wooded than the lower two reaches, with larger volumes of instream LWD and water 

 temperatures moderated by groundwater on a continuous basis (Results Part IV). 



The middle reach extends 21.1 rm from Arrastra Creek downstream to Nevada 

 Creek (rm 67.7). At this junction, the river is less wooded; the channel loses slope and 

 becomes highly sinuous and prone to the accumulation of fine sediment. Stream bank 

 erosion and active channel migrations increase in the downstream direction. No 

 tributaries enter this reach. Water temperatures in the middle reach increase during the 

 summer and decrease during the winter more extremely as compared with the upper reach 

 (Appendix I). Channel icing (including increased anchor ice formation) also 



progressively increases downstream. 



The lower reach extends 13.7 rm from Nevada Creek (a large water quality 

 impaired tributary), downstream to the mouth of the North Fork (rm 54). Below Nevada 

 Creek, the river becomes confined by moraine against the Garnet Mountain where a 

 major increase in channel slope and substrate size occurs. Once confined, the river 

 acquires a more linear longitudinal profile, sinuosity decreases and channel gradient 

 increases abruptly from 4' to ISVmile. Within this reach boulders increase, volumes of 

 instream LWD decrease and channel bedforms and velocities become more variable. 

 Riparian health also declines in this section (Marler 1997), water quality is diminished by 

 the influence of non-point runoff originating in the Nevada Creek watershed (Ingman et 

 al. 1990). Compared with the two upper reaches, summer and winter water temperatures 

 are extreme in the lower reach (Appendix I). Several small and degraded tributaries 

 (Frazier, Wales and Youmame Creeks) enter this reach, all of which are fisheries- 

 impaired (Pierce et al. 2001). This reach supports the lowest salmonid densities for the 

 Blackfoot River downstream of Lincoln (Pierce et al. 2000; this report. Results Part 11). 



Methods 



Forty-five WSCT and 10 bull trout were captured and implanted with radio 

 transmitters between March 13-April 18, 2002 and March 18-April 13, 2003. 

 Transmitters were evenly distributed within the three study reaches. Fish captures were 

 made in early spring, prior to migrations, with either hook and line or by electro-fishing 

 with a Coffelt model VVP-15 DC electroshocker mounted on an 14 ' aluminum drift boat. 



We followed surgery methods described by Swanberg (1997) and Schmetterling 

 (2001). Captured fish were anesthetized with tricaine methanesulfonate (MS-222), 

 measured (total length, mm) and weighed (g). For this report, all metrics were converted 

 to standard units. Surgical tools were sterilized in betadine and rinsed with 0.9% saline 

 solution prior to each surgery. New surgical scalpels, latex gloves, and steel surgical 

 staples were used for each surgery. Surgeries consisted of bathing the gills with diluted 

 MS-222, while radio transmitters (Lotek Wireless) were inserted internally through a 2- 

 cm incision made along the linea alba anterior to the pelvic girdle. The transmitter 

 antenna was then passed through the body wall posterior to the pelvic girdle (Ross and 

 Kleiner 1982). Transmitters weighed 7.7 grams and did not exceed 2% offish weight as 

 previously suggested (Winter 1996). Transmitter life was estimated at -454 days. 

 Incisions were closed with Reflex-One 35W surgical staples (Swanberg et al. 1999). 

 Surgeries lasted 1-15 minutes (mean 4.2 min). Following surgery, the fish were held in a 

 live car in the river until fully recovered and then released at capture locations. Each 



65 



