Djiby223/Mali_Rainfall_Prediction_Project

Rainfall and NDVI analysis for Mali (1981–2026) combining climate and remote sensing data to support agricultural and environmental forecasting. # Mali_Rainfall_Prediction_Project Rainfall and NDVI analysis for Mali (1981–2026) combining climate and remote sensing data to support agricultural and environmental forecasting. ## Project Overview This project analyzes historical rainfall patterns in Mali and investigates their relationship with vegetation dynamics using NDVI (Normalized Difference Vegetation Index). A climate analytics and rainfall forecasting foundation project using historical rainfall and NDVI data from Mali. The project performs data cleaning, exploratory analysis, and prepares datasets for future machine-learning-based rainfall prediction. The objective is to: - Clean and harmonize rainfall and NDVI datasets. - Explore long-term climate variability. - Analyze rainfall-vegetation relationships. - Build predictive models for rainfall forecasting. - Support climate resilience and agricultural planning in Mali. --- ## Datasets ### Rainfall Data - Monthly rainfall observations - Coverage: 1981–2026 - Spatial scale: Mali ### NDVI Data - Vegetation index derived from satellite imagery - Administrative-level coverage across Mali - Monthly observations --- ## Methodology ### 1. Data Cleaning - Standardized dates - Converted decimal formats - Removed metadata rows - Handled missing values ### 2. Data Alignment - Converted datasets to common monthly time series - Merged rainfall and NDVI records ### 3. Exploratory Data Analysis - Rainfall trends - Seasonal patterns - NDVI dynamics - Correlation analysis ### 4. Prediction Modeling Potential models: - Linear Regression - Random Forest - XGBoost - Time Series Forecasting --- ## Tools - Python - Pandas - NumPy - Matplotlib - Scikit-learn - Jupyter Notebook --- ## Preliminary Findings - Strong seasonality in rainfall. - Positive relationship between rainfall and vegetation greenness. - Significant interannual variability linked to climatic conditions. --- ## Future Work - Incorporate temperature data. - Develop district-level fo …