Development of Fish monitoring (Electrofishing) and Recent

Trends in the Grand Canyon Fish Community 2000-2006

History and development of long-term monitoring

• In 1999 GCMRC recognized the need for a more robust monitoring Program within the Grand Canyon. Experimental flows were being carried out with little power to detect long-term temporal or spatial changes in the overall fish community.

• In 2000, Carl Walters was hired to help develop a long term monitoring program that would better describe the fish communitywithin the Grand Canyon and with power to detect trends within this community.

• Initial analyses of GCMRC fish data suggested that electrofishing(EL) may be most effective in monitoring rainbow trout, brown trout and common carp and that netting (hoop and trammel) may most effectively capture trends in native fish.

• Arizona Game and Fish Department took over the task of developing EL monitoring and SWCA was hired to conduct netting surveys.

Data driven approach to Long-term monitoring design

(how many samples and where??)

• Carl W. Designed a visual basic program (SAMPLE.exe) to best focus our sampling efforts both in terms of sample size and spatial allocation for each of the species of concern. This program was run using data from 1991-1999.

• In 2004 we recreated sample.exe in excel and reanalyzed data from 2000-2004 to refine long-term monitoring.

reach samplesSamples per

nightSamples per

boat days1 48 24 12 22 48 24 12 23 120 24 12 54 48 24 12 25 168 24 12 76 96 24 12 47 96 24 12 48 48 24 12 29 90 30 15 3

10 60 30 15 211 40 20 10 2

sum 862 35

Actual sample design 2005-2006Total effort and allocation of effort by reach are based an data analysis

Reach River miles1 0- 29.12 29.2- 563 56.1 – 68.64 68.7 - 76.75 78.8 -108.56 108.6 -1297 130.5 - 166.68 166.7- 179.59 179.8 - 200

10 200.1 - 22011 220.1 - 225

Bootstrap data 1000 times over varying Ns

CV by sample size for CRP

0.060.070.080.090.1

0.110.120.130.14

400 500 600 700 800sample size

CV

N for each speciesCV = 0.1

BNT 800RBT 225CRP 675

Find CV by species and fish reach and adjust optimum samples per reach by CV and shoreline miles

Reality checkAdjust for logistic feasibility

% Five-year linear detectable change by sample size for CRP

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

35.00%

40.00%

400 500 600 700 800sample size

Perc

ent c

hang

e

Positive

Negative

How did we develop the current sampling strategy? (Sample.exe)

Rainbow troutMean CPUE (fish/hour) for rainbow trout at individual sample sites in the Colorado River from Lees Ferry to Diamond Creek (long-term monitoring trips [2001-2005]). Each point represents an N of 8-15)

0.00

50.00

100.00

150.00

200.00

250.00

0.00 50.00 100.00 150.00 200.00 250.00

fish

/ hou

r

RBT_2001

RBT_2002

RBT_2003

RBT_2004

RBT_2005

River Mile

Brown troutMean CPUE (fish/hour) for brown trout at individual sample sitesin the Colorado River from Lees Ferry to Diamond Creek (long-term monitoring trips [2001-2004]). Each point represents an N of 8-15)

0.00

20.00

40.00

60.00

80.00

100.00

120.00

0.00 50.00 100.00 150.00 200.00 250.00

fish

/ hou

r

BNT_2001

BNT_2002

BNT_2003

BNT_2004

BNT_2005

River Mile

Common carpMean CPUE (fish/hour) for common carp at individual sample sites in the Colorado River from Lees Ferry to Diamond Creek (long-term monitoring trips [2001-2004]). Each point represents an N of 8-15)

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

0.00 50.00 100.00 150.00 200.00 250.00River mile

fish

/ hou

r

CRP_2002

CRP_2003

CRP_2004

CRP_2005

Rainbow troutMean CPUE (fish/hour) in Grand Canyon

(RM 0 – 226, 2000-2006)

YEAR

2006200520042003200220012000

RBT

/Hou

r (95

% C

I)

60

50

40

30

20

10

0

Brown troutMean CPUE (fish/hour) in Grand Canyon

YEAR

2006200520042003200220012000

BN

T/H

our (

95%

CI)

16

14

12

10

8

6

4

2

0

-2

(RM 0 – 226, 2000-2006)

CarpMean CPUE (fish/hour) in Grand Canyon

YEAR

2006200520042003200220012000

CR

P/H

our (

95%

CI)

7

6

5

4

3

2

1

0

(RM 0 – 226, 2000-2006)

Flannelmouth suckerMean CPUE (fish/hour) in Grand Canyon

YEAR

2006200520042003200220012000

FMS

/Hou

r (95

% C

I)14

12

10

8

6

4

2

0

-2

(RM 0 – 226, 2000-2006)

0%

10%

20%

30%

Perc

ent

2000 2001 2002

2003 2004 2005

20060%

10%

20%

30%

Perc

ent

0.00 200.00 400.00 600.00TL

0%

10%

20%

30%

Perc

ent

0.00 200.00 400.00 600.00TL

0.00 200.00 400.00 600.00TL

Length histogramsBy year and percentof total catch

Flannelmouth sucker

1 2 3 4

5 6 7 8

9 10 11

-4.00

0.00

4.00

8.00

12.00

FMS/

hour

-4.00

0.00

4.00

8.00

12.00

FMS/

hour

2000 2001 2002 2003 2004

year

-4.00

0.00

4.00

8.00

12.00

FMS/

hour

2000 2001 2002 2003 2004

year2000 2001 2002 2003 2004

year

Reach River miles1 1-282 29-563 56-684 68-765 76-1086 108-1297 129-1668 166-1799 179-200

10 200-22011 220-225

Mean CPUE (fish/hour) for flannelmouth sucker (by year) in Logistic reaches. (2000-2004)

Flannelmouth sucker

Grand Canyon Colorado River Temperatures (°C) August 1988 to July 2005

-150

153045607590

105120135150165180195210225

Miles from

Lees Ferry

Lees Ferry

Abv LCRBlo LCRPapago

Phantom

Dubendorf

MatKat

National

Parashant

Diamond Gage

YEAR94 95 96 97 98 99 00 01 02 03 04 05

6 °C

7 °C

8 °C

9 °C

10 °C

11 °C

12 °C

13 °C

14 °C

15 °C

16 °C

17 °C

18 °C

19 °C

20 °C

68.0

°F

66.2

°F

64.4

°F

62.6

°F

60.8

°F

59.0

°F

57.2

°F

55.4

°F

53.6

°F

51.8

°F

50.0

°F

48.2

°F

46.4

°F

44.6

°F

42.8

°F

Glen Cyn Dam

Bluehead suckerMean CPUE (fish/hour) in Grand Canyon

YEAR

2006200520042003200220012000

BH

S/H

our (

95%

CI)

3.0

2.5

2.0

1.5

1.0

.5

0.0

-.5

(RM 0 – 226, 2000-2006)

Diamond down (RM 228-260)June-July 2005

FMS HBC SPD BHS CCF CRP FHM MOS RBT RSH SMB STB TFS

Electrofishing 56 0 32 0 8 30 6 0 0 81 3 42 0

Angling 0 0 0 0 80 5 0 0 0 0 0 3 0

Trammel netting 4 0 0 0 24 30 0 0 0 0 0 6 1

Hoop netting 9 1 10 0 1 4 1 0 0 3 1 3 0

Longlines 0 0 0 0 9 0 0 0 0 0 0 3 0

Seining 29 3 5 0 0 0 2 0 0 87 0 3 0

Backpack electrofishing 8 0 49 0 4 13 0 8 0 20 0 0 0

TOTAL 106 4 96 0 126 82 9 8 0 191 4 60 1

BHS = bluehead sucker, CCF = channel catfish, CRP = common carp, FHM = fathead minnow, FMS = flannelmouth sucker, HBC = humpback chub, MOS = mosquitofish, RBT = rainbow trout, RSH = red shiner, SMB = smallmouth bass, SPD = speckled dace, STB = striped bass, and TFS = threadfin shad

RBT (fish / hour) within and outside the LCR depletion area

YEAR

2006200520042003200220012000

RB

T / h

our (

95%

CI)

100

80

60

40

20

0

outside dep

within dep

Examples of output from early attempts at abundance and biomass modelingThis is only an example of the potential. Trends are real, numbers are estimated (some are best

guesses and place holders) and are pending further analysis and external review.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Perc

ent t

otal

abu

ndan

ce

2000 2001 2002 2003 2004 2005 2006

Year

Percent Total abundance by year

BHS FMS CRP BNT RBT

0%

20%

40%

60%

80%

100%

Perc

ent t

otal

mas

s (k

g)

2000 2001 2002 2003 2004 2005 2006

Year

Percent total mass by year

BHS KgFMS KgCRP KgBNT KgRBT Kg

RBT abundance by reach

-50000

0

50000

100000

150000

200000

0 2 4 6 8 10 12

Fish Reach

Abu

ndan

ce

2000

2001

2002

2003

2004

2005

2006

FMS Density (fish / mile) by reach

-500

0

500

1000

1500

2000

2500

3000

0 2 4 6 8 10 12

Fish Reach

Den

sity

(fis

h / m

ile)

2000

2001

2002

2003

2004

2005

2006

Conclusions I• The current electroshocking monitoring design

appears to be appropriate for RBT, BNT, FMS and BHS.

• We need to reevaluate results from CRP monitoring.

• Salmonid densities have experienced dramatic decline over the most recent years while FMS and BHS have increased significantly.

• Recent higher water temperatures are likely partially responsible for both the decline in salmonids and increase in suckers.

Conclusions II• Gear comparisons show that electroshocking may be

appropriate for detection of most non-native warm-water species likely to occur in Grand Canyon.

• Long-term monitoring offers a larger data set to act as a control for more temporally or spatially discrete experiments.

• Catch per unit effort data may be converted to abundance and biomass in future modeling exercises. This will allow us the ability to incorporate the monitoring data into larger models such as carbon budget and ecopath models.

Robust long-term monitoring of aquatic populations is important to adaptive management programs because it

characterizes a “baseline” or antecedent context in which response of biota to changing management policies or

experiments can be interpreted.

(Walters and Holling 1990; Thomas 1996; Walters 1997).

Thank You