Modelling Australian land use competition and ecosystem services with food price feedbacks at high spatial resolution

https://doi.org/10.1016/j.envsoft.2015.03.015Get rights and content

Highlights

  • We modeled Australian land use change and ecosystem service responses to global scenarios.

  • The model features a novel approach to very high resolution land heterogeneity representation.

  • To demonstrate, we model how food price feedbacks of land competition differ spatially.

  • Modest land use change and ecosystem service impacts are observed in aggregate for Australia.

  • High resolution impacts vary from large to minuscule depending on local land heterogeneity.

Abstract

In a globalised world, land use change outlooks are influenced by both locally heterogeneous land attributes and world markets. We demonstrate the importance of high resolution land heterogeneity representation in understanding local impacts of future global scenarios with carbon markets and land competition influencing food prices. A methodologically unique Australian continental model is presented with bottom-up parcel scale granularity in land use change, food, carbon, water, and biodiversity ecosystem service supply determination, and partial equilibrium food price impacts of land competition. We show that food price feedbacks produce modest aggregate national land use and ecosystem service supply changes. However, high resolution results show amplified land use change and ecosystem service impact in some places and muted impacts in other areas relative to national averages. We conclude that fine granularity modelling of geographic diversity produces local land use change and ecosystem service impact insights not discernible with other approaches.

Introduction

With growing economic globalisation, local wellbeing in agricultural regions is increasingly influenced by national, continental and global markets and policies (Meyfroidt et al., 2013). Globalisation can provide great advantages such as better access to food and lower food prices. However, it can also create adverse local economic and environmental impacts. For example, biofuel or carbon sequestration subsidy policy in the USA or Europe can reduce land available for agriculture and this can create incentive to deforest and place land under agriculture in South America and Asia with adverse local water quality, erosion, and global climate regulation impacts (Meyfroidt et al., 2010). Reducing local negatives with minimal loss of macro-scale benefits requires understanding of local environmental, land use and regional development impacts and policies to intervene in national and global context (Renwick et al., 2013). Price feedbacks are a key mechanism through which global land and market policies influence local land use. For example, significant competing demand for agricultural land arising through bioenergy policy has resulted in land competition, reducing food supply and driving up food prices (Wright, 2012). Consequently, there is a growing need for detailed understanding of the multiple global to local scale processes influencing land use change and related ecosystem services and it is essential that such modelling accounts for price feedbacks (Lambin and Meyfroidt, 2011, Rounsevell et al., 2012).

One response has been a proliferation of regional systems models at a spatial resolution commensurate with the heterogeneous factors influencing land use change (LUC) and ecosystem service supply (ES). For example, these models have been used to assess the impacts of prices and policy drivers on land use change and ecosystem services including food, fiber, carbon sequestration, soil health, biodiversity, water resources, and bioenergy (Polasky et al., 2011, Wu et al., 2004, Polglase et al., 2013, Goldstein et al., 2012, Nelson et al., 2009, Bryan et al., 2010, Bryan et al., 2008, Antle et al., 2003, Paterson and Bryan, 2012, Harper et al., 2007). These models typically identify potential land use and ecosystem services changes through bottom-up processes. A range of spatial data and models are typically combined with economic information and behavioural models to quantify land use change and the production of ecosystem services. Global drivers such as climate and market prices are usually considered as exogenous with the justification that change in supply of, say, agricultural commodities, from an individual region is unlikely to affect global market prices (Polasky et al., 2011).

In contrast, consideration of supply and demand dynamics and price interactions is de rigeur in national to global scale partial and general equilibrium models, and integrated assessment models. Partial equilibrium modelling represents market supply, demand, and price interactions with a focus on one or a few specific sectors (e.g. agriculture, forestry, energy). An advantage is the opportunity for detailed production technology representation. A national scale example is the US national forest and agriculture sector optimisation model (FASOM) (Alig et al., 2010, Mccarl and Schneider, 2001), and global scale examples include GLOBIOM (Havlik et al., 2011) and NEXUS (Souty et al., 2012). General equilibrium modelling is the other common method for considering market price feedbacks. All economic sectors are typically included in general equilibrium modelling but more coarsely and with aggregated production functions. An example is Gurgel et al. (2011), who used general equilibrium methods to model price feedbacks through land use competition in global carbon cap and trade scenarios allowing land based carbon sequestration. Integrated assessment models extend general equilibrium models by combining global climate models with global economic models including supply, demand, and price interactions for an array of resources which can include food, energy, water, and carbon. Golub et al. (2013) provides a good example of integrated assessment of global greenhouse gas emissions, energy, forest and agricultural land use markets, and carbon accounts.

Global to national scale land use change models have evolved recently. One trend has been away from very large regions (e.g. 56 for the entire USA in FASOM) toward higher spatial resolution grid representations of land use such as the 0.5 × 0.5° grid cells used in GLOBIOM and 16 km × 16 km cells in continental Europe LUC modelling by Metzger et al. (2006). Global LUC models such as GLOBIOM, GTAP Agro-Ecological Zone (AEZ) (Golub et al., 2009), and NEXUS (Souty et al., 2012) have also evolved to represent land use as shares within spatial units for categories such as crop, pasture, and forestry, and land use intensity levels have been introduced, differentiated by factors such as predominant altitude, slope, and soil properties (Havlik et al., 2011). The most sophisticated models account for market impacts as well as local land heterogeneity with a two-step, top-down process. First, an aggregate quantity of land use change is estimated with non-spatial or coarse resolution partial or general equilibrium models of agricultural and/or forest commodity production levels. Then, land use, consistent with aggregate production determined in the first step, is allocated at higher spatial resolution using a range of techniques such as statistical models based on past land use change and heuristic algorithms to allocate land consistently considering land use regulations and land suitability. Sohl et al. (2012) applied this approach at regional scale to the Great Plains, USA with land use changes at the local level inferred from bottom-up detailed mapping of current land uses, change rates inferred from top-down national to global scenarios, and expert opinion inputs. Britz and Hertel (2011) used global market modelling and detailed land allocation downscaling within the EU. Asselen and Verburg (2013) implemented a similar approach in the CLUMONDO model with global market representation and detailed small grid cell land allocation downscaling. In the downscaling process, both studies combined spatial regression and prioritisation rules to allocate land consistently with land policy and opportunity cost of change (Verburg and Overmars, 2009, Rounsevell et al., 2006, Verburg et al., 2006).

The goals of this article are to: describe a fine resolution integrated land use modelling approach for Australia; investigate how food price feedbacks from land use competition may impact land use change in aggregate and locally at high resolution; and, to assess the significance of food price feedback relative to other significant land use drivers including agricultural productivity, carbon price changes over time and inertia in landholder decisions to convert from agricultural to alternative land uses. The next section describes the Australian continental Land Use Trade-offs (LUTO) model (Bryan et al., 2014b) which is methodologically unique in modelling high resolution land use and ecosystem service processes interacting with macro level impacts of land use competition influencing agricultural commodity prices. Next, LUTO is applied to evaluation of aggregate national and small region local land use change and ecosystem service supply responses to a set of global future outlooks. This is followed by testing of aggregated national and small region local land use change and ecosystem service supply sensitivity to agricultural commodity price feedback effects of competition for land in novel non-agricultural uses such as carbon sequestration. We also compare the magnitude of price feedback effects to effects of other key influential uncertainties in LUTO including global change, agricultural productivity and investment hurdles to land use change adoption behaviour. Discussion and conclusion sections focus on how simultaneous price feedback and high resolution land heterogeneity accounting provides very different small region estimates of land use change and ecosystem service supply than coarser heterogeneity modelling would.

Section snippets

Australian land use change and ecosystem services supply model

The Australian Land Use Trade-offs model (LUTO), shown conceptually in Fig. 1, estimates land use change for the 85.3 Mha intensive agricultural area of Australia (Fig. 2) and supply of five land based ecosystem services: food, carbon, water, energy, and biodiversity. Global integrated assessment modelling of three global outlooks—internally consistent futures for the global climate, population, economy, and greenhouse gas emissions—generated trajectories for climatic conditions, carbon and,

Aggregate land use impact of land use competition food price effects

The area of land under agricultural production is greater with endogenous food price determination, especially under the global outlooks with stronger emissions abatement action (Fig. 5). While some endogenous price feedback impacts are evident in the M3 outlook above a threshold carbon price of about 50 $ tCO2−1, on aggregate this effect is moderate until later years of the L1 global scenario where carbon prices begin to exceed 100 $ tCO2−1. Under L1, price endogeneity culminates in 21%

Discussion

We have addressed the growing need to understand how local land use change and ecosystem service outcomes are influenced by global market forces. The novelty of the analysis is the partial equilibrium modelling with bottom up land use change economics and ecosystem service supply. Previous land use models that account for market price feedbacks and fine resolution heterogeneity (Van Delden et al., 2010, Verburg et al., 2008, Britz et al., 2011, Asselen and Verburg, 2013) involve two steps.

Conclusion

From a land use change modelling perspective, we have illustrated an important general point: it is possible to combine traditional market dynamics with fine resolution economic and environmental heterogeneity in modelling land use change and this is necessary to accurately understand local implications of global change for ecosystem services. For our specific case study, highly differentiated local effects suggest that not including price feedbacks could lead to very flawed local estimates in

References (56)

  • R.J. Harper et al.

    The potential of greenhouse sinks to underwrite improved land management

    Ecol. Eng.

    (2007)
  • P. Havlik et al.

    Global land-use implications of first and second generation biofuel targets

    Energy Policy

    (2011)
  • B.A. Keating et al.

    An overview of APSIM, a model designed for farming systems simulation

    Eur. J. Agron.

    (2003)
  • O. Marinoni et al.

    Development of a system to produce maps of agricultural profit on a continental scale: an example for Australia

    Agric. Syst.

    (2012)
  • M.J. Metzger et al.

    The vulnerability of ecosystem services to land use change

    Agric. Ecosyst. Environ.

    (2006)
  • P. Meyfroidt et al.

    Globalization of land use: distant drivers of land change and geographic displacement of land use

    Curr. Opin. Environ. Sustain.

    (2013)
  • D. Murray-Rust et al.

    Agent-based modelling of land use dynamics and residential quality of life for future scenarios

    Environ. Model. Softw.

    (2013)
  • R.G. Newell et al.

    Climate change and Forest sinks: factors affecting the costs of carbon sequestration

    J. Environ. Econ. Manag.

    (2000)
  • H.M. Pant et al.

    Alternative measures of output in global economic-environmental models: purchasing power parity or market exchange rates? – comment

    Energy Econ.

    (2007)
  • A. Renwick et al.

    Policy reform and agricultural land abandonment in the EU

    Land Use Policy

    (2013)
  • B. Robson

    State of the art in modelling of phosphorus in aquatic systems: review, criticisms and commentary

    Environ. Model. Softw.

    (2014)
  • M.D.A. Rounsevell et al.

    Future environmental change impacts on rural land use and biodiversity: a synthesis of the ACCELERATES project

    Environ. Sci. Policy

    (2006)
  • M.D.A. Rounsevell et al.

    Challenges for land system science

    Land Use Policy

    (2012)
  • T.L. Sohl et al.

    Spatially explicit land-use and land-cover scenarios for the Great Plains of the United States

    Agric. Ecosyst. Environ.

    (2012)
  • H. Van Delden et al.

    Integrated assessment of agricultural policies with dynamic land use change modelling

    Ecol. Model.

    (2010)
  • P.H. Verburg et al.

    Scenario-based studies of future land use in Europe

    Agric. Ecosyst. Environ.

    (2006)
  • ABARES

    Agsurf Farm Survey Database

    (2014)
  • T. Andreyeva et al.

    The impact of food prices on consumption: a systematic review of research on the price elasticity of demand for food

    Am. J. Public Health

    (2010)
  • Cited by (59)

    • A fractional land use change model for ecological applications

      2022, Environmental Modelling and Software
    View all citing articles on Scopus
    View full text