Mizer includes a number of calibrated multi-species systems:

  • North Sea (NS_param from Blanchard et al, 2014; included with the mizer package). This calibration is made with a species-specific interaction matrix and gear. Note of caution: this calibration is done with eRepro=1. This means that reproduction is close to the maximum reproduction R_max and thus the model may be less sensitive to fishing than is realistic, and the calibration should be used with caution.
  • North Sea (NorthSea_params), Baltic Sea (Baltic_params), Benguela current (Benguela_params), North East US Continental Shelf (NEUSCS_params), and the Barents Sea (Barents_params) (from Jacobsen et al, 2016). These systems are calibrated with purely size-based predation, i.e., no species-specific interactions. They are set up with three fishing gears targeting small, medium and large species.

The models are available as MizerParams objects. Running, and plotting is as simple as:

## Loading required package: mizer
sim <- project(Barents_params)
plot(sim)

The models are calibrated to specific efforts for each of a number of fishing gear. For example, the Barents Sea model has three gears: small, medium, and large, and is callibrated with the efforts:

getInitialEffort(Barents_params)
##  small medium  large 
##   1.10   0.50   0.75

To see which species are caught by these gears, we look at the gear parameters data frame:

gear_params(Barents_params)
##   species   gear       sel_func catchability sigmoidal_weight sigmoidal_sigma
## 1 Capelin  small sigmoid_weight            1              1.3               3
## 2  Golden medium sigmoid_weight            1            250.0               3
## 3 redfish medium sigmoid_weight            1            420.0               3
## 4 Haddock  large sigmoid_weight            1            800.0               3
## 5 Pollock  large sigmoid_weight            1           1125.0               3
## 6 Halibut  large sigmoid_weight            1           2400.0               3

Simulating with a different effort is done by:

  sim <- project(Barents_params, effort = c(small = 0.5, medium = 0.2, large = 0.5))
plot(sim)

The calibrated efforts are described in the help pages for each calibrated model.