NOAA Essential Fish Habitat Consultation Screener

Batch EFH screening by coordinate. Give it a list of lat/lon sites and it returns, per site, whether the point falls inside NOAA Essential Fish Habitat, which managed species and life stages are designated there, which fishery management council and FMP amendment designated it, whether the point also falls inside a Habitat Area of Particular Concern (with its CFR citation), and a plain-language audit trail explaining the result.

Screened live against the NOAA national EFH inventory (737 designation polygons) and the national HAPC layer (237 polygons). One billable row per EFH overlap; a clear site gets one billable row that says why it is clear.

Screening aid, not a legal determination. Whether a federal action requires EFH consultation under Magnuson-Stevens Act section 305(b)(2) is decided by the action agency and NOAA Fisheries, not by this screen. EFH boundaries are council-scale and generalized. Confirm with the regional NOAA Fisheries Habitat Conservation Division (habitat.noaa.gov).

Who it is for

• Offshore wind, marine construction and dredging developers who need an early read on whether a site will pull an EFH consultation.
• NEPA and permitting consultants assembling the aquatic-resources section of an EA or EIS.
• Ports, cable and pipeline routing teams comparing candidate alignments before spending on survey.
• Environmental due-diligence firms adding a marine-habitat check to a coastal property screen.
• Fisheries and marine-spatial-planning analysts who want the designation metadata, not a map tile.

What changed in v1.1 (2026-08-01)

Every existing field is preserved with its original meaning (append-only).

1. The HAPC distinction the actor advertised could never be made. v1.0 derived is_hapc from the EFH layer TYPE column. A live returnDistinctValues over all 737 EFH features returns exactly one value: EFH. There are zero HAPC records on that layer, so is_hapc was structurally false on every row ever emitted. HAPCs live on a separate NOAA layer in the same organisation, which v1.1 now queries: Flower Garden Banks (27.8756, -93.8158) returns 2 HAPCs (West Flower Garden Bank coral, Bluefin Tuna); Cape Cod (41.6, -70.7) returns 2; Denver returns 0. HAPC results are reported as fields on the existing rows, not as extra billable rows, so the cost per site is unchanged.

2. A failed lookup used to assert that no consultation was needed. The old row template initialised in_efh: false, is_hapc: false and nmfs_efh_consultation_likely: false, and the only failure path nulled in_efh. An asset whose NOAA lookup failed was therefore emitted as a billable row reading "EFH consultation not likely" for a site that had never been checked - the most expensive wrong answer this actor can give. Now every derived answer starts null, each of the four upstream sources reports its own status field, and a verdict is null, never false, when any contributing source did not answer.

3. The PHMS council code was missing from the lookup. FMC is a closed vocabulary of ten codes live, and it includes PHMS (Pacific Highly Migratory Species), which the v1.0 map did not have - so every Pacific HMS row silently carried fmc_name: null. Added, and the vocabulary is now asserted at runtime: an unknown council code fails the run rather than emitting a null name.

4. An inland clear is now explicitly labelled as valid. EFH is designated in coastal and marine waters only, so an inland zero is a legitimate answer, not a service failure. v1.1 adds a context-radius probe and a coastal_context field that distinguishes inside EFH from coastal but just outside a boundary from nothing within 25 miles, consistent with an inland site. Denver returns 0 EFH at the point and 0 within 25 miles; the Gulf canary returns 13 at the point and 29 within 25 miles.

5. Nine columns were being fetched on every request and thrown away. Recovered: FMP_REPORT (the amendment that designated the habitat), LTTD_LINK_ (the species life-stage map PDF, with a page anchor), DATACAVEAT (the NOAA caveat, populated on 634 of 737 features), DDBOX / LTTD_TIT_1 (species common name), FMC_REPORT, LTTD_TITLE, Shape__Area and OBJECTID.

6. 22 live drift assertions run before any billable row. Layer names, geometry types and field sets on both layers; feature counts inside measured bands; the closed FMC and TYPE vocabularies; positive canaries on both layers and a negative control on both; and a coordinate-magnitude check on the published centroids. Any failure stops the run and bills nothing.

7. Truncation is now detectable. Every read is cross-checked against an exact server-side count, and the rollups are nulled rather than published over an arbitrary subset if a read came up short.

8. The run fails instead of billing when every asset errors, and that gate runs before pushData.

Two things the source does not publish, stated plainly

• in_state_waters is always null. NOAA publishes the INSTATEWAT column blank on all 737 national EFH features. Blank means not published, never not in state waters, so it is emitted as null rather than guessed.
• acres is always 0. NOAA publishes ACRES as 0 on all 737 features. It is preserved exactly as published; use designation_area_sq_m instead, which v1.1 computes from Shape__Area with the Web Mercator latitude correction.

A candidate new field was also dropped before release rather than shipped empty: ZONE is blank on 737 of 737 features and is structurally impossible to populate, so no zone column is emitted.

Example input

{
  "assets": [
    { "lat": 27.8756, "lon": -93.8158, "label": "Flower Garden Banks (EFH + coral HAPC)" },
    { "lat": 28.5, "lon": -91.0, "label": "Gulf of Mexico offshore Louisiana" },
    { "lat": 39.7392, "lon": -104.9903, "label": "Denver CO (inland control)" }
  ],
  "radiusMiles": 0,
  "contextRadiusMiles": 25,
  "includeHapc": true,
  "maxAssets": 500
}

Input fields

Field	Type	Description
assets	array	Points to screen: { "lat": number, "lon": number, "label": string } in WGS84 decimal degrees.
radiusMiles	number	Buffer around each point, 0-25 statute miles. 0 (default) is a pure point-in-polygon intersect.
contextRadiusMiles	integer	Radius, 0-200 miles (default 25), used to answer why a clear point is clear. 0 skips the probe and coastal_context says so.
includeHapc	boolean	Default true. Query the separate NOAA HAPC layer as well as EFH.
maxAssets	integer	Safety cap on assets per run (default 500, max 2000).

Output fields

Every field the actor declares is listed below.

Site identity and provenance

Field	Type	Description
asset_label	string	null
asset_lat	number	null
asset_lon	number	null
radius_miles	number	null
checked_at	string	null
source_url	string	null
hapc_source_url	string	null
screen_basis	string	null
disclaimer	string	null
error	string	null

Verdict and audit trail

Field	Type	Description
in_efh	boolean	null
nmfs_efh_consultation_likely	boolean	null
coastal_context	string	null
context_radius_miles	number	null
efh_within_context_radius_count	integer	null
hapc_within_context_radius_count	integer	null
consultation_basis	string	null

This EFH designation (one row per species/life-stage overlap)

Field	Type	Description
efh_site_name	string	null
efh_species_common_name	string	null
life_stage	string	null
designation_type	string	null
is_hapc	boolean	null
fmc_code	string	null
fmc_name	string	null
fmc_report	string	null
region	string	null
fmp_amendment	string	null
council_text_url	string	null
lifestage_map_title	string	null
lifestage_map_url	string	null
data_caveat	string	null
acres	number	null
designation_area_sq_m	number	null
centroid_lat	number	null
centroid_lon	number	null
in_state_waters	boolean	null
efh_objectid	integer	null

All EFH at this point (rollup, repeated on every row)

Field	Type	Description
efh_overlap_count	integer	null
efh_species_count	integer	null
efh_species_list	string	null
efh_lifestages	string	null
efh_councils	string	null
efh_council_count	integer	null
efh_regions	string	null
efh_fmp_amendments	string	null

Habitat Areas of Particular Concern (separate NOAA layer)

Field	Type	Description
hapc_overlap	boolean	null
hapc_overlap_count	integer	null
hapc_site_names	string	null
hapc_species_list	string	null
hapc_habitats	string	null
hapc_fishery_management_plans	string	null
hapc_cfr_citations	string	null
hapc_regulation_urls	string	null
hapc_publication_dates	string	null
hapc_data_caveats	string	null

Per-source outcome (null means NOT CHECKED)

Field	Type	Description
efh_status	string	null
hapc_status	string	null
efh_context_status	string	null
hapc_context_status	string	null
sources_attempted	integer	null
sources_ok	integer	null
sources_failed	integer	null
sources_failed_names	string	null
partial_result	boolean	null

Coverage, truncation and upstream health

Field	Type	Description
efh_features_matched	integer	null
efh_features_exact_count	integer	null
efh_read_truncated	boolean	null
hapc_features_matched	integer	null
hapc_features_exact_count	integer	null
hapc_read_truncated	boolean	null
service_max_record_count	integer	null
drift_checks_passed	integer	null
drift_checks_total	integer	null

Reading the results correctly

• null is not false. A null verdict means a source did not answer, so nothing was verified. Check partial_result and sources_failed_names before acting on a row.
• One row per EFH overlap. A single offshore point commonly intersects a dozen overlapping species designations; the Gulf canary returns 13 and Flower Garden Banks returns 17. The asset-level rollups (efh_species_list, efh_councils, hapc_site_names) repeat on every row for that asset so a single row is enough to answer "what is here".
• is_hapc vs hapc_overlap. is_hapc describes the EFH row itself and is false everywhere, because the EFH layer carries no HAPC records. hapc_overlap is the real answer, from the separate HAPC layer.
• A clear inland row is the right answer. coastal_context tells you which kind of clear you have. no_efh_within_context_radius_consistent_with_an_inland_or_unmapped_site means EFH simply is not designated on land.
• hapc_cfr_citations is published on 84 of 237 HAPC features, and hapc_data_caveats on 1 of 237, so nulls there are upstream gaps, not bugs. A verified input that populates the CFR citation: { "lat": 37.0658, "lon": -74.6622 } (Norfolk Canyon, Tilefish HAPC, 50 CFR 648.296).

Use as an MCP tool

This actor is available to AI agents through mcp.apify.com. Every input and output field carries a description, so an agent can call it and reason about the answer directly. It chains cleanly with the other coordinate screeners: pass the same points to the USFWS critical-habitat screener, the NWI wetlands screener and the EPA impaired-waters screener, then join on asset_label.

Typical agent prompt: "Screen these six offshore lease-block centroids for Essential Fish Habitat and tell me which ones also fall in a Habitat Area of Particular Concern, with the CFR citation."

Pricing

Pay per result: $10 per 1,000 result rows ($0.01 per row). A site inside EFH produces one row per species/life-stage overlap; a clear site produces exactly one row. HAPC results are fields on those rows, not extra rows, so adding the HAPC layer did not change what a site costs. Graduated discounts apply on paid Apify plans (Bronze 20% / Silver 30% / Gold 45% / Platinum 60% / Diamond 70%).

A run that cannot verify the upstream is healthy, or in which every asset failed, fails and bills nothing.

FAQ

Does an EFH hit mean I have to consult? No. It means a federal action at that point is likely to trigger EFH consultation under MSA section 305(b)(2). The determination is made by the action agency and NOAA Fisheries. consultation_basis spells out what the row is actually claiming.

Why did my inland site return one row saying nothing? Because that is the correct answer. EFH is designated in coastal and marine waters only. The row carries coastal_context and a consultation_basis that says so explicitly, precisely so a legitimate clear is not mistaken for a service failure.

What is the difference between EFH and a HAPC? A HAPC is a subset of EFH that NOAA has singled out for higher conservation priority - typically rare, ecologically important, or especially vulnerable habitat. HAPCs carry a Code of Federal Regulations citation where one has been published. They live on a different NOAA layer, which is why v1.0 could not see them.

Why is acres always 0 and in_state_waters always null? Because that is exactly what NOAA publishes on all 737 features. Both are preserved verbatim rather than guessed. Use designation_area_sq_m for area.

Why did my run fail with "Upstream drift detected"? A live assertion about the NOAA service failed - a renamed layer, a collapsed feature count, an unknown council code, or a spatial filter that stopped filtering. The alternative would have been to bill you for a confident wrong answer. The message names the exact check.

Can I search near a point rather than at it? Yes. Set radiusMiles up to 25 for a buffered search. That changes what counts as a hit; radius_miles is recorded on every row.

Related actors

Same coordinate in, same asset_label out:

• USFWS Critical Habitat Screener - the terrestrial/ESA sibling of this screen.
• FWS Wetlands Proximity Screener - National Wetlands Inventory.
• EPA Impaired Waters 303(d) Screener - Clean Water Act listings.
• NHD Surface Water 404 Screener - Clean Water Act section 404 jurisdiction.
• BOEM Offshore Wind Lease Tracker - cross-sells directly for offshore siting.
• Site Due Diligence Bundle - 20 layers in a single call.

Data sources

Both keyless, both on ArcGIS Online organisation FiaPA4ga0iQKduv3 (Federal_User_Community), counts verified live 2026-08-01 and asserted at runtime:

• nationwide_efh_1/FeatureServer/0 - layer Nationwide_EFH_B, 737 EFH polygons.
• Habitat_Areas_of_Particular_Concern_1/FeatureServer/0 - layer Nationwide_HAPC_B, 237 HAPC polygons.