Stacey, M. T.. University of California,
Batchelder, H. T.. University of California,
Powell, T. M.. University of California,
Twombly, S. University of Rhode Island,

The estimation of vital rates (mortality, reproduction and growth rates) from time series of zooplankton stage abundances is explored using an Augmented Kalman Filter (AKF), where the state vector includes the parameters to be estimated. The AKF is applied to sub-windows of the data, providing parameter estimates which may vary with time. We evaluated the ability of the AKF to correctly estimate vital rates with synthetic data sets, in which the "true" vital rates are known. The method is successful at identifying step and pulse changes in vital rates in the twin experiments, although the ability to resolve the changes is sensitive to noise in both the sampled data and in the parameters. We have successfully applied the technique to a stage-structured data set on Diaptomus negrensis in a tropical lake. Variation in vital rates during the 6 month period might provide clues on extrinsic changes in the environment--e.g., changes in predation intensity or food quality. With these estimates of the parameters, a model trajectory is produced which closely mimics the observations and facilitates interpretation of the biology, including identification of cohorts.
Day: Tuesday, Feb. 2
Time: Poster
Location: Sweeney Center
Code: SS44TU0499S