Simon-Kimanzi/Eneza_Ecological_Project

Automated data pipeline and spatial visualization tool for monitoring Thrips and Fruit Fly distribution in Africa (globally) using citizen science data. 📋 Overview The Eneza Ecological Project leverages citizen-science data from platforms like iNaturalist to unravel key interactions between high-impact agricultural pests specifically Thrips and Fruit Flies and their host plants. By utilizing photo-documented, georeferenced data, this project aims to establish relationships between these insects and prevailing ecological conditions By leveraging citizen science data from the iNaturalist API, this pipeline automates the collection, cleaning, and spatial visualization of occurrence records. ## Regional Focus This project specifically analyzes ecological data within the East African corridor (Kenya, Uganda, Tanzania, Ethiopia). 🎯 Purpose of the Project Early Warning System: To identify clusters of pest activity before they reach critical agricultural zones in Kenya and neighboring countries and even globally. Trans-boundary Analysis: To understand how these pests move across borders by analyzing GPS coordinates and observation timestamps. Digital Portfolio: To demonstrate a modular, scalable approach to ecological data science, moving from raw API "blobs" to structured, actionable maps. 🛠️ The Workflow My project follows a professional ETL (Extract, Transform, Load) architecture: Extraction: I wrote a Python worker (src/fetch_data.py) that communicates with the iNaturalist servers. It specifically requests data for Taxon ID 51789 (and others), handling the JSON response and converting it into a structured format. Transformation: Using the Pandas library, I clean the data. I specifically handle cases where GPS locations might be missing (NoneType) to ensure the pipeline doesn't crash a critical step for data integrity. Loading/Visualization: I use the Folium library to generate interactive maps (notebooks/pest_map.html). These maps allow users to click on specific markers to see exactly which species was identified and where. Over the course of this project, I will be: Spatial Filtering: Restricting the data to …