PlainHazard

Original research, PlainHazard editorial

Top 10 FEMA Incident Types by Declaration Frequency

PlainHazard ranks FEMA incident types by the total number of disaster declarations attributed to each type, using the disaster_by_type aggregate. Rendered server-side from a live SELECT against the disaster_by_type table.

Research period:

Reviewed by PlainHazard Editorial on 2026-05-17

Research question

Across the full OpenFEMA disaster declaration registry, which incident types — severe storms, hurricanes, fires, floods, biological events — account for the largest number of federal disaster declarations, and how broadly distributed are these incident types across US states?

Methodology

We queried the PlainHazard disaster_by_type table at server render time and pulled the columns incident_type, total_count, state_count, earliest_year, latest_year. The query ranks records by total_count DESC and returns the top 10. Every numeric value rendered on this page derives from a live SELECT against the production disaster_by_type table — no figure is hardcoded, and the table refreshes whenever the underlying Federal Emergency Management Agency dataset is reingested.

FEMA classifies every declaration by a single primary incident type — the hazard that triggered the federal response. Broad categories like "Severe Storm" and "Flood" dominate the count because they recur across many states and seasons and frequently combine (a severe storm system that produces both tornadoes and flooding is typically logged under one incident type). Rarer categories — "Earthquake", "Volcanic Eruption", "Tsunami" — appear far down the list not because they are mild, but because they strike infrequently and in a limited geography, so they generate few separate declarations.

Reading frequency alone can mislead: a category with many declarations is not necessarily the costliest or deadliest. A single hurricane or wildfire complex can cause more loss than years of routine severe-storm declarations. To weigh severity rather than count, pair this ranking with the per-state damage and fatality data and with the FEMA National Risk Index expected-annual-loss estimates. The state-count column also matters — an incident type spread across many states reflects a nationwide hazard, while one concentrated in a few states points to a regional risk.

See the methodology page for the complete ETL pipeline, source vintage, and column lineage.

Top 10 FEMA Incident Types by Declaration Frequency

Live data — rendered from a SELECT against the portal database at request time

records
Source Federal Emergency Management Agency

The ranked top 10

Every row below is rendered from a live SELECT against the 10-row result returned by the query in the frontmatter above. Refresh the page after an ETL run to see the latest values.

# Incident type Total declarations States affected Earliest year Latest year
1 Fire 588 32 2012 2026
2 Severe Storm 150 40 2016 2026
3 Flood 83 37 1973 2026
4 Hurricane 64 21 2021 2025
5 Tropical Storm 35 16 2023 2025
6 Winter Storm 30 22 2023 2026
7 Biological 14 14 2020 2021
8 Severe Ice Storm 13 9 2021 2025
9 Tornado 13 8 2020 2024
10 Other 9 7 2021 2026

Source: Federal Emergency Management Agency — OpenFEMA Disaster Declarations Summaries V2 (incident type aggregates). Values are queried live from the PlainHazard SQLite snapshot at request time; the snapshot is refreshed by the portal ETL pipeline. Federal Emergency Management Agency — OpenFEMA Disaster Declarations Summaries V2 (incident type aggregates). Values are queried live from the PlainHazard SQLite snapshot at request time; the snapshot is refreshed by the portal ETL pipeline.

Findings

Top entity in the ranking

The top-ranked record in this dataset is Fire, with a value of 588 on the Total declarations column. The full top-10 set is rendered in the table above. Every value derives from the underlying disaster_by_type table; no number is hardcoded into this page. When the Federal Emergency Management Agency publishes a revision and our ETL pipeline reingests, the ranking and the prose around it update on the next page load.

Distribution shape

The gap between the top-ranked record (588) and the 10th-ranked record (9) characterizes how concentrated the top of the distribution is. Where the top value is many multiples of the median value of the visible set, the population is highly concentrated — a small number of entities accumulate the bulk of the measured quantity. Where the top and bottom of the visible set are close together, the distribution is relatively flat across the top end. The full distribution beyond this top-10 cut is summarized in the aggregate context section below and explored in the linked entity profiles.

Aggregate context

Across the full disaster_by_type population, the aggregate query returns the following summary statistics. These anchors situate the top-10 ranking against the underlying population: how many records exist in total, what the sum of the ranking column is across all qualifying rows, and what the mean per-record value looks like. The methodology page documents the exact filter applied by the aggregate query (records with null or zero values on the ranking column are excluded). The aggregate row is computed by the same database engine that renders the ranking above, against the same snapshot.

Source provenance

The records in this ranking originate from Federal Emergency Management Agency, specifically the OpenFEMA Disaster Declarations Summaries V2 (incident type aggregates). PlainHazard ingests the source vintage published by the agency, transforms it into a normalized SQLite schema, and serves it from a read-only snapshot. Every render of this page is a fresh SELECT against that snapshot — there is no static export carrying stale numbers, and the edge cache lifetime is bounded by the portal middleware so that a reingested dataset propagates within hours. The methodology page documents the source URL, the vintage date, and the transformation steps applied during ETL.

Why this ranking matters

Rankings like this one let a reader scan a population quickly and identify outliers, concentrations, and patterns that warrant deeper investigation. The detail pages linked from each entity in the table above give the full per-entity context: time-series history where available, related metrics from adjacent tables, and links onward to the underlying source records. The methodology page explains how an entity earns inclusion in the dataset and how the ranking column is computed at the source.

What this analysis cannot tell us

FEMA's incident type taxonomy has evolved over time — categories like 'Severe Storm(s)' and 'Severe Ice Storm' overlap in many declarations and the same physical event may be coded differently in different years. The NOAA Storm Events Database uses a separate taxonomy (e.g. 'Thunderstorm Wind', 'Hail', 'Tornado') that does not directly map to FEMA's incident types. Counts here reflect distinct declarations, not the total cost, lives lost, or geographic footprint of each event; a single major hurricane declaration may affect dozens of counties and impose billions in damage, while a smaller emergency declaration may affect a single jurisdiction. Readers comparing hazard categories should consult the methodology page for the full code mapping and the agency-published documentation.

Secondary cut from the same source

Top 10 NOAA Storm Events Database event types by raw event count (separate dataset from FEMA declarations)

records
Source Federal Emergency Management Agency

Related data

Sources