obm.qmd

# Observation Model

The goal of the observation model is to produce output similar to the empirical data available from the fishery and the fishery-independent survey. The observation model generates total coastwide catch data every year as:
(15)
Fishery and survey age compositions are also generated per year: js, jF. For the fishery, the numbers at age in the catch (Equation 14) are found by dividing biomass at age in the catch by weight at age. Compositions are:
where Ny,a,F is the abundance of individuals at age in the catch. All ages over 15 are summed up for both the fishery and the scientific survey.
The model generates observed biomass biennially at a coastwide scale to mimic the binational acoustic trawl survey:

where q is the catchability coefficient, and sa is the survey selectivity. We assume that the survey takes part in the third quarter of the year with error ∈s ~ Lognormal(C,s2survey). The standard deviation comprises two different values, s2surv and s2s,y, where s2surv is a constant variance and s2s,y is a standard deviation specific to the survey years taken from the kriging of the acoustic data (Berger et al. 2019).