Exploring satellite‑derived NDVI as indicative evidence of irrigation effects
This repository contains the Python code used to support the analysis described in
“Exploring satellite‑derived NDVI as indicative evidence of irrigation effects”.
The analysis explores whether Earth Observation (EO) data—specifically NDVI—show
indicative signals of improved dry‑season vegetation persistence following the
implementation of a small‑scale irrigation scheme.
The work is intended as a learning and feasibility exercise, not a formal impact evaluation.
The analysis uses:
- Sentinel‑2 Surface Reflectance imagery to compute NDVI time series and composites.
- CHIRPS precipitation data for contextual rainfall interpretation.
- A historical baseline (2018–2023) to express NDVI and rainfall anomalies as z‑scores.
- Pre‑ vs post‑intervention comparison, aligned to the irrigation scheme operational date (Q4 2024).