riparian restoration challenges: biocontrol and birds, genetics … · 2018. 4. 19. · riparian...

Riparian restoration challenges: Biocontrol and birds, genetics-based

restoration strategies and an understudied non-native system

Sean M. Mahoney

Department of Biological Sciences, Northern Arizona University

April 9, 2018 Riparian Restoration & Tamarisk Beetle Workshop

Cochise College Benson Center

Changing landscape: How do native species respond?

Tamarisk Defoliated tamarisk

?

?

How do we restore riparian habitats?

Native-dominated

Defoliatedtamarisk

Other players in riparian systems

Russian olive Thorns!

Outline

• Tamarisk biological control

– How do birds respond to tamarisk biocontrol?

• Genetics-based approach to restoration

– How does cottonwood architecture affect tamarisk re-growth?

• Russian olive: “The new tammy”

– What are the bird communities in Russian olive-dominated habitat?

Study Area Virgin River

Mesquite, NV

Beaver Dam, AZ

Desert Springs, AZ

Gold Butte, NV

Big Bend, AZ

Gold butte

(2)

Mormon Mesa

(2)

Big

Bend

(2)

(2)

Native-dominated site (n=3) (~50-75% native)

Native-dominated site Tamarisk-dominated site (n=5)

(~75% tamarisk)

Biocontrol-affected site (n=3) (~90% dead tamarisk)

0

0.1

0.2

0.3

0 0.25 0.5 0.75 1

Enh

ance

d v

eget

atio

n in

dex

(EV

I)

% Biocontrol-affected tamarisk

R2 = 0.56, p = 0.007 (2013/2014)

Decreases in “greenness”

Temperature/humidity decreases

20

25

30

35

25

27

29

0.1 0.3 0.5 0.7 0.9

Rel

ativ

e h

um

idit

y (%

)

Tem

per

atu

re (

°C)

% Biocontrol-affected tamarisk

Temperature: R2 = 0.75, p = 0.007

Relative humidity: R2 = 0.67, p = 0.01

How do birds respond to tamarisk-biocontrol?

• 2 visits in 2013 and 2014 each

• 6 points per site

• 0500-1100, low wind, no rain

• estimated % tamarisk biocontrol-affected

• quantified community assemblages based on presence/absence

• calculated species richness

• modeled bird densities in Distance package (R)

Community assemblages differed overall

-0.1

0

0.1

0.2

-0.1 0 0.1 0.2

NM

DS2

NMDS1

Overall PERMANOVA: p < 0.0001

Native

Dead tam

Live tam

Species richness declines

0

10

20

30

0 0.25 0.5 0.75 1

Mea

n s

pec

ies

rich

nes

s

Biocontrol-affected tamarisk (%)

R2 = 0.41, p < 0.02

Complex responses to biocontrol

0

1

2

3

0 0.25 0.5 0.75 1

Bir

ds/

ha

R2 = 0.65 p < 0.005

0

1

2

3

0 0.25 0.5 0.75 1

R2 = 0.58 p = 0.01

0

1

2

3

0 0.25 0.5 0.75 1% Tamarisk biocontrol-affected

R2 = 0.05 p > 0.5

0

1

2

3

0 0.25 0.5 0.75 1

Bir

ds/

ha

% Tamarisk biocontrol-affected

R2 = 0.05 p > 0.5

% Tamarisk biocontrol-affected

% Tamarisk biocontrol-affected

Birds like weevils, hate beetles

Available Diet

Weevil TLB Larvae L.

hopper Other

Mahoney et al. 2017 Biological

Inv.

Rel

ativ

e av

aila

bili

ty

Management implications

• Birds respond differently to biocontrol

• Nesting behavior may be an effective way predict bird response to biocontrol –Caution: densities do not assess habitat

quality (productivity, food, etc.)

• Stress the need for restoration –Minimize loss of habitat for native species

Part 2: Genetics-based restoration

US Forest Service

Part 2: Genetics-based restoration

2 challenges in restoration

• How do we deal with interspecific competition between native and non-natives? (Sher et al. 2002, Dewine and Cooper 2008) – Architecture is genetics-based (Bailey et al. 2011) – Interaction between native architecture and tamarisk

architecture is not well understood (especially in relatively older trees)

• How do we deal with changing climates? (Tom Whitham’s NAU cottonwood group) – “Genes that are adaptive today and maladapted for

tomorrow”

• So what’s the best stock to select from given these two challenges?

How does cottonwood architecture affect tamarisk re-growth?

• Chevelon Creek common garden/restoration project, Winslow AZ

• Collected and planted cottonwoods from broad elevational/latitudinal/temperature ranges

• 3 years old • Measured: cottonwood

height, canopy width, basal area, stems

• Measured: understory tamarisk height, stems

Mahoney et al. in revision Restoration Ecology

Cottonwoods exhibit different architecture

PERMANOVA: p < 0.001

Mahoney et al. in revision Restoration Ecology

Source population

within 3°C from Chevelon

0

100

200

300

400C

ott

on

wo

od

hei

ght

(cm

)

0

100

200

300

NCC OV SP AL ML

Co

tto

nw

oo

d w

idth

(cm

)

0

25

50

75

Co

tto

nw

oo

d b

asal

are

a (c

m2)

0

2

4

6

8

NCC OV SP AL ML

Co

tto

nw

oo

d s

tem

s

a a

b,c b

c

b b

b

b

a

a

b

c b,c

c

a

a,b b

b,c c

Height Basal Area

Canopy Width Stems

Source pop. Source pop.

Cottonwood architecture affects tamarisk re-growth R2 = 0.27, p < 0.001 R2 = 0.21, p = 0.003

R2 = 0.14, p = 0.02 R2 = 0.2, p = 0.004

Tam

aris

k H

T Ta

mar

isk

stem

s

Cottonwood HT Cottonwood canopy Mahoney et al. in revision Rest. Ecol.

Part 2 Conclusions

• Cottonwood architecture differed by source population at Chevelon Creek

• Cottonwood architecture negatively affected understory tamarisk re-growth – Possibly driven by response to frost

• Therefore, selecting proper cottonwood genotypes are critical for restoration success – Local stock may be best for short-term gains

– Stock within 3°C best for long-term gains

• Introduced in late 1800s (Christiansen 1963).

• Native to Europe and Asia (Little 1961).

• Occurs in riparian areas (Fischer et al. 2012).

• May be replacing native riparian vegetation (Currier 1982).

• 4th most dominant woody plant species in western US riparian areas (Friedman et al. 2005).

• Conservation implications: Riparian habitat is important for birds, mammals, and herps (Finch and Ruggiero 1993).

• Biocontrol agent has been identified and trials are being conducted to test efficacy

Russian olive

Part 3: Russian olive: “The new tammy”

Study sites • Sites varied in amount of Russian olive.

– RO site = ≥75% Russian olive cover/≤25% native cover. – Native site = ≥50% native cover/≤50% Russian olive cover.

• Tamarisk was also present at all sites, but not dominate cover type.

Mixed site (n=10) Russian olive site (n=20)

Presence/Absence surveys

• Conducted bird presence/absence surveys in May, June, July, August

• Presence=Any bird seen or heard by at least one observer

Peter J. Motyka

Willow flycatcher

Peter J. Motyka

Bullock’s oriole

Species richness similar among Russian olive- and native-dominated sites

Number of sites

Rar

efie

d S

pec

ies

rich

nes

s (±

95

% C

I)

Native Russian olive

Mahoney et al. in revision J. Arid Environs

Russian olive-dominated habitats support similar bird communities as native-dominated

PERMANOVA: p = 0.2 Dispersion: p = 0.1

Native Russian olive

Mahoney et al. in revision J. Arid Environs

Russian-olive dominated habitats support similar distribution of foraging

guilds to native-dominated

0

0.25

0.5

0.75

Insectivore Granivore Generalist Driller Frugivore

Perc

en

tage

Dis

trib

uti

on

Native Russian olive

K-S Goodness of Fit test: D = 0.4, p = 0.8

Mahoney et al. in revision J. Arid Environs

Part 3 Conclusions

• Russian olive may provide habitat and food for native species (Olson and Knopf 1986, Fischer et al. 2012).

• Like tamarisk, Russian olive biocontrol may progress faster than restoration – removing habitat and food.

• Russian olive control must be coupled with restoration (Bateman and Paxton 2010).

Final considerations

• Riparian areas are undergoing many changes – Biological control

– Multiple non-native species

• Our understanding of the ecology of some players is not well known

• Effective management should consider response of native species as well as the interaction between genetics and the environment

Acknowledgements • Part 1

– Matthew J. Johnson (NAU CPRS)

– Jennifer A. Holmes (NAU CPRS)

• Part 2

– Jesse B. Mike (NAU SOF)

– Jackie L. Parker (NAU BIO)

– Linda S. Lassiter (NAU BIO)

– Thomas G. Whitham (NAU BIO)

• Part 3

– Anna Nellis B. Smith (CPRS NAU)

– Peter J. Motyka (NAU BIO)

– Erick Lundgren (UT Sydney, AUS)

– Matthew J. Johnson (NAU CPRS)

– Raemy R. Winton (NAU CPRS)

– Bo Stevens (NAU BIO)

Yellow-breasted chat maintains densities after biocontrol

0

0.1

0.2

0.3

0

1

2

3

0 0.25 0.5 0.75 1

Enh

ance

d V

eget

atio

n In

dex

(EV

I)

Bir

d d

ensi

ty (

bir

ds/

Ha)

% Tamarisk cover

R2 = 0.21, p = 0.17