What can you tell me about the Beauford Assessment Model BAM?
**The Beaufort Assessment Model (BAM)** — note the standard spelling is *Beaufort*, not Beauford — is a statistical catch-at-age stock assessment model developed at NOAA’s Beaufort Laboratory (Southeast Fisheries Science Center) in Beaufort, North Carolina. (LINK)
It is an integrated age-structured population model that simulates a fish stock forward in time. It incorporates biological processes (growth, recruitment, natural mortality, maturity, reproductive output) and fishing processes (selectivity, fishing mortality, landings, and discards). Parameters are estimated so that the simulated population matches observed data on the real population. The population is treated as closed (no immigration/emigration) and typically runs on an annual time step with multiple age classes, including a plus-group for the oldest ages. (LINK)
### Key Features
– **Implementation**: Coded in AD Model Builder (ADMB). It is highly flexible and customizable for individual stocks (users edit the ADMB template and data files), which is both a strength and a requirement for technical expertise.
– **Structure**: Similar to other age-structured models such as Stock Synthesis and the Age Structured Assessment Program (ASAP). It fits multiple data sources simultaneously in a single analysis (landings, discards, abundance indices, age/length compositions, etc.).
– **Initialization and dynamics**: Options for starting abundance at age (equilibrium or with estimated deviations). Growth often uses the von Bertalanffy equation; natural mortality can be age-specific (e.g., Lorenzen or Charnov methods); recruitment commonly follows a spawner-recruit relationship (e.g., Beverton-Holt) with possible deviations and bias corrections.
– **Outputs**: Estimates of abundance, biomass, fishing mortality, and biological reference points such as maximum sustainable yield (MSY), \(F_{\text{MSY}}\), and spawning stock biomass at MSY (\(SSB_{\text{MSY}}\)).
– **Validation**: Simulation testing has shown it can recover true parameters accurately. It has undergone multiple independent reviews. (LINK)
### Applications
BAM (and earlier versions of the same modeling approach) has been widely used in the SEDAR (Southeast Data, Assessment, and Review) process for U.S. South Atlantic and Gulf of Mexico stocks. Examples include Atlantic and Gulf menhaden, red snapper, gag, black sea bass, vermilion snapper, red grouper, snowy grouper, tilefish, blueline tilefish, greater amberjack, Spanish mackerel, and others. Assessment reports are available through SEDAR. (LINK)
### Documentation
The primary technical reference is:
– Williams, E. H., and K. W. Shertzer. 2015. *Technical documentation of the Beaufort Assessment Model (BAM)*. NOAA Technical Memorandum NMFS-SEFSC-671. (DOI: 10.7289/V57M05W6).
It includes model equations, a flow diagram of operations, and example ADMB code (e.g., for gag). Stock-specific working papers with customized code and equations appear in many SEDAR documents.
**Note on other “BAM” acronyms**: There are unrelated models such as the Brief Addiction Monitor (a clinical progress-monitoring tool for substance use disorders), various Behaviour/Backward Assessment Models in education or pervasive computing, and others. Given the phrasing “Beauford Assessment Model BAM,” the fisheries stock-assessment model from the Beaufort Lab is the clear primary match.