Differences in lobster fishing effort before and after MPA establishment

Abstract

In April 2003, California established a network of no-take marine protected areas (MPAs) around the northern Channel Islands located within the Santa Barbara Bight. Prior to the MPAs enclosing 17% of the islands' lobster fishing grounds, 25 commercial lobster fishermen caught 50% of the regional annual landings from the Channel Islands. To best manage MPAs and affected fisheries we ask a critical question: Where did the fishermen go? Spillover theory emerging from models of MPAs and adjacent fisheries suggests displaced fishermen will concentrate their effort along MPA borders; a phenomenon called “fishing the line”. These models do not consider habitat-specific fishing effort, habitat heterogeneity, nor fixed-gear fisheries such as lobster where traps are set, soaked for 1 to 3 nights, pulled and re-set. With fixed-gear fishing, space is “marked” or occupied, and reduces the possibility of another fisherman to fish that space. Lobster trap fisheries are notoriously territorial as a result. Lobster fisheries therefore stand to experience a skewed impact based on a priori territorial distributions and habitat quality. We use a Geographic Information System (GIS) to map 10 years (5 before reserves and 5 after) of fishery-dependent logbook data assisted with fishery interviews to test if commercial lobster fishermen aggregated fishing effort at MPA borders as an adaptive fishing strategy. We found that fishermen around the Channel Islands MPAs did not concentrate effort at MPA boundaries but instead the proportion of total traps pulled in close proximity (within 1 km of reserve borders) to MPAs declined from 10% to 5%. Chi2 analysis found a significant decrease in the proportion of a season's traps pulled in areas near MPA borders (n = 157,071; p < .001). T test analysis testing the difference in CPUE between areas far from MPAs and areas adjacent to MPA borders showed a significant reduction in the difference between CPUE following MPA designation (n = 50,206; p < .001).