Spillover effects in `marlin`

[echelleburns]

Hi Dan,

I noticed in a couple of your vignettes you refer to spillover. I was wondering if you could briefly describe how marlin defines spillover and which parameters are used to model spillover. I was also wondering whether it's possible/makes sense to ever "turn off" spillover effects.

[DanOvando]

The most precise measure of spillover would be the difference in various metrics (e.g. biomass) in locations outside the actual borders of the MPA relative to that same metric in those locations in an identical simulation without the MPA.

Mechanically, the way that the MPA affects areas outside the borders in the model is through post-recruit movement and/or recruitment. In terms of post-recruit movement, the model has proportions of animals move both out from and in to MPAs based on the movement models. Setting adult_diffusion = 0 would stop this completely, i.e. post-recruits are completely sedentary. Biomass inside the MPA can also affect biomass outside the MPA through recruitment dynamics. There are a few ways to turn this off, either by setting density_dependence to local_habitat (meaning each patch has completely seperate density dependent processes) or one of the pre or post dispersal and setting recruit_diffusion = 0.

If part of the question the model is being used for is "what are the impacts outside", certainly worth playing with different levels of connectivity to see how sensitive results are to say movement rates.