Assessing the Impact of Transient High-Volume Groundwater Withdrawals on Headwater Streams

Abstract:

Use of high-volume hydraulic fracturing for natural gas extraction from unconventional reservoirs has increased in the past decade and expanded across diverse geographic regions with varying water resource constraints. Hydraulic fracturing completions can require several millions of gallons of freshwater over a 2-4 week time period. This freshwater is often acquired from nearby surface or ground water, which will exert additional pressures on local water resources, potentially resulting in unsustainable development of unconventional reservoirs. The State of Michigan has a Water Withdrawal Assessment Tool (WWAT) that serves as a screening instrument for permitting high-volume water withdrawals. Although it was not originally designed to assess transient groundwater withdrawals, it is currently being used to evaluate groundwater withdrawals associated with hydraulic fracturing activities in Michigan. As a result, it may not accurately capture stream-groundwater interactions during hydraulic fracturing water withdrawals.

In this study we developed a high-resolution groundwater flow model in MODFLOW to investigate scenarios where hydraulic fracturing water withdrawals from unconfined shallow aquifers occur in the vicinity of headwater streams. The region of study is in Michigan’s northern Lower Peninsula where the surface geology is dominated by sandy glacial till and headwater streams are primarily groundwater-fed. In this region the Utica-Collingwood shale formation is being developed via high-volume hydraulic fracturing with individual gas well completions having reported use of over 20 million gallons of water. For future well development, the State has granted permission to withdraw up to 35 million gallons per well. Stream-groundwater interactions in the area are examined under transient pumping conditions similar to those expected during such groundwater withdrawals. We also evaluate the influence of the distance from the well to the stream, withdrawal well density, pumping rates, aquifer heterogeneity, and coincident industrial activities on potential reductions in base flow to the stream. Results from our modeling effort will be compared to those of the existing WWAT to assess the adequacy of the State’s WWAT in screening hydraulic fracturing water withdrawals in Michigan.