2026 earthquake research snapshot

The year's worst earthquakes, ranked by lives and loss.

Magnitude measures the rupture—not the disaster. This reviewed snapshot separates reported human impact, documented physical damage, and USGS rapid-loss modelling so urgent numbers remain useful without becoming misleading.

Research snapshot · 11 August 2026 · 19:18 UTC

Deadliest
Venezuela · >6,300 deaths
Costliest documented
Venezuela · $19.6B direct damage
Strongest magnitude
Philippines · M7.8

Fatalities reported in 2026

This ranking reviews 2026 earthquakes with at least 10 reported deaths, or a major documented/modelled economic impact, for which a government, recognized disaster authority, or major wire source was located. Missing people are not counted as deaths. Colombia remains a fast-changing active emergency.

Deaths: ≥224 · Colombia update: the 224 figure is a consolidated local-authority count. Earlier federal reporting lagged behind it, and the total may rise as rescue teams reach more communities.
  1. 01

    Venezuela

    24 June · M7.2 + M7.5

    Reported total; still provisional

  2. 02

    Colombia

    10 August · M7.4

    Local-authority total; active rescue

  3. 03

    Mindanao, Philippines

    8 June local time · M7.8

    Latest official report located; 7 missing

  4. 04

    Kumamoto, Japan

    28 July · M6.8 USGS / M7.1 JMA

    Latest official/wire total located

  5. 05

    Jurm, Afghanistan

    3 April · M5.8

    Government-reported total

Comparable rapid economic-loss modelling

Most 2026 events do not yet have audited, like-for-like damage totals. This ranking therefore uses the same USGS PAGER empirical model for each event, then shows ground-reported evidence separately.

02

Kumamoto, Japan

28 July · M6.8

PAGER model
$12.7B
Most likely band
$10B–$100B

What has been documented

No finalized national post-event damage total located.

03

Colombia

10 August · M7.4

PAGER model
$6.48B
Most likely band
$1B–$10B

What has been documented

Model only; field damage assessment is still underway.

What the rankings mean—and what they do not

01

Magnitude is not impact

The M7.8 Philippines earthquake was the year's strongest, but Venezuela's shallower doublet caused far greater human loss. Depth, exposure, construction, and response capacity shape the outcome.

02

Reported and modelled stay separate

Fatality totals come from authorities and field reporting. Economic comparisons use PAGER until consistent post-event assessments exist. A model is never presented as observed damage.

03

Every number needs a timestamp

Rescue, identification, and damage assessment can continue for weeks. Each update should retain its source and verification time instead of silently replacing earlier evidence.

04

This is a reviewed set, not a complete catalogue

Included events meet the stated fatality or impact threshold and have usable source evidence. The ranking is only as complete as the reports located by the snapshot time; USGS remains the source for a full seismic catalogue.

Turn the evidence into reusable response tools.

The initiative began in Venezuela and now expands through Colombia. Join builders adapting open-source foundations for damage mapping, connectivity, safety, coordination, and recovery.