Spatiotemporal Recurrence Quantification Analysis of Atmospheric Solar Transmittance Indices across the West African sub-Saharan Region: Seasonal Modulation and Spatial Predictability Gradients

We investigated the spatiotemporal characteristics of the nonlinear dynamics of three solar transmittance indices from the atmosphere. Clearness Index (MT), Diffuse Fraction (Md), and Transmitting Coefficient (Mt), at sixteen locations representing different climatic zones of Nigeria. Four measures were computed from hourly MERRA-2 reanalysis for monthly and seasonal timescales and used the Recurrence Quantification Analysis (RQA) algorithm: Recurrence Rate (RR), Determinism (DET), Shannon Entropy (ENT), and Laminarity (LAM), with phase-space reconstruction using empirically determined embedding parameters (τ = 6 h, m = 8). There was a consistent seasonal modulation of all metrics across the regions under investigation: the Core Dry Season yields the highest DET (up to 0.96) and ENT (up to 2.79) at all sixteen stations, while the Monsoon Season yields the lowest values. The August Break is independently recovered as a local DET maximum at Port Harcourt in 6 of 10 study years and intermittently at other coastal stations. A north–south gradient in dry-season predictability is observed, with the Spearman correlation between station latitude and Core Dry DET reaching rs = +0.78 (p < 0.001); Jos and Maiduguri jointly record the highest Core Dry DET (0.963) and Gboko the lowest (0.754). Of the three indices, the Md is consistently the most stochastic, and LAM shows a distinct, complementary seasonal pattern to DET and ENT. The results may serve as a dynamic basis for the design and sizing of solar energy resources and storage in zone-differentiated solar energy planning across the West African sub-Saharan climatic zones.