Open AccessDoes Pond Water Reflectance Influence Double‐crested Cormorant Selection of Aquaculture Ponds?
Author(s) -
Radomski Andrew A.,
Zimba Paul V.
Publication year - 2010
Publication title -
journal of the world aquaculture society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.655
H-Index - 60
eISSN - 1749-7345
pISSN - 0893-8849
DOI - 10.1111/j.1749-7345.2010.00384.x
Subject(s) - biology , aquaculture , cormorant , reflectivity , fishery , selection (genetic algorithm) , fish pond , ecology , fish <actinopterygii> , predation , physics , optics , artificial intelligence , computer science
Double‐crested cormorants, Phalacrocorax auritus , are a frequent and major avian predator on channel catfish, Ictalurus punctatus , and other aquaculture species throughout the southeastern USA. Although cormorant movements and occurrence within the aquaculture production region are understood, no studies have focused on the mechanisms utilized by these avian predators to select foraging sites within these high‐density aquaculture systems. The objective of this study was to determine if cormorant abundance is associated with specific (ultraviolet to visible) light reflectance spectra within commercial catfish ponds. Three commercial farms were used in this study; each farm had four ponds classified as “preferred” and another four ponds classified as “avoided” based on observing cormorant numbers for 3 mo prior to reflectance measures (>100 h observation for each pond). Reflectance and pond turbidity were measured weekly for 7 wk during the period of maximal over‐wintering cormorant numbers. Water samples were collected at the surface (1 cm depth) and secchi depth (10–41 cm) at two locations (upwind and downwind sides) within each pond to determine spectral reflectance and turbidity. Average pond reflectance values were not significantly different between ponds preferred and avoided by cormorants; however, upwind reflectance significantly differed from downwind reflectance between cormorant use groups. Using cross‐validation discriminant analysis, the accuracy of separating upwind ponds preferred and avoided by cormorants was over 62%. Repeated measures ANOVA identified the reflectance at 370, 380, 460, 500, 510, and 620 nm as significantly different between ponds preferred and avoided by cormorants. The lower wavelengths (380–410 nm) correspond to known UV‐A sensing compounds in cormorants and the other wavelengths share characteristics associated with algal biomass. Cormorants utilize ponds that have lower algal biomass when compared with avoided ponds.