Optimizing land use distribution to minimize nutrient depletion: a case study for the Atlantic Zone of Costa Rica

Abstract

Nutrient depletion may be reduced by effective land use planning using a nutrient balance model and linear programming techniques to optimize the distribution of land use over different land units. This technique plans the geographical distribution of land utilization types and creates a more sustainable basis for agriculture in an area. In contrast with traditional land use planning where land utilization types are matched with land units on the basis of maximizing present agricultural production, this approach focuses on long-term effects. A case study was carried out in an area of 3340 ha in the southern part of a settlement scheme in the Atlantic Zone of Costa Rica. Land units vary between young fertile volcanic soils and relatively old, leached unfertile soils. Land utilization types vary between extensively grazed pasture and intensively cultivated annual and perennial crops. Information on land units with its land qualities and on land utilization types was stored in a geographical information system which was integrated with a nutrient balance model (NUTBAL). This model, initially developed for Kenya and calibrated for local conditions in Costa Rica, was based on separate assessments of 5 nutrient input factors: mineral fertilizers, organic manure, wet and dry deposition, nitrogen fixation and sedimentation, and 5 nutrient output factors: harvested product, crop residue removal, leaching, denitrification and erosion. For every combination of land unit and land utilization type the nutrient balance was modelled. With a linear programming model, the land utilization types were distributed over the different land units, such that nutrient depletion was minimized in the area within a set of boundary conditions. Two different scenarios were elaborated. One in which all land utilization types were redistributed and one in which the location of forest areas was fixed at their present location. The optimized land use distribution based on current land use types was compared with the actual distribution which had an average nutrient depletion of 22, 5 and 13 kg ha −1 yr −1 for respectively N, P and K. There was little difference in nutrient depletion between the two scenarios. The first scenario resulted in a nutrient depletion of 18, 3 and 9 kg ha −1 yr −1 and the second one in 19, 3 and 10 kg ha −1 yr −1 for N, P and K respectively. However, there was a large difference with the actual situation on one hand and the two scenarios on the other. In both cases there was a slight increase in the estimated agricultural production. The introduction of a nutrient depletion model in land use planning seems to constitute a useful complement to the existing procedures.