Projects

Towards an estimate of the absolute abundance of red snapper in the U.S. Gulf of Mexico

End Date: 12/5/16

Objectives

The primary objective of this study is to design a multidisciplinary survey to estimate the abundance of age-2+ red snapper in the U.S waters of the northern Gulf of Mexico (GOM). A secondary objective will be to estimate red snapper growth, mortality, site fidelity, and population connectivity rates. Ultimately, population parameters estimated during the implementation phase of the design will either challenge assumptions of rates utilized within the red snapper stock assessment model (e.g., natural mortality), or will be compared to estimates computed with the model (e.g., population abundance).

Methodology

Additional PIs: William Patterson, University of South Alabama; Gregory Stunz, Texas A&M University at Corpus Christi and David Wells, Texas A&M University at Galveston

Several different approaches will be pursued concurrently to provide estimates of red snapper population size and dynamics in the GOM. First, we have already begun a comprehensive review of habitat data in the GOM. This is key to study success given the diversity of natural and artificial reef habitats that exist on the nearly 1 million km2 GOM shelf. This includes an estimated ~27 thousand km2 of rock dominant natural reef substrate, as well as myriad manmade reef structures on the nGOM, such as oil and gas platforms (12,000 and ~20 km2), state permitted artificial reefs (0.13 km2), and ship-wrecks and obstructions (0.74 km2). Characterization of benthic substrates and habitat types will be key to stratifying survey designs briefly described below.

Metholologies to estimate red snapper abundance and population dynamics fall into two broad categories: tagging and passive technologies.  Tagging approaches will involve both conventional and genetic tagging, while passive technologies will involve remotely operated vehicles, camera rigs, and acoustics.  In the main body of the proposal we outline the major assumptions of each of these methods, the ways in which these assumptions will be tested, challenged, or estimated, and the applicability of each method to address different aspects of study objectives.  A central goal of that aspect of the study will be to characterize the accuracy and precision of the various methods for estimating population size. Specifically, we will examine the likelihood that the various methods can be employed to produce population abundance estimates within level of precision prescribed in the RFP (CV = 0.3). The power of the various approaches, as well as the likely uncertainty in estimates they produce, will be rigorously tested via simulation analysis.  A simulator has already been programmed in R and preliminary evaluations of method-specific sample sizes that would be required to estimate red snapper population size are reported in the project description.

Key aspects of the efficacy of our approach is recognition of, and experience sampling on, the high number of habitat types that exist on the nearly 1 million km2 GOM continental shelf. As such, we chose methodological apporaches will be evaluated beyond just sample sizes and likely precision estimates. Our group involves expert researchers who routinely employ the different methods under consideration. Their expertise will be tapped to estimate sampling costs and logistics, which is essential information for determining the effectiveness of a given approach. Lastly, most of our team is currently involved in cooperative research with participants in the commercial or recreational red snapper fishery and fisherman expertise will aslo be drawn upon to evaluate the feasibility of cooperative approaches to implement different aspects of the proposed study design.

The approaches to be evaluated for potential inclusion in the implementation phase of the study (RFP II) are generalized such that individual working group members can implement them in a way that is best suited to local conditions, as well as scalable based upon the funding available. Aspects of the theory and methodological constructs of individual methods and designs, that are underpinned with a great deal of scientific literature on surveys of fishes in the sea, are described throughout the text. Lastly, it should be stressed that we chose methods and survey designs that can be combined into a toolbox that is both flexible and capable of making localized estimates of abundance that can be extrapolated to reach the goal of an absolute GOM-wide estimate.

Rationale

Marine fisheries are exceedingly important to coastal economies of the northern Gulf of Mexico (GOM). On the shelf, no group is more economically important than reef fishes, despite the fact that several marquee species, including red snapper, are estimated to be overfished. Increasingly restrictive fishery regulations, following passage of the Magnuson-Stevens Reauthorization Act of 2006, have had positive impacts on red snapper recovery, but shortened recreational fishing seasons have also brought angst and mistrust, particularly among recreational fishery lobby groups  Among the issues routinely cited by these groups is the preception that artificial reefs have greatly increased the productivity of red snapper in the GOM yet that is not accounted for in stock assessment models, and the perception that red snapper population size is greater than estimates derived from those same models. Part of the persistence of these perceptions owes to the failure of  scientists to adequately convey to the fishing public that the catch at age matrix routinely has the greateset influence on estimates of stock biomass and productivity in statistical catch at age assessment models. Therefore, if the majority of the catch in a given region is taken at artificial reefs, then clearly the influence of those reefs is captured in assessment models.

Recent funding made available by Congress to estimate red snapper population size, as well as population dynamics rates, provides a unique opportunity to conduct GOM-wide sampling to challenge data input and parameter estimates of the GOM red snapper stock assessment model. This is an awesome challenge given the GOM's nearly 1 million km2 shelf and the diversity of natural and artificail reefs found there. Therefore, evaluation of the scale and methodological composition of sampling programs required to estimate absolute red snapper.  Despite the many challenges to addressing the objectives of this study, we feel the sampling approaches and methods proposed herein will make this daunting challenge tractable. In turn, this work will constitute an precedence opportunity to challenge input assumptions and outputted estimates, including estimated abundance, of the GOM red snapper stock assessment model.

Results

Experimental design final report