This forum is shutting down! Please post new discussions at community.naturalcapitalproject.org

Invest Fisheries model

Mery_khMery_kh Member
edited December 2016 in Marine Models
Hi NatCap,

I'm trying to apply the "InVEst fisheries" model for the Mediterranean hake fishery, whose objective is to test different management scenarios, in particular the closure of an area for fishing. I have some doubts about the correct design of the data required for the model (inputs) and I hope to find answers to my questions in order to carry out the right execution of the program.

1) First, I try to test the model without integrating aspects of migration and habitat dependencies. I will assume fixed recruitment (see no relationship between stock and recruits based on the historical series) and conducted a cohort analysis (via VIT) to identify the initial recruitment number, will this number be appropriate ?

2) For the exploitation  fraction, I considered the fishing effort applied at the level of each fishing zone to specify it. My sub-regions are 28 different fishing zones where the activity takes place. The zone subject to the greatest fishing effort is that to which I have assigned a value of exploitation fraction 1 and on its basis I have concluded the exploitation fraction of the other regions.

3) Finally, my big doubt is about the dispersal of larvae in the model. I tried to establish it on the basis of the density of the individuals of age 0. Having some data on the abundance of the population, I performed an interpolation (using Qgis) to have an average density of each age (0 to 5+) in each sub-region. For the density result of age 0, considering the dispersion in each subregion as fraccion (density subregion x / total density of age 0 in the 28 subregions) will it be a good method for estimating the dispersion of larvae requiered ?


Thank you so much and sorry to have been a bit long. 


Post edited by Mery_kh on

Comments

  • Hi,

    We are corresponding over email.  I have followed up with the fisheries biologists who developed this model to get a more detailed answer to your questions.  I will email you and post the reply here as well when I have it.  Thanks for your patience!  Jess
  • Hi Meryeme, 

    I just send this to you over email, but will post on the forum as well.

    1) First, I try to test the model without integrating aspects of migration and habitat dependencies. I will assume fixed recruitment (see no relationship between stock and recruits based on the historical series) and conducted a cohort analysis (via VIT) to identify the initial recruitment number, will this number be appropriate ?

    This sounds right.
     2) For the exploitation  fraction, I considered the fishing effort applied at the level of each fishing zone to specify it. My sub-regions are 28 different fishing zones where the activity takes place. The zone subject to the greatest fishing effort is that to which I have assigned a value of exploitation fraction 1 and on its basis I have concluded the exploitation fraction of the other regions.
    This makes sense to me if you're assuming that all fishing zones have the same fish populations and you're trying to utilize catch information to back calculate the exploitation fraction.  Keep in mind that the exploitation fraction is the proportion of the population vulnerable to harvest that is actually harvested from 0-1 (0-100%) in each zone, and that each zone is independent.  It might be useful to take a look at how the sample data is formatted - the spiny lobster model from Belize example might be helpful.

    3) Finally, my big doubt is about the dispersal of larvae in the model. I tried to establish it on the basis of the density of the individuals of age 0. Having some data on the abundance of the population, I performed an interpolation (using Qgis) to have an average density of each age (0 to 5+) in each sub-region. For the density result of age 0, considering the dispersion in each subregion as fraccion (density subregion x / total density of age 0 in the 28 subregions) will it be a good method for estimating the dispersion of larvae requiered ?

    This makes sense to me.  If you look at the few sentences on Larval dispersion in the user's guide chapter you'll see we have estimated it a couple different ways (by available habitat in a region and by area of the subregion).  The citation for the Arkema, et al. paper (this is the Belize lobster example) has changed slightly (see below).  A sensitivity analysis here could be useful.

    Embedding ecosystem services in coastal planning leads to better outcomes for people and nature

    Arkema, Katie K ; Verutes, Gregory M ; Wood, Spencer A ; Clarke-Samuels, Chantalle ; Rosado, Samir ; Canto, Maritza ; Rosenthal, Amy ; Ruckelshaus, Mary ; Guannel, Gregory ; Toft, Jodie ; Faries, Joe ; Silver, Jessica M ; Griffin, Robert ; Guerry, Anne D

    Proceedings of the National Academy of Sciences of the United States of America, 16 June 2015, Vol.112(24), pp.7390-5
  • ileeyaileeya Member

    Hi Jess

    I am seeking to use the fisheries InVEST model for my project. Particularly,  to value white and brown shrimp yield, and to explore how changes in mangroves under management scenarios can influence white and brown shrimps yield.  Now I watched your tutorial on the fisheries model,which version of the InVEST model contain four libraries of the single species model for fisheries please? With the 3.6.0 is only the fisheries model in its entirety without the four libraries for single species model. Although, I am able to to access their sample data for the e.g for the white shrimp, spiny lobster etc.. but not the models.

    I want to explore the shrimp model with its sample data to provide me with a good step on preparing my own data and test it as well. 

  • Hi Ileeya,

    There is one version of the InVEST Fisheries model that is flexible to accommodate different species.  You'll see on the user interface that you can use different recruitment parameters, make it an age vs. stage based model, etc.  All the parameters are either input directly into the the UI or they're in the .csv files.  

    We do have sample 'libraries' for white shrimp, blue crab and spiny lobster since we've already built these models.  You can run the white shrimp model that we built for Galveston Bay by loading the .json file (white_shrimp_galveston_bay.invs.json).  To modify the model for another area you would have to create a new AOI, and change the relevant parameters to match your new area.

    Does that make sense?

    J.



  • I would also add, there a second tool - the Fisheries Habitat Scenario Tool, this is what will let you change habitat area and evaluate the effects on survival.

  • ileeyaileeya Member

    Hi Jess, 


    Yes makes perfect sense. Thank you very much. 

    Do you mind if we exchange emails for near future inquires please? Fisheries InVEST appears appropriate and promising for what Iam trying to do but obviously I'll be needing some form of guide through the process.
  • Hi,

    Coincidentally, I am currently building a fisheries model for one of our applications and it will be my first time  parameterizing it from scratch for a new area.  So, your request is strangely well timed!  Yes, feel free to email me at jess.silver@stanford.edu.

    Jess
  • ileeyaileeya Member

    Hi Jess


    Wow. God is great. Thank you Jess.
Sign In or Register to comment.