| Summary: | Thesis (Master's)--University of Washington, 2012 As place-based frameworks are increasingly applied to fisheries management, researchers are faced with the need to define ecosystem units and collect data in a manner that will meet this management need. Place-based management approaches, such as ecosystem-based management and marine spatial planning, are explicitly tied to spatial considerations and rely on an understanding of the spatial arrangement of elements within the system and how elements interact. I explore the utility of spatial pattern analysis and spatial statistics for understanding the distribution of marine resources. I focus on the upper shelf and outer slope region of the Eastern Bering Sea continental margin. This region is highly heterogeneous in terms of geomorphology of the shelf break and system of canyons that incise the shelf. I used slope and aspect to divide this region into shelf and slope geomorphic zones and facets and divided this region into a northern and southern portion to explore spatial pattern at a range of scales. I quantified structural heterogeneity with a shape index and surface roughness metric. I used groundfish catch per unit effort (CPUE) trawl survey data to describe and quantify the degree of spatial autocorrelation in this region. I found spatial autocorrelation of groundfish CPUE in the upper shelf and outer slope region of the Eastern Bering Sea continental margin. The clustering pattern was dominated by low-value clustering at the global level. At the local level, low-value clusters were confined to the southern portion of the study area and high- value clusters varied spatially and temporally. Outliers were most commonly found in close proximity to the shelf-slope break. This explicitly spatial method demonstrates the feasibility of this approach in fisheries management.
|
|---|