10.7 Case Study 6: Agent-Based Modeling of Agricultural
Technology Adoption in a Northern Thai Watershed
10.7.1 Agent-Based Modeling of Technology Diffusion
The diffusion of technologies is widely recognized to be an important driver of land
use change in the mountainous areas of Southeast Asia, as well as elsewhere.
9
However, very few models of land use change have taken this driver explicitly into
account. The aim of this study was to address this gap using a land use simulation
model that combined whole-farm mathematical programming to simulate the
economic decision-making of farmers, with an agent based approach that individually represented each farmer and each plot in the model. Through this individual
representation, the model represented real-world heterogeneity much more rigorously than conventional models.
We used the MP-MAS software framework developed at the University of
Hohenheim in order to understand how agricultural technologies, market dynamics,
environmental change and policy interventions affected the economic and biophysical sustainability of a heterogeneous landscape and the population of farm
households (Schreinemachers and Berger 2011). The MP-MAS software had previously been applied with case studies in Chile (Berger 2001; Berger et al. 2007),
Uganda (Schreinemachers et al. 2007), Thailand (Schreinemachers et al. 2009) and
Vietnam (Marohn et al. in press). In this study we used MP-MAS to ex-ante assess
the adoption of agricultural innovations, and the impact of such innovations on
various economic and biophysical indicators.
10.7.2 Background and Objectives
Litchi is the major tree crop grown in the mountainous parts of northern Thailand,
but farm gate prices have declined significantly in recent years, mostly because
growth in the supply of litchi has exceeded the growth in demand for the fruit, as
illustrated in Fig. 10.18. Farmers in some areas have; therefore, reduced orchard
management or substituted more profitable crops for litchi trees. Although this is a
logical adjustment for market-oriented farmers, the substitution of vegetables for
fruit trees has raised environmental concerns, as seasonal crops require more
intense tillage, can worsen soil erosion, use greater amounts of agrochemicals,
and intensify the run-off from hillsides – a factor linked to the flooding of lowland
areas (Turkelboom et al. 1997; Delang 2002; Sidle et al. 2005). Scientists and
extension workers have stressed the importance of keeping hillsides covered with
9 This section was written by Pepijn Schreinemachers, Chakrit Potchanasin and Thomas Berger,
and draws on Schreinemachers et al. (2010).
10 Integrated Modeling of Agricultural Systems in Mountainous Areas
413
Technology Adoption in a Northern Thai Watershed
10.7.1 Agent-Based Modeling of Technology Diffusion
The diffusion of technologies is widely recognized to be an important driver of land
use change in the mountainous areas of Southeast Asia, as well as elsewhere.
9
However, very few models of land use change have taken this driver explicitly into
account. The aim of this study was to address this gap using a land use simulation
model that combined whole-farm mathematical programming to simulate the
economic decision-making of farmers, with an agent based approach that individually represented each farmer and each plot in the model. Through this individual
representation, the model represented real-world heterogeneity much more rigorously than conventional models.
We used the MP-MAS software framework developed at the University of
Hohenheim in order to understand how agricultural technologies, market dynamics,
environmental change and policy interventions affected the economic and biophysical sustainability of a heterogeneous landscape and the population of farm
households (Schreinemachers and Berger 2011). The MP-MAS software had previously been applied with case studies in Chile (Berger 2001; Berger et al. 2007),
Uganda (Schreinemachers et al. 2007), Thailand (Schreinemachers et al. 2009) and
Vietnam (Marohn et al. in press). In this study we used MP-MAS to ex-ante assess
the adoption of agricultural innovations, and the impact of such innovations on
various economic and biophysical indicators.
10.7.2 Background and Objectives
Litchi is the major tree crop grown in the mountainous parts of northern Thailand,
but farm gate prices have declined significantly in recent years, mostly because
growth in the supply of litchi has exceeded the growth in demand for the fruit, as
illustrated in Fig. 10.18. Farmers in some areas have; therefore, reduced orchard
management or substituted more profitable crops for litchi trees. Although this is a
logical adjustment for market-oriented farmers, the substitution of vegetables for
fruit trees has raised environmental concerns, as seasonal crops require more
intense tillage, can worsen soil erosion, use greater amounts of agrochemicals,
and intensify the run-off from hillsides – a factor linked to the flooding of lowland
areas (Turkelboom et al. 1997; Delang 2002; Sidle et al. 2005). Scientists and
extension workers have stressed the importance of keeping hillsides covered with
9 This section was written by Pepijn Schreinemachers, Chakrit Potchanasin and Thomas Berger,
and draws on Schreinemachers et al. (2010).
10 Integrated Modeling of Agricultural Systems in Mountainous Areas
413