Network-scale spatial and temporal variation in Chinook salmon (Oncorhynchus tshawytscha) redd distributions: patterns inferred from spatially continuous replicate surveys

• Published in: Canadian journal of fisheries and aquatic sciences Vol. 63; no. 2; pp. 285 - 296
• Main Authors: Isaak, Daniel J, Thurow, Russell F
• Format: Journal Article
• Language: English
• Published: Ottawa, Canada NRC Research Press 01.02.2006; National Research Council of Canada; Canadian Science Publishing NRC Research Press
• Subjects: Analysis; animal behavior; Animal populations; Animal, plant and microbial ecology; Applied ecology; Biological and medical sciences; Chinook salmon; climate change; Exploitation and management of natural biological resources (hunting, fishing and exploited populations survey, etc.); Fish; fish culture; Freshwater; Fundamental and applied biological sciences. Psychology; geographical distribution; habitats; nesting sites; Oncorhynchus tshawytscha; Refugia; River basins; Salmon; Spatial analysis (Statistics); spatial variation; Spawning; Statistics; Studies; temporal variation; Variance analysis; Wilderness; Idaho; USA, Idaho; Aquatic environment; Vertebrata; Oncorhynchus tshawytscha; Spatial distribution; Continuous; Pisces; Spatial scale; Spatial variation; Time variation; Breeding site
• ISSN: 1205-7533; 0706-652X; 1205-7533
• DOI: 10.1139/f05-214

Spatially continuous sampling designs, when temporally replicated, provide analytical flexibility and are unmatched in their ability to provide a dynamic system view. We have compiled such a data set by georeferencing the network-scale distribution of Chinook salmon (Oncorhynchus tshawytscha) redds across a large wilderness basin (7330 km2) in central Idaho for 9 years (1995-2003). During this time, the population grew at a rate of 5.3 recruits per spawner, and redd numbers increased from 20 to 2271. As abundances increased, fish expanded into portions of the stream network that had recently been unoccupied. Even at the highest escapements, however, distributions remained clustered, and a limited portion of the network contained the majority of redds. The importance of the highest density spawning areas was greatest when abundances were low, suggesting these areas may serve as refugia during demographic bottlenecks. Analysis of variance indicated that redd numbers were strongly affected by local habitats and broad climatic controls, but also revealed a space-time interaction that suggested temporal instability in spatial patterns. Our results emphasize the importance of maintaining habitats with high densities of individuals, but also suggest that broader views may be needed to accommodate the dynamics of natural salmonid populations.