Forums

Back

Growth per cohort

U
Anonymous, modified 4 Years ago.

Growth per cohort

Hello, I am trying to implement a model that changes the growth rate parameter (K) by cohort. The model that I have so far is simple (1 recruitment per year, 1 area, 1 season). It is clear how to do this PER YEAR (linking the growth parameter to a environmental variable). However, it is still unclear to me how I can do this PER COHORT (what do I have to link the environmental variable to?).

 

Thanks.

 

Giancarlo

KD
Kathryn Doering, modified 4 Years ago.

RE: Growth per cohort

Youngling Posts: 57 Join Date: 2/19/19 Recent Posts

Hi Giancarlo,

 

Thanks for the question. One way I think you could try is to create a model with multiple growth patterns (also called growth morphs) or possibly platoons.

Here is the definition of growth morphs from the SS user manual:

Number (N) of growth patterns (GP), also referred to as morphs:
These are collections of fish with unique biological characteristics (growth,
mortality,weight-length, reproduction). The GP x Sex x Settlement Events
constitute unique growth patterns that are tracked in SS. They are assigned
these characteristics at birth and retain them throughout their lifetime. At
recruitment, growth pattern members are distributed across areas (if any)
and they retain their biological characteristics even if they move to another
area in which a different cohort with different biological characteristics
might predominate. For example, one could assign a fast-growing growth
pattern to recruit predominately in a southern areas and a slow-growing
growth pattern to a northern area. The natural mortality and growth
parameters are specified for each growth pattern in the mortality-growth
parameter section in the order of females growth pattern 1 to growth
pattern N followed by males growth pattern 1 to growth pattern N in a
two sex model.

However, I can see how this approach could get too complex if you wanted to do this for each cohort in the model. Perhaps there is a simpler way that I am not aware of?

 

Hope this helps get you started, at least.

Cheers, 

Kathryn

 

Richard Methot, modified 4 Years ago.

RE: Growth per cohort

Youngling Posts: 219 Join Date: 11/24/14 Recent Posts
Dear  Giancarlo,

You will be able to do this with SS.  The feature you want to use is cohort growth deviation, which works as a multiplier on K on a cohort-specific basis.  This parameter is near the bottom of the biology parameter list:
#  Cohort growth dev base
 0.1 10 1 1 1 0 -1 0 0 0 0 0 0 0 # CohortGrowDev

It has a base value of 1.0 which has null effect and it is not to be estimated , so keep its phase negative.

I assume you are using SS 3.30, not 3.24.  What you want to do is to make it time-varying, for example:
  0.1 10 1 1 1 0 -1 0 1 1980 2017 4 0 0 # CohortGrowDev

and then add the needed standard error and autocorrelation parameters.  Note that there is a way to get SS to autogenerate these parameters; check out the manual.

Rick


On Fri, Feb 28, 2020 at 3:49 PM Anonymous <VLab.Notifications@noaa.gov> wrote:

Hello, I am trying to implement a model that changes the growth rate parameter (K) by cohort. The model that I have so far is simple (1 recruitment per year, 1 area, 1 season). It is clear how to do this PER YEAR (linking the growth parameter to a environmental variable). However, it is still unclear to me how I can do this PER COHORT (what do I have to link the environmental variable to?).

 

Thanks.

 

Giancarlo


--
Anonymous Stock Synthesis Virtual Lab Forum https://vlab.noaa.gov/web/stock-synthesis/public-forums/-/message_boards/view_message/9238066 VLab.Notifications@noaa.gov


--
Richard D. Methot Jr. Ph.D.
NOAA Fisheries Senior Scientist for Stock Assessments
Mobile: 301-787-0241
U
Anonymous, modified 4 Years ago.

RE: Growth per cohort

Hi Giancarlo,

For an example of how the cohort growth deviations work in practice, see Whitten et al. 2013 "Accounting for cohort-specific variable growth in fisheries stock assessments: A case study from south-eastern Australia" https://doi.org/10.1016/j.fishres.2012.06.021 which used Stock Synthesis. A report of a more recent assessment of that same stock by Claudio Castillo-Jordán and Geoff Tuck (which continues to use the feature) is available at https://www.afma.gov.au/sites/default/files/blue_grenadier_stock_assessment_2018.pdf. The standard r4ss figures related to the time-varying growth (which could be improved someday) are shown in Figure A.2.

-Ian

U
Anonymous, modified 4 Years ago.

RE: Growth per cohort

Note: some of us didn't get notice of Rick's message on this thread. If you didn't see it, you can view it on the VLab forums website at 

https://vlab.noaa.gov/web/stock-synthesis/public-forums/-/message_boards/message/9263116#_com_liferay_message_boards_web_portlet_MBPortlet_message_9238443