❓ FAQ
Frequently asked questions
Straight answers about the data, accuracy, and what these numbers really mean.
Analysis uses the most recent low-cloud Sentinel-2 L2A scene available for the location.
Sentinel-2 revisits most places every ~5 days, but cloud cover means the actual "freshest"
usable image can be older. The NDVI time-series tab lets you inspect change over a date range.
They are index-derived proxies, not validated ecological measurements. Habitat health is a
direct transform of mean NDVI, and risk is its complement (100 − habitat). They are useful for
intuition and comparison, but they don't include species data or ground truth. Treat them as a
starting point, not a verdict.
The most common causes are persistent cloud cover, imagery gaps over very remote areas or open
ocean, or a date range with no usable scenes. Try a nearby point, a wider date range, or a
different season.
Habitat Health = clamp((mean NDVI + 1) × 50, 0, 100), and Bird Decline Risk = 100 − Habitat Health.
Healthier, greener vegetation implies better habitat and lower risk. Full details, including
land-cover thresholds, are on the Methodology page.
No accounts or tracking. Your theme preference and saved locations are kept only in your
browser (session/local storage). Coordinates you analyze are sent to the serverless API to fetch
imagery and weather, but they aren't persisted server-side by the app.
No. The dashboard is open to everyone — no account, sign-up, or login is required.
Just open it and start analyzing locations.
Not on its own. It's an educational demonstration of remote-sensing techniques. Real conservation
decisions need field data, expert review, and peer-reviewed methods. See the caveats on the
About page.