Today’s water resources face intense challenges. Continued population growth and urbanization will lead to further increases in demand for high quality water. Meanwhile, in drier regions the risk of temporary and long-term drought is increasing and water levels from over-drafted groundwater basins continue to fall. Existing supplies and systems will be taxed as the competition for water increases.
In an attempt to address these challenges, centralized institutional governance arrangements generate policies aimed at improving the state of our water resources. For instance, British Columbia’s Living Water Smart Plan commits to increasing water efficiency by 33 percent, developing new approaches to water management to safeguard against increased drought risk and climate change, and developing conservation initiatives to meet fifty percent of new municipal water needs by 2020. These types of goals are often developed in consultation with experts using water models that overemphasize reductionist science to solve water supply issues, infrastructure issues, or demand allocation problems. These models operate inside a range of known conditions and are unable to make very accurate predictions beyond what is revealed by actual water data. Such constraints limit the models’ ability to adapt in the face of uncertainty, as well as their ability to represent the complete possible range of water user preferences and behaviours in response to policy options. The centralized approach of this type of governance often means analytical support systems fail to include local knowledge and preferences. Complexity arises from the continual need to negotiate tradeoffs between conflicting resource uses and changing interests, which are poorly represented by obsolete policies unable to endure changing circumstances. In short, centralized institutional governance does not properly address the real-world of natural resource governance, which requires consideration of human-environment systems that are inherently complex and unpredictable.
Alternatively, adaptive governance systems accommodate behavioural complexities and recognize the importance of integrating local contexts. Adaptive governance systems are capable of incorporating context specific knowledge and preferences into policy goals, readily navigating the constant negotiation of tradeoffs between competing resource interests. Adaptive governance is also realized when science is used to support system-level outcomes in which the collective values and visions for resource use are expressed. Classical water models lack this capacity, however stated preference choice models are ideal for evaluating the preferences and attitudes of water users in relation to future or hypothetical scenarios. In these approaches, respondents are asked to make trade-offs by evaluating competing scenarios described in the form of policy options. Choice modeling techniques permit predictive modeling of stakeholders’ response to future policies, thus providing crucial information for integrative models and adaptive decision-making. A synthesis of civil engineering and social science modeling approaches represents an opportunity to incorporate water user preferences into predictive water models, thereby increasing the sensitivity of the resulting water model. Such an integrated approach would help address gaps in both the theoretical and applied dimensions of existing water models and facilitate adaptive governance.
I, In collaboration with Dr. Murray Rutherford, and Dr. David Yates – NCAR – conducted research in the Okanagan Valley Water Basin about the potential of integrated engineering-behavioural water modelling to foster adaptive governance. I focused on behavioural evaluations derived from stated choice surveys to include the preferences of a representative sample of the residential and agricultural water users in the Okanagan. Utilizing the existing Okanagan water demand model I then integrated these stated preference data with revealed preference data for a more complete behavioural model, and finally created a new integrated scocio-hydrological model. This research represents the first known attempt to integrate engineering based demand models with stated choice models to this degree of sophistication.