A process-based agricultural model for the irrigated agriculture sector in Alberta, Canada

Connections between land and water, irrigation, agricultural productivity and profitability, policy alternatives, and climate change and variability are complex, poorly understood, and unpredictable. Policy assessment for agriculture presents a large potential for development of broad-based simulation models that can aid assessment and quantification of policy alternatives over longer temporal scales. The Canadian irrigated agriculture sector is concentrated in Alberta, where it represents two thirds of the irrigated land-base in Canada and is the largest consumer of surface water. Despite interest in irrigation expansion, its potential in Alberta is uncertain given a constrained water supply, significant social and economic development and increasing demands for both land and water, and climate change. This paper therefore introduces a system dynamics model as a decision support tool to provide insights into irrigation expansion in Alberta, and into trade-offs and risks associated with that expansion. It is intended to be used by a wide variety of users including researchers, policy analysts and planners, and irrigation managers. A process-based cropping system approach is at the core of the model and uses a water-driven crop growth mechanism described by AquaCrop. The tool goes beyond a representation of crop phenology and cropping systems by permitting assessment and quantification of the broader, long-term consequences of agricultural policies for Alberta's irrigation sector. It also encourages collaboration and provides a degree of transparency that gives confidence in simulation results. The paper focuses on the agricultural component of the systems model, describing the process involved; soil water and nutrients balance, crop growth, and water, temperature, salinity, and nutrients stresses, and how other disciplines can be integrated to account for the effects of interactions and feedbacks in the whole system. In later stages, other components such as livestock production systems and agricultural production economics will be integrated to the agricultural model to make the systems tool. It will capture feedback loops, time delays, and the nonlinearities of the system. Moreover, the model is designed for quick reconfiguration to different regions given parametrized crop data.