Geospatial Data Analysis

Author: Naziru Halilu

A hands-on course/project package covering vector and raster GIS, optical and radar (Sentinel-1/2) remote sensing, and unsupervised/supervised land-cover classification in Python and Google Earth Engine.

Contents

• 01_Vector_Data_Analysis.ipynb — vector data manipulation (agricultural plots).
• 02_Raster_Data_Analysis.ipynb — raster data manipulation; digital elevation model, slope map generation, orthophotos.
• 03_Optical_Data_Sentinel2_NDVI.ipynb — working with a downloaded Sentinel-2 image: NDVI, time-series generation from CSV, plot/crop comparison.
• 04_Unsupervised_Classification.ipynb — unsupervised classification with a single Planet image and a 3-date time series (April, August, November).
• 05_Supervised_Classification.ipynb — supervised classification in Python.
• 06_Sentinel1_Download_Processing_TimeSeries.ipynb — Sentinel-1 download, processing, and time-series analysis.
• GoogleEarthEngine_Class3_CloudMasking.js — Google Earth Engine script (filters, cloud masking, image collection handling).
• GIS_Vector_Layers_Soil_Hydrology_AgriculturalPlots.7z — compressed shapefiles: soil polygons (EDAFOL_Pol_Suelos25m), main-river hydrology polygons (HIDROG_Pol_RioPrincipal), and the 2017 agricultural plot register (PAC_2017).
• Class3_Intersection_Plots.7z — compressed shapefile of plot intersections used in the Google Earth Engine exercise.
• Unsupervised_Classification_PlotResults.7z — compressed CSV of per-plot unsupervised classification results.
• Sentinel2_Composite_2017.tif — Sentinel-2 2017 raster composite.
• Sentinel2_PlotStatistics_Raw.csv — Sentinel-2 zonal statistics per agricultural plot (raw).
• Sentinel2_PlotStatistics_MovingAverage.csv — same, smoothed with a moving average.

All .7z archives are kept compressed because several of the underlying shapefiles individually exceed 100 MB uncompressed. Extract them locally with 7-Zip or p7zip before use.

How to Run the Code

1. Clone the repository

git clone https://github.com/halilunaziru73-creator/Geospatial-data-analysis.git
cd Geospatial-data-analysis

2. Install dependencies

This is a Jupyter notebook-based package. At minimum you'll need:

pip install jupyter numpy pandas matplotlib geopandas rasterio scikit-learn earthengine-api

(Some notebooks may need additional packages depending on which cells you run — install any missing package as prompted.)

3. Extract the compressed data archives

Several shapefiles/results are kept as .7z archives because they exceed 100 MB uncompressed. Extract them first:

7z x GIS_Vector_Layers_Soil_Hydrology_AgriculturalPlots.7z
7z x Class3_Intersection_Plots.7z
7z x Unsupervised_Classification_PlotResults.7z

(Install p7zip-full / p7zip first if 7z isn't available on your system.)

4. Launch and run the notebooks in order

jupyter notebook

Then open and run, in this order (each is self-contained but file names/paths are aligned across notebooks):

1. 01_Vector_Data_Analysis.ipynb
2. 02_Raster_Data_Analysis.ipynb
3. 03_Optical_Data_Sentinel2_NDVI.ipynb
4. 04_Unsupervised_Classification.ipynb
5. 05_Supervised_Classification.ipynb
6. 06_Sentinel1_Download_Processing_TimeSeries.ipynb

5. Google Earth Engine script

GoogleEarthEngine_Class3_CloudMasking.js runs separately in the Google Earth Engine Code Editor — sign in with a GEE-enabled Google account, paste the script, and click Run.

Notes

• Institutional logos embedded in the notebooks reflect only the funding partners relevant to this package (UPNA and EU co-funding).
• File names follow a consistent numbering scheme, and all in-notebook data references are aligned with these file names so the notebooks run correctly end to end.

License

Released under the MIT License.

Citation

If you use this repository, please cite it using the metadata in CITATION.cff (GitHub renders a "Cite this repository" button on the repo's main page, in the top-right "About" panel).

Related work

Part of a broader body of research on GIS, remote sensing, and machine learning for agronomic and environmental applications:

• Digital Twin for Gully Biocontrol
• Geometry-Agnostic Contrastive Learning (GACL)
• GIS-Based Delineation for Livestock Slurry Application
• Hybrid CNN-BiLSTM-Attention for Sediment Transport
• Operationalizing GIS and ML across Cropping Systems
• Geospatial Data Analysis