Disasters are not isolated events. Many U.S. counties experience federally declared disasters year after year, and that cumulative exposure reshapes how local organizations operate, plan, and serve their communities. This interactive map visualizes county-level disaster exposure over time, drawing on FEMA declaration data from 2000 to 2024. Use the slider to see how exposure accumulates, and explore which areas face the most frequent and recent disruptions. This is the context in which the community-based organizations I study do their work.
This interactive map draws on two primary datasets:
Counties are identified by their 5-digit FIPS code, constructed by zero-padding the 2-digit state FIPS code and 3-digit county FIPS code from the FEMA records. County boundaries are rendered using the U.S. Atlas TopoJSON dataset (Albers USA projection).
Records associated with tribal areas are detected and excluded based on keywords in the designated area name (e.g., “reservation,” “tribe,” “pueblo,” “rancheria”). This exclusion is not because the tribal data is incorrect. Rather, ACS and FEMA often treat tribal areas differently from counties (using distinct FIPS codes or geographic boundaries), and tribal nations operate under different governance structures than county governments. Mixing tribal and county records in a strictly county-based analysis frame would introduce inconsistencies, so tribal areas are excluded to keep the unit of analysis uniform.
Counts the number of individual FEMA disaster declaration records for each county. In per-year mode, only events with an incident begin date in the selected year are counted. In cumulative mode (default), all events from the start of the record through the selected year are counted.
A single disaster event (e.g., Hurricane Katrina) may generate multiple declaration records for the same county if it triggers separate declaration types, so the count reflects the total number of federal disaster declarations rather than distinct weather events.
Color scale: Sequential yellow-orange-red. Counties with zero events are shown in gray. Darker red indicates higher counts.
Measures how frequently a county experiences disasters by computing the mean number of years between distinct disaster years. The calculation through the selected year is:
Avg Interval = (last disaster year − first disaster year) / (number of unique disaster years − 1)
Counties with fewer than two distinct disaster years through the selected year have no computable interval and are shown in gray. A lower interval (closer to 0) means disasters occur nearly every year; a higher interval means they are more spread out over time.
Color scale: Diverging red-yellow-green, ranging from 0 to 5 years. Red indicates very frequent disasters (interval near 0); green indicates rare disasters (interval of 5+ years).
Identifies the most common disaster type for each county through the selected year. All disaster declarations through that year are tallied by incident type (e.g., Hurricane, Severe Storm, Flood, Fire, Tornado, Snow). The type with the highest count is selected.
In the case of a tie, the type with the most recent occurrence is chosen. Counties with no events through the selected year are shown in gray.
Color scale: Categorical palette with distinct colors per disaster type. The legend dynamically shows only the types present in the current view.
Classifies each county into one of four exposure strata based on two dimensions: frequency (how often disasters occur) and recency (how active the county has been in the recent past). This stratification follows the methodology established in prior sampling and analysis work on this project.
Frequency is derived from the average interval (described above), computed over all disaster years through the selected year:
Recency counts the number of disaster declarations in the 5-year window ending at the selected year (e.g., for 2024, the window is 2020–2024):
Recency is evaluated first. If a county has low recency (LR), it is assigned to the LF/LR stratum regardless of its frequency. If a county has high recency (HR), its frequency bin determines the final stratum:
| Stratum | Criteria | Color | Interpretation |
|---|---|---|---|
| HF/HR | Avg interval ≤ 1.0 yr and ≥ 3 recent events | Red | Persistently high disaster exposure |
| MF/HR | Avg interval 1.0–1.5 yr and ≥ 3 recent events | Orange | Moderate long-term frequency with recent activity |
| LF/HR | Avg interval > 1.5 yr and ≥ 3 recent events | Blue | Historically infrequent but recently active |
| LF/LR | Fewer than 3 events in last 5 years | Gray | Low recent exposure regardless of history |
For the Avg Interval and Exposure Strata metrics, the year slider is replaced by a period toggle offering three views:
The 2010–2013 gap between periods is intentional to avoid overlap and create two clean comparison windows.
| Category | Color | Meaning |
|---|---|---|
| Worsened | Red | County moved to a higher-severity stratum in 2014–2024 compared to 2000–2009 |
| Improved | Green | County moved to a lower-severity stratum in 2014–2024 compared to 2000–2009 |
| No change | Gray | Same stratum in both periods |
Shows the difference in average interval between periods (period 2 − period 1). Negative values (red) indicate worsening — disasters becoming more frequent. Positive values (green) indicate improving — disasters becoming less frequent. Counties with data in only one period are folded into the worsening or improving direction accordingly.