Enhancing the Brazilian land use module in Aglink-Cosimo

The primary objective of this report is to document the improvements made to the Brazilian module of the Aglink-Cosimo model (Enciso et al., 2015; OECD-FAO, 2015) related to the inclusion of land use dynamics. Aglink-Cosimo is an agricultural partial equilibrium model developed by the Organisation for Economic Co-operation and Development (OECD) and the Food and Agriculture Organization of the United Nations (FAO) in partnership with some member countries of these organisations. It is a partial equilibrium model used to simulate the annual developments of supply, demand and prices for the main agricultural commodities produced and traded worldwide over a 10- year period. It is a recursive-dynamic model, since current economic decisions are reached by taking into account lagged information on prices and quantities. Aglink-Cosimo is central to the integrated Modelling Platform for Agro-economic Commodity and Policy Analysis (iMAP) of the Joint Research Centre (JRC) of the European Commission. Its main objectives are the implementation of quantitative tools for commodity market and agricultural policy analysis, and the preparation of a benchmark for the assessment of agricultural policies in a medium- to long-term horizon (so-called 'baseline'). Models in iMAP are used in stand-alone mode or in combination, to address a broad range of topics linked to the economic assessment of agricultural and rural development policies, as well as those concerning related topics such as trade, energy, the environment and climate change. The model improvements to the Brazilian module reported here are important to assess the effects of domestic policy changes on the economic performance of the Brazilian agricultural sector, but also to analyse how agricultural production and land use in Brazil would respond to changes in world market commodity prices. The utilised agricultural area in Brazil has increased in recent years, mainly driven by increases in multiple cropping and livestock density on pastureland. Here we propose a methodology to capture the multi-cropping aspect of Brazilian agricultural in Aglink-Cosimo and expand the crop coverage of the model to also include pasture land and beans. Moreover, to capture the changing livestock density on pastureland, a beef cow intensity function (i.e. cow inventory divided by land use) is estimated using historical data. Last but not least, the movement and allocation of crops within Brazil are discussed in relation to their dist