Withdrawals from the deep sandstone aquifers of northeastern Illinois have exceeded sustainable supplies for over 100 years. Many communities who use these aquifers will be forced to switch sources, and the time-frame is short: 10-15 years in parts of the region. For more information on the issue facing northeastern Illinois, check out Devin Mannix’s recent blog post and short paper. In the coming decade, water supply planning in these at-risk areas must account for complexities in the aquifers, which will require both analysis of data and well-calibrated groundwater flow models.

To help communities best assess future water supply issues, the Illinois State Water Survey has developed a large regional groundwater flow model to simulate future water demand scenarios. To test the reliability of this model, we compare the impact of simulated historic withdrawals to actual water level data collected from wells, which the ISWS has been collecting since the late-1800s. This large data set is indispensable for improving the quality of our groundwater flow models. However, speaking from experience, such a wealth of data can also be enormously frustrating, as models rarely match all real-world data on the 1st (or 2nd or 60th) attempt. The sandstone aquifers in northeastern Illinois posed a particular problem to our models, as the three-dimensionality of the system is not easy to conceptualize, and assumptions that worked 100 years ago may not be valid today.

My colleague, Dr. George Roadcap, recently developed a new approach to improve our ability to visualize aquifer behavior. He has taken a set of water level data collected from the Mahomet Aquifer observation well network in central Illinois and assigned it to the groundwater flow model, an approach contrary to the traditional modeling method. This so-called “head-specified” model simulates a water level surface that preserves mass balance and honors all water levels exactly. While limited in their ability to simulate future conditions, head-specified models allow us to precisely investigate how the groundwater system has changed through time. We adopted this approach for the sandstone aquifers of northeastern Illinois using the entire database of water levels that we have amassed at the ISWS. The results were eye-opening, allowing us to visualize changing water levels through time, as you can see in the animation at the top of this post. The model has other benefits, such as helping us understand where recharge to the aquifers is occurring and how geologic structures such as faults affect groundwater flow in the aquifers. The results and implications of this work are published in a Groundwater journal article, “Developing potentiometric surfaces and flow fields with a head-specified MODFLOW model”. (May need a subscription to access.)

Good data are critical to enhancing our understanding of aquifers, which in turn greatly improves our modeling capabilities. We strongly encourage municipal, industrial, and agricultural water users to submit measured water levels during their annual reporting to the Water Survey’s Illinois Water Inventory Program (IWIP). Furthermore, we strongly encourage local water authorities to contact the ISWS about developing a real-time monitoring network. In a future blog post, I will discuss how we are utilizing dedicated monitoring data from the Imperial Valley Water Authority in central Illinois to assess changing conditions in their aquifer, particularly important during droughts.

Abrams, D. B., Roadcap, G. S. and Mannix, D. (2017), Developing Potentiometric Surfaces and Flow Fields with a Head-Specified MODFLOW Model. Groundwater. doi:10.1111/gwat.12540