omarmaalej12/predicting-the-Future-Evolution-of-Vegetation-Cover-in-Tunisia-Using-Advanced-Machine-Learning

Domain:

environment and energyclimategeospatial

Record type:

project

Creator:

oma

Host:

predicting the Future Evolution of Vegetation Cover in Tunisia Using Advanced Machine Learning Methods # 🌱 Vegetation Cover Prediction in Tunisia under Climate Change Scenarios This repository contains the code and resources developed during a research internship at the Laboratory of Ecology, Systematics, and Evolution (ESE) – Université Paris-Saclay. ## 🧠 Objective The goal of this project is to predict the **future evolution of vegetation cover in Tunisia** under climate change (SSP245 scenario), using advanced **machine learning techniques** and large-scale **geospatial and bioclimatic data**. --- ## 📦 Key Data Sources and Preprocessing ### 🌍 Vegetation and Climate Data - **Vegetation data** comes from **GLAD 2019 maps** (Global Land Analysis and Discovery Lab, University of Maryland), based on **Landsat imagery** with a **30m spatial resolution**. Each pixel is labeled with one of 20 land cover classes (0 to 19). For this project, only the two tiles covering Tunisia (`40N_010E` and `40N_000E`) were used and processed via QGIS. - **Climate data** (BIO1–BIO19) from WorldClim was used for both the current period (1970–2000) and future projections (2021–2040) under **SSP245** scenario. ### 🗺️ Geographic Processing with QGIS Using **QGIS**, we carried out the following: - Extracted the **geographic boundary of Tunisia**. - Generated a **1km² resolution pixel grid** over the country. - Overlaid and fused the **vegetation maps** using zonal statistics ("majority" rule). - Assigned the **dominant vegetation class** to each 1km² grid cell. - Aligned 19 climatic variables (BIO1–BIO19) with each pixel. - Produced two final datasets (Excel/CSV) with: - Coordinates (lat, lon) - Dominant vegetation class - 19 bioclimatic variables These datasets served as the inputs for all machine learning models. --- ## 📊 Workflow Overview ### 1. **Data Processing** - Removal of missing values. - Dimensionality reduction via **correlation analysis**. - **Standardization** of selected variables using z-score scaling. - Aggregation of 20 original vegetation classes into 3 simplif …

Visit

github.com

Predicting Food Price Trends in Nigeria Using Advanced Machine Learning Techniques: LSTM and XGBoost

Food price volatility poses significant challenges to food security, poverty reduction, and economic

Vegetation Mapping of Africa Using Machine Learning

AMS 2024 104th annual meeting, 28 January- 1 Feburary 2024, Baltimore, MD The accurate monitoring an

Ensemble Methods for Time Series Forecasting in Nigeria: Predicting Agricultural Yields Using Advanced Machine Learning Approaches

International audience Accurate forecasting of agricultural yields is essential for m

Predicting Brain Stroke Using Machine Learning

On the basis of the GBD (Global Burden of Disease) 2013 Study, this article provides an overview of

Predicting malaria outbreak in The Gambia using machine learning techniques

Malaria is the most common cause of death among the parasitic diseases. Malaria continues to pose a

Advanced machine learning for predicting groundwater decline and drought in the Rabat–Salé–Kénitra region, Morocco

ABSTRACT The Rabat–Salé–Kénitra region of Morocco faces critical g