58-7 Comparative Analysis of Stressors Affecting Delta Smelt Population Dynamics: Implications of a Spatially-Explicit Individual-Based Model

Kenneth A. Rose , Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA
Wim Kimmerer , Romberg Tiburon Center, SFSU, Tiburon, CA
Karen Edwards , National Centre for Ocean Forecasting, UK Met Office, Exeter, United Kingdom
William A. Bennett , Bodega Marine Laboratory, Bodega Bay, CA
Delta smelt has become increasingly controversial due to its long-term decline and the more recent pelagic organism decline (POD).  To address some of the questions related to the causes for the decline, we developed an individual-based population dynamics model.  The model tracks thousands of individuals on the same spatial grid as used by Department of Water Resource’s DSM2 hydrodynamics model. For computational reasons, we use a super-individual approach, whereby each model individual represents a number of identical population individuals. Daily water temperature, salinity, turbidity, and the densities of six zooplankton prey types are represented on the spatial grid. The model follows the reproduction, growth, mortality, and movement of individuals over their entire life cycle. Reproduction is evaluated daily and egg cohorts are tracked until hatching. New model individuals are introduced as individual yolk-sac larvae. Growth of later stages is based on bioenergetics and zooplankton densities. Mortality includes a constant rate plus mortality due to starvation, turbidity, and entrainment. Movement of individuals is by particle tracking for the larval stages and behavioral algorithms for juveniles and adults. We simulated the population decline using 1995 to 2005 conditions, and explored the relative influence of historical changes in food and entrainment on delta smelt population dynamics. Historical food was simulated by matching each year between 1995 and 2005 to earlier years from the 1970’s to 1980’s based on similarity of monthly X2 values. We then repeated the 1995 to 2005 simulation but with food from the matched earlier years. Simulating entrainment effects was done by repeating the 1995 to 2005 simulation but with the entrainment mortality from the particle tracking movement and from negative flows in the Old and Middle Rivers set to zero. We also preformed additional simulations of the effects of food and entrainment using the 1995 to 2005 period with alternative assumptions about natural mortality (size-based or not) and the degree of density-dependent mortality. Simulations indicate that while both factors affected delta smelt population dynamics, changes in the food over the past few decades have had a larger effect on delta smelt population dynamics than entrainment. Increased understanding of how changes in food and entrainment affect delta smelt population dynamics will inform the protection and restoration of delta smelt. The modeling can be used to filter the possible management actions that could be taken, helping to identify effective and efficient options from an ecological perspective.