Interview with Donald Keefer

May 16, 2017

Donald Keefer is a Research Affiliate and retired Senior Hydrogeologist with the Illinois State Geological Survey (ISGS), a division of the Prairie Research Institute. His research applies knowledge management, machine learning, and data mining technologies to geology to improve methods of characterizing uncertainties in geologic models. These models are used for the management of many different natural resources, from groundwater to petroleum to aggregate, and much of his research is targeted to problems of groundwater security.

What is the importance of having reliable geologic maps?

All models contain uncertainties which, if uncharacterized, can lead to significant risk in any decision that depends on the reliability of predictions from the models. If these uncertainties can be well characterized, however, this insight can lead to reductions in risk through improved decision making. If the uncertainties can be further organized or reduced, this can lead to a still lower risk and better decisions.

Geologic models of aquifers and non-aquifers—such as the Mahomet Aquifer in central Illinois—are the basic framework within which a groundwater flow model of the system is constructed. Typically, engineers use the approximate distributions and properties of aquifers and non-aquifers to mathematically model groundwater flow. These calculations are necessarily presumptive since they are based on limited data points and do not account for detailed spatial variations in the geologic parameters that control groundwater flow, such as permeability. Insights on higher-resolution estimates of permeability can provide important knowledge both on how to vary these parameter distributions inflow models, and of where resulting groundwater flow predictions might be more or less reliable. This knowledge about model reliability can be used by decision-makers to better constrain and even reduce the risk associated with specific groundwater management decisions. This type of improvement in the reliability of decisions exemplifies why it can be beneficial for geologists and hydrologists to communicate and ensure that relevant variations in geologic deposits and properties are not overlooked.

Analysis of uncertainties in different scales of geologic variability shows that uncertainties in the distribution of geologic deposits can have the largest impact on the reliability of groundwater modeling results. Groundwater management decisions that don’t benefit from insights on geologic model reliability will be more likely to lead to unsustainable management choices.

What are the consequences of unsustainable water use?

When an aquifer is pumped for many years at rates significantly faster than the groundwater can recharge, several bad changes can occur in the system that can have long-term or even permanent consequences. Of primary importance, unsustainable pumping can render an aquifer unusable—water will no longer flow into wells in the deposit at rates fast enough to meet the needs of the well owner. The more wells being unsustainably pumped, the larger the area that is likely to be affected, and the longer the system will take to recover. In deep aquifers with thick overlying aquitards (non-aquifer deposits), it can so long to recover from over-pumping that it effectively makes the system permanently unavailable. Another potential problem of long-term unsustainable groundwater pumping is that air can get introduced into the aquifer, allowing undesired chemical reactions to take place. These changes can either introduce natural toxins to the pumped water by dissolving and mobilizing these toxins from natural minerals in the aquifer (e.g., arsenic) or permanently reducing or eliminating the ability to obtain water from the well by enabling the precipitation of previously-dissolved minerals in the pore spaces of the aquifer. In geologic settings where the aquifer is a sand and gravel deposit that has no significant cement between the grains, long-term over-pumping of groundwater can also allow the weight of overlying geologic materials to partially collapse the aquifer, permanently reducing the amount of water that the deposit can hold.

What are some major issues that you see impacting groundwater security? Are there better solutions?

In addition to the need for improved methods for understanding geologic and groundwater flow model uncertainties, I think the largest threat to groundwater security is overuse due to a deliberate, non-conservation mindset. This doesn’t necessarily mean anti-conservation, but much of the U.S. culture seems to be willfully ignorant of the consequences of any of our water-use choices. The increasing use of multiple shower heads in our bathrooms; farming in the desert; irrigation of row crops in temperate environments as a hedge to minor losses in yield; ubiquitous quests for the perfect turf-grass lawn; bottled water demanded at any venue: all are examples of a non-conservation mindset. There needs to be a cultural shift to a conservation mindset for water security to fundamentally improve.

Drinking water, as a resource and commodity, is nearly free to most individuals in the U.S. Water bills typically pay for administrative costs and the continued maintenance and expansion of the infrastructure. Ready access to unrestricted amounts of water is often seen as almost a fundamental right. Costs for cleanup, protection, and replenishment are often passed to government agencies and not reflected in water pricing. Some people feel that these attitudes and behaviors encourage poor water-use decisions and foster a non-conservation perspective. Many communities are beginning to explore water pricing as a way to encourage water conservation. While pricing is still low and these structures are beginning to be implemented, it is unclear how effective this will be for encouraging sustainable-use decisions.

In the end, grassroots involvement, education of children, and repeated exposure of the general population to understandable science-based information seem to be the most effective combination of tools for building a conservation mindset and ensuring water security. We need to realize that we are all responsible for our water-use choices and the long-term consequences they will bring.