Who made this
Built by Anchit Tandon as an independent project; more at anchit-tandon.com.
Why
Disaster information is spread across many agencies, each with its own map, words and alert levels. We want to make it easier to find, compare and check, in one place, without making it sound more certain than it is.
We keep three things apart: events that have happened, official warnings, and our own forecasts. Every event and warning shows where it came from and how recent it is. We lock each forecast before the outcome is known, and check it afterwards against rules set in advance, so the track record shows how often we were right.
How the data flows
- SourcesOfficial agencies publish their reports: the Global Disaster Alert and Coordination System (GDACS), the US Geological Survey (USGS), the European-Mediterranean Seismological Centre (EMSC), NASA, the US National Oceanic and Atmospheric Administration (NOAA), ReliefWeb and others. See every source.
- Our serverReads every source, removes duplicates and puts each report into one format. It keeps one shared copy, refreshed about every minute.
- Your browserEach page asks our server for that copy, never the agencies directly, and refreshes about every minute while it is open.
Our forecasts run daily. The weather models start from Open-Meteo’s forecast and judge it against each place’s 30-year record from NASA POWER. Each forecast is locked when it is issued and graded when its window ends. How it works.
What this is not
This is not an official warning service. It does not replace official emergency warnings, evacuation orders or instructions from your local authorities. Our forecasts are experimental and can be wrong.
For real alerts, follow your national agency: the India Meteorological Department (IMD) and the National Disaster Management Authority (NDMA) in India, the National Weather Service (NWS) and USGS in the US, and GDACS worldwide.
Feedback and corrections
Seen a wrong event, a broken page or a misleading word? Please open an issue on GitHub, with the address of the page and what looks wrong. The code is public, so you can also see how every figure is worked out.
Privacy
- No account. You do not sign in, and we do not ask for your name.
- Kept in your browser. Your time zone, saved places, sort and map choices are stored in your own browser (local storage). They are not sent to us, and clearing your browser data removes them.
- Place searches. What you type in a place search goes from your browser to OpenStreetMap Nominatim to find the place. When a page shows details for an event or a place you looked up, our server sends its coordinates to BigDataCloud to name the nearest town; nothing else about you is sent with them.
- Maps and the globe. Map pictures load from Esri (satellite photos and place names) and NASA (the daily satellite photo). The 3D globe loads its code from unpkg or jsDelivr and its terrain from Re:Earth. Like any website, these services see your IP address when your browser asks them for a picture or a file.
- Alerts for your places. Only if you turn them on: our server keeps your browser’s push address and your saved places, rounded to about 1 km, so it can tell you about reports near them. Turning alerts off deletes them.
- Location. Your device’s location is used only if you choose to share it, to check the place you are in.
Updated October 2026.