Summary: | The Laguna Madre is a hypersaline lagoon. Despite harsh conditions, the upper Laguna Madre (ULM) is a highly productive ecosystem and a popular sportfishing area, especially for spotted seatrout and red drum. It is also the most important Texas bay for commercial fishing of black drum. TPWD’s Coastal Fisheries division began conducting routine monitoring of coastal fishery resources in 1977 to guide management. The goal of the present study was to improve understanding of spatiotemporal trends in relative abundance of selected fish and macro-invertebrate species in the upper Laguna Madre. I used TPWD’s bag-seine and water-quality data from the years 1985-2004 to examine variation in species’ relative abundances and relationships to several environmental factors. I hypothesized that one or more of these variables, alone or in combination, were related to spatial and temporal trends in community composition. I used detrended correspondence analysis (DCA) to measure species turnover (beta diversity) and to determine which model (linear or unimodal) of species response along a gradient to apply. I used canonical correspondence analysis to relate species abundances directly to explanatory variables. The explanatory variables were tested for significance and the variance partitioned among three groupings: temporal, spatial and environmental. DCA indicated complete species turnover along two dimensions: seasonal and spatial. It also indicated that a unimodal method such as CCA was appropriate for further analysis. The CCA model included 39 variables. The included variables explained 14% of the variation in species abundance in the data set. Since the first four axes explained 67% of the variation contained in the first two DCA axes, the chosen explanatory variables were sufficient to explain the majority of the tractable variation in species abundance. The variance partitioning procedure indicated that temporal effects were the most important in explaining species variation in the Upper Laguna, followed by the spatial ...
|