RE: Recruitment Deviation Setup - Public Forums - Stock Synthesis

Hi Marta Cousido,

The most common approach to set these input values is to run the r4ss::SS_plots() function in R (described at https://r4ss.github.io/r4ss/articles/r4ss-intro-vignette.html) which calls on the SS_fitbiasramp() function and provides some estimated values to set these 5 inputs. An example of that output is at the bottom of this page: https://pfmc-assessments.github.io/petrale/r4ss_plots_pre-review/_SS_output_RecDev.html. The "estimated alternative" inputs provided by that function are not always better than what a person can improve the fit over the simple algorithm (as was the case in that example), but are a useful starting place.

Let me know if you have more questions about this.

-Ian

Thanks for your reply Ian. As you point out, the Recdev page of the model output does indeed provide suggestions for the values of these parameters. However, we would like to know more about the rationale behind this in order to solve questions that we have such as the specific ones described in our first post, for example the role of the Early Recruitment Deviation Start Year.

Thank you for your help.

Hi Marta.  Here are some thoughts.  First off, the long development of SS3 emphasized flexibility of options to meet situations as they arose.  The corollary of that is that there are many features for which definitive advice on good practices is not available.  Early recruitment devs is one of them.

The situation in which use of early recruitment devs is most useful is when:

1. long time series of catch and sorter time series of data that are informative about recdevs
2. desire to include uncertainty about early recdevs even if they cannot be estimated individually
3. desire to use the recdev  option 1 for later recdevs so their mean value will closely match conditions of the estimated MSY

That lead to creation of the early recdevs which are a simple dev vector without a sum to zero constraint.  An advantage of use of early recdevs is that they can be turned on in a later phase to capture their variance without slightly extending runtime associated with estimating them in early phase.

Over time, we have learned, but not definitely demonstrated, that the option 1 dev-vector has disadvantages, particularly when doing MCMC, and that the sum-to-zero constraint can produce illogical patterns in the recdevs.

The current status is that I generally recommend using option 2 for all rec_devs and not using early recdevs.

Love to see someone do a good comparison of options 1 and 2 (and 3 and 4).

Rick

0

Following up on Rick's response.

I am currently undertaking two assessments and have been exploring a number of these issues. The species (NZ snapper) is relative long lived (M ~ 0.07). Both stocks have shown a marked increase in productivity over the last 10 years. The stocks have a good time series of age composition data from the mid 1970s onwards. There was a substantial amount of catch from the preceding decades although the magnitude of the catch is not well known.

Previously, we conducted a full catch history model starting from unexploited conditions. However, the change in productivity means that the B0 is no longer considered appropriate for management purposes. Starting the model under exploited conditions (in 1975) means that we no longer need to include the less reliable large catches from the earlier years. 

Previously, we have tended to estimate recruitments with the constrained dev vector (option 1). Some simulation work indicated that this approach had the potential to bias estimates of stock status. I also encountered issues with the stock crashing when running MCMC - when the stock was at very low biomass the constrained rec devs were unable to generate enough recruits (for a relatively small proportion of the MCMC draws, but frustrating). This was resolved when switching to unconstrained devs (option 2). Other comparisons did not result in appreciable differences in the estimated recruitment/biomass from the two options.

Starting the model in 1975, I am estimating an initial equilibrium fishing mortality - informed by the early age composition data. However, those data exhibit considerable variability in recruitment over the 20-30 y age range. So, I have also been estimating early rec devs to better fit those age comps and incorporate the uncertainty in the initial F. If those data aren't very informative then you can get confounding between the initial F and the early rec devs. The attached plot of the rec devs for the two periods (early and main) is attached. The very early (1960-65) recruitments are highly uncertain, while the 1970-74 deviates are quite well informed by the age comp data - there is good progression of some very strong year classes. Estimating the early rec devs allowed a good fit to those data, but didn't greatly change the estimate of initial F or initial biomass.  

I hope these observations are useful - this could serve as a potential case study.

Adam Langley

Thank you both Rick and Adam for your comments and detailed explanations. They are very helpful. On the basis of these ideas we will work on the recdevs of our model.

 Keep in touch,
Marta