Demonstrating conservation impacts in California Marine Protected Areas using large-scale participatory science data.

Measuring human interactions with protected areas is a key need in conservation social science, both to assess the potential for continuing negative impacts on sensitive resources and to support positive relationships with important places. Our paper investigates human activities in California Marine Protected Areas (MPAs) using an emerging approach of large-scale participatory data collection coupled with statistical analysis. Twelve organizations across the state developed, standardized, and implemented a data collection protocol for volunteers to record human activities, called "MPA Watch." From 2012-2020, approximately 1,900 surveyors conducted more than 31,700 surveys, observing more than 1.2 million activities at 104 different sites across California's coast, both inside and outside MPAs. We analyzed these data using generalized linear mixed models (GLMMs) with zero inflation to account for sampling bias associated with volunteer-driven data. We included covariates regarding weather and tides; time of day, week, and year; and beach type and amenities. We found statistically significantly lower consumptive activities inside MPAs with the strictest protections as compared with non-MPAs, and no significant difference between MPAs and non-MPAs for non-consumptive recreational activities. Some consumptive activities significantly declined over time. We also detected expected use patterns (e.g., for all activity categories, weekend counts were significantly higher than weekday counts). In addition, we noted patterns of interest for future study, including much higher incidence of non-consumptive activities than consumptive activities at a statewide level; higher popularity of some activities at certain individual sites; and the beginnings of the effects of the COVID-19 pandemic in 2020. Our analysis demonstrates that consumptive activities are lower inside California MPAs (indicating compliance with MPA rules), and results at particular sites can inform site-specific management strategies. Further, we show the potential of large-scale participatory science, coupled with appropriate statistical modeling, to monitor and inform conservation policy and management for protected areas.