How to Scrape FEMA Disaster Declarations

FEMA declaration pages are authoritative, but recurring analysis needs consistent records rather than manual browsing. The FEMA Disaster Scraper turns OpenFEMA declaration summaries into rows with identifiers, geography, dates, incident labels, and assistance flags.

Start with a focused declaration query

Use official state codes and FEMA incident labels. A narrow first run makes validation straightforward.

{"states":["TX"],"incidentTypes":["Flood"],"fromDate":"2025-01-01","maxResults":50}

Inspect the disaster number, designated area, declaration date, incident window, declaration type, and program fields. Open several source_url values and confirm that the state and title match. Use the OpenFEMA row ID as the durable record identifier; do not deduplicate solely by disaster number because the same declaration covers multiple areas.

When loading results into a warehouse, keep FIPS codes as strings so leading zeros survive. Treat null program values separately from false values if your analysis depends on source completeness. The declaration date is not the incident start date, and either can differ significantly.

After validation, save the input as a Task and schedule it. For incremental monitoring, choose a recent fromDate and upsert known IDs. Re-reading an overlap window protects against delayed records and official corrections.

This dataset supports risk research and response workflows, but it is not a live warning feed. Never use it instead of official emergency instructions, forecasts, or local alerts.

Keep the original declaration type codes in derived tables because major disasters, emergencies, and fire-management declarations represent different administrative actions.