Assessing multi-criteria approaches with environmental, economic and social attributes, weights and procedures: A case study in the Pampas, Argentina

Soil erosion control is a major issue in agriculture. The no till system of soybean production has been widely adopted; however, soil erosion may increase due to pasture and forestry land convertion to cropland in the marginal Pampas, Argentina. The aim of this paper is to assess the conflicts and trade-off among environmental, economic and social interests by using three continuous multi-criteria approaches and a set of different weights. Different land uses, crops, pastures, forestry and soil and water conservation practices at the basin scale in the marginal Pampas were assessed. The basin (423 km²) was discretised into 176 sub-basins to focus the management strategy on 5th and 6th order streams. Minimum basic information was obtained using intensive field observations and satellite images. The basin hydrology, soil erosion, sediment delivery and vegetated filter strip models and GIS were used to quantify the technical coefficients. Thirteen decision factors and six criteria (peak run-off, annual erosion, sediment, investment, gross margin and employment) were used in the optimisation trials. Weighted goal programming, lexicographic goal programming, compromise programming and a sensitivity analysis of weights were performed. The results showed a high impact of soil and water management practices on the environmental factors and a strong conflict between environmental and economic interests. The three multi-criteria approaches also showed that it is possible to obtain a good level of goal achievement with different plans. These plans should include: soil conservation practices; crop rotation (of a 2:1 soybean-to-corn ratio); gully and channel erosion control; regulation ponds; and pasture and agro-forestry areas. This goal achievement is mainly limited by public and private investment. Sensitivity analysis of the decision-maker weights shows differences among the values achieved by the criteria and their trade-off. Thus, the weight value of each criterium should be supported by the negotiation-consensus process.