Abstract

While the southeastern United States and Caribbean generally suffer from a paucity of fisheries data with which to carry out formal stock assessments, management advice can sometimes be obtained with limited data, provided they contain sufficient resolution over space or time. Here we present two case studies, which were originally motivated by the need to generate management advice under severe data limitations. The first example focuses on the impact of the Haitian fishery at Navassa Island, where a SCUBA fish monitoring survey was carried out over the span of a decade (Karnauskas et al. 2011). Species-level and community-level indicators of ecosystem status were estimated using a framework which accounted for the artifacts of sampling. Despite the small sample size and the relatively short time series, significant trends emerged and these were in agreement with anecdotal observations of the level of fishing pressure. The second example relates to the spatial management of two rarely-encountered grouper species in the South Atlantic. While a plethora of data sources are available for this region, these data sets are largely incoherent in both space and time, and detection rates of the study species are extremely low. By incorporating all data into a generalized linear modeling framework, we were able to produce a map of probability of occurrence across the entire South Atlantic (Farmer and Karnauskas 2013). These results are being used to guide the design of marine reserves intended to protect these species. Both of the methodologies presented here could be applied to other fisheries in the region where similar data limitations exist.

PDF Preview