The finding
Restoration seasonality is the month-by-month pattern in when disaster work arrives. Across 1,422 restoration-relevant FEMA disaster declarations from 2015 to 2026, incidents peak in August (244 disasters) and bottom out in November (43), a 5.7 times swing. Weighted by the number of counties declared, the peak moves to September and the swing widens to 11.1 times, because hurricanes cover many counties at once. Each state has its own calendar.
Headline figures
In this report
Key findings
- August is the busiest month for restoration-relevant incident starts in FEMA data from 2015 to 2026, with 244 disasters, and November the quietest with 43.
- Counting one row per disaster, the peak month carries 5.7 times the trough month (FEMA OpenFEMA).
- In FEMA data, fire incidents account for 61% of August disasters, and fire grants make up 42% of all disasters in the sample.
- Weighted by FEMA county-level rows, the peak moves to September and the peak-to-trough swing widens to 11.1 times; tropical system incidents supply 91% of September county rows.
- In FEMA data from 2015 to 2026, tropical systems peak in September, winter storms in January, fires in August and severe storms in May.
- Of the 6 states with the most disasters, 3 different peak months appear, so no single national calendar fits them.
- Among states with at least 20 disasters, Oklahoma has the sharpest season, with 46.0 times as many disasters in its busiest three months as its quietest three, and Tennessee the flattest at 2.8 times.
- Using complete incident years 2015 to 2025 only, the peak is still August and the trough November, a 5.1 times swing.
The national calendar
Restoration seasonality shows up clearly in the FEMA record. Counting each disaster once by the month its incident began, activity builds from spring, climbs through summer to an August peak, stays high in September and then drops away to a November low. The busiest three-month window in the FEMA data, Jul to Sep, holds 42% of all 1,422 disasters from 2015 to 2026.
Which hazards drive which months
| Hazard group | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Fire | 5 | 12 | 59 | 37 | 24 | 72 | 117 | 148 | 80 | 39 | 25 | 16 | 634 |
| Severe storm | 30 | 27 | 31 | 37 | 41 | 33 | 29 | 17 | 9 | 15 | 6 | 39 | 314 |
| Flood | 6 | 17 | 22 | 27 | 24 | 24 | 20 | 16 | 10 | 9 | 5 | 11 | 191 |
| Tropical system | 0 | 3 | 2 | 4 | 5 | 3 | 14 | 63 | 70 | 43 | 4 | 1 | 212 |
| Winter storm | 30 | 17 | 7 | 4 | 0 | 0 | 0 | 0 | 0 | 1 | 3 | 9 | 71 |
The hazards take turns. In FEMA data from 2015 to 2026, the busiest three months for each group are: tropical systems Aug to Oct, winter storms Dec to Feb, fires Jul to Sep, severe storms Apr to Jun and floods Apr to Jun. Severe storms and floods are the least concentrated, with 35% and 39% of their disasters in their busiest window, against 83% for tropical systems. Fire is the largest group by count, at 634 disasters.
That fire weighting matters. Most fire declarations are Fire Management Assistance grants, which are numerous and usually cover one county. Removing the 594 fire grants leaves a FEMA calendar whose busiest three months, Jul to Sep, hold 33% of disasters, against 42% with them included. August is still the busiest single month, but the summer bulge is largely a fire-grant effect.
Counting counties instead of disasters
One row per disaster treats a single-county fire grant the same as a hurricane declared across dozens of counties. For anyone sizing workload, that understates large events. The second measure counts every distinct county-level row instead: 17,942 rows in FEMA data from 2015 to 2026.
| Month | Disasters | Disasters excl. fire grants | County rows | County rows, DR only |
|---|---|---|---|---|
| January | 71 | 67 | 1,756 | 704 |
| February | 76 | 64 | 1,980 | 1,440 |
| March | 121 | 64 | 1,204 | 1,077 |
| April | 109 | 74 | 1,549 | 1,214 |
| May | 94 | 70 | 1,161 | 1,056 |
| June | 132 | 63 | 837 | 685 |
| July | 180 | 70 | 1,028 | 510 |
| August | 244 | 107 | 2,356 | 900 |
| September | 169 | 95 | 3,363 | 1,328 |
| October | 107 | 70 | 1,474 | 733 |
| November | 43 | 21 | 304 | 191 |
| December | 76 | 63 | 930 | 782 |
The two weightings disagree for a clear reason. The disaster count is driven by many small fire grants in summer, while the county count is driven by a few very wide events. In FEMA data, tropical system incidents supply 91% of September county rows. Winter months also gain weight, because ice and winter storm declarations often cover most of a state. The widest single declarations in the file show the effect. Our frozen pipe damage statistics show what one such winter storm did to homes in Texas.
| Disaster number | State | Incident type | Declaration type | Incident start | County rows |
|---|---|---|---|---|---|
| 3554 | TX | Severe Ice Storm | EM | 2021-02 | 254 |
| 4586 | TX | Severe Ice Storm | DR | 2021-02 | 254 |
| 4338 | GA | Hurricane | DR | 2017-09 | 159 |
| 3616 | GA | Tropical Storm | EM | 2024-09 | 159 |
| 3387 | GA | Hurricane | EM | 2017-09 | 152 |
Emergency declarations made ahead of a storm can cover many counties that see little eventual damage, which is why the DR-only column is the more conservative county measure. Neither weighting is the true one; they answer different questions.
Every state has its own season
| State | Disasters | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| California | 155 | 12 | 11 | 2 | 0 | 3 | 7 | 26 | 30 | 23 | 22 | 8 | 11 |
| Washington | 93 | 3 | 0 | 0 | 0 | 2 | 5 | 27 | 35 | 11 | 1 | 3 | 6 |
| Oregon | 82 | 2 | 3 | 1 | 1 | 0 | 8 | 20 | 21 | 23 | 0 | 0 | 3 |
| Oklahoma | 71 | 2 | 11 | 26 | 9 | 8 | 2 | 0 | 1 | 0 | 3 | 5 | 4 |
| Texas | 53 | 1 | 5 | 12 | 5 | 6 | 2 | 9 | 5 | 3 | 2 | 0 | 3 |
| Florida | 47 | 0 | 0 | 2 | 6 | 4 | 0 | 2 | 9 | 11 | 9 | 4 | 0 |
Read across the rows and the national calendar falls apart. In FEMA data from 2015 to 2026, California peaks in August, when fire declarations lead; Washington peaks in August, when fire declarations lead; Oregon peaks in September, when fire declarations lead; Oklahoma peaks in March, when fire declarations lead; Texas peaks in March, when fire declarations lead; Florida peaks in September, when tropical system declarations lead. A firm working one state should plan from that state’s row, not the national chart.
Fire grants explain why some inland states peak earlier than the national pattern. Grass and range fires in the southern plains tend to come in late winter and spring, before green-up, so a fire season there can run months ahead of the western forest fire season. That is a reading of the incident types in the FEMA file, not a forecast, and a single dry spring can move a state’s peak. Our wildfire damage by state report ranks the recorded fire losses behind those grants.
Peak-to-trough ratio by state
To compare how seasonal states are, we divide the busiest three-month window by the quietest three-month window, with windows wrapping across the year end. Single months are too thin in most states: many record zero disasters in some months, which would make a month ratio undefined. States with fewer than 20 disasters are left out.
| State | Disasters | Busiest 3 months | Share in busiest | Quietest 3 months | Peak-to-trough ratio |
|---|---|---|---|---|---|
| Washington | 93 | Jul to Sep | 78% | Feb to Apr | quiet window empty |
| Florida | 47 | Aug to Oct | 62% | Dec to Feb | quiet window empty |
| Alaska | 34 | Jun to Aug | 47% | Feb to Apr | quiet window empty |
| New Mexico | 30 | Apr to Jun | 83% | Nov to Jan | quiet window empty |
| Colorado | 26 | Jun to Aug | 65% | Jan to Mar | quiet window empty |
| South Carolina | 26 | Aug to Oct | 65% | May to Jul | quiet window empty |
| Utah | 26 | Jun to Aug | 77% | Oct to Dec | quiet window empty |
| Nebraska | 23 | Mar to May | 65% | Sep to Nov | quiet window empty |
| West Virginia | 22 | Feb to Apr | 45% | Oct to Dec | quiet window empty |
| Missouri | 21 | Mar to May | 57% | Aug to Oct | quiet window empty |
| Alabama | 20 | Aug to Oct | 50% | May to Jul | quiet window empty |
| Oklahoma | 71 | Feb to Apr | 65% | Jul to Sep | 46.0x |
| Oregon | 82 | Jul to Sep | 78% | Mar to May | 32.0x |
| Nevada | 37 | Jun to Aug | 78% | Mar to May | 29.0x |
| Arizona | 40 | May to Jul | 70% | Nov to Jan | 28.0x |
| Montana | 38 | Jul to Sep | 63% | Jan to Mar | 24.0x |
| Kentucky | 30 | Feb to Apr | 53% | Aug to Oct | 16.0x |
| Hawaii | 23 | Jul to Sep | 70% | Nov to Jan | 16.0x |
| California | 155 | Jul to Sep | 51% | Mar to May | 15.8x |
| Georgia | 24 | Aug to Oct | 54% | May to Jul | 13.0x |
| New York | 20 | Jul to Sep | 65% | Jan to Mar | 13.0x |
| Arkansas | 21 | Mar to May | 57% | Sep to Nov | 12.0x |
| Kansas | 35 | Mar to May | 63% | Oct to Dec | 11.0x |
| North Carolina | 26 | Sep to Nov | 54% | Dec to Feb | 7.0x |
| Louisiana | 36 | Aug to Oct | 56% | Nov to Jan | 6.7x |
| Idaho | 23 | Jul to Sep | 57% | Sep to Nov | 6.5x |
| Texas | 53 | Mar to May | 43% | Nov to Jan | 5.8x |
| South Dakota | 24 | Apr to Jun | 46% | Sep to Nov | 5.5x |
| Mississippi | 31 | Feb to Apr | 35% | May to Jul | 3.7x |
| Tennessee | 32 | Feb to Apr | 34% | Jun to Aug | 2.8x |
11 states record no disasters at all in their quietest three-month window in FEMA data from 2015 to 2026, so their ratio is undefined and they are the most seasonal of all. Among the rest, Oklahoma is sharpest at 46.0 times and Tennessee flattest at 2.8 times. The states with an empty quiet window are Washington, Florida, Alaska, New Mexico, Colorado, South Carolina, Utah, Nebraska, West Virginia, Missouri, Alabama. For a firm in one of them, the quiet window is a real possibility of no declared work at all, which is the planning case for a second line of business such as routine water losses or mould work that does not depend on declarations.
Why it matters for planning
What follows is our reading, not FEMA data. A seasonal calendar is a staffing and cash-flow tool. If a firm’s state peaks in late summer, the months before are the time to hire and train, and the trough months are the natural window to service dehumidifiers, air movers and vehicles. A quiet quarter can be budgeted for rather than discovered.
The calendar also says when to expect travelling crews. Firms that follow catastrophe work tend to move with the tropical season, so local competition in a coastal state can rise in the same months as local demand. Our restoration demand volatility index covers the year-to-year swings that sit on top of this monthly pattern, and the restoration demand index ranks states by total volume.
Method
How we counted. We retrieved 26,032 FEMA disaster declaration summary rows with a declaration date of 1 January 2015 or later from the OpenFEMA API on 23 September 2026; the latest declaration is dated 2026-09-15. We kept the 13 incident types that generate restoration work (17,974 rows). For the disaster measure we kept one row per disaster number (1,422 disasters). For the county measure we kept distinct combinations of disaster number, state, place code and designated area (17,942 rows). Each record is assigned to the calendar month of its incident begin date. State seasonality uses three-month windows that wrap across the year end.
Reproduce this
Reproduce this
- Endpoint
https://www.fema.gov/api/open/v2/DisasterDeclarationsSummaries- Query
$filter=declarationDate ge '2015-01-01'- Rows retrieved
- 26,032 rows
- Retrieved
- 23 September 2026
No API key is required. Use incidentBeginDate, not declarationDate, for the month, or the calendar shifts later by the declaration lag.
What this does not show
What this does not show
- Declared disasters only. Routine water losses from burst pipes, appliance failures and freeze-thaw damage follow their own seasonal drivers and never reach a federal declaration, so most everyday restoration work is absent.
- Incident begin month only. A long incident period is assigned to the month it started, even if the damage ran into the next month.
- The 2026 year is partial. Declarations run to 2026-09-15, so late-2026 months are under-counted. We check the result on complete years 2015 to 2025 in the findings.
- The window is set by declaration date. 3 disasters declared in 2015 began in late 2014 and are included.
- A declaration is not a job count. County rows measure geographic reach, not the number of damaged buildings.
Most quotable figures
- Restoration-relevant FEMA disasters peak in August and bottom out in November, a 5.7 times swing (2015 to 2026, n = 1,422, FEMA OpenFEMA).
- Weighted by county, FEMA disaster declarations peak in September, with 11.1 times the November level (2015 to 2026).
- Jul to Sep holds 42% of restoration-relevant FEMA disasters from 2015 to 2026.
- Among states with at least 20 FEMA disasters, Oklahoma has the sharpest restoration season at 46.0 times busiest to quietest quarter (2015 to 2026).
Questions this report answers
What is the busiest month for disaster restoration work?
By number of FEMA disasters from 2015 to 2026, August. Weighted by counties declared, September.
What is the quietest month?
November, with 43 restoration-relevant FEMA disasters beginning in that month across 2015 to 2026.
When is hurricane season for restoration firms?
In FEMA data from 2015 to 2026, tropical system disasters peak in September and cluster from August to October.
Why do the two measures give different peak months?
Counting disasters gives weight to many small single-county fire grants in summer. Counting counties gives weight to wide events such as hurricanes and statewide winter storms.
Does my state follow the national pattern?
Probably not exactly. The 6 busiest states in the FEMA file show 3 different peak months. Use the state table above.
Does this include burst pipes and freeze damage?
No. Only federally declared disasters are counted. Everyday water losses have their own seasonal pattern that this data does not capture.
Can I reuse this data?
Yes. The tables and derived figures are released under CC BY 4.0 with attribution to Remediation Insider. The underlying declarations data is public and available from FEMA.
How to cite
Remediation Insider. The Restoration Seasonality Calendar: When Disaster Work Arrives. https://remediationinsider.com/restoration-seasonality-calendar/ (retrieved 2 October 2026).
Data and tables in this report are released under CC BY 4.0. Attribution: Remediation Insider.
Sources
- Federal Emergency Management Agency. FEMA OpenFEMA Disaster Declarations Summaries, v2. Declarations dated 1 January 2015 or later. 26,032 rows retrieved 23 September 2026.
- FEMA OpenFEMA Terms and Conditions.
- No survey, interview or site visit was conducted for this report. Every figure is computed from the retrieved declarations file.
This product uses the Federal Emergency Management Agency’s OpenFEMA API, but is not endorsed by FEMA. The Federal Government or FEMA cannot vouch for the data or analyses derived from these data after the data have been retrieved from the Agency’s website(s).