Habitat restoration in degraded rural streams: The Granite Creeks Project

Nick Bond, P.S. Lake & Alena GlaisterCRC for Freshwater Ecology &

School of Biological Sciences, Monash University

The granite creeks system

• Sand slug formation in the early1900’s

• Agents of major geomorphic change and habitat loss

• Major ecosystem changes - especially fish communities

Degradation by sand slugs

• Greatly reduced habitat diversity

• Loss of stable substrates (burial of large timber)

• Decreased retention of organic material.

• potential habitat and energy limitations on populations.

Decreases in geomorphic complexity

0 200 400 600 800 1000

0.0

0.5

1.0

0.0

0.5

1.0

Distance (m)

Dep

th (m

)

Clay

Sand

N

GoulburnRiver

0 5 km

Castle

CreightonsPranjipMuddy W

aterhole

AVENEL

Strathbogie

Ranges

Creek

HUME FREEWAY

Worm

angal

EUROACreek

Creek

Creek

Creek

Sand slug locations

• A dramatic, and widespread formof stream degradaton

Sand Slugs

Research program

• Project established against a strong backdrop of earlier work on the ecology and geomorphology of these streams. (Nick O’Connor, Jenny Davis, Brian Finlayson, Barbara Downes et al.)

• A multidisciplinary project.– Geomorphology (Dan Borg & Ian Rutherfurd)– Fish & Invertebrates (Bond, Lake and Glaister)– Metabolic processes (Bonnie Atkinson, Mike Grace &

Darren Baldwin)– Nutrient cycling (Kellie Vanderkruk)– Genetics and connectivity among populations (Ben Cook)– Large-scale disturbances & refugia (Bond & George Perry)

• Focus for todays talk is on localised faunal response to timber, and the meaning of these results from a streamscapeperspective.

Granite Creeks Project overview

• The project has centered around a manipulative experiment, in which timber structures were added to sites on 2 streams.

• Control, 1-structure and 4-structure siteshave been monitored over time.

AIMS:– To test the “field of dreams” hypothesis, which

underpins much stream restoration work, especially the reintroduction of timber.

– To bring together local and regional factors in understanding habitat-biota relationships in the context of stream restoration

100m

N

Goulburn River

0 5 km

Castle

CreightonsPranjipMuddy W

aterhole

AVENEL

Strathbogie

Ranges

Creek

HUME FREEWAY

Worm

angal

EUROACreek

Creek

Creek

Creek

control1 sleeper4 sleepers

Site locations

Timber addition –geomorphic change

• Sleeper addition caused scour pool development at most sites

• Most scour pools were dynamic – infilled at some lower lows but re-scoured again at high flows.

• Scour and fill patterns unpredictable in space and time

• Pools generally were smaller and less persistent than predicted from flume experiments

0 1 4No. Structures

0

10

20

30

40

50

No.

fish

/ si

te

G. olidus

0 1 4No. Structures

0

10

20

30

40

50

No.

fish

/ si

te

G. olidus

Habitat creation and fish response

• But, small scour pools complemented by debris build-up and the creation of cover.

• Resulted in a positive response by G. olidus and G. marmoratus

• No colonisation by exotics such as carp.

See Bond & Lake (in press) restoration ecology

Colonisation of introduced red gum

• 20 week colonisation exp.

• Rapid colonisation of algae (diatoms and blue-greens) closely tracked by invertebrates

• Some evidence of nutrient limitation in Castle Creek.

• Loss of algae due to summer drying

• Positive GPP on redgumsubstrates*

*Whole-stream GPP strongly negative (B. Atkinson) – ie. li it d l l d ti

0 4 8 12 16 200

5

10

15

20

25

No

taxa

/ sam

ple

Invertebrate richness

Colonisation of introduced red gum

• 20 week colonisation exp.

• Rapid colonisation of algae (diatoms and blue-greens) closely tracked by invertebrates

• Some evidence of nutrient limitation in Castle Creek.

• Loss of algae due to summer drying

• Positive GPP on redgumsubstrates*

*Whole-stream GPP strongly negative (B. Atkinson)

0 4 8 12 16 20

10

100

1000

No

indi

vidu

als

/ sam

ple

invertebrate abundance

0 4 8 12 16 200

20

40

60

80

100

120

µg-C

hl-a

/ cm

2 Chlorophyll-a

0 4No. Structures

0

5000

10000

15000

Abu

ndan

ce (p

er s

ite)

Before - Dec 2000

0 430

40

50

60

70

80

90

Taxo

n ric

hnes

s

After - Dec 2001

Response by benthic fauna

No. Structures

Summary of local response

• Population increases for two fish species –mountain galaxias and river blackfish

• A weak response by benthic invertebrates around timber structures.

• Rapid colonisation of timber by algae and invertebrates.

• Increase in algal production in a system that is otherwise strongly heterotrophic.

Drought and stream drying

Diversity of refuge habitats

Loss of refuge habitats

139sanded

325210Clay

Total No. Fish

Habitat area (m2)

Creightons Creek 20 November 2002

Creightons Creek 22 November 2002

<2<5<10<20>20

Pool Volume (Ml) 500mN

Habitat loss and fragmentation -5/12/02

Key

<2<5<10<20>20

<2<5<10<20>20

Pool Volume (Ml)

Newly split pools

500mN

17/12/02

Key

<2<5<10<20>20

Newly split pools

New rain-filled pools

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

3/1/03

<2<5<10<20>20

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

15/1/03

<2<5<10<20>20

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

27/1/03

<2<5<10<20>20

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

11/2/03

<2<5<10<20>20

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

23/2/03

<2<5<10<20>20

<2<5<10<20>20

Key

Pool Volume (Ml) 500mN

7/3/03

Castle Ck

distance (km)

Freq

uenc

y

0 2 4 6 8

02

46

810

12

• ••••

•••

•

• ••••••••

3 0 3 6

Refugia at the landscape scale

km

Dispersal patterns of fish

• Genetic and stable isotope data used to look at dispersal by Ben Cook, Griffith University– Both techniques show strong population differentiation even

within creeks, indicating very limited dispersal.

• Implications for response to restoration and use of refugia.

Castle Creek

0.0

5.0

10.0

15.0

-35.0 -33.0 -31.0 -29.0 -27.0 -25.0

C13

N15

Pranjip Road

Bloss

Telfords Bridge

Local manipulations in a landscape context.

• Drought dramatically overrode the positive short-term response.

• Creation of refuge habitats a possible future target?

• Lack of dispersal runs counter to dominant belief about connectivity.

• Are many localised manipulations better than a small number of large ones?– Providing critical thresholds (e.g. permanent

water) are crossed.

Conclusions

– Local fish populations responded positively when habitat was created by sleeper addition.

– Sleeper addition did not always create new habitat – both temporally and spatially variable.

– Drought and lack of water a major constraint on the likely success of habitat manipulations.

– Dispersal may be much more limited than one might expect – constrains population recovery rates.

– Refuge habitats of critical importance in these streams - often threatened by water extraction & stock access.

Future Research

• Continue macroinvertebrate and fish monitoring; both restoration and drought recovery.

• Assess responses to restoration of solute retention capacity, POM storage and retention, metabolism (production + respiration), microbial diversity and DOC processing.

• Modelling of future restoration strategies—scaling-up of structures, restoring riparian sustainability.