Technical, Logistical, and Economic Considerations for the Development and Implementation of a Scottish Salmon Counter Network: Scottish Marine and Freshwater Science Vol 7 No 2

Appendix 2. Questionnaire Results
Summary of results from counter questionnaire from various counter users.

General information

Watershed	Counter	Type	Structure	Cost ( GBP)	Main Activity Required	Main Operations Cost	person-day	Annual operating costs ( GBP)
Unknown	Vaki	Optical	Crump Weir
Unknown	Logie	Resistivity	Crump Weir
Unknown	Logie	Resistivity	Crump Weir
River Dee, Scotland	Vaki	Optical	Fishway	£25,000 (2008)	Maintenance	salary	20	£3000- £10 000
Kirkcudbrightshire Dee, Scotland	Vaki	Optical	Fishway	£20,000 (2007)	Maintenance	salary	27	£3000- £10 000
Galloway, Scotland	Vaki	Optical	Fishway		Maintenance	salary		£3000- £10 000
Deveron, Scotland	Logie	Resistivity	Crump & fish Weir		Infrastructure	salary	25	£0- £3000
Deveron, Norway	DVR	Video	None		Analysis	salary	60	£20000-£50000
River Shin, Scotland	not Logie	Resistivity	Fishway
North Esk, Scotland	Logie	Resistivity	Crump Weir	£250 000 (1980)-Logie <£10000 (1990)-Westwater	Maintenance	Equipment	50/site	>£50000
Tweed, Scotland	Vaki	Optical	Fishway	£25 000 (2010)	Data collection and processing	salary	7	£0- £3000
Spey, Scotland	Split-beam	Acoustic Sonar	None	£80,000 for five year (1996-2001)	Data collection and processing	salary	>200	£0- £3000
Kamloops, Canada	Didson / Aris	Acoustic Sonar	Deflection Weir		Data collection and processing			>£50000

1. What type of technology do you use? (Q3 - 13 answered)

Technology	Number	Overall Percentage
Optical Beam (Vaki)	5	38%
Sonar (Didson, Aris, Splitbeam)	2	15%
Resistivity (Logie, Mark)	5	38%
Real-time Video	1	8%

2. Infrastructure used to deploy and operate counter in river? (Q4 - 13 answered)

Structure	Number	Overall Percentage
Fishway	5 --- (4 for optical, 1 for resistivity)	38%
Purpose-built weir (Crump weir)	5 --- (1 for optical, 4 for resistivity)	38%
Deflection fence	1 --- (1 for acoustic sonar)	8%
Fish weir	1 --- (1 for resistivity)	8%
None	1 --- (1 for split beam)	8%

It should be noted that deflection fences are only used in North America.

3. What quantitative methods do you use to evaluate data quality? (Q16 - 6 answered, 7 skipped)

Vaki

• Check with biological staff
• don't know
• Video analysis to check counter accuracy (number)

Video

• Data from video systems running in "time lapse" modus in itself a control. In video surveillance with sufficient frame rate, no object can pass cameras without being detected.

Acoustic Sonar

• Video cameras placed under acoustic beam to film fish for periods

Resistivity

• Use a fish counter signal processing key

4. Are your data analyzed in season (e.g. estimates of fish passage over the counter are produced and updated during fish migrations)? (Q20 - 11 answered 2 skipped)

Answer Choices	Number	Overall Percentage
Yes	9 --- (3 for Resistivity, 2 for Acoustic, 3 for Optical)	81.8%
No	2 --- (1 for Resistivity, 1 for Optical)	18.2%

5. Which additional method(s) would improve your data processing? Pick all that apply. (Q21 - 7 answered 6 skipped)

Vaki

• Automating quality control
• Visualization of data
• Summary of tables of data
• Error elimination
• Automating species assignment (applying known %s of each spp to unidentified fish)
• Don't know
• Having a Scotland wide network of users to provide support when needed

Acoustic Sonar

• Automating quality control
• Visualization of data
• Summary of tables of data
• Error elimination
• Eliminating sporadic Ethernet disconnections

Resistivity

• Automating quality control
• Visualization of data
• Summary of tables of data
• It is difficult to say as there are a lot of trade-offs. We would want to at least maintain current level of data processing and so if any of the above methods would result in less time being spent to achieve the same results then this would be welcome

6. Do you validate fish passage detection rates? This refers to validating the counters up and down count accuracy. For example, this could include comparing observations of fish moving over the counter with counter output. If 10 fish were observed passing up over the counter but the counter only detected 9 the counter would only be 90% accurate. (Q22 - 12 answered 1 skipped)

Answer Choices	Number	Overall Percentage
Yes	10 --- (3 for Resistivity, 2 for Acoustic, 4 for Optical, 1 for video)	80.0%
No	2 --- (2 for Optical)	20.0%

7. If "no" was answered to previous question "Do you validate fish passage detection rates?", why not? (Q23 - 2 answered 11 skipped)

Optical

• not enough time but will do sometime soon. Confident that counter is reliable
• staffing issues too date and validation planned for 2015

8. During which part of the fish migration do you validate fish passage detection rates? Pick all that apply. (Q26 - 10 answered 3 skipped)

Answer Choices	Number	Overall Percentage
Early in migration	0	0.0%
Late in migration	3 --- (2 for Resistivity, 1 for Optical)	30.0%
Peak of migration	0	0.0%
Entire migration	6 --- (1 for Resistivity, 2 for Acoustic, 2 for Optical, 1 for video)	60.0%
Not applicable	2 --- (1 for Resistivity, 1 for Optical)	20.0%

9. How do you validate your fish passage detection rates? (Q27 - 8 answered 5 skipped)

Acoustic

• Compare individual video observations to counter produced fish passage records
• Compare individual visual observations to sonar produced fish passage records

Resistivity

• Not applicable
• Compare individual visual observations to counter produced fish passage records
• Compare individual video observations to counter produced fish passage records
• Compare signals with signal validation key which was constructed using in-river calibration

Video

• Video surveillance can be evaluated continuously through the season, based on observation of water turbidity

Optical

• applicable
• Compare individual visual observations to counter produced fish passage records
• Compare individual video observations to counter produced fish passage records

10. How do you use your validation information? Pick all that apply. (Q28 - 8 answered 5 skipped)

Answer Choices	Number	Overall Percentage
To calculate absolute abundance	3 --- (1 for Resistivity, 1 for Optical, 1 for video)	37.5%
For making sure the counter accuracy is consistent from year to year	5 --- (3 for Resistivity, 2 for Optical)	62.5%
Not applicable	1 --- (1 for acoustic)	12.5%

Acoustic

• To ensure no fish are going past the sonars ( DIDSON and ARIS) undetected

11. Are there more than one species passing over the counter at the same time of year? (Q29 - 11 answered 2 skipped)

Answer Choices	Number	Overall Percentage
Yes	10 --- (3 for Resistivity, 2 for Acoustic, 4 for Optical, 1 for video)	90.9%
No	1 --- (1 for resistivity)	9.1%

12. Do you validate species identification? (Q30 - 11 answered 2 skipped)

Answer Choices	Number	Overall Percentage
Yes	7 --- (2 for Resistivity, 2 for Acoustic, 2 for Optical, 1 for video)	63.6%
No	3 --- (2 for Resistivity, 1 for Optical)	27.3%
Not applicable	1 (1 for optical)	9.1%

13. If yes to the question "do you validate species identification?" How do you determine different species? (Q31 - 6 answered 7 skipped)

Answer Choices	Number	Overall Percentage
Using video	1 --- (1 for video)	16.7%
Using size discrimination	2 --- (1 for Resistivity, 1 for Acoustic)	33.3%
Both	3 --- (1 for Resistivity, 1 for Acoustic, 1 for Optical)	50.0%

Acoustic

• Fish behaviour (e.g. tail beat frequency; traveling in groups), and spatial separation of species (e.g. Chinook normally migrate down the middle while Sockeye are more off to the side riverbanks.)

Resistivity

• We are fortunate in that species can be discriminated by size and the counter is set up to only count the target species
• Run timing

14. How often do you validate species identification? (Q32 - 8 answered 5 skipped)

Answer Choices	Number	Overall Percentage
Every year	4 --- (1 for Resistivity, 1 for Acoustic, 1 for Optical, 1 for video)	50.0%
1 out of 2 years	1 --- (1 for Resistivity)	12.5%
1 out of 3 years	0	0.0%
1 out of 5 years	0	0.0%
Only when there are large # of non-targeted species	1 --- (1 for Resistivity)	12.5%
Only when river conditions permit	1 --- (1 for Acoustic)	12.5%
Never	1 --- (1 for Optical)	12.5%

Acoustic

• Difficult to validate species identification in more turbid locations which can make live counting very difficult.

Resistivity

• Qualitative assessment of in-river fishery catch of other species (e.g. sea trout)

15. What type of abundance estimates do you produce with the counter data? (Q35 - 9 answered 4 skipped)

Answer Choices	Number	Overall Percentage
Minimum estimate - this estimate is produced using just the raw counter data with no validation	3 --- (1 for Resistivity, 2 for Optical)	33.3%
Absolute estimate - this estimate would incorporate the counter accuracy values (fish passage detection rates) from each year to account for fish missed or erroneous counts produced by the counter	5 --- (3 for Resistivity, 1 for Acoustic, 1 for Optical)	55.6%
Index - this would be applying a correction factor (similar to a counter accuracy estimate) to the raw data. The difference between and index and absolute count estimate is the index applies the same correction factor each year whereas the absolute abundance estimate corrects the raw data with accuracy estimate for each year.	1 --- ( for Resistivity, 1 for Acoustic)	11.1%

Video

• Wide systems give direct counts and are not estimates. That does not mean there are errors, but they are evaluated.

16. If there are more than one species, do you produce species-specific estimates of fish passage abundance? (Q36 - 10 answered 3 skipped)

Answer Choices	Number	Overall Percentage
Yes	7 --- (2 for Resistivity, 4 for Optical, 1 for video)	70.0%
No	3 --- (2 for Resistivity, 1 for Acoustic)	30.0%

17. Do you account for uncertainty in your estimates of fish passage abundance by including uncertainty in counter accuracy or species identification? (Q38 - 8 answered 5 skipped)

Answer Choices	Number	Overall Percentage
Uncertainty is not accounted for	6 --- (2 for Resistivity, 2 for Acoustic, 2 for Optical)	75.0%
Bootstrapping	1 --- (1 for Resistivity)	12.5%
Simulations (e.g. Monte Carlo)	1 --- (1 for Resistivity)	12.5%

Video

• Uncertainty is evaluated from different parameters: water flow, turbidity, etc.

Optical

• Uncertainties are classed as sea trout or later brown trout rather than salmon due to previous experience and video analysis.

18. Which would help improve your analyses and generation of fish passage abundance estimates? Pick all that apply. (Q41 - 7 answered 6 skipped)

Answer Choices	Number	Overall Percentage
Accounting for uncertainty	4 --- (1 for Resistivity, 1 for Acoustic, 2 for Optical)	57.1%
Automation of estimating fish passage abundance	5 --- (2 for Resistivity, 2 for Acoustic, 1 for Optical)	71.4%
Species-specific fish passage abundance estimates	5 --- (2 for Resistivity, 1 for Acoustic, 2 for Optical)	71.4%
More accurate fish passage abundance estimates	6 --- (2 for Resistivity, 2 for Acoustic, 2 for Optical)	85.7%
More precise fish passage abundance estimate	3 --- (1 for Resistivity, 1 for Acoustic, 1 for Optical)	42.9%

19. Do you know approximately total installation cost in GBP and what year they were incurred? (Q45 - 5 answered 8 skipped)

Optical

• £25,000 (2010)
• £20,000 (2007)
• £25,000 (2008)

Resistivity

• £250,000 (1980) -- Logie, <£10,000 (1990) -- Westwater

Acoustic

• £80,000 for five year (1996 to 2001)

20. Please rank the types of counters operation costs, where the most costly operating cost is ranked 1 ^st and the least costly operating cost is ranked 5 ^th. (Q47 - 9 answered 4 skipped)

21. Please rank the main annual counter operator activities in order of cost, where the most costly operating activity is ranked 1 ^st and the least costly operating activity is ranked 7 ^th. (Q48 - 9 answered 4 skipped)

22. How many person-days are used each year to operate the counter? (Q49 - 8 answered 5 skipped)

Acoustic

• - Approx. 60 person days for one sonar project/year
• - 200+ days

Optical

• - 7 days V
• - 25 days V
• - 27 days V
• - 20 days V

Resistivity

• - 25 days
• - 50 days per site

Video

• - 60 days

23. What are the approximate annual operating cost of the counter? Values are in GBP. (Q50 - 9 answered 4 skipped)

Answer Choices	Number	Overall Percentage
0-3000	3 --- (1 for Resistivity, 1 for Acoustic, 1 for Optical)	33.3%
3000-10000	3 --- (3 for Optical)	33.3%
10000-20000	0	0.0%
20000-50000	1 --- (1 for Video)	11.1%
50000+	2 - (1 for resistivity, 1 for acoustic)	22.2%

24. What non-annual cost are associated with counter operations? (Q51 - 7 answered 6 skipped)

Answer Choices	Number	Overall Percentage
Structure upkeep	1 --- (1 for Optical)	14.3%
Equipment replacement	3 --- (1 for Acoustic, 1 for Optical, 1 for video)	42.9%
Both	3 --- (1 for Resistivity, 2 for Optical)	42.9%