Ombutora/hydrosense-kenya

# HydroSense-Kenya: Scientific Computing for Irrigation HydroSense-Kenya is a scientific computing and simulation framework designed to optimize agricultural water usage in Kenya. By modelling soil moisture dynamics using difference equations and weather data, the system evaluates and recommends irrigation strategies that conserve water while maintaining crop health. ## Project Structure The project is structured into six progressive Jupyter notebooks: 1. **Level 1**: Problem Framing & Conceptual Model 2. **Level 2**: Vectorization & Error Analysis 3. **Level 3**: Numerical Methods (Euler's Forward Method) 4. **Level 4**: Exploratory Data Analysis & Visualization 5. **Level 5**: Simulation Scenarios (Testing multiple irrigation strategies) 6. **Level 6**: Final Integration, Optimization, & Decision Support ### Core Modules (`src/`) - `numerical_methods.py`: Core physical engine simulating soil moisture over time. - `simulation.py`: Executes various operational irrigation scenarios. - `optimization.py`: Contains mathematical optimizers (JIT Greedy and Constrained) for water minimization. - `visualization.py`: Helper functions for generating comparative charts and heatmaps. ## Installation Ensure you have Python 3.10+ installed. ```bash python -m venv .venv source .venv/bin/activate # On Windows: .venv\Scripts\activate pip install -r requirements.txt ``` ## Running Tests Automated tests are written using `pytest`. ```bash pytest tests/ ``` ## Key Findings The Final Integration notebook (Level 6) demonstrates that mathematical optimization (JIT Greedy strategy) can significantly reduce water usage compared to fixed daily schedules by timing irrigation events precisely based on crop-specific evaporation and uptake coefficients.