Wavelet Coherence Between Large-Scale Climate Indices and Precipitation Variability over Nigeria (1961–2021)

Abstract

Precipitation variability across Nigeria is strongly modulated by remote climate teleconnections, yet existing studies rarely examine multi-index interactions across localized time–frequency scales. This study applies wavelet coherence and cross-wavelet phase analysis to examine the relationships between six large-scale climate indices — the Atlantic Meridional Mode (AMM), Atlantic Multidecadal Oscillation (AMO), North Atlantic Oscillation (NAO), Niño 3.4 (NINA), Pacific Decadal Oscillation (PDO), and Southern Oscillation Index (SOI) — and monthly precipitation across Nigeria's five climatic zones (Sahel, Sudan savannah, Guinea savannah, tropical rainforest, and coastal) from 1961 to 2021. Monthly rainfall data were extracted from the ERA5 reanalysis across twenty-five representative locations, and spatial correlation structures were mapped using ordinary kriging interpolation. The AMM exhibited the strongest and most persistent coherence with precipitation across all five zones, concentrated predominantly at near-annual periodicities (~0.0625 – 0.125), identifying it as the primary large-scale driver of Nigerian rainfall variability. The AMO and NAO demonstrated intermittent, seasonally gated coherence; Niño 3.4 displayed distinct anti-phase coherence at 2–8-year periodicities concentrated in the tropical rainforest zone; while the PDO and SOI exhibited weak, inconsistent associations. Spatially, negative correlations for the AMM, AMO, and NAO strengthened toward the southern coastline, whereas Niño 3.4 showed an inland-to-coastal gradient of increasingly negative correlation. These findings demonstrate the priority of the AMM for seasonal rainfall forecasting in Nigeria and provide zone-specific empirical evidence to support agricultural planning and water resource management.