The vulnerability of Australian rural communities to
climate risk
P. Kokic, philip.kokic@csiro.au, CSIRO,
R. Nelson, S. Crimp, S. M. Howden and H. Meinke
ABSTRACT
Vulnerability is a term used frequently to describe the potential threat to rural communities
posed by climate risk. Despite a growing use of the term, analytical measures of vulnerability
that are useful for prioritising and evaluating policy responses are yet to be formally
developed. The demand for research capable of prioritising adaptation responses has grown
rapidly with the increasing awareness of climate change and its current and potential future
impacts on rural communities. Whilst much of the resulting research has been hazard or
impact modelling, we argue that these approaches alone can lead to entirely erroneous
conclusions about the vulnerability of rural communities, with potential to significantly
misdirect policy interventions. In addition, an approach that relies entirely on a biophysical
assessment of impact does not provide an assessment that enables targeted policy responses.
In this paper we summarise research (Nelson et al., 2010) which shows how hazard/impact
modelling can be combined with an holistic measure of adaptive capacity to analyse the
vulnerability of Australian rural communities to climate variability and change. Bioeconomic
modelling was used to explore the exposure of Australian rural communities to climate risk,
whilst a rural livelihood approach was used as a conceptual framework to construct a metric
of adaptive capacity.
Key words: Adaptive capacity, Bioeconomic modelling
1. Introduction
There is growing interest in understanding the future challenges resulting from humaninduced climate change to rural communities in order to prioritise strategic adaptation
responses. The type of science traditionally used to inform rural adaptation in Australia and
elsewhere has been based on a long heritage of biophysical hazard/impact modelling (Cline,
2007; Nelson et al., 2010). This mode of analysis has resulted in vulnerability analysis being
highly disciplinary-specific with little ability to inform the emergent properties of
vulnerability. We consider that there is an urgent need to combine hazard/impact modelling
with methods capable of identifying and enhancing diverse sources of adaptive capacity and
present a framework to undertake such analysis.
2. Impact modelling of climate risk
The observed exposure of rural communities to climate variability was measured using a
coefficient of variation for each of annual rainfall, simulated pasture growth and historical
farm income data over the 10-year period from 1996–97 to 2005–06 (Figure 1). Rainfall
variability was calculated using historical climate data from the Australian Bureau of
Meteorology. The Aussie GRASS model (Carter et al., 2000) was used to simulate annual
pasture growth across Australia. Aussie GRASS integrates the impact of climate with regional
1
differences in soils, pasture types and livestock management. The combined
influence of agricultural input and output prices, farm management, climate and
other drivers of physical and economic productivity were captured through
historical farm income data collected in the Australian Agricultural and Grazing Industry
Survey (AAGIS) (ABARE, 2003).
Meinke et al., (2006) provided evidence that Australian rural communities most exposed to
climate variability are often also highly adapted to it. This is confirmed in Figure 1, which
shows that the rural communities that have experienced the most variable rainfall and pasture
growth are not necessarily those that have experienced the most variable farm incomes. This
provides tangible evidence that farmers in regions with severe climate variability are not
necessarily most vulnerable to current climate risk as they have developed appropriate
farming systems to manage this variability. Figure 1 also demonstrates how misleading it can
be to substitute or confuse hazard/impact modelling with more integrated approaches to
vulnerability assessment when even highly integrated biophysical measures of exposure, such
as simulated pasture growth, may provide few insights into the adaptive capacity of rural
communities.
Figure 1: The exposure of Australian broadacre farm households to climate variability over the 10 years
from 1996–97 to 2005–06 measured using historical data for rainfall (left), pasture growth (centre) and
farm incomes (right)
3. Adaptive capacity using rural livelihood analysis
Moving beyond biophysical impact modelling to understanding the adaptive capacity of rural
communities requires a process which enables one to compare and contrast multiple sources
of information and combine them into a formal measure. The rural livelihoods framework
(Ellis, 2000) was used as the conceptual framework underpinning the construction of an
adaptive capacity index. This framework conceptualises adaptive capacity as an emergent
property of the diverse forms of human, social, natural, physical and financial capital from
which rural livelihoods are derived. Farm households with a greater diversity of assets and
activities are likely to have greater adaptive capacity because of a greater capacity to
substitute between alternative livelihood strategies in response to external pressures.
A composite index of adaptive capacity was constructed using data provided by farmers
through AAGIS (Nelson et al. 2010). Three variables were selected to represent each of the
five capitals suggested by Ellis (2000), in order to preserve their statistical relevance and
transparency of interpretation (Table 1). Conceptually important dimensions of the five
capitals were supplemented using data from the Australian Bureau of Statistics and the
National Land & Water Resources Audit. The three variables selected were weighted to form
2
a farm level measure of each capital type. These measures were then weighted
together to form an overall index of adaptive capacity.
Table 1: Variables selected to measure adaptive capacity. Extracted from Table A4, Nelson et al. (2010)
Operator education
Spouse education
Self-assessed health
Landcare membership
Number of business partners
Internet usage
Pasture growth index
Dams per hectare
Remnant vegetation
Plant and machinery
Structures
Livestock
Capital
Total cash income
Access to finance
Human capital
Ordinal variable measuring education level
Same as above but for the operator’s spouse
Ordinal variable measuring the operator’s level of health
Social capital
An indicator of the bridging form of social capital
An indicator of the bonding form of social capital
An indicator for the linking form of social capital
Natural capital
Average pasture productivity from the Aussie GRASS model
A measure of the potential of the farm to access stored water
Ordinal variable measuring future productivity and ecosystem services
Physical capital
Value-weighted Fisher index of the quantity of plant and machinery
The corresponding Fisher index for fixed structures on the farm
Fisher index of the number of livestock held during the year
Financial capital
The total capital value of the farm including the value of leased land
Total farm and off-farm income to meet living and other expenses
Access to credit plus liquid assets
4. Vulnerability to climate variability and change
Vulnerability to climate risk was assessed by combining the theory of hazard/impact
modelling with adaptive capacity assessment. The potential impact on Australian rural
communities from climate change was measured using models to project expected changes in
rainfall, pasture growth and farm incomes to 2030. Nelson et al., (2010) used the MPIECHAM5 model (Roeckner et al., 2003) to project annual changes in climate expected by
2030 under the A1FI emission scenario as this model and scenario best represents the
observed changes in Australia (Nelson et al., 2010). Following the approach outlined by
Crimp et al. (2002) these projections of future climate were then used to project annual
changes in pasture growth with the Aussie GRASS model, and changes in farm incomes using
the AgFIRM bioeconomic model (Kokic et al., 2007).
Figure 2: The current vulnerability of Australian rural communities to climate variability, with respect to
pasture growth variability (left), and to farm income variability (right)
3
The current exposure and adaptive capacity were separately classified into three
ordinal categories using the 10th and 25th percentiles of each variable as cutoffs.
Rural communities vulnerable to climate variability and change were defined as
those for whom high exposure coincides with low or moderate adaptive capacity, and vice
versa. Even if biophysical impact modelling is integrated with adaptive capacity, the resulting
analysis of vulnerability can be misleading. Figure 2 shows the stark differences in
vulnerability assessed using biophysical impact modelling of pasture growth (left), compared
to the broader risk-encompassing economic measure of farm incomes (right). Confining the
analysis to the impacts on pasture growth would lead to a conclusion that inland Australia is
most vulnerable due to high exposure to a variable climate, and low adaptive capacity (Figure
2, left). Assessments of vulnerability based solely on climate impacts such as rainfall (Figure
1, left) would be even more misleading. When farm incomes are used as a more integrative
measure of exposure to climate variability, the spatial vulnerability of rural communities
becomes considerably more complex (Figure 2, right).
5. Conclusion
In this paper we have demonstrated, to have any potential for policy relevance, hazard/impact
modelling needs to be combined with holistic measures of adaptive capacity to provide
insights into the emergent dimensions of vulnerability.
References
ABARE, 2003. Australian Farm Survey Report. Australian Bureau of Agriculture and Resource Economics.
Carter, J., Hall, W., Brook, K., McKeon, G., Day, K., Paull, C., 2000. Aussie GRASS: Australian grassland and
rangeland assessment by spatial simulation. In: Hammer, G.L.,Nicholls, N., Mitchell, C. (Eds.), Applications of
Seasonal Climate Forecasting in Agricultural and Natural Ecosystems—the Australian Experience. Atmospheric
and Oceanographic Sciences Library, vol. 21. Kluwer Academic Publishers, Dordrecht.
Cline, W., 2007. Global Warming and Agriculture: Impact Estimates by Country, Peterson Institute for
International Economics, ISBN paper 978-0-88132-403-7.
Crimp, S., Flood, N., Carter, J., Conroy, J., McKeon, G., 2002. Evaluation of the Potential Impacts of Climate
Change on Native Pasture Production—Implications for livestock Carrying Capacity. Final Report for the
Australian Greenhouse Office, 60 pp.
Ellis, F., 2000. Rural Livelihoods and Diversity in Developing Countries. Oxford University Press, Oxford.
Kokic, P., Nelson, R., Meinke, H., Potgieter, A., Carter, J., 2007. From rainfall to farm incomes—transforming
advice for Australian drought policy: Part I. Development and testing of a bioeconomic modelling system.
Australian Journal of Agricultural Research 58 (10), 993–1003.
Meinke, H., Nelson, R., Kokic, P., Stone, R., Selvaraju, Baethgen, W., 2006. Actionable climate knowledge—
from analysis to synthesis. Climate Research 33, 101–110.
Nelson, R., Kokic, P., Crimp, S., Martin, P., Meinke, H., Howden, S.M., de Voil, P., Nidumolu, U., 2010. The
vulnerability of Australian rural communities to climate variability and change: Part II--Integrating impacts with
adaptive capacity. Environmental Science & Policy 13, 18-27.
Roeckner, E., Bauml, G., Bonaventura, L., Brokopf, R., Esch, M., Giorgetta, M., Hagemann, S., Kirchner, I.,
Kornblueh, L., Manzini, E., Rhodin, A., Schlese, U., Schulzweida, U., Tompkins, A., 2003. The atmospheric
general circulation model ECHAM5, Part I. Model description. In: Tech. Rep. 349, Max-Planck-Institute for
Meteorology, Hamburg, Germany.
4