<p>Rainfall is vital for sustaining water reservoirs and mitigating climate-related disasters like floods and droughts. However, climate change has made rainfall patterns increasingly unpredictable, which can lead to water shortages or reservoir overflows. This study analysed the relationship between ground-based and gridded rainfall estimates across the diverse eco-climatic regions of Nigeria over 39 years (1981–2019), using data from 48 ground-based stations, 11 gridded rainfall products, and Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations under three Representative Concentration Pathway (RCP) scenarios (2.6, 4.5, and 8.5). It evaluates predictive models, including stepwise regression, Auto-Regressive Integrated Moving Average (ARIMA), and Global Circulation Models (GCMs). Stepwise regression emerged as the most accurate method, with strong correlations (R² = 0.88–0.98) and low error rates, outperforming ARIMA (which under-predicted rainfall by 9–34%) and GCMs (which over-predicted by 6–33%). The Global Precipitation Climatology Centre (GPCC), Global Precipitation Measurement (GPM), Integrated Multi-Satellite Retrievals for GPM (IMERG), and TAMSAT African Rainfall Climatology And Time Series (TARCAT) were identified as the most reliable datasets for predicting ground-based observations. These findings highlight the importance of region-specific dataset selection and robust predictive modelling, with direct implications for water resource management, agricultural planning, and disaster preparedness in Nigeria.</p> Graphical Abstract <p>This Study Enhances Rainfall Predictability in Nigeria by Integrating Ground-based measurements, satellite-derived Precipitation Datasets, and Climate Model projections. The Research Evaluates Rainfall Estimation across Diverse eco-climatic Regions Using 48 Ground stations, 11 Gridded Rainfall products, and CMIP5 Model Outputs (RCP 2.6, 4.5, and 8.5). The Methodology Involves Spatial mapping, Data Consistency checks, and Model Ranking Using Stepwise Regression, ARIMA, and CMIP5 simulations. Results Indicate that Stepwise Regression Outperforms ARIMA and GCMs, Showing the Highest Predictive Accuracy (R² = 0.88–0.98), while ARIMA Underestimates Rainfall (9–34%) and GCMs Overestimate It (6–33%) in some Regions. The GPCC, GPM-IMERG, and TARCAT Datasets Emerged as the Most Reliable Precipitation sources. These Findings Provide a Data-driven Approach for Optimizing Rainfall forecasting, Essential for Nigeria’s Water Resource management, Agricultural planning, and Climate Adaptation Strategies.</p>

Optimising Rainfall Prediction in Nigeria: Integrating Ground-Based, Satellite, and Climate Model Data

期刊论文
发表日期：

2025年10月30日

查看全文

Afeez Alabi Salami

摘要

Rainfall is vital for sustaining water reservoirs and mitigating climate-related disasters like floods and droughts. However, climate change has made rainfall patterns increasingly unpredictable, which can lead to water shortages or reservoir overflows. This study analysed the relationship between ground-based and gridded rainfall estimates across the diverse eco-climatic regions of Nigeria over 39 years (1981–2019), using data from 48 ground-based stations, 11 gridded rainfall products, and Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations under three Representative Concentration Pathway (RCP) scenarios (2.6, 4.5, and 8.5). It evaluates predictive models, including stepwise regression, Auto-Regressive Integrated Moving Average (ARIMA), and Global Circulation Models (GCMs). Stepwise regression emerged as the most accurate method, with strong correlations (R² = 0.88–0.98) and low error rates, outperforming ARIMA (which under-predicted rainfall by 9–34%) and GCMs (which over-predicted by 6–33%). The Global Precipitation Climatology Centre (GPCC), Global Precipitation Measurement (GPM), Integrated Multi-Satellite Retrievals for GPM (IMERG), and TAMSAT African Rainfall Climatology And Time Series (TARCAT) were identified as the most reliable datasets for predicting ground-based observations. These findings highlight the importance of region-specific dataset selection and robust predictive modelling, with direct implications for water resource management, agricultural planning, and disaster preparedness in Nigeria.

Graphical Abstract

This Study Enhances Rainfall Predictability in Nigeria by Integrating Ground-based measurements, satellite-derived Precipitation Datasets, and Climate Model projections. The Research Evaluates Rainfall Estimation across Diverse eco-climatic Regions Using 48 Ground stations, 11 Gridded Rainfall products, and CMIP5 Model Outputs (RCP 2.6, 4.5, and 8.5). The Methodology Involves Spatial mapping, Data Consistency checks, and Model Ranking Using Stepwise Regression, ARIMA, and CMIP5 simulations. Results Indicate that Stepwise Regression Outperforms ARIMA and GCMs, Showing the Highest Predictive Accuracy (R² = 0.88–0.98), while ARIMA Underestimates Rainfall (9–34%) and GCMs Overestimate It (6–33%) in some Regions. The GPCC, GPM-IMERG, and TARCAT Datasets Emerged as the Most Reliable Precipitation sources. These Findings Provide a Data-driven Approach for Optimizing Rainfall forecasting, Essential for Nigeria’s Water Resource management, Agricultural planning, and Climate Adaptation Strategies.