Split beam and DIDSON in position for sampling Mill Creek in spring 2008. Note overhead camera above the white plates. (CDFG photo: M. Johnson)

Mill Creek site showing the results of a snow melt event. (CDFG photo: M. Johnson)

by Brenda Olson, Red Bluff FWO

A two-year pilot study was initiated on Mill Creek, tributary to the Sacramento River, Tehama County, California to evaluate two different hydroacoustic technologies in estimating escapement of spring-run Chinook salmon.  The two technologies were a BioSonics split-beam system and a dual-frequency identification sonar (DIDSON) system.  In 2008, California Department of Fish and Game (CDFG) also installed a video weir at the site.  This gave us an opportunity to test all three systems.  In addition to testing system performance, we also were interested in comparing the acoustic results to population estimates based on redd surveys.

The 2006 and 2008 water years were very different, giving  us a great opportunity to test performance under both high and low flows. During both study periods the DIDSON out-performed the BioSonics split-beam.  The split-beam system was saturated by either fish milling (difficult to discern Chinook targets from smaller resident fish) in the sample area during 2008 low flows or turbulence and air entrainment during 2006 high flows.
 

The DIDSON system worked reliably and effectively throughout the 2006 and 2008 sampling periods.  During 2008, the video weir was an integral part of the study.  The weir increased the opportunity for both acoustic systems to detect all upstream migrating Chinook salmon.  The video data allowed comparison and validation of the DIDSON count data during periods when water clarity allowed for video imaging of passing fish.

The DIDSON-video comparison provided information regarding the relative advantages and disadvantages of the two systems for estimating salmon escapement.  The primary advantage of DIDSON over video is its ability to sample effectively regardless of water clarity or turbidity condition.  In addition, the DIDSON does not require artificial light for sampling during periods of darkness.  The primary advantage of video over DIDSON is the ability to identify the species of each fish detected.  The video also allows the ability to count small-sized salmon that are otherwise missed by DIDSON due to size-thresholding of the DIDSON data. 

To estimate the number of spring Chinook that spawned in Mill Creek during 2008, CDFG conducted redd surveys over 58 km of stream.  Their results indicate an estimated 362 Chinook spawned in 2008.  DIDSON counts estimate escapement for 2008 was 383. 

During 2006 the DIDSON escapement estimate was 1,447 whereas the estimate derived form redd counts was 1,002.  We speculate a large part of the difference was due to one 3-mile reach was not surveyed for redds.  This reach was surveyed in 2007 and 2008 and has a fair amount of spawning by spring Chinook salmon.

Results in 2008 indicated that the video and DIDSON systems operating simultaneously provided a better count of Chinook passage than either system operating alone.  However, for streams with high turbidity, the DIDSON performs effectively and reliably in determining salmon escapement.  The BioSonics split-beam did not perform well in Mill Creek and it is not recommended for streams in the Central Valley of this size.