Land-use modeling for exploring alternative agricultural futures : linking choices and consequences

The Lower Mainland of British Columbia is the province's most concentrated farming region and contributes about 60% of the provincial gross farm receipts. However, agriculture here is faced with significant environmental, social and economic challenges. Rapidly increasing population and urbanization, coupled with intensification of agriculture, are resulting in landuse shifts that are posing a serious threat to its sustainability. Understanding the dynamics of land-use changes and their future evolution, and quantifying their impacts are central to the debate on sustainability. Land-cover and land-use changes result from complex interactions between environmental and socio-economic factors. Although land-use research commonly has focused on biophysical variables, it is also necessary to understand and incorporate the human factors, particularly the role of human choices and decisions made at individual and institutional levels when predicting land-use changes. In this thesis I develop a model, AgFutures, which links human decision-making with land-use. changes and their consequences. AgFutures is an integrated land-use model that is capable of generating a wide range of possible agricultural futures in response to users' choices related to land-use policy in the region, agricultural management practices, and food consumption preferences. The model forecasts urban and agricultural land-use changes - considering different food production systems comprising extensive cropping, livestock and greenhouse operations - up to 2040, and quantifies their impacts on sustainability, under different userdefined scenarios. It uses a number of social, economic, environmental, agricultural and institutional variables and parameters as input, for projections of land conversions and their impacts. The key novel aspects of this model include: a) the integration of cellular automata-based dynamic spatial simulation methods and interactive multi-criteria decision-making methods with GIS to explicitly address the spatially-dependent multi-scale linkages between land-use changes and the human decision-making process; and b) its design, which represents a balance between the need for rigour and the need for a tool that can be used by a wide array of users. A suite of indicators that measure the impacts of land-use changes are produced and the multidimensional information is presented through an impact matrix, cobwebs and hierarchically aggregated indices to satisfy the different needs for details by different users. Different indices have been developed, an Outcomes Index, a Relative Sustainability Index, and a Relative Benefits Index, all of which are used for quantitative comparison of different scenarios. Four different scenarios - 'Baseline', 'Agribusiness', 'Protectionist', and 'Vegebusiness' - were constructed for the Lower Mainland using this model. Each of these scenarios were developed for two policy options related to preservation of agricultural land - preserve and do not preserve agricultural land. By 2040, due to urbanization, around 3600 hectares of agricultural land will be lost under the preservation of the Agricultural Land Reserve (ALR) scenario, while more than 25000 hectares will be lost in the scenario that does not preserve the ALR. Hot-spots of land-use changes have been identified. There is a positive linear relationship between economic agricultural output and its impacts. As expected, the agribusiness scenario results in the highest impacts on the environment, while the protectionist scenario results in the least impacts. However, the relative benefits of increases in agricultural production over its impacts are highest for the vegebusiness scenario. The benefits are, in general, less for those scenarios which do not preserve agricultural land. The overall impacts on sustainability are less for the protectionist and vegebusiness scenarios than for the baseline and agribusiness scenarios. However, these results change when different weights are applied, indicating the critical role of value judgments of people in derivation of sustainability indices. The model helps understand the complex trade-offs associated with different choices and is a valuable tool for policy planners, stakeholders as well as the general public that assists in making informed land-use decisions for a sustainable future.Arts, Faculty ofGeography, Department ofGraduat