search

trophia / sna1

A model of the SNA1 fishery
2 stars 2 forks source link

Please provide example of exponential growth parameterisation in Readme file #22

Closed mckenziej closed 7 years ago

nokome commented 7 years ago

See https://github.com/trophia/sna1#growth . BUT there seems to be a problem when using exponential growth - I'm working on that now.

nokome commented 7 years ago

The issue I was seeing occurred when there is a high fishes_linf_sd. For some reason, with the exponential model that is causing a nan in biomass (a few fish with undefined growth ?). I found that reducing fishes_linf_sd to 1 avoids this issue i.e:

    "fishes_growth_model": "e",
    "fishes_k_mean": 0.1,
    "fishes_k_sd": 0.02,
    "fishes_linf_mean": 60,
    "fishes_linf_sd": 1,

I have added some fish growth trajectories to the standard outputs and a script in reports/fishes/growth.R which produces some diagnostic plots e.g for the above pars

image

You'll notice that the model runs slower with exponential model because of the extra calcs involved. We can optimise this.