Improvement of Meadow Stream Health due to Livestock Distribution Efforts K.W. Tate, T.A. Becchetii, C. Battaglia, N.K. McDougald, D.F. Lile, H.A. George,

Improvement of Meadow Stream Health due to Livestock Distribution Efforts

K.W. Tate, T.A. Becchetii, C. Battaglia, N.K. McDougald, D.F. Lile,

H.A. George, and D.L. Lancaster

University of California

• Grazed streams across CA range

from excellent to poor health.

• What practices are associated with

excellent to poor health?

• Conduct a cross-sectional survey of

health and management of grazed

stream systems.

Grazing – Stream Health

Initial Survey of 128 Rangeland Streams

• Summer of 2000 and 2001.

• Cross-section of the grazed

rangeland stream population.

• Cascades, Sierra Nevada, and

Coast Range.

• Gradient of stream health and

grazing management conditions.

Grazing ManagementEPA – CDFG Stream Health Assessment

stability

fish habitat

macroinverts.

Overall Health Score

0 – 5 poor

6 – 10 marginal

11 – 15 suboptimal

16 – 20 optimal

Off-site water, herding, season, frequency, fencing, etc.

Site Characteristics

Elevation, channel slope, substrate, watershed disturb., past disturb., etc.

• Regression analysis – correlate

stream health score (0-20) to

grazing management variables

• Analyzed by substrate type:

bedrock/boulder, cobble/gravel,

fines/alluvial

• Associations v. Cause-Effect

Grazing and Stream Health Associations

Grazing – Stream Health Score Correlations: Meadow Streams

+Time maintaining off-stream

attractants (days/yr).

+Herding to reduce time near

stream (days/yr).

– Cattle density (AU/ac) .

– Frequency (times/yr).

Follow-up Survey of 58 Streams

• Summer of 2003 and 2004.

• Collect stream macroinvertebrate

assemblages.

• Meadow streams (Rosgen C and E).

• Cascades and Sierra Nevada

• 3,500 and 8,500 ft elevation.

• Current management in place >10 yr.

Macroinvertebrates

• Integrate stream conditions and

disturbance over space and

time.

• EPA – CDF&G protocols

– 3 transects per 100m reach, d-

ring kick-net, 0.25 m2 sample

area, 3 minute sample effort

• ID to genus – species

Sub-sample edge and mid stream substrates at 3 transects each stream

Site Summary

Min. Mean Max.

Elevation (ft) 3,400 5,900 8,500

Pasture / Allotment Area (ac)

800 18,000 105,000

Herd Size (AU) 10 150 670

Off-site Attractants (d/yr)

0 3 34

Total Dist. Effort (d/yr)

0 9 41

Substrate Type

(% sites)

Fines Gravel Cobble

53 27 20

Data Analysis

• Negative binomial regression (back-step).

• Count-based data, skewed towards low values, etc.

• Dependents:

– 11 macroinvertebrate metrics (11 nbreg models)

– no. taxa, no. stoneflies, %EPT, etc.

• Independents:

– Grazing: dist. effort, rest, AUM/ac/yr, freq., etc.

– Substrate: fines, gravel, cobble

• Cluster Variable: stream reach

• 11 metrics significantly assoc. with total dist. effort

(p<0.10).

• No other grazing variables were significant (p>0.10).

• Sensitive metrics increased as dist. effort increased.

• Substrate a major determinant of 10 metrics.

Select Metrics Dist. Effort (d/yr)

Substrate

(cobble to fines)

No. EPT taxa + (p=0.01) – (p=0.07)

No. Total Taxa + (p<0.01) – (p=0.02)

No. Intolerant Taxa + (p<0.01) – (p=0.02)

% Intolerant Taxa + (p=0.06) – (p=0.06)

% Dominant Taxa – (p=0.03) n.s.(p>0.10)

Livestock Distribution Effort (days/yr)

0 10 20 30 40

Num

ber

of T

axa

15

20

25

30

CobbleGravelFines

Total Taxa

Total livestock distribution effort v. richness for

streams with fine, gravel, and cobble substrate.

Streambed Substarte

Cobble Gravel Fines

Nu

mb

er

of T

axa

0

5

10

15

20

25

30

Total Taxa

EPT Taxa

Intolerant Taxa

Sensitive metrics decreases with substrate size.

Must be accounted for in cross stream comparisons.

• Distribution effort to reduce time livestock spend

near-stream is positively associated with stream

health.

• Relevant to mixed conifer – meadow landscapes,

where distribution is inherently a problem.

You got to put in the time to see the benefit…

Get along little doggie!