Skip to content

Developing a new model

Grant Adams edited this page Jan 21, 2026 · 6 revisions

Potential workflow for building a new CEATTLE model. Data can be updated in R, but it may be easier to update in excel.

Single species (start here)

  1. Use the the BS2017SS excel file (written from code below) as a basis to bridge to the new model.

library(Rceattle) data(BS2017SS) # ?BS2017SS for more information on the data

# Write data to excel Rceattle::write_data(data_list = BS2017SS, file = "BS2017SS.xlsx")

  1. Update the control for the specifications of population model

  2. Update the fleet_control for the specifications of observation model (surveys/fisheries). This includes parameterizations for catchability/selectivity.

  3. Update data on index_data,catch_data, comp_data

  4. If comp_data are length based, update the age transition matrix (time-invariant currently)

  5. If ageing error is present update age_error (time-invariant currently)

  6. Update life history bits (weight, maturity, sex_ratio, M1_base sheets for weight(kg), maturity-at-age, proportion female at age, natural/residual mortality at age, respectively)

  7. Fit in R and compare to the SS model

Multi-species This may depend on how the bioenergetics and diet data end up looking like!

  1. Update the bioenergetic parameterization on the bioenergetics_control and ration_data sheets

  2. Update residual M1 on M1_base sheet if fixed (or can estimate)

  3. Update bottom/sea-surface temperature on env_data sheet

  4. Update diet data. Proportion of prey-at-age by weight in the stomach of a predator at-age on the diet_data sheet

  5. Fit in R and celebrate

Clone this wiki locally