Simulation of Streamflow in White Sturgeon Habitat

The braided-straight reach and meander-straight reach models were used to simulate depths, depth-averaged velocities, bed shear stresses, and water-surface elevations in the Kootenai River for streamflows of 170, 566, 1,130, 1,700, and 2,120 m³/s (tables 2 and 4). The range of simulated streamflows was selected to span river conditions typical of those before and after the construction of Libby Dam. The highest streamflow (2,120 m³/s) represents a discharge approximately equal to the annual median peak streamflow (2,240 m³/s) prior to construction of Libby Dam in 1972. The water-surface elevation at the downstream boundary of both models (table 5) was set for each streamflow condition using a one-dimensional model developed by Berenbrock (2005). Each simulated streamflow corresponds to the computed median water-surface elevations in Kootenay Lake at the Queens Bay gaging station (08NH064; fig. 1). A complete description of these conditions is in a report by Berenbrock (2006).

A few simple simulations are presented to demonstrate how the model can be used to link physical characteristics of streamflow to biological or other habitat data. Some measures of depth, velocity, and substrate composition generally are considered when assessing sturgeon spawning habitat (Parsley and others, 1993). Discussions of the model simulation include relating the model simulation of depth and velocity to observed patterns of spawning and egg locations. Adult sturgeon in the braided reach generally avoid minor channels and sloughs; therefore, the average depth and average velocity computed for the braided reach is based on model nodes only in the main channel and excludes any nodes in inundated minor channels and sloughs. Additionally, the average and maximum values of simulated depth and velocity were computed for cross-sections positioned every 100 m along the model reach.

River Depth

River depths are considerably shallower upstream of RKM 245.6 in the upper part of the straight reach and in the braided reach as compared to the lower and middle part of the straight reach and to the meander reach. Simulated depths averaged across cross sections positioned every 100 meters along the stream are shown in figure 9; maximum depths at each cross section are shown in figure 10. Simulated average depths in the main channel of the braided reach for the lowest and highest streamflow simulations were 2.4 and 6.7 m (fig. 9) and the average depths along the deepest part of the thalweg are 3.3 and 8.5 m (fig. 10). Downstream of Ambush Rock (fig. 2) in the meander reach, the average depths of the channel for the lowest and highest streamflow simulations were 4.8 and 9.7 m, respectively. The average depths along the deepest part of the thalweg are 8.00 and 15.0 m (fig. 10).

The river channel depth is greater than 13 m for the highest simulated flow at one location in the braided reach, at the downstream end of the straight reach, and throughout most of the thalweg in the meander reach (fig. 10). Near the upper end of the braided reach at RKM 253.7 the river impinges against a lone bedrock outcrop along the right bank resulting in a deep scour hole. The deepest points in the straight and meandering reaches are at Ambush Rock (RKM 244.5) and the Klockmann Ranch gaging station (12314000) at RKM 224.7 (fig. 2) where the river impinges against a bedrock outrcop and forms a scour hole. The maximum river depth at Ambush Rock and at the Klockmann Ranch gaging station is 20 and 21 m for the lowest simulated flow and 27 and 28 m for the highest simulated flow (fig. 10). Fishing for spawners for the KTOI Sturgeon hatchery has shown that white sturgeon use the scour hole at the Klockmann Ranch gaging station as a staging area prior to the spawning season (Jack Siple, Kootenai Tribe of Idaho, oral commun., 2007). During the spawning season, the farthest known upstream congregation of adult white sturgeon was observed in the scour hole at Ambush Rock. Fish telemetry has shown that some sturgeon use this hole as a staging area to swim into the straight reach and into the lower part of the braided reach (Pete Rust, Idaho Fish and Game, oral commun. 2006). Other deep areas in the meander reach are at the outside of channel bends where the curvature is highest such as the S-shaped meander near Myrtle Creek.

During the 1994–2002 spawning seasons, egg collection in the Kootenai River indicated that sturgeon eggs tend to accumulate in deep water in the thalweg of the meander reach (Vaughn Paragamian, U.S. Fish and Wildlife Service, oral commun., 2005). Results from studies of white sturgeon spawning locations and simulated water depths (McDonald and others, 2004) and microhabitat criteria curves Parsley and Beckman (1994) can be interpreted that sturgeon may avoid spawning in the braided reach due to shallow water; however, other factors such as velocity patterns, hydraulic complexity, and site fidelity may influence sturgeon spawning patterns.

Simulated Streamflow Velocity

Depth-averaged streamflow velocities in the braided reach were greater than velocities in the straight and meander reaches (fig. 11). In the main channel of the braided reach, the average depth-averaged velocities for the lowest and highest streamflow simulations were 1.1 and 1.3 m/s, respectively. The maximum depth-averaged velocities in the main channel for the lowest and highest streamflow simulations were 2.3 and 3.3 m/s. In the meander reach, average velocities for the lowest and highest streamflow simulations were 0.2 and 0.9 m/s. The maximum velocities for the lowest and highest streamflow simulations in the meander reach were 0.3 and 1.6 m/s.

The braided, straight, and meander reaches can be divided into four velocity zones based on simulated maximum depth-averaged velocity along the length of the model reach (fig. 11; Barton and others, 2005). The predominance of spawning in the downstream zone (zone D) and the model simulation results indicate that white sturgeon may be focusing on a specific velocity signature. Zone A, the main channel of the braided reach, is a region of high velocity except near RKM 248. Zone B, the straight reach, is a region of transition from the high velocity braided reach to the low velocity meander reach. At streamflows greater than about 566 m³/s, backwater conditions extend upstream more than 1 km into the lower part of the braided reach resulting in decreased velocities that are somewhat similar to velocities in the straight and meander reaches. Between RKMs 249.3 and 250.0, the velocity increases as streamflow increases to 1,130 m³/s. At higher streamflows, the velocity decreases due to backwater conditions extending farther upstream. Since 1994, IDFG has detected less than a few percent of the overall white sturgeon spawning in zone B. Maximum streamflow velocities are uniform compared with the rest of the meander model reach in zone C between Ambush Rock and the first sharp meander downstream of the mouth of Deep Creek at RKM 237.5. One notable exception is near the deep scour hole and lateral recirculation eddy upstream of Deep Creek located along the left bank. Less than 15 percent of the overall white sturgeon spawning has been detected in zone C (Paragamian and others, 2002). Maximum velocity increased with increasing streamflow and formed distinct zones of relatively high velocity in zone D, downstream of zone C, and extending to the downstream model boundary at RKM 222.1. Most white sturgeon spawning has been detected in zone D. McDonald and others (2004) reported that areas of relatively high velocity and white sturgeon spawning locations are correlated. Studies on the Columbia River indicate that sturgeon seek areas of high velocity for spawning (Parsley and others, 1993).

Streamflow velocities equal to or greater than 1.0 m/s are believed to greatly reduce or eliminate predation of white sturgeon eggs (Bob Hallock, U.S. Fish and Wildlife Service, oral commun., 2005). The simulated average velocity in the spawning reach within the meander reach does not approach 1.0 m/s until streamflow is about 2,120 m³/s (fig. 11). Based on maximum simulated velocity in the meander reach, velocities in parts of the channel approach 1.0 m/s for streamflows roughly equal to or greater than 1,130 m³/s. However, velocities in the main channel of the braided reach generally exceed the 1.0 m/s threshold.

Flow structure at the white sturgeon spawning locations in the meander reach differs substantially from the flow structure in the braided reach (fig. 12). This difference could have some influence on the location of spawning events. McDonald and others (2006a) reported that areas of relatively high velocity and white sturgeon spawning locations are correlated. These areas of relatively high velocity occur in the meander reach at the same locations regardless of the discharge magnitude as modeled over a range of pre- and Libby Dam era flow conditions. In the braided reach, locations of regions of higher velocity are variable and shift on the order of kilometers in the upstream or downstream direction (fig. 11) under different backwater and streamflow conditions. This flow structure behavior can be attributed to the single-threaded meander reach having uniform channel geometry compared to the multi-threaded braided reach. As stage declines or the streamflow decreases to less than about 1,130 m³/s in the braided reach, the various minor channels and sloughs of the river begin to dewater causing the regions of higher velocity to change location. It is not known if white sturgeon can perceive the evolving velocity structure of the braided reach as too complex for spawning site selection.