Charity-Githogora/kenya-climate-data

A comprehensive dataset and analysis of weather data from Meteostat weather stations across Kenya, covering multiple counties and historical trends. Includes data preprocessing, exploration, and mapping of county coverage. # 🇰🇪 Kenya Climate Data (2015–2025) ## 📌 Overview This repository contains a comprehensive **10-year climate dataset for Kenya**, covering the period from **January 1st, 2015 to October 29th, 2025**. The dataset was built to address the challenge of **fragmented and inaccessible climate data in Kenya**, providing a structured and analysis-ready resource for researchers, data analysts, and developers. --- ## ⚙️ Data Collection & Processing ### 1. Data Collection The raw data was collected using the **Meteostat API**, which provided historical weather data from **31 weather stations across Kenya**. ### 2. Data Cleaning The dataset underwent thorough preprocessing to ensure: * Consistency across all stations * Handling of missing or null values * Standardized formats for all variables ### 3. County Mapping Each station’s data was mapped and grouped into its respective **Kenyan counties**, enabling: * Regional analysis * County-level insights * Better usability for policy and research applications --- ## 📊 Dataset Features The dataset includes key climate variables such as: * 🌧️ Rainfall * 🌡️ Temperature * 🌬️ Wind * 📈 Pressure * 🌍 Overall Climate Indicators --- ## 📅 Monthly Climate Index To enhance usability and analytical value, a **monthly index** was calculated for each of the following: * Rainfall Index * Temperature Index * Wind Index * Pressure Index * Composite Climate Index These indices provide: * Simplified trend analysis * Easier comparison across months and years * A foundation for predictive modeling --- ## 📁 Structure The dataset is organized into structured files, making it easy to: * Access specific regions (counties) * Work with time-series data * Integrate into analysis pipelines --- ## 🚀 Use Cases This dataset can be used for: * Climate trend analysis * Time-series forecasting * Drought and rainfall prediction models * Agricultural planning insights * Data visualization projects (Power BI, Python, etc.) * Machine learnin …