Methods

Selection of Sediment Sampling Sites

Three sampling sites, two in the braided reach and one in the meander reach, were selected to represent the white sturgeon critical habitat (fig. 2). Bedload-sediment and suspended-sediment samples were collected during each sampling period.

The two sampling sites farthest upstream were sampled in previous years, and the sampling site farthest downstream was established in 2008 (fig. 2). Each site is representative of the three distinct geomorphic and hydrologic features throughout the critical habitat. The sediment-sampling sites were not co-located with real-time USGS gaging stations, so following methods described in Mueller and Wagner (2009), an acoustic Doppler current profiler (ADCP) was used to obtain accurate hydrologic data at the sampling sites.

Kootenai River below Moyie River

The Kootenai River below Moyie River sampling site (4842231161104) is near the upstream end of the Kootenai River white sturgeon critical habitat at RKM 257.1, 1.0 km downstream of the confluence of the Moyie River and 11.3 km upstream of the city of Bonners Ferry (fig. 2). The sampling site is near the bottom of the canyon reach. Substrate across the sampling cross section ranged from coarse gravel to fine cobble. Particle-size diameter at this site ranged from 20.8 to 78.2 mm (D₁₆ and D₈₄), with a median particle-size (D₅₀) of 39.9 mm (appendix B, fig. B1). Velocities from ADCP measurements during sediment sampling ranged from 1.1 to 1.9 m/s. Widths at the sampling site ranged from 138.0 to 142.0 m; average depths ranged from 3.8 to 5.0 m. Discharge at this location is completely free flowing and is not affected by Kootenay Lake backwater, even when lake levels are high.

Kootenai River below Fry Creek

The Kootenai River below Fry Creek sampling site (12309490) is 1.0 km upstream of Bonners Ferry at RKM 246.7 (fig. 2). This sampling site is near the transition zone of the braided reach to the meander reach. Particle-size diameter at this site ranged from 12.8 to 55.6 mm (D₁₆ and D₈₄) with a median particle-size (D₅₀) of 29.8 mm (appendix B, fig. B2). Velocities from ADCP measurements during sediment sampling ranged from 1.0 to 1.1 m/s. Widths at the sampling site ranged from 200.0 to 209.0 m; average depths ranged from 3.0 to 5.6 m.

The backwater effect from Kootenay Lake plays an important role in sediment transport at the Fry Creek sampling site. Early in the year when Kootenay Lake levels are low, this section of the reach is less affected by backwater. As Kootenay Lake stage increases, the backwater extends upstream of the sampling site, resulting in decreased velocity and increased stage, causing decreased sediment transport capacity of the river. Berenbrock (2005) provided a three-parameter relation that can be used to locate the transition between backwater and free-flowing river.

Kootenai River above Shorty’s Island

The Kootenai River above Shorty’s Island sampling site (4845421162319) is 14.8 river kilometers downstream of the U.S. Highway 95 bridge in Bonners Ferry at RKM 231.7 (fig. 2). The site is approximately 3.5 river kilometers upstream of the downstream boundary of the white sturgeon critical habitat. The substrate at the sampling cross section consists of “fine to coarse-grained alluvial sand” corresponding to particle-size diameters of 0.2 to 1.4 mm (Barton, 2004). Velocities from ADCP measurements during sampling times ranged from 0.6 to 0.8 m/s. Widths and depths at the sampling site ranged from 228.0 to 237.0 m and 6.5 to 7.0 m, respectively. This area is affected year-round by the backwater of Kootenay Lake.

Sampling Frequency

The rate of discharge in the Kootenai River is regulated primarily by operations at Libby Dam. Sediment samples were collected during the three primary peaks during water year 2008 (fig. 3). Two of these three sampling periods coincided with regulated discharge from Libby Dam—the winter ramping and the sturgeon pulse—that conform to levels prescribed in the 2006 biological opinion (U.S. Fish and Wildlife Service, 2006). The third sampling period occurred during the spring freshet, a combination of snowmelt runoff from local tributaries downstream of Libby Dam and controlled discharge from the dam.

Winter Ramping

To meet power demands, discharge from Libby Dam is increased and decreased by about 142 m³/s on a daily or weekly basis. This fluctuation of discharge is known as winter power ramping, load shaping, or winter ramping (U.S. Fish and Wildlife Service, 2006). River discharge during winter ramping fluctuates from about 283.0 to 708.0 m³/s. Sediment samples were collected during the winter ramping in December 2007.

Spring Freshet

Spring freshet flows in the Kootenai River downstream of Libby Dam are dominated by snowmelt runoff in the Cabinet and Purcell Mountains of northern Idaho and northwestern Montana. Snowmelt runoff is limited to the drainages that have their confluence downstream of Libby Dam. The Fisher, Yaak, and Moyie River drainages are the major contributors to the freshet.

In 2008, the spring freshet flows were during the weeks of May 8–30. Two USGS gaging stations (12301933 and 12301000) were used to approximate runoff contribution between Libby Dam and the white sturgeon critical habitat. The peak flow of the freshet was on May 19. The flow at the Tribal hatchery gaging station was about 1,460 m³/s, and sustained flows from Libby Dam were about 396 m³/s. The difference in flow from these two locations at the peak was 1,064 m³/s. The combined flow from the Fisher and Yaak Rivers was about 379 m³/s.

Sturgeon Pulse

The spring sturgeon spawning season flow augmentation pulse, or sturgeon pulse, is carried out to simulate the spring river flows that occurred before the construction of Libby Dam. These flow conditions are determined by the guidelines set forth in the biological opinion to create flow conditions in the river during a period favorable to sturgeon spawning in the spring (U.S. Fish and Wildlife Service, 2006). The sturgeon pulse for water year 2008 occurred on June 2, followed the spring freshet, and nearly equaled the spring freshet in total discharge. During the sturgeon pulse, 57 percent of the flow was from Libby Dam in contrast to a 27 percent contribution during the spring freshet. The peak flow was about 1,310 m³/s at the Tribal hatchery gaging station and flow out of Libby Dam was about 745 m³/s. The difference in flow from these two locations at the peak was about 565 m³/s. Flow contributions from the Fisher and Yaak Rivers during the peak of the sturgeon pulse were about 150 m³/s.

Sampling Methods

A jet boat equipped with a hydraulic boom was used for all sediment sampling. Both suspended- and bedload-sediment sampling followed the equal width increment (EWI) and single equal width increment (SEWI) methods described by Edwards and Glysson (1999). The SEWI sampling method was modified as the widths of the first and last sampled zones were not equal to the width of the other sampled zones. This modification was necessary to establish 20 equally spaced sampling stations. Sampling stations were created using digital imaging software that included the left and right bankfull locations and 20 equally spaced vertical sampling stations. Using this technique, each vertical was sampled in the same spatial location per sampling sequence. Figure 4 shows an example diagram of this sampling method. Fugawi™ navigation software with positioning from a mapping grade TrimbleAg232™ global positioning system (GPS) was displayed on a laptop computer to locate each equally spaced vertical sampling station spatially.

Suspended Sediment

Suspended sediment was collected using a D-96 depth-integrated sampler (Davis, 2001) using the EWI method. Sampling methods at each site followed USGS-approved methods described by Edwards and Glysson (1999). The D-96 samples the water column from the free-flowing surface to approximately 102.0 mm above the riverbed. Vertical samples were composited from each pass across the reach.

Bedload Sediment

Bedload sediment was collected using 102 × 203 mm Elwha (Holnbeck, 2005) and 77 × 77 mm Helley-Smith (Emmett, 1980) style samplers. The Elwha, also known as the US-ER1, is a smaller version of the US-TR2, or Toutle River Sampler (Childers, 1999). This sampler was suitable for the Kootenai River because the entrance size was twice as large as the average surface bed sediment. Three Elwha samplers were used to collect bed samples. The samplers were identical size and proportion, but the weights varied (about 27, 36, and 59 kg). The mesh diameter of the sampling bag for the Elwha samplers was 0.50 mm. A Helley-Smith sampler was used for a short time during the sampling. The mesh diameter of the sampling bag for the Helley-Smith sampler was 0.25 mm. Samples were collected at each bedload sampling site using two traverses (A and B sets) of the channel with 20 equally spaced verticals for each traverse as prescribed by USGS sampling standards (Edwards and Glysson, 1999). Each vertical sample was collected for data analysis if a quantifiable amount of material was captured. Some samples were composited after five verticals due to lack of appreciable material available for each vertical.

In February 2008, Wolman pebble counts were completed near the sampling sites in the braided reach. A gravelometer was used following procedures outlined in Wolman and Leopold (1954) to measure coarse riverbed particle-size diameter. Samples were collected from exposed substrate and in shallow water depths across the entire cross section.

River Discharge, Stage, and Slope

An ADCP discharge measurement was made at each sampling site after the suspended and bedload sampling per USGS standards (Mueller and Wagner, 2009). A Trimble™ real-time kinematic (RTK) GPS was used for positioning of the ADCP measurements. Stage and slope measurements were made using a combination of USGS streamflow-gaging stations, and stage recorders positioned near the sampling sites.