Evaluating policies for sustainable land use: a sub-regional model with farm types in Costa Rica

An interdisciplinary methodology for land use analysis at the sub-regional level is presented using data for the Neguev settlement in the perhumid tropical lowlands of Costa Rica. A linear programming (LP) model is employed for the maximisation of farm household income, given a flexible set of resource and sustainability related criteria. A large number of different land use systems with fixed input-output coefficients, developed on the basis of crop growth simulation and expert systems, are offered to the LP model for the optimisation of land use in terms of crop selection and technology choice. A geographical information system is used for the presentation of the results. The matrix of the LP model includes five sub-matrices, each encompassing a different farm type. Farm types are distinguished on the basis of their land-labour ratios and the availability of three different soil groups. Sustainability is explicitly taken into account through estimating environmental damage in terms of soil nutrient depletion using nutrient balances (for N, P and K) and through an index for biocide use. Different land use scenarios for the effects on land use of a number of policy interventions including changes in output, input and factor prices, capital availability, and regulatory measures, are analysed. For a number of these policy instruments, the results indicate trade-offs between sustainability and income objectives. The methodology, which is currently being upscaled towards the regional level, provides a valuable tool for ex ante assessment of different land use options and should become an integrated part of policy design and execution