Home | Recent data | Photo archive | Elver pass |2006 |2007 |2008 |2009 |2010 |2011 |2012 |2013 |2014 |2015 |2016 |Temperature | Flow data | DIDSON | Summary Data | Contact | Avon Temperature Monitoring Site

DIDSON (Dual frequency IDentification SONar) trials

Click on the image to view a short clip of DIDSON footage.

A DIDSON is an acoustic camera which gives near video quality images for inspection and identification of objects underwater (www.soundmetrics.com). When applied to fish biology it can be used to count fish (either manually by the operator or by using motion detection software supplied with the unit). In this case the footage was recorded and saved onto external hard drives and reviewed manually.
Under high flows, hatches 11 to 16 on the Great Weir (which are not fitted with Resistivity Counters) are opened to maintain the upstream water level. The proportion of fish which pass through these hatches and therefore undetected is referred to as leakage. To estimate leakage, a DIDSON acoustic camera was temporarily deployed on two occasions in 2007 to evaluate the proportion of fish passing through the hatches without counters.

August 2007
As the flows were not particularly high in August (but the DIDSON was available) a trial was staged and hatches 17 and 18 were closed and 11 and 12 were opened. This meant that roughly equal amounts of water were passing through the hatches with counters on (19 and 20) as the hatches without counters (11 and 12).
During the study, 20 fish (including 2 Sea Trout) were found to move upstream through the counters on hatches 19 and 20 (10 fish per hatch). During the same period, 29 fish were observed moving through hatches 11 and 12 on the DIDSON footage (14.5 fish per hatch).

December 2007
During this trial, hatches11, 12, 17, 18, 19 and 20 were open fully.
16 fish were detected moving upstream through the counter hatches 17, 18, 19 and 20 (4 fish per hatch). During the same period, 16 fish were observed passing through hatches 11 and 12 on the DIDSON footage (8 fish per hatch).
Date of trial
No. of fish per hatch (counter hatches)
No. of fish per hatch (no counter hatches)
August 2007
December 2007

This data indicates that the counter is potentially missing a significant proportion of fish under high flows. Due to this we are currently installing extra resistivity fish counters on hatches 11 and 12 (as soon as the water level drops sufficiently). We are also using the DIDSON to further evaluate this leakage in 2008.

In order to allow for leakage we monitor which hatches are open and calculate it proportionally on a day by day basis. For example, if 12 fish go through the 4 hatches with counters on and 2 hatches without counters on are open, we estimate that 6 fish will have gone through these other hatches. In 2007 this increased the run estimate by 19.68%.