Updated Age Structure & Calibration

[jcaracappa1]

Summary

Updated age structured functional groups to properly reflect life history and calibrated groups to resemble v2.0.2. Parameters changed include:

neus_groups.csv: NumCohorts and NumAgeClassSize at_biology.prm: rescaled mum, C, & FSPB; age_mat at_harvest.prm: CatchTS_agedistrib neus_init.nc: initial ResN, StructN, and Numbers

Also includes improvements to cloud calibration workflow & various calibration parameter changes. Groups with different age structure: MAK, HER, WHK, BLF, WPF, SUF, WIF, WTF, HAL, PLA, BUT, ANC, GOO, MEN, COD, SHK, OHK, POL, RHK, TYL, RED, OPT, SAL, STB, TAU, WOL, TYF, DOG, SMO, SSH, DSH, BLS, POR, WSK, LSK, SB, REP, RWH, BWH, and TWH

NOTE: MAK is crashing but needs to be addressed in a future update

Individual Changes

Organization

• .gitignore: exempt cloud setup files for easy transfer

Parameter Files

• at_biology.prm -- SB recruit type 1 -- tuned C for : LOB, MAK, HER, WHK, BLF, WPF, SUF, WIF, WTF, BFT, BUT, ANC, GOO, MEN, COD, SHK, OHK, POL, TYL, OPT, SAL, TAU, WOL, YTF, DOG, SMO, SSH, POR, and LSK -- tuned mum for : SCA, MAK, HER, WHK, BLF, WPF, SUF, WIF, WTF, BIL, BUT, ANC, GOO, MEN, COD, SHK, OHK, POL, RHK, TYL, RED, OPT, SAL, TAU, WOL, FDF, YTF, DOG, SMO, SSH, POR, and LSK -- tuned mL and/or mQ for : SB, ZL, COD, SSH, PSH, RWH, and BWH -- changed age_mat for most species -- changed FSPB for most species -- tuned KDENR for : SMO, BLS< POR, WSK, SB, SWH, and TWH -- tuned BH recruitment for : MAK, WHK, BLF, WPF, SUF, WTF, HAL, BFT, BUT, ANC, BPF, MEN, OHK, and RHK
• at_harvest.prm : changed CatchTS_agedistrib for most spp
• at_run.prm : Changed init scalars to match current dev
• neus_init.cn : Changed init ResN, StructN, Num by age

R Scripts

• R/Calibration_Tools/call_mumC_age.R : fix file reference
• R/Calibration_Tools/compare_init_numbers.R: Script to compare initial numbers between two runs to re-parameterize new age classes
• R/Calibration_Tools/compare_weight_age.R : Compare weight at age between survdat reference and model run
• R/Calibration_Tools/edit_param_C_age.R: fixed function to be compatible with +- 10 age classes
• R/Calibration_Tools/edit_param_assim_eff.R : function to edit assimilation efficiency parameters in at_biology.prm
• R/Calibration_Tools/edit_param_catch_age.R : Fixed age class size calculations
• R/Calibration_Tools/edit_param_init_scalar.R : Fixed age class size calcuations
• R/Calibration_Tools/edit_param_invert_c_mum.R : Fixed bug in grep
• R/Calibration_Tools/edit_param_mortality_age.R : Fixed bug in grep
• R/Calibratrion_Tools/edit_param_mum_age.R : now works with +/- 10 age classes
• R/Calibration_Tools/general_calibration.R : script to call on many types of parameter changes
• R/Calibration_Tools/get_age_mat.R : Turned into function
• R/Calibration_Tools/get_init_agedist.R : Turned into function
• R/Calibration_Tools/get_init_spatial_dist.R : Turned into function
• R/Calbriation_Tools/get_param_assim.R : deprecated and absorbed into edit_param_assim_eff.R
• R/Calibration_Tools/make_init_nofill.R : creates "nofill" version of initial conditions for use in atlantistools
• R/Calibration_Tools/make_setup_file_cloud.R : creates a template setup file for general_calibration.R
• R/Calibration_Tools/rescale_age_params.R : changes mum/C parameters between old and new age classes
• R/Calibration_Tools/rescale_init_age_params.R : changes initial conditions between old and new age classes (ResN and StructN)
• R/Calibration_Tools/rescale_init_nums.R : rescales initial abundance based on reference run
• R/Calibration_Tools/subset_setup_cloud.R : rerun a partial cloud setup run
• R/Post_Processing/get_atl_paramfiles.R : fixed bug reading init_nofill.nc
• R/Post_Processing/make_atlantis_diagnostic_figures.R : fixed age-based bugs and naming issues
• R/Post_Processing/make_post_process.R : missing argument
• R/Post_Processing/plot_calibration_set.R : Plots comparisons after general_calibration.R
• data-raw/pull_age.r : Script to pull survdat age-based parameters

data files

• data/survey_lenagewgt.rds : survdat length, weight, and age data
• diagnostics/fishbase_len_wgt_ref.csv : data collected from fishbase for calibrating age classes
• diagnostics/group_matrure_age.csv : pull of maturity-at-age from biology.prm

Other

• diagnostics/cloud_calibration_setup_example.csv : example for using general_calibration.R on cloud
• diagnostics/max_age_comparison.xlsx : worksheet for age class changes

[jcaracappa1]

@RGamble1 @andybeet Please take time to review somewhat carefully. There are a lot of changes made throughout parameters and calibration scripts to accommodate the new age structured groups. Attached is the run comparison between dev. 6536_cloud_new_age_PRModel_Comparison_Biomass.pdf

[RGamble1]

Joe, this was a lot of work. Thanks, and it looks good to me. @andybeet , @jcaracappa1

[andybeet]

looks good. great documentation. thanks for that. i will need to update the postprocessing package to include some of those changes. Maybe this might be a good time for the group to understand how to update/develop R packages. maybe we need to combine packages. maybe we need to thing about adding calibration tools also.