Dam Safety Risk Monitor — USACE National Inventory of Dams (NID)

Search and monitor 92,000+ US dams from the US Army Corps of Engineers National Inventory of Dams and get a normalized, analysis-ready record per dam with a composite risk score built from three signals regulators track: hazard potential, latest condition assessment, and inspection staleness — plus flags for dams with no Emergency Action Plan (EAP) on file.

No API key, no scraping fragile HTML: data comes straight from the official USACE NID ArcGIS service.

Who this is for

• P&C insurers and reinsurers — underwrite flood and property exposure downstream of high-hazard dams; screen a book of locations against poor-condition, overdue-inspection dams.
• Dam-safety and civil engineers — build state watchlists of High-hazard dams in Poor/Unsatisfactory condition or with stale inspections.
• Infrastructure due-diligence and real-estate risk teams — check dams near an asset, portfolio, or development site before acquisition.
• Journalists, researchers, and emergency managers — track dams missing Emergency Action Plans, county by county.

What it does

1. Queries the USACE NID FeatureServer (keyless, official) with your state / hazard / condition / owner filters.
2. Normalizes each dam into a clean snake_case record (trimmed names, ISO dates, parsed county).
3. Computes risk_score = hazard weight (High 10 / Significant 5 / Low 1) × condition weight (Unsatisfactory 3.5 / Poor 3 / Fair 2 / Satisfactory 1) × inspection-staleness factor (1.0 fresh → 3.0 at 20+ years), + 5 when no current EAP. Range ≈ 1–110.
4. Adds machine-readable risk_flags such as high_hazard_poor_condition, inspection_stale_10yr, eap_missing.
5. Optional client-side screens: minRiskScore and requireMissingEap.

Example input

{
  "states": ["CO"],
  "hazardPotential": ["High"],
  "conditionAssessment": ["Poor", "Unsatisfactory"],
  "minRiskScore": 40,
  "requireMissingEap": false,
  "maxResults": 500
}

States accept postal codes (CO) or full names (Colorado). Leave states empty to scan the whole country (bounded by maxResults).

Output fields (one record per dam)

nid_id, federal_id, name, state, county, city, river, latitude, longitude, owner_type, primary_purpose, primary_dam_type, dam_height_ft, max_storage_acre_ft, year_completed, hazard_potential, condition_assessment, condition_assess_date, last_inspection_date, inspection_age_years, inspection_frequency_years, eap_status, eap_last_rev_date, operational_status, risk_score, risk_flags[], data_updated, source_url

Use as an MCP tool

This actor works as a clean, chainable MCP tool for AI agents (Claude, Cursor, etc.) via mcp.apify.com — the input and every output field are fully described, so an agent can query "high-hazard poor-condition dams in Texas without an EAP" unassisted.

Monitoring

Schedule it (e.g. weekly per state) and diff risk_score, condition_assessment, and eap_status between runs to catch downgraded dams the day the inventory updates.

FAQ

How many high hazard dams are in my state? Run with states=["<your state>"], hazardPotential=["High"] — the dataset count is the answer, each row a dam.

Which dams near me are in poor condition? Filter your state with conditionAssessment=["Poor","Unsatisfactory"], then sort by risk_score or filter by county.

Which dams have no emergency action plan? Set requireMissingEap: true. High-hazard dams without an EAP are the classic regulatory red flag.

Is the National Inventory of Dams free / do I need an API key? Yes, free and keyless — this actor reads the official USACE public ArcGIS service directly.

How is the dam risk score calculated? Hazard × condition × years-since-inspection, plus an EAP penalty — see "What it does" above; risk_flags explains every contributing factor per dam.

How current is the data? The NID is updated continuously by USACE and state regulators; each record carries its data_updated date.