Identifying Sturgeon Spawning Locations Through Ensembles of 1-Dimensional Hydraulic Models

Abstract:

The Upper Missouri (UMR) and Yellowstone Rivers (YR) provide habitat for the endangered pallid sturgeon but fragmentation by dams is a threat to population persistence. Pallid sturgeon larvae disperse downstream passively for 5-7 days following hatch. Understanding advection and dispersion of these larvae is critical to ongoing management actions for species recovery. Past evaluations have been based on 1-D HEC-RAS-based advection-dispersion models to link reservoir management to larval drift duration in a 400 km river segment. These implementations specified spawning locations to evaluate probable downstream transport. We present an expanded usage of these models to answer related biophysical questions. We seek to identify the probability of spawning at unknown upstream locations given capture of a larva of known age at a downstream location. Within the HEC-RAS water quality module, we calculate advection and dispersion of a known mass of conservative material introduced within one 0.8-kilometer cell. For each discharge, we simulate an array of possible spawning points and breakthrough curves. Given capture of a larval pallid sturgeon at a known location and its time since hatch (larval age), we extract data from these scenarios to back-calculate possible spawning locations. Proportion of total mass at a capture location, summed over breakthrough curves from all potential spawning locations, corresponds to probability of each spawning location. From the probabilities we identify a range of locations within the river where spawning may have occurred accounting for physical processes of dispersion. On the UMR, this dispersion is confounded by backwater effects from both the confluence with the YR and a downstream reservoir, Lake Sakakawea. Backwater effects from the YR are documented in the 1-D model as reduced velocities and flattened water-surface slope; backwater extends as much as 6 kilometers upstream from the confluence at median flow-exceedance. Our assessment includes evaluation of the interaction the two rivers in influencing probable dispersal distance.