fisheries-survey-finder

name	fisheries-survey-finder
description	Use when the user asks about fisheries data — phrases like "bottom trawl survey", "stock assessment", "ichthyoplankton", "groundfish", "longline survey", "{species} catch records", or any NMFS science center mention (AFSC, NEFSC, NWFSC, PIFSC, SEFSC). Encodes which science center covers which region and what survey types live where.

Fisheries survey finder

The five NOAA NMFS regional science centers operate distinct survey programs. Picking the right one is essentially a geography question.

See science-centers.md for the full survey-type matrix.

Region → Science center

Region	Center	Collection ID
Alaska / Bering Sea / Aleutians	AFSC	AFSC
Northeast US (Maine → Cape Hatteras)	NEFSC	NEFSC
Northwest US / Pacific (CA, OR, WA)	NWFSC	NWFSC
Pacific Islands / Hawaii	PIFSC	PIFSC
Southeast US / Gulf of Mexico / Caribbean	SEFSC	SEFSC

If the user names a region, look up the matching center first. If they name a species, use multiple centers (most species are surveyed by 2+ centers).

Survey type quick reference

Each center runs different survey types. The most common are:

Survey type	What it samples	Typical centers
Bottom trawl	Demersal fish, crabs, groundfish abundance	AFSC, NEFSC, NWFSC, SEFSC
Longline	Sharks, swordfish, tunas, large pelagics	NEFSC, SEFSC, PIFSC
Acoustic / hydroacoustic	Pelagic fish biomass (anchovy, herring, hake)	AFSC, NWFSC, NEFSC
Ichthyoplankton (egg/larvae)	Spawning ecology, recruitment indices	NEFSC, SEFSC, AFSC
Reef Visual Census (RVC)	Coral reef fish abundance	SEFSC (Caribbean), PIFSC
Marine mammal / seabird	Pinniped, cetacean, seabird abundance	NEFSC, SEFSC, AFSC, PIFSC
Trap / pot	Crabs, lobsters, fish traps	AFSC, NEFSC, SEFSC

Pattern: "what fisheries data exists for X"

1. Identify region from user → pick center collection
2. aggregate(collections=["AFSC"], group_by="datetime", interval="1Y", bbox=<region bbox>)
   to see survey time coverage
3. search_datasets(collections=["AFSC"], bbox=<bbox>, datetime=<window>, q="<survey type>", limit=25)

Pattern: species-specific query

"Find Pacific cod data from Alaska bottom trawls in 2023."

search_datasets(
  collections=["AFSC"],
  bbox=[-180, 51, -130, 65],   # Bering Sea + Gulf of Alaska
  datetime="2023-06-01/2023-09-30",   # AFSC bottom trawl summer survey
  q="Pacific cod bottom trawl",
  limit=25
)

The q field can take both common and scientific names. When the user uses a common name, search with both:

q="Pacific cod"             # common name
q="Gadus macrocephalus"     # scientific name

Try the common name first — most NMFS metadata is written for non-specialists.

Pattern: stock assessment data

For population estimates / stock status:

• AQUAVIEW exposes the survey data, which feeds stock assessments — not the assessments themselves
• For published stock-status outputs, point users to NOAA Fisheries (fisheries.noaa.gov) or NMFS-specific portals
• Use AQUAVIEW for the input data: catch-per-unit-effort, length distributions, abundance indices

Multi-region species

Many species are surveyed by multiple centers (e.g., bluefin tuna by NEFSC + SEFSC; salmon by NWFSC + AFSC). Search across centers:

search_datasets(
  collections=["NEFSC", "SEFSC", "PIFSC"],
  q="bluefin tuna",
  datetime="2020-01-01/2024-12-31",
  limit=50
)

Common variables in fisheries items

Variable	Notes
catch_kg or catch_n	Catch weight or count (per haul / station)
species_code	NMFS species code (e.g., 21720 = Pacific cod)
bottom_depth_m	Trawl depth
tow_duration_min	Standardized to compute CPUE
length_cm	Per-fish length, where individual records are reported
sex, age	Sex/age data, sparse

Output formats

Most NMFS fisheries data in AQUAVIEW comes as CSV (per-haul records) or NetCDF (gridded survey indices). Tell the user that direct download links resolve to either format depending on the survey program.

Worked example

"Show me bottom trawl groundfish data from the Bering Sea for 2023."

1. AFSC region; bbox = [-180, 53, -158, 66] (eastern Bering Sea)
2. search_datasets(
     collections=["AFSC"],
     bbox=[-180, 53, -158, 66],
     datetime="2023-06-01/2023-08-31",
     q="bottom trawl groundfish",
     limit=25
   )
3. Returned items represent stations / hauls; each has CSV asset with per-tow catch