Skip to contents

Get Catch and Discard Fits Plot

get_catch_fits_plot.Rd

Plots observed catch and discard time series alongside model-predicted values for one or more SPoRC model runs, for both pooled (region-level) and population-specific data streams.

Usage

get_catch_fits_plot(data, rep, model_names)

Arguments

data

List of length n_models, where each element is a SPoRC data list. ObsCatch [n_regions × n_yrs × n_seas × n_fish_fleets] provides pooled observed catch, and ObsDiscard provides pooled observed discards. Wt_Catch and Wt_Discard, together with ln_sigmaC, are used to reconstruct observation-error standard deviations for confidence interval construction.

When UseCatch_pop or UseDiscard_pop contain active elements, population-level data (ObsCatch_pop, ObsDiscard_pop) and corresponding weights are also used.

rep

List of length n_models, where each element is a SPoRC model report (output of obj$report() after optimisation). PredCatch and PredDiscard [n_pop × n_regions × n_yrs × n_seas × n_fish_fleets] are summed across populations for pooled trajectories and used directly for population-specific trajectories. ln_sigmaC and ln_sigmaC_pop are used for confidence interval construction.

model_names

Character vector of length n_models giving display names for each model run. Used in the legend for predicted trajectories.

Value

A list of ggplot objects:

[[1]] catch_fit_rg_plot

Pooled catch fits (region-level). Produced when UseCatch == 1. Observations shown as geom_pointrange with 95% lognormal confidence intervals; predictions shown as lines colored by model. Faceted by (Season × Fleet) × Region with free y-scales.

[[2]] catch_fit_pop_plot

Population-specific catch fits. Produced when UseCatch_pop == 1. Faceted by (Population × Season × Fleet) × Region. Returns NULL if not used.

[[3]] discard_fit_rg_plot

Pooled discard fits (region-level). Produced when UseDiscard == 1. Same structure as catch plots.

[[4]] discard_fit_pop_plot

Population-specific discard fits. Produced when UseDiscard_pop == 1. Returns NULL if not used.

Details

Pooled predictions are summed across populations before comparison with observed data. Years where observed values are zero are excluded from both observed and predicted layers to avoid issues in lognormal confidence interval construction.

This function produces diagnostic plots for both catch and discard data. Observed and predicted time series are shown for each, with lognormal confidence intervals derived from ln_sigmaC (or population-level equivalents) and sampling weights.

Observations equal to zero are excluded prior to plotting to avoid issues in log-space confidence interval construction.

Predicted catch and discard are aggregated across populations for pooled diagnostics, while population-specific plots use unaggregated outputs.

See also

Other Model Diagnostics: do_jitter(), do_likelihood_profile(), do_retrospective(), do_runs_test(), get_comp_prop(), get_idx_fits(), get_idx_fits_plot(), get_model_rep_from_mcmc(), get_nLL_plot(), get_osa(), get_retrospective_plot(), get_retrospective_relative_difference(), plot_resids()