| one publication added to basket [33852] | Fish and macro-crustacean response surfaces to environmental gradients in the Westerschelde estuary
Hostens, K. (2003). Fish and macro-crustacean response surfaces to environmental gradients in the Westerschelde estuary, in: Hostens, K. The demersal fish and macro-invertebrate assemblages of the Westerschelde and Oosterschelde estuaries (Southern Bight of the North Sea) = De demersale vis- en macro-invertebraten gemeenschappen van de Westerschelde en Oosterschelde estuaria (Zuidelijke Bocht van de Noordzee). pp. 105-116
In: Hostens, K. (2003). The demersal fish and macro-invertebrate assemblages of the Westerschelde and Oosterschelde estuaries (Southern Bight of the North Sea) = De demersale vis- en macro-invertebraten gemeenschappen van de Westerschelde en Oosterschelde estuaria (Zuidelijke Bocht van de Noordzee). PhD Thesis. Universiteit Gent. Faculteit Wetenschappen: Gent. XVI, 205, 1 cd-rom pp., more
|  |
| Available in | Author |
|
Document type: Dissertation
|
| Keywords |
Crabs
Crabs
Developmental stages > Juveniles
Fauna > Aquatic organisms > Aquatic animals > Fish
Fauna > Aquatic organisms > Aquatic animals > Shellfish
Invertebrates > Arthropods > Crustacea > Malacostraca > Decapoda > Shellfish > Crabs
Invertebrates > Arthropods > Crustacea > Malacostraca > Decapoda > Shellfish > Shrimps
Modelling
Shrimps
Spatial variations
Temporal variations
ANE, Netherlands, Westerschelde [Marine Regions]
Marine/Coastal; Brackish water |
| Abstract |
Juvenile fish and macro-crustaceans of the Westerschelde were intensively sampled with a 3-metre beam trawl in the periods 1988-91 and 1999-2001. Models were developed to predict the occurrence and density of the 15 commonest species in response to a limited set of environmental variables. Single logistic regressions yielded good descriptions of the occurrence of any one species along one environmental gradient. This was related to the maximum likelihood of presence in the field, although several species showed a broad tolerance to- wards one or more of the 4 environmental variables used. The response curves should only be interpreted as actual distribution patterns of juvenile fish and macro-crustaceans as a function of the four environmental variables. Multiple logistic regressions and normal regressions gave insight into the relative importance of each environmental variable for every single species. AII response surfaces, either based on presence/ absence or on density data, were highly significant when combining data on temperature, salinity, turbidity, dissolved oxygen concentration and for their quadratic effects. The addition of other, extrapolated variables (current velocity, mysid prey density, chlorophyll a or suspended particulate matter) did not improve the predictions. For most species the prediction of presence/ absence was relatively successful (60-9(J % correct I y predicted). Sensitivity (% present predicted as present) and specificity (% absent predicted as absent) were equally high in most models, and validation proved the models to be accurate and robust. The models that predicted density patterns could only explain between 20-55% of the variance. Best 'density' models were built for those species that were present in the estuary during a longer period with only one clear density peak, i.e. Limanda limanda, Pomatoschistus microps, Carcinus maenas, Liocarcinus holsatus, Platichthys jlesus, Sprattus sprattus and Pomatoschistus minutus. The least models concerned species belonging to the ecological guild of 'marine juveniles ' (e. g. Trisopterus luscus, Merlangius merlangus, Clupea harengus, Solea solea and Pleuronectes platessa). Also, the 'density' models for Crangon crangon, Syngnathus rostellatus and Pomatoschistus lozanoi were less successful. Temperature (and dissolved oxygen concentration) mainly reflected seasonal effects, while salinity and turbidity reflected spatial effects. Still, it is argued that the interaction between several environmental variables (e.g. temperature and salinity) was even more important in predicting species occurrence and density. |
|
