Statistical Correlation of Some Meteorological Variables for Solar Energy Generation Forecasting and Grid Integration in Abuja, North-Central, Nigerian
A. B. Gbenro *
National Space Research and Development Agency (NASRDA), Abuja. Nigeria.
Ene Denis George
National Space Research and Development Agency (NASRDA), Abuja. Nigeria.
Ozoemelam Onyebuchi *
National Space Research and Development Agency (NASRDA), Abuja. Nigeria.
Iloanusi Nkiru Lilian
National Space Research and Development Agency (NASRDA), Abuja. Nigeria.
Mustapha Shaibu
National Space Research and Development Agency (NASRDA), Abuja. Nigeria.
*Author to whom correspondence should be addressed.
Abstract
Accurate forecasting of solar photovoltaic (PV) generation is important for reliable grid integration in Nigeria, where dedicated solar irradiance measurement infrastructure is largely absent. Six years of monthly observational data (2015–2020, n = 72) from the Nigerian Meteorological Agency (NiMet) and the Tropospheric Data Acquisition Network (TRODAN) were used in the analysis, comprising 3,149–3,513 hr/yr of sunshine (mean: 3,324 hr/yr), annual rainfall of 1,087 mm, relative humidity, and rainfall rate. Descriptive statistics, one-way analysis of variance (ANOVA) for both monthly and annual groupings, linear regression for temporal trend detection, and Pearson correlation analysis were applied to evaluate seasonal patterns, inter-annual variability, and inter-variable relationships. The results demonstrate that all four variables exhibit highly significant monthly variability (p < 0.001), confirming a well-defined seasonal pattern comprising a wet season (April–October) and a dry season (November–March). Annual ANOVA revealed no significant inter-annual differences in rainfall amount, sunshine, or rainfall rate, indicating stable year-to-year climatological patterns. Relative humidity displayed the strongest seasonal coupling with rainfall. Pearson correlation coefficient (PCC) analysis revealed statistically significant strong negative correlations between sunshine duration and rainfall (r = -0.7387, p < 0.01), rainfall rate (r = -0.7463, p < 0.01), and relative humidity (r = -0.7291, p < 0.01). Rainfall and rainfall rate were virtually collinear (r = 0.9969), while rainfall and relative humidity showed a strong positive association (r = 0.8134). The results confirm that moisture-related parameters (rainfall, rainfall rate, and humidity) collectively constitute reliable inverse proxies for solar resource availability, providing a low-cost, operationally actionable basis for generation forecasting for the Nigerian grid. The six years of ground-based data used in this study serve as a first step in a feasibility study for solar farm development at the location and will be extended when funding becomes available. The study provides a local empirical basis for assessing solar potential in FCT and helps to address the gap in local-level solar resource studies in North-Central Nigeria.
Keywords: Solar forecasting, Pearson correlation, rainfall, relative humidity, grid integration, PV generation prediction, meteorological variables, sunshine duration, ANOVA, seasonal variability