Description of Model Report
d_model_report.RmdThe following tables describe all elements contained within
obj$rep, which is generated after a model is run. Note that
when n_sexes > 1, the first dimension will always be
females and the second dimension will always be males. Similarly, when
n_pop > 1, populations are indexed in the order they are
defined.
Biological Processes
| Name | Description | Dimensions |
|---|---|---|
R0 |
Global mean or virgin recruitment for each population | n_pop |
R0_yr |
Global mean or virgin recruitment by year, through its
R0_blocks time blocks; R0 itself is the
R0_ref_block value |
n_pop × n_years |
rinit |
Global recruitment scalar for initialization for each population | n_pop |
rec_region_prop |
Recruitment regional apportionment proportions | n_pop × n_regions |
rec_seas_prop |
Recruitment seasonal apportionment proportions | n_pop × n_seas |
stray_rate |
Proportion of each population’s spawning biomass contributing to
non-natal regions (only relevant when n_pop > 1) |
n_pop × n_years |
sexratio |
Recruitment sex ratio (female proportion) | n_pop × n_regions × n_years × n_sexes |
h_trans |
Steepness parameter by population and region; ignored under mean recruitment unless a stock-recruit penalty is applied | n_pop × n_regions |
SR_pred |
Stock-recruit curve prediction under
rec_model = "mean_rec" with sr_penalty,
evaluated alongside the dynamics but never advancing them; ones when no
such penalty is applied |
n_pop × n_regions × n_years |
sr_R0 |
Scale of the stock-recruit curve under sr_penalty.
Equals ln_global_R0 exponentiated under
sr_R0_spec = "shared", ln_sr_R0 under
"est", and rinit under "rinit",
so the three settings are distinguishable here. |
n_pop |
NAA |
Numbers-at-age (fished) | n_pop × n_regions × (n_years + 1) × n_seas × n_ages × n_sexes |
NAA0 |
Numbers-at-age (unfished) | n_pop × n_regions × (n_years + 1) × n_seas × n_ages × n_sexes |
NAA_bef |
Numbers-at-age at the start of the season, before movement is applied | n_pop × n_regions × (n_years + 1) × n_seas × n_ages × n_sexes |
NAA_aft |
Numbers-at-age after movement is applied. Under
move_timing = 0 this is the post-movement, pre-mortality
state; under move_timing 1 and 2 movement is resolved at
the end of the season, so it is the post-season state (movement and
mortality together) |
n_pop × n_regions × (n_years + 1) × n_seas × n_ages × n_sexes |
NAA_pred |
Deterministic numbers at the start of a season that the mortality
and ageing step predicts, before the state-space overwrite. Zero unless
NAA_re is on, and zero in seasons outside
NAA_re_seasons. The realized innovation is
log(exp(ln_NAA)) - log(NAA_pred) over the active cells |
n_pop × n_regions × n_years × n_seasons × n_ages × n_sexes |
NAA_scalar |
Multiplicative factor the state applied to that prediction, one wherever it did not apply. Tag cohorts and the unfished numbers take the same factor | n_pop × n_regions × n_years × n_seasons × n_ages × n_sexes |
NAA_state_nLL |
Penalty on the state-space numbers-at-age innovations. Zero when
NAA_re is "none"
|
scalar |
ZAA |
Total instantaneous mortality | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes |
natmort |
Instantaneous natural mortality per year, in force during each
season; multiply by seasdur for the amount accumulated
within a season |
n_pop × n_regions × n_years × n_seas × n_ages × n_sexes |
bias_ramp |
Vector of bias ramp correction values applied to recruitment deviations | n_est_rec_devs |
Movement |
Movement (transition) matrix | n_pop × n_regions × n_regions × (n_years + n_proj_yrs_devs) × n_seas × n_ages × n_sexes |
Mrate |
Instantaneous movement rate matrix (CTMC movement only;
NULL otherwise) |
n_pop × n_regions × n_regions × (n_years + n_proj_yrs_devs) × n_seas × n_ages × n_sexes |
sgl_seas_spawning_movement |
Spawning movement matrix applied during the spawning event when
n_seas == 1 and n_pop > 1
|
n_pop × n_regions × n_regions × n_years × n_ages × n_sexes |
Fishery Processes
| Name | Description | Dimensions |
|---|---|---|
init_F |
Initial fishing mortality applied to equilibrium age structure | n_regions × n_seas × n_fish_fleets |
ln_sigmaC |
Log-scale standard deviation for the region-aggregated catch likelihood | n_regions × n_years × n_seas × n_fish_fleets |
ln_sigmaC_pop |
Log-scale standard deviation for the population-specific catch likelihood | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
Fmort |
Fishing mortality scalar | n_regions × n_years × n_seas × n_fish_fleets |
dmr |
Discard mortality rate | n_regions × n_years × n_seas × n_fish_fleets |
tot_FAA |
Total fishing mortality at age (retained + dead discards) | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_fish_fleets |
ret_FAA |
Retained fishing mortality at age | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_fish_fleets |
disc_FAA |
Dead discard fishing mortality at age | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_fish_fleets |
CAA |
Retained catch-at-age | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_fish_fleets |
DAA |
Dead discarded catch-at-age | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_fish_fleets |
CAL |
Retained catch-at-length | n_pop × n_regions × n_years × n_seas × n_lens × n_sexes × n_fish_fleets |
DAL |
Dead discarded catch-at-length | n_pop × n_regions × n_years × n_seas × n_lens × n_sexes × n_fish_fleets |
PredCatch |
Predicted retained catch (population-specific; summed across populations for aggregated likelihood) | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
PredDiscard |
Predicted discards (population-specific; summed across populations for aggregated likelihood) | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
PredFishIdx |
Predicted fishery index (population-specific; summed across populations for aggregated likelihood) | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
fish_sel |
Total fishery selectivity (age-based) | n_pop × n_regions × (n_years + n_proj_yrs_devs) × n_seas × n_ages × n_sexes × n_fish_fleets |
ret_sel |
Retention selectivity (age-based) | n_pop × n_regions × (n_years + n_proj_yrs_devs) × n_seas × n_ages × n_sexes × n_fish_fleets |
fish_q |
Fishery catchability | n_regions × n_years × n_fish_fleets |
Survey Processes
| Name | Description | Dimensions |
|---|---|---|
PredSrvIdx |
Predicted survey index (population-specific; summed across populations for aggregated likelihood) | n_pop × n_regions × n_years × n_seas × n_srv_fleets |
srv_sel |
Survey selectivity (age-based) | n_pop × n_regions × (n_years + n_proj_yrs_devs) × n_seas × n_ages × n_sexes × n_srv_fleets |
srv_q |
Survey catchability | n_regions × n_years × n_srv_fleets |
SrvIAA |
Survey index-at-age | n_pop × n_regions × n_years × n_seas × n_ages × n_sexes × n_srv_fleets |
SrvIAL |
Survey index-at-length | n_pop × n_regions × n_years × n_seas × n_lens × n_sexes × n_srv_fleets |
Note: Length-based selectivity arrays (fish_sel_l,
ret_sel_l, srv_sel_l) are only reported when
the corresponding fish_selex_type,
ret_selex_type, or srv_selex_type equals 1,
and share the same dimension structure as their age-based counterparts
but replacing n_ages with n_lens and without
the population and season dimensions (since length-based selectivity
does not vary across those dimensions directly, population and season
specificity enters only upon conversion to age via the size-age
transition matrix).
Tagging Processes
| Name | Description | Dimensions |
|---|---|---|
pred_conv_tag_fish_recap |
Predicted conventional tag recaptures | conv_tag_max_liberty × n_seas × n_conv_tag_cohorts × n_pop × n_regions × n_ages × n_sexes × n_fish_fleets |
conv_tag_fish_avail |
Number of tagged fish available for recapture at each time step | (conv_tag_max_liberty + 1) × n_seas × n_conv_tag_cohorts × n_pop × n_regions × n_ages × n_sexes |
conv_tag_fish_reporting |
Tag reporting rate by region, year, and fleet | n_regions × n_years × n_fish_fleets |
Parameter Deviations
| Name | Description | Dimensions |
|---|---|---|
ln_RecDevs |
Log-scale recruitment deviations by population, region, and year (including projection years) | n_pop × n_regions × (n_est_rec_devs + n_proj_yrs_devs) |
move_devs |
Movement deviations by population, region pair, year, age, and sex (including projection years) | n_pop × n_regions × (n_regions - 1) × (n_years + n_proj_yrs_devs) × n_ages × n_sexes |
ln_fishsel_devs |
Log-scale fishery selectivity deviations by region, year, bin, sex, and fleet | n_regions × (n_years + n_proj_yrs_devs) × n_bins × n_sexes × n_fish_fleets |
ln_srvsel_devs |
Log-scale survey selectivity deviations by region, year, bin, sex, and fleet | n_regions × (n_years + n_proj_yrs_devs) × n_bins × n_sexes × n_srv_fleets |
Likelihoods
Age-disaggregated observation likelihoods
Reported whenever the model is built, and zero everywhere unless that data source is fit. Each has an age dimension the aggregated counterparts do not, and is summed over ages within a cell before its weight is applied.
| Object | Dimensions |
|---|---|
CatchAA_nLL |
region, year, season, age, sex, fishery fleet |
DiscardAA_nLL |
region, year, season, age, sex, fishery fleet |
SrvIdxAA_nLL |
region, year, season, age, sex, survey fleet |
CatchAA_pop_nLL, DiscardAA_pop_nLL,
SrvIdxAA_pop_nLL
|
as above, with a leading population dimension |
PredCatchAA, PredDiscardAA,
PredSrvIdxAA (and _pop) |
shaped like their likelihood arrays, on the natural scale |
A data source summed over regions or sexes writes into slot one of
those dims and leaves the rest at zero, so the array’s shape does not
tell you what the fleet reported; its Type does. A cell
whose ages are correlated contributes its whole density to the first age
present, so the array still sums correctly but is not readable age by
age.
The parameters behind them are also reported:
ln_sigmaCAA, ln_sigmaDAA, and
ln_sigmaSrvIdxAA, each over age, sex and fleet. The age
shape of catchability is not among them: an index fit age by age has it
in selectivity, through the "nonparfree" form.
Region-aggregated likelihoods combine contributions across all
populations; population-specific likelihoods (_pop_) are
only computed when the corresponding UseCatch_pop,
UseDiscard_pop, UseFishIdx_pop,
UseSrvIdx_pop, or composition _pop use-arrays
contain active observations.
| Name | Description | Dimensions |
|---|---|---|
Catch_nLL |
Negative log-likelihood for region-aggregated catch | n_regions × n_years × n_seas × n_fish_fleets |
Catch_pop_nLL |
Negative log-likelihood for population-specific catch | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
Discard_nLL |
Negative log-likelihood for region-aggregated discards | n_regions × n_years × n_seas × n_fish_fleets |
Discard_pop_nLL |
Negative log-likelihood for population-specific discards | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
FishIdx_nLL |
Negative log-likelihood for region-aggregated fishery index | n_regions × n_years × n_seas × n_fish_fleets |
FishIdx_pop_nLL |
Negative log-likelihood for population-specific fishery index | n_pop × n_regions × n_years × n_seas × n_fish_fleets |
SrvIdx_nLL |
Negative log-likelihood for region-aggregated survey index | n_regions × n_years × n_seas × n_srv_fleets |
SrvIdx_pop_nLL |
Negative log-likelihood for population-specific survey index | n_pop × n_regions × n_years × n_seas × n_srv_fleets |
FishAgeComps_nLL |
Negative log-likelihood for region-aggregated fishery age compositions | n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishAgeComps_pop_nLL |
Negative log-likelihood for population-specific fishery age compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishLenComps_nLL |
Negative log-likelihood for region-aggregated fishery length compositions | n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishLenComps_pop_nLL |
Negative log-likelihood for population-specific fishery length compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishAgeComps_discard_nLL |
Negative log-likelihood for region-aggregated discard age compositions | n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishAgeComps_discard_pop_nLL |
Negative log-likelihood for population-specific discard age compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishLenComps_discard_nLL |
Negative log-likelihood for region-aggregated discard length compositions | n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
FishLenComps_discard_pop_nLL |
Negative log-likelihood for population-specific discard length compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_fish_fleets |
SrvAgeComps_nLL |
Negative log-likelihood for region-aggregated survey age compositions | n_regions × n_years × n_seas × n_sexes × n_srv_fleets |
SrvAgeComps_pop_nLL |
Negative log-likelihood for population-specific survey age compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_srv_fleets |
SrvLenComps_nLL |
Negative log-likelihood for region-aggregated survey length compositions | n_regions × n_years × n_seas × n_sexes × n_srv_fleets |
SrvLenComps_pop_nLL |
Negative log-likelihood for population-specific survey length compositions | n_pop × n_regions × n_years × n_seas × n_sexes × n_srv_fleets |
M_nLL |
Negative log-likelihood (prior) for natural mortality | scalar |
Fmort_nLL |
Negative log-likelihood (penalty) for fishing mortality deviations. Zero wherever the deviation is not estimated | n_regions × n_years × n_seas × n_fish_fleets |
dmr_nLL |
Negative log-likelihood (penalty) for discard mortality rate deviations. Zero wherever the deviation is not estimated | n_regions × n_years × n_seas × n_fish_fleets |
Rec_nLL |
Negative log-likelihood (penalty) for recruitment deviations | n_pop × n_regions × n_est_rec_devs |
Init_Rec_nLL |
Negative log-likelihood (penalty) for initial age deviations. Zero for any sex whose copy of a shared parameter is not the penalized one | n_pop × n_regions × (n_ages - 1) × n_sexes |
Init_Sex_nLL |
Negative log-likelihood (penalty) tying each later sex’s initial age
deviations to the first sex’s (Use_init_sex_pen). Zero for
the first sex and throughout when the tie is off |
n_pop × n_regions × (n_ages - 1) × n_sexes |
Rec_level_nLL |
Negative log-likelihood (penalty) on the log recruitment series
itself (Use_rec_level_pen). Zero in years outside
rec_level_pen_yrs, and throughout when the penalty is
off |
n_pop × n_regions × n_years |
SR_pen_nLL |
Negative log-likelihood (penalty) for the stock-recruit residual,
the log difference between realized recruitment and
SR_pred, under rec_model = "mean_rec" with
sr_penalty. Zero in years outside sr_pen_yrs,
and zero throughout when no such penalty is applied |
n_pop × n_regions × n_years |
conv_fish_tag_nLL |
Negative log-likelihood for conventional tag recaptures | conv_tag_max_liberty × n_seas × n_conv_tag_cohorts × n_regions × n_fish_fleets |
h_nLL |
Negative log-likelihood (prior) for steepness | scalar |
fish_q_nLL |
Negative log-likelihood (prior) for fishery catchability | scalar |
sel_nLL |
Negative log-likelihood (penalty/prior) for selectivity | scalar |
srv_q_nLL |
Negative log-likelihood (prior) for survey catchability | scalar |
Movement_nLL |
Negative log-likelihood (penalty/prior) for movement | scalar |
TagRep_nLL |
Negative log-likelihood (prior) for tag reporting rate | scalar |
rec_prop_nLL |
Negative log-likelihood (prior) for recruitment proportions | scalar |
jnLL |
Joint negative log-likelihood | scalar |
Derived Quantities (also reported via ADREPORT)
| Name | Description | Dimensions |
|---|---|---|
Total_Biom |
Total biomass | n_pop × n_regions × n_years |
SSB |
Spawning stock biomass by population and region | n_pop × n_regions × n_years |
eff_SSB |
Effective spawning stock biomass at each population’s natal region, including stray contributions from other populations | n_pop × n_years |
Dynamic_SSB0 |
Unfished spawning stock biomass by population and region | n_pop × n_regions × n_years |
Aggregated_SSB |
Spawning stock biomass summed across populations and regions | n_years |
Dynamic_Aggregated_SSB0 |
Unfished spawning stock biomass summed across populations and regions | n_years |
Rec |
Recruitment by population and region | n_pop × n_regions × n_years |
Note: All derived quantities are additionally reported in log space
via ADREPORT (e.g., log_SSB,
log_eff_SSB, log_Total_Biom, etc.) to
facilitate delta-method standard error estimation. A small constant
(1e-5) is added before log transformation to avoid numerical issues in
regions or populations with zero biomass.