Linking Fine-Scale Groundfish Distributions withLarge-Scale Seafloor Maps: Issues and Challengesof Combining Biological and Geological Data

摘要

Groundfishes are an important fishery resource on the West Coast of the United States, but their population sizes have undergone dramatic declines in recent years. A number of area-based management and assessment strategies have been suggested to help rebuild and monitor these populations. Most groundfish species have strong affinities with specific substratum types, resulting in spatially patchy distributions. Hence, incorporating information on the types and amounts of seafloor substrata present (i.e., habitat availability) into sample design and biomass assessment of groundfish populations could increase the precision and accuracy of fish density and, consequently, population abundance estimates. The success of using habitat availability as a proxy for fish abundance, however, is contingent on the ability to identify those measurable habitat characteristics (e.g., substratum type, depth, relief, etc.) that fish respond to, precisely estimating fish densities within those habitats, and accurately characterizing and delineating these same characteristics across large areas (i.e., seafloor substratum maps). Characterizing seafloor substratum over a large area is not an exact process, but rather, it commonly uses remotely collected information (e.g., acoustic data, sediment samples, and local geology) to infer the seafloor characteristics. As a consequence, combining estimates of fine-scale fish density per unit area of habitat and the amount of each habitat type to generate a population abundance estimate will reflect the combination of the uncertainty and error in both estimates. If sampling uncertainty or error is large for either estimate (error and uncertainty around the largest mean will be the most critical), then the final population abundance estimate might be of little use to managers. We examine a case study in which an in situ groundfish survey, conducted in an area where a detailed seafloor substratum map was available, suggested that mapsven with suboptimal resolution could be used to increase precision in estimates of fish density. In considering the issues and challenges encountered in linking geological and biological data, it is vital to determine the level of resolution required in the seafloor substratum map, which will depend on the degree of habitat specificity to which the organism responds. Further considerations include whether the mapping technology and methodology can achieve this level of resolution and, finally, whether this sampling approach is cost effective.