|Abstract or Summary
- All animals that interact with fishing gear are not necessarily captured, and all animals that are captured are not necessarily retained. Fishing practices and gear configuration, management regulations, and markets dictate which animals ultimately are retained or discarded. The impact of a fishery and the efficacy of management regulations can depend on the mortality rate of the animals that interact with the gear or are discarded. The Reflex Action Mortality Predictor (RAMP) is a simple, non-invasive, and inexpensive approach that has been used to evaluate this component of fishing mortality. The RAMP approach relates the degree of reflex impairment in an animal to the probability the animal will die. Since its introduction in 2006, the RAMP approach has been utilized in the U.S. and abroad to evaluate mortality for a variety of species, fishing gear types, and stressors. Although there have been numerous applications of the RAMP approach in mortality estimation studies, there has been limited research to directly evaluate RAMP estimates and some skepticism remains in the fisheries science and management communities about the reliability and accuracy of the approach. The goal of this dissertation was to conduct research to assess RAMP and to synthesize findings from previously completed RAMP studies.
The three research studies described in this dissertation consider: (1) the accuracy of applying an established relationship between reflex impairment and mortality probability to predict overall mortality attributed to novel stressors; (2) the development and utilization of a RAMP relationship to evaluate discard mortality in a fishery with management regulations that mandate discarding of certain categories of animals; and (3) whether the RAMP approach produces accurate estimates of mortality if survival is determined through laboratory captive holding.
The first study estimated a relationship between reflex impairment and mortality probability for Tanner crab (Chionoecetes bairdi) discarded from the groundfish bottom trawl fishery in the Gulf of Alaska. This relationship was then compared to one previously established for Tanner crab in the Bering Sea bottom trawl fishery that encountered the fishing gear, but remained on the seafloor ('unobserved bycatch'). While mortality probabilities were similar between the two studies for crab with no or full reflex impairment, discarded crab with intermediate levels of reflex impairment had lower mortality probabilities than those from the unobserved bycatch study. Results from this study indicate the importance of describing all stressors to which animals are exposed and detailing the study methodology when initially creating a RAMP relationship. Failure to do so may result in inaccurate mortality estimates when the RAMP is applied to animals exposed to stressors not included in the original calibration.
The second study developed a RAMP relationship using laboratory captive holding for Dungeness crab (Cancer magister) discarded in the Oregon commercial and recreational Dungeness fisheries and estimated that the discard mortality rate is lower than previously determined. This supports the goal of the '3-S' management strategy currently employed for these fisheries to protect sub-legal males (Size), females (Sex), and soft-shell (Season) crab by discarding them from the catch. For the commercial ocean Dungeness fishery, the estimated overall discard mortality rates (five days after release) varied by sex and shell-hardness, and reflex impairment was a significant predictor of mortality for both the commercial and recreational fisheries. In addition, results indicated that, when evaluating the role of discard mortality in '3-S' management with respect to fishery impact and sustainability, it is important to look not only at mortality rates, but also at the mortality- and bycatch- per retained ratios, and temporal trends relative to changes in effort, animal condition, and catch composition. This study also highlighted the (i) importance of evaluating the influence of biological, environmental, and fishing variables on mortality, (ii) complications that arise when establishing a RAMP relationship for a low impact fishery, and (iii) limitations of determining mortality through laboratory captive holding.
The third study used mark-recapture methods to evaluate the reliability of results generated using the RAMP relationship established in the second study, which was based on the survival of crab held in captivity in the laboratory. Given the unnatural conditions for determining survival in captivity and the short-term duration of the experiment, mortality probability estimates may be biased. Similarities in patterns of relative survival rates between the studies lend support to the ability of the RAMP relationship to estimate discard mortality rates using captive holding. The laboratory-based RAMP approach was superior in its ability to provide direct estimates of mortality rates, whereas the mark-recapture study was limited to providing relative survival rates between reflex impairment levels that were imprecise due to low numbers of recaptured crab. This study highlighted the complications associated with tagging discarded animals and conducting a RAMP study with a fishery that has highly variable seasonal fishing effort.
A synthesis of the research described in this dissertation and published work by other researchers highlights the limitations of the RAMP approach so that future researchers can avoid pitfalls in its application, and leads to suggestions on how to standardize some of the methodological steps. This analysis aims to increase the reliability of future RAMP studies and their production of high quality estimates of discard mortality rates that promote sustainable fisheries.