Trade-off analysis in the Northern Andes to study the dynamics in agricultural land use J.J. Stoorvogel a, * , J.M. Antle b , C.C. Crissman c a Laboratory of Soil Science, Department of Environmental Sciences, Wageningen University, Duivendaal 10, P.O. Box 37, 6700 AA Wageningen, Netherlands b Department of Ag. Economics and Economics, Montana State University, 312 Linfield Hall, 59717 Bozeman, MT, USA c International Potato Center, P.O. Box 25171, Nairobi, Kenya Received 25 August 2003; revised 4 March 2004; accepted 10 March 2004 Abstract In this paper we hypothesize that land use change can be induced by non-linearities and thresholds in production systems that impact farmers’ decision making. Tradeoffs between environmental and economic indicators is a useful way to represent dynamic properties of agricultural systems. The Tradeoff Analysis (TOA) System is software designed to implement the integrated analysis of tradeoffs in agricultural systems. The TOA methodology is based on spatially explicit econometric simulation models linked to spatially referenced bio-physical simulation models to simulate land use and input decisions. The methodology has been applied for the potato-pasture production system in the Ecuadorian Andes. The land use change literature often describes non-linearity in land use change as a result of sudden changes in the political (e.g. new agricultural policies) or environmental setting (e.g. earthquakes). However, less attention has been paid to the non-linearities in production systems and their consequences for land use change. In this paper, we use the TOA system to study agricultural land use dynamics and to find the underlying processes for non-linearities. Results show that the sources of non-linearities are in the properties of bio-physical processes and in the decision making-process of farmers. q 2004 Elsevier Ltd. All rights reserved. Keywords: Tradeoff analysis; Agricultural land use change; Ecuador 1. Introduction The agricultural sector has to deal with rapidly increasing population pressures, with highly dynamic world markets, with a changing biophysical environment through land degradation and climate change, with rapid technological advancements, and with many changes in the political arena through, e.g. the introduction of structural adjustment programs. All these factors influence, directly or indirectly, the land allocation and land management decisions by farmers. As a result, land use is highly dynamic and rarely reaches equilibrium. Despite the complexity and variability of land use change, it is extremely important to get insight in possible trajectories of land use change and the driving factors behind these changes. Only then can we evaluate the need for intervention and the feasibility and impact of new agricultural technologies and policies. This explains the increasing research activities in the field of land use and land cover change (Veldkamp and Lambin, 2001). Land use data often show that changes do not occur linearly but gradual changes are punctuated with rapid changes. In some cases, these rapid changes may occur as a result of easily explainable factors like the introduction of certain agricultural policies, technological innovations or natural disasters. In other cases, however, these changes and their causes are not obvious. Reasons can be found in, e.g. gradual changes in soil quality as a result of soil degradation or prices. Non-linear relationships between, e.g. soil quality and crop choice may subsequently cause a rapid change in land use patterns. In this paper we will study the relationships and thresholds in production systems that may lead to non-linear or abrupt changes in agricultural land use by impacting the decision making processes of farmers. Different tools have been developed to analyze land use dynamics varying from statistical approaches to 0301-4797/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvman.2004.03.012 Journal of Environmental Management 72 (2004) 23–33 www.elsevier.com/locate/jenvman * Corresponding author. Tel.: þ 31-317-484043; fax: þ 31-317-482419. E-mail addresses: jetse.stoorvogel@wur.nl (J.J. Stoorvogel), jantle@ montana.edu (J.M. Antle), c.crissman@cgiar.org (C.C. Crissman).