Reproductive ecology, habitat associations, and population dynamics of two imperiled cyprinids in a Great Plains river
I studied reproductive ecology, recruitment, habitat associations, and population dynamics of the sharpnose shiner Notropis oxyrhynchus and smalleye shiner Notropis buccula in the Brazos River, Texas between April 2003 and March 2005. Both species are endemic to the Brazos River and were historically common throughout the entire 1700-km length of the river, but currently are restricted to approximately the upper third of the river basin above Possum Kingdom Reservoir. In 2002, both species were designated as Candidate Species for protection under the United States Endangered Species Act. The paucity of basic biological information available for either species has made it difficult to identify and mitigate specific factors responsible for their imperilment. Thus, the objective of my study was to obtain basic information on multiple aspects of the life history of sharpnose shiner and smalleye shiner, and to use the information obtained to evaluate current hypotheses about the reproductive ecology of other Great Plains cyprinids. I collected adult and juvenile sharpnose shiner and smalleye shiner monthly from four sites on the upper Brazos River for two years, except during the spawning season, May through September, when samples were collected twice each month. The population structure and reproductive ecology of both species was similar. Length-frequency analyses and inspection of annual growth increment formation in otoliths showed that populations of both species consisted of three age groups: age 0, age 1, and age 2. Reproduction occurred over a six-month period between April and September during which individual fish spawned multiple times. Patterns in oocyte development, oocyte size distribution, and gonadosomatic indices indicated that sharpnose shiner and smalleye shiner spawn asynchronously except during periods of elevated stream discharge when spawning is synchronized within populations. Mean annual fecundity of sharpnose shiner was 379.3 for age-1 females and 1379.9 for age-2 females. Mean annual fecundity of smalleye shiner was 443.3 for age-1 females and 2175.4 for age-2 females. I studied daily growth increments in the otoliths of larval and juvenile sharpnose shiner and smalleye shiner to assess recruitment during the reproductive season. Successful production of young for both species occurred throughout much of the six-month spawning season. The study revealed that recruitment by sharpnose shiner and smalleye shiner populations in the Brazos River, Texas are related to stream discharge in two principle ways. First, the absence of stream discharge appears to preclude any successful production of young. Second, the greatest proportion of young produced during the reproductive season is associated with an ascending hydrograph. I used logistic regression models to assess the relationship between presence of adult sharpnose shiner and smalleye shiner and depth, current velocity, temperature, conductivity, dissolved oxygen, and turbidity. Of the six variables evaluated, depth and current velocity were the only significant predictors of sharpnose shiner and smalleye shiner habitat use. In months when only a narrow range of depths and current velocities were available, fish were more frequently found in deeper and faster areas. However, in months with the greatest depths and current velocities, the presence of sharpnose shiner decreased as depths and current velocities increased. Using information obtained from reproductive and recruitment analyses, I constructed an age-specific population dynamics model for sharpnose shiner and smalleye shiner populations. I included stream discharge as a driving variable and used maximum likelihood methods to model age-0 survival as a function of stream discharge. I compared the observed abundance of sharpnose shiner and smalleye shiner from available catch data to the abundance predicted by the discharge-based model and two alternative null parameterizations using the Akaike Information Criterion. Among the three alternative models, the discharge model was the best predictor of smalleye shiner abundance but one of the null parameterizations was the best predictor of sharpnose shiner abundance. Elasticity analysis and sensitivity simulations of the projection matrix indicated that age-0 survival and age-1 fecundity are the most influential parameters affecting population dynamics of both species. The model also indicated that a mean summer discharge of 2.61 m3.s-1 for sharpnose shiner and 5.52 m3.s-1 for smalleye shiner is required in the Brazos River to maintain positive population growth rates.