White et al. (2020, Conservation Letters: article press release) used a size-structured integral projection models to estimate harvest rates inside and outside of MPAs in California’s northern Channel Islands. The models were fit to length-based survey data collected by PISCO divers after the MPAs were established in 2003. The estimated harvest rates were effectively zero inside the MPAs (indicating no poaching) but much higher at some sites outside the MPAs, suggesting that perhaps fishing effort has been displaced from the MPAs to those sites.

This is the first example of using a population model to test for evidence of poaching in the population size structure itself, rather than using indirect assessments such as observations of fishing boats inside MPAs or game warden enforcement records.

These results, and the general approach, should be useful as California develops a state-wide assessment of its MPA network in 2022.