Variable F with Fixed Effort

[dorleta]

#---------------------------------------

SUMMARY OF THE PROBLEM & SOLUTION

#---------------------------------------
In DAMARA simulations and afterwards in Iberian Waters we run an scenario with fixedEffort
and we obtained an increasing trend in the fishing mortality of some stocks.
We were expecting a fairly constant F and we thought there was a bug somewhere.

The problem was that the discards and landing weigth was different and that at fixedEffort
was combined with a restrictive TAC. In the simulations due to the TAC restriction the
ratio of landings and discards was changing over time and as the weights were different
the number of fishes died because of fixing was varying.

If the weigths of dicards and landings are very different maybe the conditioning should
be revisited.

#---------------------------------------

EMAIL CHAIN with more information

#---------------------------------------

Yes, this is exactly the reason, I think at the beginning of DAMARA we had a similar problem related with the discard weight but I did not remember about it until discovering that the problem was on discards again.

Dorleta

-----Mensaje original-----
De: Paul Dolder [mailto:[email protected]] Enviado el: miércoles, 28 de septiembre de 2016 7:49
Para: Dorleta Garcia
CC: Coilin Minto
Asunto: Re: Fixed Effort and time varying F?¿

Hi Dorleta,

Thanks - I think I understand what you're saying and it certainly could be the case for our case as well.

If I understand correctly: because the TAC is restrictive on landings, any over-quota catch due to effort being higher than required for the quota is assigned to discards based on X tonnes over-quota. Because generally discards have a higher N to weight ratio (smaller fish etc..), this extra discards translates to disproportionately larger numbers of fish and hence results in higher discard Fishing mortality... Is that correct ?

Paul

On Tue, Sep 27, 2016 at 08:59:50PM +0000, Dorleta Garcia wrote:

Hi,

After some time checking and comparing data I've found the reason for the odd behaviour with fixed effort, at least in our case study, I hope the reason is the same in yours.

In our CS "discards.wt != landings.wt" and although we use fixed effort we also restrict the landings by the TAC. As the TAC is variable, the ratio between landings and discards also varies along time. Furthermore, as weight at age of landings and discards is different the selection pattern also changes along time and hence the fishing mortality also does.

Lots of interactions within the model that are difficult to understand....

Cheers,

Dorleta

-----Mensaje original-----
De: Paul Dolder [mailto:[email protected]] Enviado el: jueves, 22 de
septiembre de 2016 18:48
Para: Dorleta Garcia
CC: Coilin Minto
Asunto: Re: Haddock in DAMARA

Hi,

I guess we can rule out one of my fears - it was a data issue! I'll also try and take another look at it and try and work out what's happening...

Hope the ASC was fun!

Thanks,
Paul

On Thu, Sep 22, 2016 at 01:55:55PM +0000, Dorleta Garcia wrote:

Hi,

Back from the ASC I've started working in the PhD again and realized that in some scenarios I'm running I've the same problem you have for Haddock!! You can see the problem in the file attached, there are 3 scenarios and I use fixed effort in all but the problem only appears in one of the scenarios, strange, and only for two of the stocks HKE and LDB... At the moment I've not clue about what is going on but I hope to find it asap, I suppose that the problem (or the reason) in both cases will be de same. I'll keep you informed.

Cheers,

Dorleta

-----Mensaje original-----
De: Paul Dolder [mailto:[email protected]] Enviado el: miércoles,
07 de septiembre de 2016 16:54
Para: Dorleta Garcia
CC: Coilin Minto
Asunto: Re: Haddock in DAMARA

Hi Dorleta,

Now attached...

Thanks,
Paul

On Wed, Sep 07, 2016 at 07:40:19AM +0000, Dorleta Garcia wrote:

Hi Paul,

You forgot the file, let's see if I can find the problem

Dorleta

-----Mensaje original-----
De: Paul Dolder [mailto:[email protected]] Enviado el: martes,
06 de septiembre de 2016 23:24
Para: Coilin Minto
CC: Dorleta Garcia
Asunto: Re: Haddock in DAMARA

Hi Cóilín, Dorleta,

Hope its going well - think you've the better of the weather :)

Here is a zip of a single iteration run to show the issue. When you unpack it, under the 'run' folder there is a copy where the problem of increasing F under 'previous' effort (with the LO settings) and 'fixedEffort' (without LO) shows a divergence from the expected broadly stable F (see the plot under the plots folder).

You should just be able to run the two scenarios by running the '08_Scenarios_loop.R' code in the 'code' folder. Though if this causes problems let me know as I haven't run it in a while...

I'm really at a loss as to what is causing this, but happy to do some digging if you can point me in the right direction...

Paul

On Tue, Sep 06, 2016 at 03:36:32PM +0100, Coilin Minto wrote:

Hi Paul,

Greetings from Barcelona. Hope you are enjoying Seattle.

Dorleta and I were discussing the case of the haddock F's in the
baseline run in DAMARA. Would it be possible to send Dorleta the
RData files or a link to the scenario that isn't working?

Thanks!

Cóilín

[marchtaylor]

Might I ask what is the status with this issue? It seems to me from the discussion, that this is expected behavior given the explanation that Paul proposes. I have a similar conditioning situation for many stocks where weights of landings and discards are different.

[dorleta]

Hi Marc, You are right it is an expected behavior not a coding bug. We got surprised by the results initially but then when we analyzed the results in deep we realized it was produced by the difference in the weight of land and disc and the time dependent landing and discards proportion. Cheers, Dorleta De: marchtaylor [mailto:[email protected]] Enviado el: jueves, 09 de febrero de 2017 5:17 Para: flr/FLBEIA CC: Dorleta Garcia; Author Asunto: Re: [flr/FLBEIA] Variable F with Fixed Effort (#11) Might I ask what is the status with this issue? It seems to me from the discussion, that this is expected behavior given the explanation that Paul proposes. I have a similar conditioning situation for many stocks where weights of landings and discards are different. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#11 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AEz9Mtd3m-AfR1kZN1L9yclCMXht1RjIks5rapNBgaJpZM4KupjK>.

[marchtaylor]

Hi Dorleta,

That's good to hear. Thanks for the update.

-Marc