Flatfish-habitat associations in Alaska nursery grounds: use of continuous video records for multi-scale spatial analysis
Stoner, A.W.; Spencer, M.L.; Ryer, C.H. (2007). Flatfish-habitat associations in Alaska nursery grounds: use of continuous video records for multi-scale spatial analysis. J. Sea Res. 57(2-3): 137-150. https://dx.doi.org/10.1016/j.seares.2006.08.005
In: Journal of Sea Research. Elsevier/Netherlands Institute for Sea Research: Amsterdam; Den Burg. ISSN 1385-1101; e-ISSN 1873-1414, more
Also appears in:Yamashita, Y.; Nash, R.D.M.; van der Veer, H.W. (Ed.) (2007). Proceedings of the Sixth International Symposium on Flatfish Ecology, Part 1, held at Maizuru, Kyoto, Japan from 20-25 October 2005. Journal of Sea Research, 57(Spec. Issue 2-3). Elsevier: Amsterdam. IV, 89-235 pp., more
|  |
| Keywords |
Audiovisual materials > Videotape recordings
Habitat selection
Information systems > GIS
Nursery grounds
Sediments
Lepidopsetta polyxystra Orr & Matarese, 2000 [WoRMS]
INE, USA, Alaska, Kodiak I. [Marine Regions]
Marine/Coastal |
| Author keywords |
habitat complexity; sediment; rock sole; generalized additive models;GIS |
| Authors | | Top |
- Stoner, A.W.
- Spencer, M.L.
- Ryer, C.H.
|
|
|
| Abstract |
Flatfish distributions have traditionally been described in terms of depth, temperature, and sediment characteristics, but other environmental variables may be important depending upon spatial scale. Surveys for age-0 northern rock sole (Lepidopsetta polyxystra) were conducted in five near-shore nursery sites at Kodiak Island, Alaska (USA), using a towed camera sled integrated with navigational data. The continuous record of fish density and habitat features made possible a spatially comprehensive analysis of fish-habitat associations at several spatial scales, ranging from tens of kilometres to less than 1m. A combination of multivariate statistical interpretation and geographic information systems (GIS) revealed that the distribution of juvenile rock sole was associated with environmental variables and spatial scales that are not normally detectable with usual flatfish - and habitat - sampling methods (i.e., trawls and grabs). Generalized additive models (GAM) incorporating habitat variables determined from video provided large improvements over models using only the traditional variables such as depth and sediment type. At the broadest (regional) scale of analysis, combinations of sediment composition, surface bedform, temperature, and density of worm tubes provided the best model for rock sole density. Within-nursery variation in fish density was modelled best with depth, habitat structural complexity created by emergent fauna and macroalgae, and worm tube density. At the microhabitat scale (< 1m), there was little evidence of direct contact between rock sole and structures such as shell or algae. Rather, they were loosely associated on a scale tens of metres. This study showed that spatially comprehensive surveys can be conducted with towed camera systems and without the need for sediment grab samples. This approach yields detailed habitat information for fishes and the opportunity for landscape analysis of spatial patterns that will be important in conserving critical habitats for flatfishes and other fish species. |
|
