search

sizespectrum / mizer

Multi-species size-based ecological modelling in R
https://sizespectrum.org/mizer
37 stars 42 forks source link

Integration with FLR project? #165

Open gustavdelius opened 4 years ago

gustavdelius commented 4 years ago

Do you have any view about what kind of integration with FLR would be useful? In @iagomosqueira FLR has a very responsive maintainer. I imagine that some integration would make it easier to integrate mizer with some stock assessment, bioeconomic modelling and fisheries management tools. I am sure @drfinlayscott will have considered this in the past.

Most of the mizer output arrays could easily be wrapped up as FLQuant objects. However I am not yet familiar enough with FLR to know whether there are FLR packages that will be able to consume those size-resolved outputs. But if there aren't, they might begin to exist in the future.

FLR objects might be interesting as mizer inputs when people set up mizer models and want to use stock assessment outputs for that purpose.

I will update this issue as I learn more about the FLR universe. But in the meantime, please let me already know your ideas.

astaaudzi commented 4 years ago

This is a great idea @Gustav Deliusmailto:gustav.delius@gmail.com. I think one output would be to export length frequencies (with flexible binning categories perhaps) so these could be used to compare with expectations in length based assessments. If we knew how these length frequencies should look like we could assess fishing mortality and stock status. mizer could help generate these frequencies under different scenarios and assumptions. Some of this has been explored by Ken and others in Kokkalis et al papers Another of course is natural mortality. Not sure if any of FLR methods use size based nat mortality but if data was available they could perhaps use it

From: Gustav W Delius notifications@github.com Sent: Friday, June 12, 2020 9:26 pm To: sizespectrum/mizer Cc: Subscribed Subject: [sizespectrum/mizer] Integration with FLR project? (#165)

Do you have any view about what kind of integration with FLRhttps://flr-project.org/ would be useful? In @iagomosqueirahttps://github.com/iagomosqueira FLR has a very responsive maintainer. I imagine that some integration would make it easier to integrate mizer with some stock assessment, bioeconomic modelling and fisheries management tools. I am sure @drfinlayscotthttps://github.com/drfinlayscott will have considered this in the past.

Most of the mizer output arrays could easily be wrapped up as FLQuant objects. However I am not yet familiar enough with FLR to know whether there are FLR packages that will be able to consume those size-resolved outputs. But if there aren't, they might begin to exist in the future.

FLR objects might be interesting as mizer inputs when people set up mizer models and want to use stock assessment outputs for that purpose.

I will update this issue as I learn more about the FLR universe. But in the meantime, please let me already know your ideas.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/sizespectrum/mizer/issues/165, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AHXRNUB7U2E2YLRIO6IY6T3RWIGFNANCNFSM4N4HZCWA.

University of Tasmania Electronic Communications Policy (December, 2014). This email is confidential, and is for the intended recipient only. Access, disclosure, copying, distribution, or reliance on any of it by anyone outside the intended recipient organisation is prohibited and may be a criminal offence. Please delete if obtained in error and email confirmation to the sender. The views expressed in this email are not necessarily the views of the University of Tasmania, unless clearly intended otherwise.

Kenhasteandersen commented 4 years ago

Hi Gustav,

It sound like a good idea to integrate with FLR. I have no experience with FLR or about how much it is actually used, so unfortunately I cannot give any useful advice in that direction.

I’ve CCed Nis who might know more about FLR.

Ken

Ken H. Andersen, http://ken.haste.dkhttp://ken.haste.dk/, twitter: @69kno Professor in theoretical marine ecology, head of section and phd school, and deputy director of Centre for Ocean Life http://www.oceanlifecentre.dkhttp://www.oceanlifecentre.dk/

Try my Corona Quarantine Simulator: http://oceanlife.dtuaqua.dk/Corona/

New book: Fish Ecology, Evolution, and Exploitation - A New Theoretical Synthesis https://press.princeton.edu/titles/13516.html

[cid:image001.jpg@01D09493.A377DFB0]

On 12 Jun 2020, at 13.26, Gustav W Delius notifications@github.com<mailto:notifications@github.com> wrote:

Do you have any view about what kind of integration with FLRhttps://flr-project.org/ would be useful? In @iagomosqueirahttps://github.com/iagomosqueira FLR has a very responsive maintainer. I imagine that some integration would make it easier to integrate mizer with some stock assessment, bioeconomic modelling and fisheries management tools. I am sure @drfinlayscotthttps://github.com/drfinlayscott will have considered this in the past.

Most of the mizer output arrays could easily be wrapped up as FLQuant objects. However I am not yet familiar enough with FLR to know whether there are FLR packages that will be able to consume those size-resolved outputs. But if there aren't, they might begin to exist in the future.

FLR objects might be interesting as mizer inputs when people set up mizer models and want to use stock assessment outputs for that purpose.

I will update this issue as I learn more about the FLR universe. But in the meantime, please let me already know your ideas.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/sizespectrum/mizer/issues/165, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ADFHLAOSOIWIM33AEIALLLDRWIGFLANCNFSM4N4HZCWA.

iagomosqueira commented 4 years ago

FLCore classes can handle length-based data or outputs, but the number of methods employing length data is limited. But as you say more are now available, and some of them they could be wrapped for FLR easily.

When it comes to simulations and MSE, the population and fisheries projections are age-based, using the FLasher package, which can handle projections involving multiple fleets and stocks.

There are some examples of using length-based indicators, by generating samples of length distributions from abundances at age and an ALK, for example.

Maybe it would be useful for you to identify one initial application and we can look into how it could be implemented.

juliablanchard commented 4 years ago

Really nice to see the idea of linking with FLR resurfacing again (hi @iagomosqueira!). We had originally planned to use North Sea model as an operating model of the MSE process as part of EU-Image project. @nielshintzen set up an observation model (virtual survey) to generate catches from the operating model, which were then used to calculate indicators. @drfinlayscott and I ended up using process error on recruitment instead in that work, but the idea was eventually to come back to FLR ( with harvest control rules as well being a possibility ). I agree length based indicators, or simple abundance index would be a good starting point!

From: Iago Mosqueira notifications@github.com Sent: Monday, June 15, 2020 5:10:09 PM To: sizespectrum/mizer mizer@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: Re: [sizespectrum/mizer] Integration with FLR project? (#165)

FLCore classes can handle length-based data or outputs, but the number of methods employing length data is limited. But as you say more are now available, and some of them they could be wrapped for FLR easily.

When it comes to simulations and MSE, the population and fisheries projections are age-based, using the FLasherhttps://github.com/flr/FLasher/ package, which can handle projections involving multiple fleets and stocks.

There are some examples of using length-based indicators, by generating samples of length distributions from abundances at age and an ALK, for example.

Maybe it would be useful for you to identify one initial application and we can look into how it could be implemented.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/sizespectrum/mizer/issues/165#issuecomment-643946507, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABGD7ORLQMR3IN3QPEDPM63RWXCNDANCNFSM4N4HZCWA.

University of Tasmania Electronic Communications Policy (December, 2014). This email is confidential, and is for the intended recipient only. Access, disclosure, copying, distribution, or reliance on any of it by anyone outside the intended recipient organisation is prohibited and may be a criminal offence. Please delete if obtained in error and email confirmation to the sender. The views expressed in this email are not necessarily the views of the University of Tasmania, unless clearly intended otherwise.