Stock Assessments use data from fishery dependent and fishery independent sources to accurately assess marine fish populations.
Fishery-Dependent Data Sources
Fishery-dependent data are collected directly from the commercial and recreational fisheries. Types of information may include fishing effort, total amount of fish removed from the ocean (landings and discards), species (target and incidental), and biological information. Fishery dependent data may be collected through use of dockside monitors, at-sea observers, logbooks, electronic monitoring and reporting systems, telephone surveys, and vessel-monitoring surveys. For more information or to download fishery dependent data, please visit our Fisheries Statistics Division, which provide automated data summaries of commercial fisheries landings and recreational fisheries landings.
Fishery-Independent Data Sources
Fishery independent data are collected from at-sea surveys where scientists from NOAA Fisheries science centers and partner agencies/institutes gather information on fish stock abundance, biology and their ecosystem for inclusion in stock assessments. These surveys are conducted on NOAA ships operated through NOAA’s Marine and Aviation Operations (NMAO) or fishing vessels chartered by NOAA Fisheries. The choice of survey mode is driven principally by specific survey requirements, regional availability of suitable vessels, and maintaining continuity of survey time series. NOAA Fisheries is working to improve survey capabilities with wise investments in the evaluation, development and implementation of advanced sampling technologies. Click here to access the Fishery Independent Survey System (FINSS) for more information on surveys conducted by NOAA Fisheries.
Stock assessments use data collected from various sources including fishery-dependent (collected directly from the fishermen or fishery) and fishery-independent (collected independently of the fishermen or fishery).
Three primary categories of data inputs are required for stock assessments: abundance, biology and catch (often called the ABCs of stock assessments). The following demonstrates how these ABC data are collected and how they work together in stock assessment models to produce scientific advice for fishery management.
Measures trends in relative abundance, used to reflect historical changes in population size. One source of these data is through trawl or acoustic surveys done from research ships. NOAA Fisheries conducts many core surveys on NOAA ships NOAA Fisheries also works with the fishing industry by chartering fishing vessels for additional survey work, and collaborates with states that use their vessels to survey inshore waters.
Fishery Independent Survey System (FINSS)
Advanced Sampling Technologies
Provides information on stock productivity and rates of change.
Samples are collected from surveys, dockside monitoring, observers, and targeted research studies
Measures fish size, age, reproductive rates, movement and natural mortality
Fish ear bones (otoliths) are used to determine the ages of fish
Partnerships with industry, States, academia, OAR are important
Provides the amount of fish removed from a stock by all types of fishing and bycatch.
Types of catch data:
Recreational Effort Surveys
Partnership with state agencies and commissions are common
Fish stock assessment models compute the historical levels of fish stock abundance and fishing mortality that are most consistent with observed trends in abundance data, the fish’s biology, and historical catch levels. Assessment models produce estimates of important fishery management factors and provide the technical basis for managers to determine stock status and set annual catch limits and other fishery management measures to maintain sustainable fishing and healthy ecosystems.
Conceptually, fish assessment models are similar to the dynamic atmospheric models used by NOAA’s National Weather Service to forecast the weather. Weather models combine a variety of weather observations to calibrate complex atmospheric models that forecasters can then use to make informed predictions on whether it will be sunny or rainy tomorrow. Even though stock assessments operate on much longer time scales than weather models – months and years rather than hours and days – they are similar in how they incorporate complex observations from many different sources into a holistic representation of the situation.
A variety of assessment modeling packages are available, ranging from basic models to complex models that are able to represent a range of fish stock dynamics, including ecosystem factors, to models developed specifically to represent an individual species or stock. Scientists choose the model that offers the features best suited for a stock’s life history and data availability, and may try multiple models to more fully explore what we know, and don’t know, about the status of a fish stock. Read more about the assessment models available to NOAA Fisheries scientists...
Today’s stock assessment models run as computer simulations of fish populations. The most complete kind assessment model is called an integrated analysis model. Integrated analysis models are composed of three sub-model layers: 1) population sub-model; 2) observation sub-model; and 3) statistical sub-model.
Population sub-model: Computes the essential population factors such as stock abundance, mortality, growth, reproduction, and movement for each year, typically going back several decades.
Observation sub-model: Translates estimates from the population model into predictions for each observed data input to the model, including survey abundance index, fish size and age composition, and others as available.
Statistical sub-model: Uses advanced statistical methods and powerful computers to compare data predictions to the data observations and adjusts factors in the population and observation models to achieve as good a match as possible to all the available data.
In addition to producing estimates of important management factors, stock assessments also need to quantify the level of uncertainty associated with assessment results. By measuring how closely a stock assessment model fits to the available observed data, scientists provide resource managers with information on how reliable (i.e. the level of statistical confidence) the historical estimates and future predictions from the model are for a fish stock.