To what extent will climate and land use change affect EU-28 agriculture? A computable general equilibrium analysis Martina Sartori Ca Foscari University, Venice Joint study with Davide Geneletti, Stefano Schiavo and Rocco Scolozzi University of Trento
Motivation EU is one of the largest producers of agricultural goods à 10% of world output, 13% of global cereal production (EUROSTAT, 2015) Agricultural output is highly affected by climate change and land use change (IPCC, 2014; Bindi and Olesen, 2011) High uncertainty about the interactions in the climate-land-food system In addition, local demography and local economy act as exogenous drivers in combination with EU common agricultural policy Most existing research Focus on direct climate impacts on crop yields and agricultural output only (e.g., Jones and Thornton, 2003; Hertel et al., 2010; Ciscar et al., 2011; Lavalle et al., 2010) Study other regions (Schlenker et al., 2005; Deschênes and Greenstone, 2007;) Employ other techniques (van Passel et al., 2016)
The goal We simulate the structural joint consequences of both climate change and land use change on agriculture in the European Union, at the year 2050 First novelty: the combination of different modeling frameworks, each delivering key inputs for and output of the analysis Second novelty: the counterfactual simulations are conducted on a 2050 baseline, where the economic structure of the EU28 is consistent with the population and Gross Domestic Product (GDP) levels projected under the Middle of the Road Shared Socioeconomic Pathway, SSP2
Data sources and models GTAP CGE model EU28 countries, Rest of Europe and Rest of the World 4 agricultural sectors: Rice, Wheat, Cereals and Other crops Temperature projections: GAEZ dataset, HadleyCM3 GCM model, B2 SRES scenario Land-use change projections: LUISA Territorial Modeling platform (Land-Use-based Integrated Sustainability Assessment, Baranzelli et al., 2014, maintained by the JRC) Population and GDP projections: SSP2 (Kriegler et al., 2012; KC and Lutz, 2014; O Neill et al., 2017) from IIASA and OECD
Modeling strategy 2050 SSP2 scenario: GDP and Population projections CGE model baseline economic scenario at 2050 Impact of Climate Change (CC) Impact of Climate Change and Land Use Change (CCLU) % var. in agric. productivity and land availability CGE model Impact of Land Use Change (LU) Impact on agriculture
1. Climate change impact on agriculture The impact of climate change on agricultural productivity is estimated by applying the methodology proposed by Roson and Sartori (2016) Two damage functions relate increase in average temperature to agricultural productivity: One function estimates sectoral productivity variations in the yields of rice, wheat and maize only (IPCC 2014) à applied to Rice, Wheat and the broader category Cereals a second function estimates productivity changes of the aggregated agricultural sector, elaborating on Cline (2007) à applied to an aggregated residual category Other Crops
Predicted variations in average temperature The largest absolute increases will likely be experienced by Northern-Eastern countries (Finland, Sweden, Estonia, Hungary, Latvia, Lithuania) Among the smallest increases there are some Mediterranean and central Europe countries (Greece, Cyprus, Austria and Luxembourg)
1. Climate change impact on agricultural productivity - results Wheat is the most affected crop, with a predicted average decrease of -6.5%, followed by rice (-3.9%) and cereals (-1.2%) The impact on the residual sector other crops is negative for all the countries located in the South, whereas for the remaining countries the impact is small or even positive (0.3% on average) Warmer temperatures turn out to be beneficial especially in the northern countries (Finland, Sweden) and for residual crops category On average, the most affected countries are those located in Central and Eastern Europe, characterized by a continental climate and for which larger rises in mean temperature are predicted
2. Land-use change impact on agriculture Here land-use change means more or less land available for crops cultivation ( changes in the type of crops cultivated!) e.g. less land for agriculture means more land for urban, industrial and other uses The LUISA platform integrates a suite of models, considering the demand and supply of resources and socio-economic activities and infrastructures, and merges both top-down and bottom-up dynamics to simulate land use changes These estimates are available only in terms of generic agriculture land-use class (i.e., are applied uniformly to each crop category)
Actual and estimated changes in agricultural land (km 2, from LUISA platform) - results Largest absolute reductions occur in central Europe (Poland, France and Germany) A larger availability is predicted for North-Eastern countries and (surprisingly) for some Southern economies as well (such as Cyprus, Greece, Portugal and Spain)
80 60 Wheat CC LU UK Sweden Spain Slovenia Slovakia Romania Portugal Poland Netherlands Malta Luxembourg Lithuania Latvia Land Use Change LU Italy Ireland Hungary Greece Germany France Finland Climate change CC Estonia Denmark Czech Rep. Cyprus Croatia Bulgaria 5 Belgium Austria % variations % changes in agricultural output (crop yields) by country RICE Climate Change and Land Use Change CCLU Rice 0 5 10 15 CCLU
8 6 Cereals CC LU UK Sweden LU Spain Slovenia CC Slovakia Romania Wheat Portugal Poland Netherlands Malta Luxembourg Lithuania Latvia Italy Ireland Hungary Greece Germany France Finland Estonia Denmark Czech Rep. Cyprus Croatia Bulgaria Belgium Austria % variations % changes in agricultural output (crop yields) by country WHEAT UK Sweden Spain Slovenia Slovakia Romania Portugal Poland Netherlands Malta Luxembourg Lithuania Latvia Italy Ireland Hungary Greece Germany France Finland Estonia Denmark Czech Rep. Cyprus Croatia Bulgaria Belgium Austria 10 15 CCLU 80 60 40 20 0 20 40 CCLU
UK Sweden LU Spain Slovenia Slovakia CC Romania Portugal Cereals Poland Netherlands Malta Luxembourg Lithuania Latvia Italy Ireland Hungary Greece Germany France Finland Estonia Denmark Czech Rep. Cyprus Croatia Bulgaria Belgium Austria % variations % changes in agricultural output (crop yields) by country CEREALS UK Sweden Spain Slovenia Slovakia Romania Portugal Poland Netherlands Malta Luxembourg Lithuania Latvia Italy Ireland Hungary Greece Germany France Finland Estonia Denmark Czech Rep. Cyprus Croatia Bulgaria Belgium Austria 20 40 CCLU 8 6 4 2 0 2 4 6 8 10 12
% changes in agricultural output (crop yields) by country OTHER CROPS 20 Other Crops CC LU CCLU 15 % variations 10 5 0 5 10 Austria Belgium Bulgaria Croatia Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden UK
Absolute change in agricultural output by crop (M$) CCLU scenario Rice Wheat Cereals OtherCrops TOT Tot prod. in the baseline % Austria -4.6-78.7 10.9-64.2-136.6 6992.1-2.0% Belgium -60.2-153.8-5.2-791.3-1010.5 21897.9-4.6% Bulgaria -17.7-77.8 9.3-31.1-117.3 10171.7-1.2% CroaBa -19.0-18.5 6.3 76.8 45.6 7499.8 0.6% Cyprus -0.4 5.9 0.3-37.3-31.5 1494.0-2.1% CzechRep -1.7-125.2-26.8-118.3-272.0 11407.4-2.4% Denmark -3.2-85.5 7.9 54.5-26.3 8526.0-0.3% Estonia 0.3 45.8 4.9 66.4 117.4 1152.2 10.2% Finland -4.5 186.4 15.8 365.6 563.2 4324.8 13.0% France -82.9-1913.5 56.5-197.6-2137.5 98591.3-2.2% Germany -26.5-745.7-110.7-963.1-1846.0 57592.9-3.2% Greece -5.2 23.0 13.0-1.9 29.0 16906.3 0.2% Hungary 2.6-121.9-224.4-155.0-498.8 14742.3-3.4% Ireland -4.2-10.3 4.1 21.5 11.1 6971.7 0.2% Italy -131.4-217.6 26.1-1112.2-1435.1 73516.9-2.0% Latvia -0.3 4.9 7.2 74.3 86.2 1446.0 6.0% Lithuania -0.5-133.3-3.8-30.2-167.8 5048.5-3.3% Luxembourg -1.5-34.1-5.7-66.2-107.5 1557.9-6.9% Malta 0.0-0.1-0.3-12.3-12.8 309.6-4.1% Netherlands -11.4-44.2 4.0-72.6-124.2 34594.8-0.4% Poland -8.0-458.3-174.2-1335.6-1976.2 40410.5-4.9% Portugal -10.5-1.6-10.6-365.9-388.7 9468.5-4.1% Romania -10.1-1189.7-144.0-105.7-1449.5 49822.6-2.9% Slovakia -0.6-8.8 1.8 14.6 6.9 4834.8 0.1% Slovenia -0.7-7.2-3.7-73.3-85.0 2999.3-2.8% Spain -66.5-87.1-35.9-1335.8-1525.4 80401.8-1.9% Sweden 0.0 255.0 59.2 1543.1 1857.3 12269.4 15.1% UK -76.3-581.0-12.2-258.0-927.5 38559.9-2.4%
Impact on agriculture summing up The overall impact on crop yield may vary substantially across countries -38% in Luxembourg (driven by land-use change), +65% in Finland (driven by climate change) Climate change drives the results in some countries, land-use change in others The expected variations in the yield of the various crop categories (with the exception of rice) are qualitatively similar within the same country with some exceptions agricultural production in Sweden, Finland and Estonia is expected to grow, regardless the crop type under consideration; the reverse holds true for Hungary, Germany, Spain, Slovenia, UK and Belgium In Italy, France and Cyprus changes in crop productivity are positive for some product categories and negative for others
Impact on agriculture summing up Consistently with the literature (e.g., Alexandrov et al., 2002; Ewert et al., 2005; Audsley et al., 2006; Olesen et al., 2007; Richter and Semenov, 2005): climate-related increases in crop yields are expected above all in northern European countries Unlike other studies: among the countries expected to suffer the largest reductions in wheat and cereals yields we find non-mediterranean countries (Germany, Poland, Hungary, Belgium and Luxembourg) The negative impact is often driven by changes in land use, indeed these countries are likely to experience the largest percentage reductions in land available for agriculture
% variations in national income 0,15 CC LU CCLU % variations 0,10 0,05 0,00 0,05 0,10 0,15 0,20 Austria Belgium Bulgaria Croatia Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden UK
Take Home Messages - 1 Climate and land-use change are likely to affect agricultural systems very differently across Europe Northern countries are expected to benefit from climate change impacts, whereas other areas in Europe will suffer negative consequences in terms of reduced agricultural output, real income and welfare The most vulnerable region will not be the Mediterranean Europe, but central Europe: Poland, Romania, Bulgaria, Hungary à negative impacts are majorly driven by land use change Southern Europe is negatively affected, and impacts are driven by climate change Small impacts on national income à These outcomes may exacerbate existing regional difference across EU28 countries
Take Home Messages - 2 Adaptation strategies could, at a cost, effectively curb some negative effects of climate change or exploit positive effects, thus reducing regional disparities In our simulation exercises we deliberately abstract from any adaptation/mitigation policy In this respect, improving the understanding of the potential effects of climate change on crop yields is central to design appropriate adaptation strategies, and current European agricultural policies need to be reformed in a way that encourage flexible land use and crop production
Thank you Martina Sartori martina.sartori@unive.it