Mutual analyses of agriculture land use and transportation networks: The future location of soybean and corn production in Brazil.

Abstract

Abstract CONTEXT Given the need to expand food production to satisfy an increasing world population; regional land-use planners must direct this expansion to available, permitted areas (i.e., not legally protected forests) where production can be maximized and transportation costs, production costs and environmental impacts can be minimized. In this balancing act, the capabilities and layout of the current and future transport system are crucial in determining the spatial distribution of future agricultural production. OBJECTIVE The objective of the present study was to develop a modeling framework to determine the optimal spatial distribution of new soybean and corn production areas in Brazil and transport flows needed to move these commodities to maximize economic gains and limit or reduce negative environmental impacts. METHODS Initially, we applied Data Envelopment Analysis (DEA) to classify regions suitable for expanding soybean and corn production according to an area s potential for superior agricultural performance. Subsequently, we used a Network Equilibrium Model (NEM) to determine the best spatial distribution of new soybean and corn production areas so that their production added to existing production meets the estimated 2050 demand and the optimal configuration of existing and proposed transportation infrastructure required to move this production to appropriate locations. RESULTS AND CONCLUSIONS When compared to a 2018 baseline, the complete operation of planned railroads and the positioning of future grain production to occupy regions indicated by the modeling could increase soybean and corn production by 66% without simultaneous deforestation, reduce carbon dioxide emissions due to interregional soybean and corn transport by about 14%, and reduce combined transport and production costs by about 2%. SIGNIFICANCE The proposed modeling framework should be an extremely useful tool when determining sites for future agriculture production and the network matrix needed to transport this production, especially when considering the relationship between land use and transportation infrastructure.