Create Your First Project
Start adding your projects to your portfolio. Click on "Manage Projects" to get started
Geospatial Analysis of Crop Yield vs. NDVI
Project type
Geospatial Interpolation
Geospatial Interpolation
This project investigated the relationship between agricultural yield and vegetation health (NDVI) in the Jolanda Di Savoia plains. Key stages included:
- Preparing and imputing missing values in yield data (GPKG format) using Python (scikit-learn KNNImputer).
- Generating a spatially interpolated yield map (SciPy, rasterio) and an average NDVI map from Sentinel-2 satellite imagery (Planetary Computer STAC, xarray, Dask).
- Conducting a correlation analysis which revealed a statistically significant but very weak negative linear relationship, indicating NDVI was a poor linear predictor of yield in this context.
- Technologies: Python, GeoPandas, scikit-learn, SciPy, rasterio, xarray, rioxarray, pystac_client, stackstac, Dask, Matplotlib.