establish a spatially explicit, statistically robust baseline of rainfall stability and depth dependent soil moisture memory for a data sparse tropical setting, with direct relevance for drought early warning, planting window guidance, and flood and gully erosion risk management in southern Nigeria.

Keywords: CHIRPS, SMAP, soil moisture, rainfall trend analysis, Mann Kendall test, false discovery rate correction, land atmosphere coupling