Land Use Dynamics Simulator (LUDAS)
The Land Use Dynamics Simulator (LUDAS) is a spatially-explicit, integrated biophysical-socio-economic modelling approach that aims to predict the longer-term impacts of land use/cover changes under different management/planning/policy scenarios. It has been described by Le at al. (2008) and by Le et al. (2010) and applied in Vietnam and Ghana at watershed level.
The approach taken by LUDAS for achieving these aims is to capture the complexities of coupled human-environment processes through multi-agent simulation processes that mimic causal mechanisms, feedback loops, and interactions within the human-landscape system.
LUDAS uses two types of agents: the ‘household agent’, which represents the socio-economic state and changes of the population, and the ‘landscape agent’, which models biophysical changes of the landscape through changes in land use/cover (LUCC). The household agent is the smallest entity within a human community, at which the integration of household, environmental and policy information leads to decision-making regarding land use. The landscape agent is the ‘land unit’, and represents the smallest spatial entity, at which natural processes and changes in local conditions can be identified. Both types of agents are subject/may respond to eventual land-use related policies and policy changes. Feedback loops connect agents and policies.
The household agent has two components: (1) information on the household state, developed from a database of household profiles and perceived spatial organization, and (2) a decision-making sub-model that takes in information contained in the landscape agent, particularly with regard to benefits obtained from the land and actions needed to maintain or modify its state, as well as changes in the policy environment. These data can be obtained through a representative sample of interviews, based on a locally relevant household stratification.
The landscape agent contains spatial information on the biophysical state of the land, and can incorporate diverse biophysical sub-models, such as soil erosion or crop production models, that help to quantify the yield/benefit from different crop options and maintenance costs.
By parameterizing the decision-making model for local land use and socio-economic conditions the impact of specific policy interventions on land use decisions can be simulated. System dynamics simulation software, such as NetLogo, offers a multi-agent programmable modeling environment that provides libraries of decision models that can be adapted to local conditions.
Simulation modeling, as implemented in LUDAS, can be a valuable tool to understand the dynamics of LUCC and the associated driving forces, analyze their impact at different time steps, and contribute to design appropriate policies for sustainable management of land and water resources. Given the complexity of the possible interactions between diverse land users, biophysical environments and policy platforms, and associated uncertainties the LUDAS approach is applicable only at relatively small scale, e.g. a watershed or district, and requires very substantial data collection.