Elsevier

Agricultural Systems

Volume 150, January 2017, Pages 99-108
Agricultural Systems

Climate change impacts and farm-level adaptation: Economic analysis of a mixed cropping–livestock system

https://doi.org/10.1016/j.agsy.2016.10.013Get rights and content

Highlights

  • Whole-farm analysis of climate change impacts on a mixed cropping–livestock system.

  • Impacts varied. In general, potential losses are much greater than potential gains.

  • Benefits of adaptation were substantial, but in adverse scenarios costs are still high.

  • Not allowing for adaptation inflated the cost of climate change by 15–35%.

  • Yield or price increases could offset much or all of the impact of climate change.

Abstract

The effects of climate change on agricultural profitability depend not just on changes in production, but also on how farming systems are adapted to suit the new climatic conditions. We investigated the interaction between production changes, adaptation and farm profits for a mixed livestock–cropping farming system in the Western Australian Wheatbelt. Crop and pasture production was simulated for a range of plausible rainfall, temperature and CO2 concentrations for 2030 and 2050. We incorporated the results of these simulations into a whole-farm bio-economic optimisation model. Across a range of climate scenarios, the impact on farm profit varied between − 103% and + 56% of current profitability in 2030, and − 181% and + 76% for 2050. In the majority of scenarios profitability decreased, and the magnitude of impacts in negative scenarios was greater than the upside in positive scenarios. Profit margins were much more sensitive to climate change than production levels (e.g., yields). Adaptive changes to farm production under extreme climate scenarios included reductions in crop inputs and animal numbers and, to a lesser extent, land-use change. The whole-farm benefits of these adaptations were up to $176,000/year, demonstrating that estimating the impact of climate change without allowing for adaptation can substantially inflate costs. However, even with adaptation, profit reductions under the more negative scenarios remained large. Nevertheless, except for the most extreme/adverse circumstances, relatively minor increases in yields or prices would be sufficient to counteract the financial impacts of climate change (although if these price and/or productivity increases would also have occurred without climate change then the actual cost of climate change may still be high).

Introduction

The effect that climate change has on the productivity and economic viability of agriculture will depend on how much it is possible to adapt to reduce the change's impact (Lobell, 2014). Therefore, estimates of the economic impact of climate change will likely be overstated if adaptation is not allowed for. Nonetheless, in many existing projections of climate change impacts adaptation is not considered (White et al., 2011).

We investigate the impact of climate change, allowing for adaptation, in the Wheatbelt region of Western Australia. In this region the agricultural growing season is limited by moisture availability and as the region is predicted to warm and dry with climate change (e.g., Moise and Hudson, 2008, Turner et al., 2011) the dryland agriculture practiced there is potentially vulnerable. Climate change may already be affecting the region: average growing-season rainfall (May to October) has declined by more than 10% since the 1970s (Ludwig et al., 2009). Interestingly, despite this, farms in the region experienced high yield and productivity growth in the 1980s and 1990s (Islam et al., 2014). However, more recently, average yields appear to have stabilised (Stephens et al., 2012, Turner et al., 2011).

Studies of the economic impacts of climate change that incorporate agricultural adaptation need to encompass: (a) the impacts of climate change on the production of outputs in various possible production systems, and (b) an economic assessment of the impact of these production changes and the options for adaptation that are available to the farmer. Aspect (a) is often addressed using detailed plant and/or animal simulation models, and there have been a number of studies of this type for the case-study region (Anwar et al., 2015, Asseng et al., 2004, Asseng and Pannell, 2013, Farre and Foster, 2010, Ludwig and Asseng, 2006, Ludwig et al., 2009, Moore and Ghahramani, 2013, van Ittersum et al., 2003).

Aspect (b) has been much less thoroughly researched for the study area. There are two main approaches that can be used to investigate it. The first is to identify packages of adaptations that are of interest and then simulate the economic consequences of each package (e.g., Crimp et al., 2012, Ghahramani et al., 2015). An advantage of this approach is that the modeller has complete control over which adaptations are simulated, allowing transparent analysis of particular strategies that are of interest. Deciding which packages of adaptations to simulate can be problematic though (White et al., 2011), particularly in complex mixed farming systems such as those found in the case-study region. The modeller may not be able to anticipate which of the many potential combinations of adaptations are most likely to be worth assessing.

The second approach is to use optimisation to automatically assess all of the available combinations of adaptations. The obvious advantage is avoiding the need for numerous simulations to identify the adaptations that best meet the farmers' economic objectives (Klein et al., 2013). However, the analysis may be less transparent than under the simulation approach, and the objective function used in the optimisation model may not match that of all farmers.

In this study, we utilise process-based simulation models for the first phase, and extensively modify an existing bioeconomic whole-farm optimisation model for the second. We judged that the very large number of production options available in our case-study region means that the advantages of the optimisation approach outweigh its disadvantages. Also, previous analyses of climate change impacts on the case-study region have tended to consider impacts on a solitary crop or enterprise in isolation. Our use of a whole-farm model allows the simultaneous consideration of impacts on all elements of a typical farming-system in the region. Amongst other things, this allows adaptation in the form of changing land use to be represented in our study (Reidsma et al., 2015).

Our aim is to explore potential impacts of future climate change on production and profitability in the West Australian Wheatbelt. Specifically we address the following questions: 1) What is the impact on farm production and profits under a range of realistic climate scenarios over the next 15 to 35 years?; 2) Which currently available adaptations are most effective in moderating any adverse effects or exploiting positive effects, and to what extent do they improve farm profits?; Finally, 3) What increase in prices or yields would be needed to maintain profits equivalent to the no-climate-change scenario?

Section snippets

Study area

The Western Australian Wheatbelt region accounts for approximately 40% of the wheat and 11% of the wool exported by Australia (around 5% and 7% of the wheat and wool traded internationally—ABARES, 2013). Our study area is the central part of this Wheatbelt region, around the township of Cunderdin (Fig. 1). This area has a Mediterranean-type climate with long, hot and dry summers and cool, moist winters. Historically annual rainfall is between 330 and 400 mm, approximately 75% of which falls

Impact of climate change on profitability

The analysis indicates that farm profitability is sensitive to changes in annual rainfall, temperature and CO2 even after allowing for the most beneficial adaptations (Fig. 2, Fig. 3). Of the 36 scenarios selected to represent the range of possible circumstances for 2030 (Fig. 2), six result in profit increasing by more than 10% relative to the base case, four give profits within 10% of the base case, and 26 result in profits falling by more than 10%. The potential for losses is much greater

Discussion

Given the high level of uncertainty about the details of future climate change, the plausible range of financial outcomes for farmers in the case-study area is very wide. In both the medium term (2030) and the longer term (2050), financial outcomes from the modelled scenarios range from moderately positive to highly negative. Results suggest that the more extreme climate scenarios would likely see sizeable reductions in the economic activity generated by agriculture in the study area. Though

Conclusions

Our estimation of climate-change impacts at the system/whole-farm level is unlike most analyses that instead focus on a single crop or enterprise, and thereby ignore the interactions between the various enterprises that can have a large impact on the performance and make-up of a farming system. Unlike some studies, we also allowed for adaptation with existing management strategies when projecting climate impacts, showing that failing to allow for this adaptation would exaggerate estimates of

Acknowledgements

We are grateful to John Finlayson for his contribution at the early stages of this analysis, particularly model development. We thank the University of Western Australia, Grains Research and Development Corporation, Future Farm Industries CRC and Centre of Excellence for Environmental Decisions for their funding toward this research. We also thank Marta Monjardino and Roger Lawes for providing feedback on draft versions of this manuscript.

The views Donkor Addai expresses in this article do not

References (54)

  • D.B. Lobell

    Climate change adaptation in crop production: beware of illusions

    Glob. Food Sec.

    (2014)
  • F. Ludwig et al.

    Climate change impacts on wheat production in a Mediterranean environment in Western Australia

    Agric. Syst.

    (2006)
  • M. Monjardino et al.

    The potential contribution of forage shrubs to economic returns and environmental management in Australian dryland agricultural systems

    Agric. Syst.

    (2010)
  • A.D. Moore et al.

    GRAZPLAN: decision support systems for Australian grazing enterprises. III. Pasture growth and soil moisture submodels, and the GrassGro DSS

    Agric. Syst.

    (1997)
  • D.A. Morrison et al.

    A mathematical programming model of a crop–livestock farm system

    Agric. Syst.

    (1986)
  • E.I. Teixeira et al.

    Global hot-spots of heat stress on agricultural crops due to climate change

    Agric. For. Meteorol.

    (2013)
  • M.K. van Ittersum et al.

    Sensitivity of productivity and deep drainage of wheat cropping systems in a Mediterranean environment to changes in CO2, temperature and precipitation

    Agric. Ecosyst. Environ.

    (2003)
  • J.W. White et al.

    Methodologies for simulating impacts of climate change on crop production

    Field Crop. Res.

    (2011)
  • ABARES

    Agricultural Commodity Statistics 2013

    (2013)
  • R.G. Allen et al.

    Crop Evapotranspiration—Guidelines for Computing Crop Water Requirements—FAO Irrigation and Drainage Paper 56

    (1998)
  • S. Asseng et al.

    Adapting dryland agriculture to climate change: farming implications and research and development needs in Western Australia

    Clim. Chang.

    (2013)
  • S. Asseng et al.

    Uncertainty in simulating wheat yields under climate change

    Nat. Clim. Chang.

    (2013)
  • S. Asseng et al.

    Rising temperatures reduce global wheat production

    Nat. Clim. Chang.

    (2015)
  • K.J. Boote et al.

    Putting mechanisms into crop production models

    Plant Cell Environ.

    (2013)
  • S.J. Crimp et al.

    Developing climate change resilient cropping and mixed cropping/grazing businesses in Australia

  • T.L. Delworth et al.

    Regional rainfall decline in Australia attributed to anthropogenic greenhouse gases and ozone levels

    Nat. Geosci.

    (2014)
  • G.J. Doole et al.

    Labour scarcity restricts the potential scale of grazed perennial plants in the Western Australian wheatbelt

    Anim. Prod. Sci.

    (2009)
  • Cited by (0)

    View full text