Drivers and rates of stock assessments in the United States

Fisheries management is most effective when based on scientific estimates of sustainable fishing rates. While some simple approaches allow estimation of harvest limits, more data-intensive stock assessments are generally required to evaluate the stock's biomass and fishing rates relative to sustainable levels. Here we evaluate how stock characteristics relate to the rate of new assessments in the United States. Using a statistical model based on time-to-event analysis and 569 coastal marine fish and invertebrate stocks landed in commercial fisheries, we quantify the impact of region, habitat, life-history, and economic factors on the annual probability of being assessed. Although the majority of landings come from assessed stocks in all regions, less than half of the regionally-landed species currently have been assessed. As expected, our time-to-event model identified landed tonnage and ex-vessel price as the dominant factors determining increased rates of new assessments. However, we also found that after controlling for landings and price, there has been a consistent bias towards assessing larger-bodied species. A number of vulnerable groups such as rockfishes (Scorpaeniformes) and groundsharks (Carcharhiniformes) have a relatively high annual probability of being assessed after controlling for their relatively small tonnage and low price. Due to relatively low landed tonnage and price of species that are currently unassessed, our model suggests that the number of assessed stocks will increase more slowly in future decades.