| dc.description.abstract | The degree of relationship between trawl catch and environmental variables was assessed in a bay-estuarine system of Goa, west coast of India using multivariate techniques. The demersal fish assemblage was dominated by the families, Leiognathidae, Sciaenidae, Clupeidae, Cynoglossidae and Stromateidae and were considered to be typical for the Indo-Pacific. Patterns in community-structure were analyzed by principal component analysis (PCA) which identified five major species-groups that explained 66% of the variation in absolute fish biomass. PCA was also performed using 13 environmental variables to reduce data set variability to four components which accounted for 75% of environmental variation. The model that best explained variation in absolute biomass based on maximum R u2 (adjusted) and minimum Mallows' Cdp statistic was 71% and included six variables such as dissolved oxygen, macrobenthic density, sediment pH, photosynthetic pigment (Chl a), particulate organic carbon (POC) and seston. The regression coefficients were significant (P is less than 0.05) and small values of Cdp indicated the preciseness of the developed model. Path analysis was used to construct a hypothetical causal path diagram to depict the interaction between fish biomass and environmental variables. The study demonstrated the most important variables with regard to environmental-biotic interactions, although the measured variables did not account for all the variation in trawl catches. Further studies elucidating ecologically meaningful relationships should be helpful in bay-estuarine fisheries management. | |
