Transcription
Walt Duffy, Matt Metheny,Zack Larson and Kerrie Pipal11/18/20101
OverviewWhat DIDSON can do.What DIDSON cannot do.Accuracy and precision.11/18/20102
Need for assessing escapement has becomemore urgent. Responding to ESA requirements Supporting assessment of restoration andmanagement. Cost effective methods for estimatingescapement are needed. Redwood Creek alone has 196 km of streamaccessible to anadromous fishes, traditional surveymethods for this and other basins are costly.11/18/20103
DIDSON uses high frequency sound toproduce near video quality images of objects,fish and anything else.DIDSON is versatile and relatively easy to use,with training.Technology is advancing - Blueviewtechnologies is developing software that mayallow species separation(http://www.blueview.com).11/18/20104
DIDSON can: Provide 24-hour monitoring of spawningescapement. Estimate net hourly upstream passage at rates of upto 7,300 fish/hour. Establish run-timing curves for spawning areas. Measure and record lengths of passing fish.11/18/20105
DIDSON cannot: Detect fish beyond ranges of 15 m (high frequency)or 40 m (low frequency). Collect biological information other than length. Detect external tags on fish. Reliably identify different salmon species in a mixedgroup. Automatically count fish at a rate of more than1,250 fish/hour.11/18/20106
Ln (Tower Count)Comparison of video and DIDSON counts of migrating sockeye salmoncompared with tower counts. Maxwell and Gove (2007) J. Acoust. Soc. Am.122:3364- 3377.11/18/20107
Tower, DIDSON, split-beamand video counts ofmigrating sockeye salmonat 15 min intervals.Maxwell and Gove (2007)J. Acoust. Soc. Am.122:3364- 3377.11/18/20108
TABLE I. The concordance correlation coefficient CCC and related statisticsfor sockeye salmon counts in the Wood River made by an observer in a towerpaired with counts from the DIDSON, Bendix counter, split-beam sonar, endix0.967Tower—Split-beam sonar 0.951Tower—Video0.99495% 4Accuracy0.9980.9990.9851.000Maxwell and Gove (2007) J. Acoust. Soc. Am.122:3364- 3377.11/18/20109
DIDSON Length (cm)12010080605060708090100110120130True Total Length (cm)Size of salmon measured at distances of 9 m from transducer (solid circles)and salmon measured 9 m from transducer (open circles).Burwen, D. L., S. J. Fleischman and J. D. Miller. 2007.AK DFG, Fish Data Series No. 07-44.11/18/201010
Burwen and colleagues concluded: Length measurements of free-swimming fish werenot biased. Reasonably good estimates of fish length can beextracted from images at distances of 12 m athigh frequency. Cronkite reported a DIDSON bias Salmon measured on DIDSON were, on average,1.6 cm larger than salmon measured at countingfence.11/18/201011
11/18/201012
Walt Duffy, Matt Metheny, Zack Larson and Kerrie Pipal. 11/18/2010 1
