ashleymiser/HydroSense-Kenya

Smart irrigation system using numerical methods and sensor data # HydroSense-Kenya Smart irrigation decision support system for a Nairobi demonstration farm. Built for ICS 2207 Scientific Computing capstone project. ## What this project does Uses daily weather data and soil sensor readings from three farm zones to: - Estimate daily evapotranspiration - Model soil moisture using the discrete water balance equation - Simulate future soil moisture using Euler and Runge-Kutta methods - Quantify rainfall uncertainty using Monte Carlo simulation - Recommend an optimized irrigation schedule that minimizes water use while keeping crops above the minimum moisture threshold ## Project structure ``` HydroSense-Kenya/ ├── data/ │ ├── raw/ # original datasets from project brief │ └── processed/ # cleaned dataset ├── notebooks/ # six level notebooks ├── src/ # reusable Python modules ├── tests/ # pytest test files ├── reports/ # plots and final report ├── AI_USE_LOG.md ├── README.md └── requirements.txt ``` ## Installation ```bash python3 -m venv venv source venv/bin/activate pip install -r requirements.txt ``` ## Running the notebooks Run notebooks in order from Level 1 to Level 6: ```bash jupyter notebook ``` Open each notebook in the `notebooks/` folder and run all cells top to bottom. ## Running the tests ```bash pytest tests/ -v ``` ## Dataset source All datasets are synthetic and included in `data/raw/`. They were provided in the project brief appendices (Appendix A, B, and C). Additional live weather data can be accessed via the JHub Africa Conduit API at conduit.jhubafrica.com ## Group members - Ashley Miser