Some things to look for in SS and AGEPRO

[efletcherPIFSC]

From @marcnadon

This online document is meant for us to remember some of the pitfalls we fell into by mis-specifying Stock Synthesis or Agepro during the latest assessments. Please feel free to add to and share this document as you see fit.

Stock Synthesis

1. SS control file: make sure fecundity is set to option 1 and a and b parameters are 1 and 0. This links fecundity to female body weight. Option 2 looks like it is doing this (Egg=a*L^b), but the a-b parameters here refer to extra fecundity parameters and not the a-b parameters used in the Length-Weight relationship.

2. SS data file: If running the population model assuming 2 genders with no biological differences (single growth curve, max age, etc.), we need to enter “-1” in the gender option, not “1”. If you enter “1”, the stock-recruitment curve will be in terms of total SSB, not female-only SSB, which is how SSB is typically defined.

3. For a 2-sex model, you need to you need to change Nsexes = 2 and include female and male parameter lines for M, growth, and LW in SS control file. If you want to assume the same growth parameters for males as for females (Lmin, Lmax, k) set INIT to 0 for male parameters. In the SS data file, composition data needs to include female and male data. If you do not have sex-specific comp data, duplicate the data you have (cbind(lengthcomp, lengthcomp[,lbin1:lbinmax])) and set sex = 0.

4. SS control file: to include catch uncertainty and have SS fit to catch data, need to set F method = 2 (instantaneous). To help the model converge, set the F estimation phase = 4 (i.e. the model will use the F=3 hybrid method to estimate F in early phase and adjust later).

5. SS control file: to include hermaphroditism, set hermaphroditism_option = 1 for female to male transition or -1 for male to female transition. For hermaphroditism_season specify after which season the transition occurs (-1 after every season, positive integer for a specific season, and can use a fractional value (1.2) to specify the season and first age of transition, otherwise age is assumed to be 1). For hermaphroditism_malesinssb specify how to include males in SSB (0.5 includes them using sex ratio, 0 no males, 0<n<1 fraction of males to include in SSB, 1 adds males to females). Also, between male growth parameters and recruitment distribution parameters include 3 hermaphrodite-specific parameters, 1. inflection point (in age) when fish begin to change, standard deviation in age, and asymptote of the max proportion that will transition to the other sex.

6. SS sends warning about low gradient on Fmsy: Open the Forecast-report.sso file and look at the “find_fmsy_to_maximize_catch” table. Look at the “SSB/SSB0” column and the “gradient” column. You can see where the search was heading from a starting “SSB/SSB0” value and where the search ended, at a high gradient. Adjust the “SSB/SS0” starting value in the forecast file (“biomass target”) based on where the search for Fmsy ended in the “forecast-report.sso” file.

Tips:

• Maturity slope = -2.9444 / (L95-L50)

AGEPRO

7. AGEPRO defines SSB as total mature biomass. Not just mature female biomass, as SS is typically parameterized to use. For the Agepro stock-recruitment curve to match the one in SS, we need to double the Beta parameter of the stock recruitment curve used in Stock Synthesis and double the SSBmsst limit in AGEPRO.

8. Make sure that the SSB in the final year of the SS model is close to the median SSB in the first projected years coming out of AGEPRO. This is a sure way to figure out if there is some parametrization mismatch between the two.