Premium
Regional variability in Atlantic salmon ( Salmo salar ) riverscapes: a simple landscape ecology model explaining the large variability in size of salmon runs across Gaspé watersheds, Canada
Author(s) -
Kim M.,
Lapointe M.
Publication year - 2011
Publication title -
ecology of freshwater fish
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.667
H-Index - 55
eISSN - 1600-0633
pISSN - 0906-6691
DOI - 10.1111/j.1600-0633.2010.00471.x
Subject(s) - salmo , habitat , ecology , spawn (biology) , watershed , streams , fish migration , peninsula , fishery , spatial ecology , geography , biology , fish <actinopterygii> , computer network , machine learning , computer science
Kim M, Lapointe M. Regional variability in Atlantic salmon ( Salmo salar ) riverscapes: a simple landscape ecology model explaining the large variability in size of salmon runs across Gaspé watersheds, Canada.
Ecology of Freshwater Fish 2011: 20: 144–156. © 2010 John Wiley & Sons A/S Abstract – Atlantic salmon ( Salmo salar ) rivers in the Gaspé Peninsula, Quebec, present a 20 to 1 variability in the average numbers of returning adult salmon per km 2 of watershed area (the ‘specific run size’). These variations are very poorly explained by interbasin differences in total stream length ( R 2 = 0.033, P = 0.533) or in estimates of total area of salmon habitat that fail to take into account complementarity and interconnections across life stage habitats ( R 2 = 0.065, P = 0.448). The relative spatial distribution of three complementary habitat types (adult holding pools, spawning beds, parr habitats) is hypothesised to be an important factor controlling the production of salmon at any given watershed area or given total stream length, and by extension the numbers of adults returning each year to spawn. We developed a simple riverine landscape ecology (or riverscape) model that uses easily accessible topographic map sources to identify optimally productive river segments. These segments were identified based on large‐scale river and valley features and the associated spatial organisation of complementary salmon habitats. We tested the ability of this model to predict salmon run sizes for 14 watersheds in the Gaspé Peninsula. The aggregate length of optimally productive segments, as defined in our model, is a strong predictor of the average size of the annual salmon runs for these watersheds ( R 2 = 0.913, P < 0.0005). Furthermore, specific salmon run sizes were accurately predicted ( R 2 = 0.771, P < 0.0005) after removal of the obvious scale effects of watershed size.