Forest degradation is quantifiable through synchronized declines in vegetation indices and spikes in exposed-land signals.

Research Framing

Transforming the core hypothesis into an interactively auditable evidence trail.

The working hypothesis posits that major land-cover disruptions in Hasdeo manifest as measurable shifts across satellite-derived telemetry. Intact canopy ecosystems reliably sustain higher NDVI, EVI, SAVI, LAI, and GCI values. Conversely, disturbed or excavated sectors exhibit pronounced BSI, NDBI, and thermal signatures.

This interactive dashboard tests that hypothesis in real-time: isolate a coordinate region, define a chronological before/after window, toggle feature indicators, and cross-reference nearest-neighbor curated imagery against the numeric timeline.

Evidence Architecture

The underlying components exposed to the application layer to guarantee auditable results.

Study Area
Hasdeo Arand and adjacent control regions form the primary monitoring perimeter for deforestation and mining impact analysis.
Observation Stack
Fusing RGB inspection imagery, multispectral Sentinel bands, monthly feature tables, and QGIS-labeled training sectors.
Expected Signal
Vegetation indices rapidly decline during clearance windows, inversely mirroring spikes in bare-soil and built-up indicators.
Event Focus
The March–April 2022 disruption window serves as the baseline disturbance period for our comparative workflow.