Evaluation of Modeling Framework and Social Perspectives Regarding Sustainable Groundwater Management in Michigan

Ensuring sustainable management of groundwater resources is important even in a water-rich state such as Michigan. Large Quantity Withdrawals (LQW) associated with hydraulic fracturing and industrial activities (e.g., Nestle’s water withdrawal request in Evart, MI) in the Northern Lower Peninsula as well as LQW related to irrigation wells in the Southern Lower Peninsula have become an issue in recent years. The objectives of this dissertation were to evaluate the groundwater management in Michigan from two different perspectives: (1) from the modeling framework where the approach currently used to evaluate LQWs were assessed and possible strategies to improve it were explored, and (2) from the social perspectives where the mental model and degree of understanding between the ‘experts’ and ‘non-experts’ stakeholders regarding groundwater management in Michigan were evaluated. As part of its groundwater management framework, the State of Michigan developed the Water Withdrawal Assessment Tool (WWAT) to screen proposed LQWs. However, it has been criticized by the public as both too conservative and not conservative enough. To evaluate the WWAT’s groundwater model, a 3-D numerical groundwater model was developed and the estimates of streamflow depletion calculated using this model were compared with those generated by the WWAT. Two issues were found with the groundwater model of the WWAT: conceptualization of the hydrogeologic setting and assignment of parameters. The modified Hunt (1999) analytical solution used by the WWAT assumed a single layer, unconfined aquifer, and it does not capture important hydrogeological heterogeneities. As a result, Ward and Lough (2011) streamflow depletion analytical solution was evaluated as an alternative approach that could be implemented as part of the online tool. Furthermore, streambed parameters used in the WWAT were poorly defined. A new approach to improve this parameter was developed by using soil and land cover data to estimate the streambed hydraulic conductivity (KS) and the resulting KS values from this approach were statistically compared against 23 locations with known KS values, resulting in an improved KS¬ values estimation using publicly available soil and land cover data. To understand the perspectives of groundwater resource management, cognitive maps of ‘experts’ and ‘non-experts’ were developed using a Conceptual, Content, Cognitive Mapping (3CM) approach. The study found that ‘experts’ have a more structured and detailed knowledge than ‘non-experts’. It was also apparent that ‘non-experts’ lacked knowledge about the regulatory framework. ‘Non-experts’ were also found to consider the users of the water in determining the impact of LQW, whereas ‘experts’ were more interested in the environmental impacts of these LQWs regardless of the users. Co-orientation surveys were utilized to assess the degree of understanding between ‘experts’ and ‘non-experts’ pertaining to water resource management in Michigan. It was found that ‘experts’ and ‘non-experts’ shared many similar views except on topics related to Great Lakes water diversions and LQWs associated with hydraulic fracturing and industrial activities. It was also found that ‘experts’ were more concerned about the impacts on the environment than they were given credit for, while ‘non-experts’ were more knowledgeable than they were given credit for. The areas of disagreement between ‘experts’ and ‘non-experts’ found in these studies could be used to foster better communication between the public and government agencies in moving towards a common goal of sustainably managing groundwater by encouraging more public involvement through community forums or meetings.PhDEnvironmental EngineeringUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttps://deepblue.lib.umich.edu/bitstream/2027.42/149860/1/ijayawan_1.pd