ArrenOnom/Eritrea-SPI-CMIP6-Analysis

Multi-model ensemble analysis of future SPI and drought/wetness characteristics in Eritrea using CMIP6 (2015–2100) # Evaluating NEX-GDDP-CMIP6 Rainfall Projections in Eritrea’s Agricultural Heartlands **Authors:** * **Tesfalem W. Ghebretnsae** (RUDN University, Russia / Hamelmalo Agricultural College, Eritrea) * **Elsayed S. Mohamed** (RUDN University, Russia / NARSS, Egypt) * **Umar Joseph James** (Niger State Hydrocarbon and Solid Mineral Development Agency, Nigeria) --- ## 📌 Project Overview This repository provides the computational framework for evaluating 34 **NEX-GDDP-CMIP6** Global Climate Models (GCMs) across Eritrea's diverse agroecological zones (Moist Highlands, Moist Lowlands, and Arid Lowlands). Using a **Comprehensive Ranking Index (CRI)** based on nine statistical metrics, this project identifies top-performing models to project future precipitation trends, seasonal concentration (PCI), and reliability (CV) under SSP2-4.5 and SSP5-8.5 scenarios. ## 🚀 Interactive Analysis The Python analysis and visualization suite can be executed interactively in Google Colab: ## 🛠 Methodology * **Dataset:** NASA NEX-GDDP-CMIP6 (0.25° spatial resolution). * **Reference Period:** Historical simulations (1981–2014) validated against CHIRPS observational data. * **Projections:** Near-term (2026-2050), Mid-term (2051-2075), and Far-term (2076-2100). * **Top-10 Multi-Model Ensemble (MME):** TaiESM1, GFDL-ESM4, FGOALS-g3, EC-Earth3, NESM3, MIROC6, MIROC-ES2L, MRI-ESM2-0, GFDL-CM4, and BCC-CSM2-MR. ## 📊 Key Research Findings * **The Wetting Paradox:** Projections indicate a substantial increase in mean annual rainfall (rising from ~320 mm to ~520 mm by 2100 under SSP5-8.5). * **Intensification:** Rainfall increases are driven primarily by the intensification of the **Kiremti (JJAS)** season. * **Vulnerability:** The Precipitation Concentration Index (PCI) rises from 22.04 to 24.55, suggesting that "wetting" will manifest as more erratic and concentrated events, escalating risks for flash floods and soil erosion. * **Reliability:** Interannual reliab …