Transcript Can agroforestry reduce risk in subsistence agriculture

Can agroforestry increase reliability of
subsistence agriculture under future
climate in southern Africa?
Amber Kerr
Ph.D. Qualifying Examination
Energy and Resources Group
November 16, 2007
Talk outline
1. Motivation and background
-
Climate change and agroforestry
Why is southern Africa a good case study?
2. Agroforestry systems in southern Africa
3. Research task 1: Field work in Malawi & Zambia
-
Potential field sites
Hypotheses, plot design, and manipulations
Metrics, statistics, possible results
4. Research task 2: Proxy data and modeling
5. Next steps and long-term plan
1 of 38
1. Motivation and background
2 of 38
Climate change in the developing world
“Poorer, developing countries are the least
equipped to adapt to the potential effects of
climate change, although most of them have
played an insignificant role in causing it;
African countries are amongst the poorest.”
~ Pak Sum Low, Climate Change and Africa, 2005
“The international community must urgently
scale up its support to climate-proof the
farming systems of the poor, particularly in
Sub-Saharan Africa.”
~ World Bank, World Development Report 2008
3 of 38
Predicted changes in T and precip, 2080-2099
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Assemblage of 21 climate models using the A1B scenario. (From Figure 11.2
in IPCC AR4 WG1, 2007).
4 of 38
Southern Africa’s vulnerability
“Assessments of water availability, including water
stress and water drainage, show that parts of southern
Africa are highly vulnerable... Food security, already a
humanitarian crisis in the region, is likely to be
further aggravated by climate variability and change.”
~ IPCC AR4 WG 2, Chapter 9, “Africa,” 2007.
“Climate change poses a serious threat to ecosystems
and human well-being in southern Africa, both in
the medium- and long-term... Warming and drying
at the regional scale will have serious consequences
for agricultural production.”
~ Southern African Millennium Ecosystem Assessment, 2004
5 of 38
Central question
Which systems
are most
flexible on a
short time
Which
scale?
Are trees
better
intercrops
than
annuals?
What biophysical
mechanisms are
important?
agroforestry systems
will maximize reliability of
subsistence agriculture under
southern Africa’s likely
future climate?
What can these systems tell
us about agricultural climate
adaptation in general?
What is their
sensitivity to
changes in
climate?
Are these systems
economically, culturally, and
biologically appropriate for
widespread adoption?
6 of 38
Vine support
Homegardens
Alley cropping
Malagasy tavy
Pigs grazing, Spain
Cuban conuco
Swidden
Effet de vent, Monet, 1891
Leucaena and maize
Vanilla in Réunion
Examples of agroforestry
Dehesa
Windbreaks
7 of 38
Taungya
Sugarcane, Australia
Living fences
Parklands
Shade plantations
Baobab, Senegal Shea & cassava, Uganda
Costa Rica, sp. unknown
Riparian buffers
Beehive, Ethiopia
Coffee, Columbia
Mahogany & maize, Indonesia
More examples of agroforestry
Insect husbandry
8 of 38
The intellectual appeal of agroforestry:
Using principles of ecology and
economics to design complex (multispecies) agricultural systems that are
efficient in their resource use,
profitable, reliable, and sustainable.
The practical appeal of agroforestry:
Harnessing the power of different
species combinations to allow farmers
to achieve the greatest benefit from their
limited resources.
9 of 38
Special challenges of agroforestry
The same complexity that can make agroforestry
systems more efficient can also make them:
• Harder to design and model
• More time-consuming to establish
• A slower return on investment
• More difficult to harvest
a dna ™emiTk ciuQ
ro sse rpmo ced )de s serpmo cnU ( FFIT
.erut cip sih t ee s ot dedeen era
• More knowledge-intensive
• More labor-intensive
• More exacting in implementation
Biophysical advantage alone is not enough to spur
adoption of an agroforestry technology.
10 of 38
What does agroforestry have to do with
climate change?
Verchot et al. propose that “agroforestry... has a
role to play in helping smallholder farmers adapt
to climate change.” They suggest that trees in the
agroecosystem may help buffer against both
production risk and income risk.
However: “Questions about the adaptation potential
of agroforestry systems are very important and
poorly studied. There is a great need for studies that
are specifically designed to address this question.”
~ L. Verchot, pers. comm., 2/26/2007
Verchot, L.V.; M. van Noordwijk; S. Kandji; T. Tomich; C. Ong; A. Albrecht; J. Mackensen;
C. Bantilan; and C. Palm. (2007). “Climate change: linking adaptation and mitigation
11 of 38
through agroforestry.” Mitigation and Adaptation Strategies for Global Change 12(5): 901-918.
How agroforestry could contribute to
climate adaptation
• Improve microclimate
– Shade  reduce temperatures
– Reduce wind speed  reduce evapotranspiration
• Improve soil quality
– More soil carbon  greater water-holding capacity
– Nitrogen fixation  reduce multiple stresses
• Access alternate resources
– More efficient interception of rainfall; less runoff
– Hydraulic lift: bring water to surface from depth
– Nutrient pumps: access nutrients at depth
• Provide stable income source
12 of 38
How agroforestry could hinder
climate adaptation
• Tree-crop water competition
– Reduces crop yield, and/or
– Inhibits tree growth, reducing tree benefits
• Lower rates of seedling germination
and establishment
– Reduces returns to labor
• Loss of trees as a long-term investment
– Tree death due to water or temperature stress
– Abandonment of trees due to temporary or
permanent migration
13 of 38
2. Agroforestry in Southern Africa
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
14 of 38
Why is southern Africa a good region to
study climate change and agroforestry?
• Much experience in agroforestry over past
several decades; established research
centers and extension networks
• Existing needs are considerable (e.g. food
security, income generation)
• Climate change impacts expected to be
severe (though the region’s contribution to
climate change has been negligible)
• Climate variability is an ongoing problem
even in the absence of climate change
15 of 38
Agriculture in southern Africa
• Maize is the staple crop (~80% of calories),
supplemented by legumes and tubers
• Soil fertility inherently low, and diminishing;
N limitation widespread
• Rainfall erratic; droughts and floods common
• Farm sizes very small (most are 0.2 - 2 ha)
• Most farmers cannot afford fertilizer
or other external inputs.
• Farmers engage in local cash
transactions, but infrastructure limits
access to broader markets.
16 of 38
Improved fallows
Year 1
Year 4
Year 2
Year 3
(In some systems, it is possible to grow three or more continuous years
of maize without a decline in yields, but two years is more common.)
17 of 38
Relay intercropping
November
September
January
July
March
May
18 of 38
Hedgerow intercropping
November
September
January
July
March
May
19 of 38
Intercropping in Zambia
Photo: ICRAF
20 of 38
Southern Africa agroforestry legumes
Leucaena leucocephala
Gliricidia sepium
Sesbania sesban
Tephrosia vogelii
These species are often
21 of 38
trained as hedgerows. Cajanus cajan
3. Proposed fieldwork
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
22 of 38
Four questions
As climate change is imposed upon southern
Africa’s agroforestry systems,
Q1: Will these systems still
confer yield benefits?
Q2: Will any one system design
outperform the others?
Q3: Will tree-crop water
competition increase?
Q4: Will establishment success
of seedlings decrease?
Outcomes
Mechanisms
23 of 38
Experimental manipulations
• Factors that could affect agroforestry include:
–
–
–
–
Increased average and maximum temperature
Decreased rainfall
Fewer and later rainfall events
Elevated CO2
• Ideal to manipulate all of these over years to
look for long-term and interaction effects.
• However, due to limited time and resources, I
have chosen to focus on rainfall manipulations.
• Rainfall may be the factor with the greatest
effect on agricultural production in this region.
24 of 38
Rainout shelters
Gutters to
intercept and
remove rainfall
Buffer zone
(control for edge
effects, roots, etc.)
Sampled subplot
(usually ~0.5 of
total area)
Reduction in
rainfall should be
proportional to area
covered (in this picture, ~30%)
Rain gauges to
ensure desired
decrease occurs
25 of 38
Treatments and controls
• Treatment 1: Improved fallows
• Treatment 2: Relay intercrop
• Treatment 3: Hedgerow intercrop
• Control 3: Unfertilized maize
• Control 2: Fertilized maize
• Control 1: Annual legume
Each of these treatments will be subjected to
ambient and reduced (-30%) precipitation.
I intend at least 5 replicates per treatment.
26 of 38
Plot design
90 cm
between
rows
90 cm
within
rows
M = maize
T = tree
(absent in
controls)
Whole plot:
9 x 9 grid,
(7.2 m)2,
81 indivs
per species
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
Furrow
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
Ridge
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
M
T
Sampled
subplot:
5 x 6 grid,
(4 m)2,
30 indivs
per species
27 of 38
Data to collect
• Grain yield
• Tree biomass
What
happened?
• Temperature and precipitation
• Evapotranspiration
• Soil moisture
• Soil nutrients and soil carbon
• Below-ground biomass
Why did
it happen?
• Rooting profile
• Germination and establishment
28 of 38
Proposed field sites
Msekera Research Station,
near Chipata, Zambia
Makoka Research Station,
29 of 38
Thondwe, Malawi
4. Proxy data and modeling
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
30 of 38
Proxy data: Interannual variability
• There is an abundance of multi-year yield data
for various agroforestry systems under ambient
precipitation.
• With these existing data, I will attempt a meta-
analysis using interannual variability as a
proxy for the effect of rainfall on the
productivity of agroforestry technologies.
• Another useful question might be to compare
the water-use efficiency of agroforestry systems
with monocultures in dry and wet years.
31 of 38
WaNuLCAS
Water, Nutrient and Light Capture in Agroforestry Systems
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
WaNuLCAS is a
multi-species crop
model that
simulates aboveand belowground
plant architecture,
competitive
interactions, and
environmental
parameters.
Image: Figure 1.2 in WaNuLCAS manual, v. 3.0 (van Noordwijk et al., 2005). Citation:
van Noordwijk, M. and B. Lusiana (1999). “WaNuLCAS, a model of water, nutrient
and light capture in agroforestry systems.” Agroforestry Systems 43: 217-242.
32 of 38
Modeling goals
• Work on extending WaNuLCAS to include
– effects of extended drought
– effects on phenology
• Identify and improve other aspects of the
model that need to be modified to realistically
simulate long-term climate change
• Use empirical data (from fieldwork, and from
interannual variability) to extend the model
• Carry out simulations for fertilizer trees in
Southern Africa: under what conditions and
locations will these systems remain viable?
33 of 38
Uapaca kirkiana, wild loquat
6. Next steps and long-term plan
34 of 38
Next steps
• Finish initial climate change simulations with
WaNuLCAS (December 2007) and submit to
Agroforestry Systems (February 2008)
• Write first chapter of dissertation, reviewing
relevant work, and draft second chapter
analyzing proxy data (Spring 2008)
• Apply for grants and fellowships: Rocca
(February 2008), Switzer (February 2008),
Lindbergh (June 2008), Fulbright-Hays (11/08)
• Use ERG block grant funds for a preliminary
fieldwork trip (Spring/Summer 2008)
35 of 38
Timeline
2008
2009
Fieldwork
scoping trip
QE
First field
season
2010
Second field
season
Modeling and
Greenhouse
proxy work work (optional)
Lit review; initial modeling
2011
Data
analysis
Dissertation writing
36 of 38
Dissertation chapters
1. Introduction: Unanswered questions on
agroforestry and climate change
2. A meta-analysis of agroforestry performance
under current climate variability
3. Methods for fieldwork
4. Results from fieldwork
5. Modifications to, and results from,
WaNuLCAS
6. Synthesis of experimental, proxy, and
modeling results
7. Policy implications; future work needed 37 of 38
Acknowledgements
I am grateful for the generous advice and
support I have received from:
My committee members: Margaret Torn, Dan Kammen, Lynn
Huntsinger, Todd Dawson, and Carol Shennan.
Other faculty members: John Harte, Louise Fortmann, Isha Ray.
Researchers: Louis Verchot, Meine van Noordwijk, Jayant
Sathaye, Asmeret Berhe.
Graduate students: Adam Smith, Barbara Haya, Rob Bailis,
Naïm Dargouth, Tracey Osborne, Dorothy Sirrine, Malini
Ranganathan, Mike Kiparsky, Eric Hallstein, Teresa Chuang,
Erin Conlisk, Danielle Svehla, Abby Swann, Michal Shuldman.
ERG staff: Bette Evans and Donna Bridges.
Friends, family and others: Kay Kerr, Rex Kerr, Jeremy Manson,
April Kerr, Carol Childs, Zachary Mason, Hal Hatch.
38 of 38
Dicrurus adsimilis, drongo
Auxiliary slides follow!
39 of 38
Goals of my dissertation work
• To contribute to the resilience and prosperity of
•
•
•
•
•
smallholder agriculture.
To elucidate mechanisms of interspecific resource
competition in mixed annual/perennial systems.
To further the development of agroforestry
simulation models.
To provide a case study on climate change
adaptation in the agricultural sector.
To gain experience with ecological methods, and
with fieldwork in developing countries.
To prepare for a career in research and teaching.
40 of 38
Not goals of my dissertation work
• Economic, anthropological, sociological, or political
analyses of agroforestry use and adoption.
• An integrated assessment of climate change risks
and adaptation options.
• Agroforestry for climate mitigation (i.e. carbon
sequestration).
• Land surface modeling using remote sensing data.
These topics are all very interesting, worthwhile, and
relevant! But to keep the scope of this project realistic,
I intend to become familiar with them only to the extent
that they are necessary to answer my central question.
41 of 38
Predicted
changes in
soil
moisture,
2080 - 2099
QuickTime™ and a
ompressed) decom
eded to see this pict
(annual
averages)
Assemblage of 21QuickTime™ and a
TIFF
(Uncompressed)
decompressor
climate
models
using
are needed to see this picture.
the A1B scenario.
(From Figure 10.12 in
IPCC AR4 WG1, 2007).
Black dots indicate
lack of data.
42 of 38
Change in agricultural output, 2070-2099
Assuming CO2 fertilization effect
QuickT i me™ and a
T IFF (Uncompressed) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
QuickT i me™ and a
T IFF (Uncompressed) decom pressor
are needed to see this picture.
% change from present
Assuming no CO2 fertilization effect
From Cline, William (2007), Global Warming: Impact Estimates by
Country. Assumes some adaptation and SRES A2 scenario.
43 of 38
Human population density and NDVI
Where do people live who might be most affected by climate change?
44 of 38
Agroforestry defined
The art and science of
growing woody and nonwoody plants together on
the same unit of land for a
range of benefits.
- Huxley, 1999
All practices that involve
a close association of
trees or shrubs with
crops, animals, and/or
pasture.
- Rocheleau et al., 1988
An old and widely practiced
land use system in which
trees are combined spatially
and/or temporally with
agricultural crops and/or
animals.
- Farrell and Altieri, 1995
Agroforestry is an
indistinct concept within
a broader agroecological
context... [it] defies
precise definition.
- Wojtkowski, 1998
45 of 38
Central agroforestry hypothesis
The trees must acquire resources that the crop
would not otherwise acquire.
increase in crop yield
(goal is to be >0)
I = F-C
competitive effect of trees
fertility effect of trees
or, more specifically,
I = Fnoncomp - Ccomp,nonrecycled
non-competitive fertility effect
of trees (resources that would
not otherwise be acquired)
competitive non-recycled effect
of trees (resources taken from crop
and not returned)
Cannell, M. G. R.; M. van Noordwijk and C. K. Ong (1996). “The central agroforestry
hypothesis: the trees must acquire resources that the crop would not otherwise acquire.”
Agroforestry Systems 33: 1-5.
46 of 38
Agroforestry as ecosystem mimicry
In theory, agroforestry systems can confer some of
the benefits of natural ecosystems by mimicking
their structure and function. For example:
• Complementarity of resource use may make
polycultures more productive than monocultures.
• Biodiversity may increase resistance to disease.
• The physical environment may favor functional
traits not usually found in food crops (e.g.
perennial habit, deep-rootedness, deciduousness).
Ecosystem mimicry remains a relatively
unexplored concept!
47 of 38
Diversity, stability, and covariance
• In theory, the yield variance
of a multispecies system
will not exceed the variance
of a single-species system.
Cassava
Maize
• If two species have a
perfectly negative
covariance ( = -1), in
theory, the system
variance can go to zero.
QuickT i me™ and a
T IFF (Uncompressed) decompressor
are needed to see this picture.
From van Noordwijk, M. and C. K. Ong (1999). “Can the ecosystem mimic hypotheses be
applied to farms in African savannahs?” Agroforestry Systems 45(1-3): 131-158.
48 of 38
In summary, agroforestry is...
• A set of land-use systems with potential to
enhance the economic and ecological viability
of agriculture.
• Not a panacaea.
The wrong system in the wrong place can be
disastrous. The right system in the right place
can bring tremendous benefits to smallholders.
49 of 38
Failure of hedgerow intercropping
Calliandra and maize, Katuk-Odeyo, Western Kenya, August 2004.
50 of 38
Unimodal precipitation regimes
Summer
maize
planted
Winter
maize
harvested
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
From Chirwa, P.W.; C.R. Black; C.K. Ong; and J. Maghembe (2006). “Nitrogen
dynamics in cropping systems in southern Malawi containing Gliricidia sepium,
pigeonpea and maize.” Agroforestry Systems 67: 93 - 106.
51 of 38
Agroforestry interventions
• Could agroforestry provide solutions to some
of southern Africa’s agricultural constraints?
• ICRAF has been active in the region since the
mid-1980s, and several national governments
actively support agroforestry research.
• Adoption is increasing but not yet widespread.
• Three of the most widely tested systems are:
– Improved fallows
– Relay intercropping
– Hedgerow intercropping
Main goal: Improve
maize yields through
enhanced soil fertility.
52 of 38
Two slightly different questions:
• Will agroforestry still provide benefits under
future climate?
• Will agroforestry be more or less useful
under future climate than under current
climate?
My data should allow me to investigate both of
these questions.
53 of 38
What does it mean to “increase
climate resilience”?
Under adverse climate conditions, would a resilient
agricultural system have...
• Less variability in yield?
• Higher average yield?
• Higher minimum yield?
I will investigate
these three metrics.
• Faster recovery after a bad year?
Agroforestry trees may also produce minor food
crops or valuable non-food products; it is difficult
to compare these directly with maize yield.
54 of 38
Unintended effects of rainout shelters
• Reduced insolation
• Reduced wind speed
• Increased air temperature
• Changed precipitation chemistry
For these reasons, it will be important to
include equipment control plots: plots over
which similar, but non-rain-capturing,
structures are placed.
from Sala, O. E.; R. B. Jackson; H. A. Mooney and R. W. Howarth, eds. (2000).
Methods in Ecosystem Science. New York: Springer-Verlag.
55 of 38
Improved fallows: methodological dilemma
I would like to perform climate manipulations on
improved fallows, but they require a four-year
minimum to complete a cycle. What should I do?
• Do two years of manipulations on each stage?
• Manipulate only years 1 and 2 (tree establishment)
and measure the biophysical changes?
• Compmletely omit this system from the
experimental design?
56 of 38
Fieldwork: open questions
• If only using one species, what inferences can
•
•
•
•
be made about cropping system?
What conclusions could be drawn about 4-year
improved fallows with only two years of data?
Will the rainout shelters be effective and
affordable?
What is the best way to quantify the degree of
tree-crop water competition?
What would happen if there is significantly
above-average rainfall for both years of the
experiment?
57 of 38
Simulating climate change with
WaNuLCAS
“If you want to use a specific time-series of local
climate change on the model, it will allow you
technically to do so, but processes of tree
mortality under prolonged drought are not
included or tested yet.
“Also, in an agroforestry context we think that
tree phenology (flowering/fruiting) may be
particularly responsive to climate change, and
this aspect will definitely need more attention
before the model will give anywhere near realistic
results.”
~ Meine van Noordwijk, pers. comm., 9/29/07
58 of 38
Image credits
Title slide: Acacia albida growing with sorghum and maize.
Drawing by Terry Hirst from Agroforestry in Dryland Africa, D.
Rocheleau et al., ICRAF, 1988.
Motivation: www.metoffice.gov.uk/research/hadleycentre
Developing world: from www.dkrz.de; see tinyurl.com/24zycc.
Appeal of agroforestry: Rainforest: A.M. Stacey; tinyurl.com/39eqot
Maize: Mrs Banda in Malawi; from ICRAF’s Image Database
59 of 38
Image credits, continued (2)
Types of agroforestry systems:
Dehesa: www.ibergour.com/images/dehesa_extremena.jpg
Vine support: from Vanilla planifolia on en.wikipedia.org
Windbreaks: Poplars - Wind Effect by Monet; worldmasterpieces.jp
Swidden: photos.wildmadagascar.org/Deforestation.html
Alley cropping: from ICRAF, via Cornell Hort 400 webpage
Homegarden: www.bioversityinternational.org (publication 753)
Parkland (shea): www.fao.org/docrep/008/y5918f/y5918f11.htm
Parkland (baobab): edcintl.cr.usgs.gov/senegal2/sine.html
Living fence: Emma Young, tripsource.com. Use by permission.
Riparian buffer: davidwallphoto.com/searchresults.asp?g=85
Taungya: www.fao.org/docrep/008/af335e/af335e03.htm
Coffee: Café Mesa de los Santos from www.new-ventures.org
Beehive: www.bridgebie.org/Village Farm/Village Farm.html
Dead acacia: www.wanderingnomads.com/gallery/bw13.jpg
60 of 38
Image credits, continued (3)
Africa political map: www.mongabay.com/images/african.gif
Africa population and NDVI: CIESIN and NASA.
Southern Africa title: from CTA (2002), “Agroforestry in Malawi
and Zambia: summary report of a CTA/MAFE study visit.”
Southern Africa agroforestry species:
Sesbania: www.css.cornell.edu/ecf3/Web/new/AF
Cajanus: vinehillcashmere.bigblog.com.au
Tephrosia: mybirds.ru/groups/popug/tephrosia.jpg
Leucaena: from forum.ctu.edu.vn; see tinyurl.com/3xgpcp
Leucaena hedgerow: instruct1.cit.cornell.edu/courses/hort400/
Gliricidia: accesscom.com/~jfinger/trees/
Field work title: Photo by the author. Graphic from van Noordwijk,
M. (1999). “Scale effects in crop-fallow rotations.” Agroforestry
Systems 47: 239 - 251.
Zambia and Malawi maps: World Book Multimedia Encyclopedia,
2004 Edition, version 8.2.1.
61 of 38
Image credits, continued (4)
Conclusions: www.vumba-nature.com/habitat-woodlands.htm
Drongo: commons.wikimedia.org
Guava: www.ilivetocook.com/images/cooking/guava.jpg
Quoll (photo and caption): www.epa.qld.gov.au, Bulletin, 9 Nov 04
62 of 38
5. Relevance to theory and policy
63 of 38
Change in global agricultural output (1)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
From Cline, William (2007), Global Warming: Impact Estimates by Country.
Washington, DC: Center for Global Development. Assumes some adaptation and
SRES A2 scenario. These data assume a CO2 fertilization effect.
64 of 38
Change in global agricultural output (2)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
From Cline, William (2007), Global Warming: Impact Estimates by Country.
Washington, DC: Center for Global Development. Assumes some adaptation and
SRES A2 scenario. These data assume no CO2 fertilization effect.
65 of 38
Building blocks of agroforestry
Loose
Stable
66 of 38
Quolls must be passed with care
Ranger Martin Fingland with a juvenile spotted-tailed quoll.
Photo by Anthony Weate, courtesy of The Courier-Mail.
67 of 38