What actually happens in El Salvador
Built from 182 records collected between 9 August 2026 and 11 October 2026. These are detections, not occurrences: what our monitoring recorded, not everything that happened.
When
Incidents recorded by calendar month, all years combined. The tall months are when this country's risk concentrates — usually weather, sometimes a single large event.
| Month | Records | |
|---|---|---|
| January | 6 | |
| February | 4 | |
| March | 7 | |
| April | 7 | |
| May | 5 | |
| June | 13 | |
| July | 9 | |
| August | 33 | |
| September | 51 | |
| October | 32 | |
| November | 7 | |
| December | 8 |
The busiest month on record is September. Before treating that as the dangerous month, check how many years the archive covers: a single severe season can create a peak on its own.
What
The types recorded most often here. The right-hand column is the important one — it marks the types that have no official real-time feed anywhere in the world, so those counts depend entirely on what local journalism reported.
| Type | Records | Data source reality |
|---|---|---|
| Earthquake | 176 | Official feed exists |
| Drought | 2 | Official feed exists |
| Road Accident | 2 | News-only — undercounted |
| Flood | 1 | Official feed exists |
| Explosion | 1 | News-only — undercounted |
How to read this without fooling yourself
- Compare this country against its own history, not against another country. For the news-only types you would mostly be comparing newsroom density.
- Earthquake counts are detection counts. Seismic networks record down to magnitudes nobody feels, so they dominate any raw ranking.
- A quiet month is not a safe month. It can be a quiet news period, or a feed that was failing — which is why source health is public.
- Casualties are not counted here unless an official source stated them.
Full method: how we track every incident on Earth. Per-type feed reality: what the world does not publish. Raw data: CSV and JSON, CC BY 4.0.