IWRA Proceedings

The agricultural sector could be one of the most vulnerable economic sectors to the impacts of climate change in the coming decades. Climate change impacts on crop production are related to changes in temperature and precipitation patterns, the frequency and magnitude of extreme weather events, and changes in seasonality and growing period, among others. All of these impacts may have consequences on agricultural production (Bates et al., 2008) The assessment of the economic impacts of climate change on the agricultural sector requires an approach aimed to provide a detailed picture of the sector and the relationships within it. In this regard, bottom-up approaches (i.e., in particular models applied at local level, but driven by global forces) could be an effective tool to evaluate the economic impacts of climate change on the agricultural sector. The main objective of this paper is to analyze the economic impacts of changes in yields, due to climate change, on the Chilean agricultural sector. The analysis is conducted using a non-linear agricultural supply model (ASM). The model is designed specifically for the analysis of the Chilean agricultural sector, and it accounts for uncertainty about agricultural yields through the use of Monte Carlo simulations The Agricultural Supply model (ASM) is a mathematical programming model designed to analyze the agricultural sector with high geographical disaggregation. It includes the major agricultural activities within the area, and differentiates between water provision systems (rainfed and irrigated), among other features. The model is calibrated to a single reference period using Positive Mathematical Programming (Howit, 1995). The model reproduces the activity levels observed for the base year and allows us to simulate hypothetical climate change scenarios. The ASM anticipates farmer's responses, in particular changes in cropland allocation and water provision systems, motivated by the differentiated effect of climate change on crop productivity, across crops and across regions. Uncertainty is included in the modeling framework using the Monte Carlo method. In this specific case, the model assumes that the agricultural yields are random variables following a Gamma distribution. Thus, several sets of agricultural yields are simulated using both uniform pseudo-random numbers and the inverse probability distribution function (Hardaker et al., 1997). The area being analyzed here included Regions Atacama in the north to Los Lagos in the south. This area included 265 communes, grouped into 36 provinces, and 10 regions. The agricultural sector was represented by 22 activities, aggregated according to the following categories: Crops (10), Fruits (10), and Forestry (2); the model considers irrigated and rainfed activities, accounting for 3.3 million ha. The crops considered were: rice (irrigated), oats (rainfed), common beans (irrigated), maize (irrigated), potatoes (irrigated and rainfed), alfalfa (irrigated), sugar beet (irrigated), and wheat (irrigated and rainfed). The fruits considered were: cherries, plums, peaches, apples, oranges, walnuts, olives, avocadoes, pears, grapes, and vine grapes, all of them irrigated activities. Finally, the model also included the area devoted to forestry, including: pine and eucalyptus, both rainfed activities. The potential agricultural yields are those computed by Santibañez at al. (2008) At the national level, the expected changes in agricultural yields have a minor impact on the total land allocation, with total agricultural land decreasing by 46600 ha. However, as expected, the estimated impacts across regions are uneven, with the largest impacts in the northern region. Agricultural production suffers from large changes due to the new land allocation across the country, with the largest negative changes faced by grapes (-86%), pears (-54%), and walnut (-38%). On the other hand, most of the increase in production is associated to rainfed activities, such as: oat (125%), potatoes (84%), and wheat (38%). In general, the total agricultural production changes from 10.6 million tons to 10.5 million tons. Results by zone and activity show that the impact on crop production is unevenly distributed across the country, with crop production decreasing by 37% in the Northern zone, while in the Southern zone it increases by 38%. All the changes described above drive a 2.7% decrease in the agricultural net income, from USD 2235 million to USD 2176 million (equivalent to USD 59 million). At the regional level, 6 out of 10 regions show a decrease in net incomes, from Atacama to Maule. Only the regions within the Southern zone could have benefits due to climate change. All the results were presented so far as crisp values, without consideration of probabilities and uncertainty. In order to account for the uncertainty associated to the change in agricultural yields, a series of Monte Carlo simulations were developed. The objective was to determine the probability of a certain income level's occurrence, depending on the yield scenario analyzed. As it was established before, our model assumes that the agricultural yields follow a Gamma distribution. The analysis of the net agricultural income distribution shows that the 25th percentile was USD 627 million, the 50th percentile was USD1155 million, and the 75th percentile was USD2083 million. Considering these figures, the income reported for the climate change scenario, USD2176 million, was above the 75th percentile, thus supporting the robustness of results obtained, even when consideration of yield variability is included in the calculations. In general, the results reported here are consistent with those reported by previous studies for Chile, showing large economic impacts on the northern zone. However, the ASM does not predict large economic consequences at the country level as previous studies did. Previous studies quantified the economic impacts of climate change, under the A2-2040 scenario, with losses between 10% (Bárcena et al., 2009) to 5% (ODEPA, 2010) of the agricultural income, while our results quantified those impacts in -3% of the agricultural income. This difference is related to the methodology used, in which the farmer could reallocate land in order to maximize the net income under different yield conditions. Bates, B., Z. Kundzewicz, S. Wu, and J Palutikof. (eds.) 2008. Climate change and water. 210 p. Technical Paper of the Intergovernmental Panel on Climate Change (IPCC), Geneva, Switzerland. Bárcena, A., A. Prado, J.L. Samaniego, yS. Malchik. 2009. La economía del cambio climático en Chile. Síntesis. Colección Documentos de proyectos, CEPAL. Naciones Unidas, Santiago, Chile. Available at http://www.eclac.cl/publicaciones/xml/8/37858/W288.pdf (accessedAugust, 2013). Gonzalez, J. and Velasco R. Evaluation of the Impacts of Climatic Change on the Economic Value of Land in Agricultural Systems in Chile. 2008. Chilean Journal of Agricultural Reserarch 68: 56-68. Hardaker, J., R. Huirne, and J. Anderson. 1997. Coping with risk in agriculture. CAB International, Oxon, UK. Howitt, R. 1995.Positive mathematical programming. American Journal of Agricultural Economics 77:329-342. ODEPA. 2010. Estimación del impacto socioeconómico del cambio climático en el Sector Silvoagropecuario de Chile. Oficina de Estudios y Políticas Agrarias (ODEPA), Santiago, Chile. Available at http://www.odepa.cl/wp-content/files_mf/1369774423Impacto_socioeconomico_cambio_climatico_sector_silvoagropecuario.pdf (accessed 9 July 2012). Santibáñez, F., P. Santibáñez, R. Cabrera, L. Solís, M. Quiroz, and J. Hernández. 2008. Capítulo I. Resumen Ejecutivo. Impactos productivos en el sector silvoagropecuario de Chile frente a escenarios de cambio climático.In Análisis de vulnerabilidad del sector silvoagropecuario, recursos hídricos, edáficos de Chile frente a escenarios de cambio climático. Centro de Agricultura y Medioambiente (AGRIMED), Facultad de Ciencias Agronómicas, Universidad de Chile, Santiago, Chile.