The buildings on Cornell University’s central campus all require heat during the cold winters in Ithaca, N.Y.
The dormitories, century-old stone buildings, post–World War II classroom buildings, and 21st-century lab facilities on Cornell University’s central campus all require heat during the cold winters in Ithaca, N.Y. As part of Cornell’s plan to become carbon-neutral by 2035, the university is planning a project that combines geophysical research with geothermal energy production. Credit: Mary C. Colomaio, Cornell Facilities and Campus Services

In 2013, Cornell University adopted a target for its Ithaca, N.Y., campus operations to become carbon-neutral by 2035. New York State followed suit in 2019, pledging to zero out net carbon emissions by 2050. However, when they adopted these targets, neither Cornell nor New York had a plan for removing carbon-based energy sources from what they consumed to heat residential and commercial buildings (which accounts for 31% of total energy consumption in the state).

One possible approach, tapping geothermal energy, could help replace fossil fuels for direct use (as distinct from electrical power generation) for heating not only the Cornell campus but also the northern United States, Canada, and much of northern Europe and Asia. Such increased deployment of geothermal district heating—for example, providing a centralized supply of geothermal heat to multiple buildings via insulated pipes—could significantly lower global carbon emissions.

Block diagram and subsurface temperature estimates at the Cornell site.
Fig. 1. Block diagram and subsurface temperature estimates at the Cornell University site. At a depth near 2,800 meters, subhorizontal Paleozoic strata (horizontal layers of varying color) overlie Grenville high-grade metamorphic basement rocks (brown with patterns). A borehole schematic shows extraction of geothermal heat by pumping high-temperature fluids (pink) from one borehole, transfer of heat into a campus heating loop, and return of cooler fluids (blue) to the subsurface via an injection borehole. The thermal profile on the left [Smith, 2019] is based on data from bottomhole temperatures taken from more than 13,000 hydrocarbon boreholes in New York, Pennsylvania, and West Virginia. The diagonal black curve extrapolates, at 500-meter intervals, the most likely temperature, accounting for uncertain variables; the curve’s gray envelope encloses the 25th–75th percentile estimates.

At Cornell, the Earth Source Heat (ESH) approach evolved to meet the need for carbon-fuel alternatives. ESH involves extracting water from hot rock using one set of wells, transferring the heat into an existing campus district heating loop, using industrial heat pumps to maximize extraction of heat from the geothermal resource, and returning the water to below ground through another set of wells (Figure 1).

At a scientific borehole planning workshop sponsored by the International Continental Scientific Drilling Program (ICDP) convened at Cornell University last January, attendees considered an emerging plan to drill a pair of test wells to evaluate the potential of heating Cornell’s campus buildings with geothermal heat extracted from a depth exceeding 2 kilometers. That pragmatic test plan spawns the opportunity to piggyback basic research studying the workings of continental crust onto the drilling operation. The ICDP-sponsored workshop allowed Earth scientists to explore how to turn this opportunity into an experiment of wide-ranging value.

We anticipate finding Paleozoic sedimentary and Precambrian metamorphic basement rocks that lie below Cornell. These rocks, like those at numerous other locations, are complex, heterogeneous, and anisotropic (Figure 1). Counterintuitively, the lack of anything geologically “special” about central New York—where levels of natural seismicity are also low—is a compelling attribute of this borehole project. To date, continental scientific drilling sites have been selected to investigate active tectonic or volcanic features, rare events like meteorite impacts, or climate history. Hence, there is uncommon value in investigating a “boring” (i.e., ordinary) location, whose results will be widely applicable.

Practical Problems with Geothermal Energy

Depth itself is not a major technical problem for the ESH project—across most stable North American continental crust, subsurface temperatures in the range of 50°C–100°C occur less than 3 kilometers below the surface, shallow enough to be reached economically with conventional drilling technology. However, geothermally heated water is used for district heating at only a few locations currently. Boise, Idaho, on the Snake River Plain, and some locations in Iceland are geologically unusual, with hot water situated near the surface. A few more thermally mundane places, like the Paris sedimentary basin in France, have rock structures at suitable depths through which fluid flows easily [Lopez et al., 2010].

Cornell is positioned to advance an initial demonstration project for extracting 100 gigawatt hours of geothermal heat per year.

Even though the rocks at suitable depths beneath most of New York are sedimentary (and thus relatively porous), low permeability is unfortunately still an obstacle to implementing geothermal technology for district heating [Camp et al., 2018]. In theory, the rocks can be stimulated to transmit fluids by interconnecting pores along natural networks of preexisting microfractures, through the use of fluid pressure to slightly disturb those fractures. However, such engineering approaches involve high financial risks because they may fail to produce adequate permeability to achieve energy extraction targets. Furthermore, drilling, artificial permeability construction, and fluid cycling all cause perturbations in the subsurface that may be associated with seismic risk. Overall, the lack of fundamental scientific and engineering understanding of the rocks and the coupled risks result in high costs and slow progress.

Following years of economic and technical feasibility analysis of its ESH approach, Cornell is positioned to advance an initial demonstration project for extracting 100 gigawatt hours of geothermal heat per year, enough to serve about 20% of campus needs. However, until deep boreholes can access the subsurface, we lack key data needed to design a geothermal system and to assess financial, technical, and seismic risks.

As a research-focused land grant university, Cornell has three integrated motivations for drilling test boreholes: to achieve its carbon neutrality goal, to demonstrate and lower the risks of a technology with potential for wide deployment, and to foster basic research.

The ICDP workshop assembled 35 visitors and 26 Cornell faculty, technical staff, and students from diverse specialties (but with little shared background knowledge), including borehole engineering, regional geology, induced earthquakes, geothermal engineering, and hydrology. About 90% of the workshop was spent in group discussions of key science questions and borehole design considerations, rather than on individual presentations.

A New Generation of Geological Investigation

Five broad research themes that an Ithaca borehole would facilitate emerged from the workshop discussions:

  • fluids and elemental cycling
  • poromechanical behavior across a range of length scales
  • controlling subsurface fractures and fluid flow
  • deep life
  • subsurface evolution of the Appalachian basin and its basement.

The first three themes are tightly coupled and, because improved knowledge is needed for societally critical activities including energy production and seismic risk assessment, they are likely to be at the core of successful efforts to raise funds for multimillion-dollar drilling projects. Exploration of the latter two themes would be enabled by accessing subsurface fluid and rock samples from a borehole, and findings in these themes might influence the other topics.

The common root of the first three themes is complex hydrologic, thermal, chemical, and mechanical relationships and processes—natural and human-induced—acting over timescales ranging from very short (days) to geological. Workshop participants focused on describing the experiments, measurements, and samples needed to underpin scientific advances in understanding the hydrogeology of old, tectonically inactive crust and of the physical controls affecting seismic hazard within continental plate interiors.

One hypothesis posits that the continental crust is at critical failure condition everywhere [e.g., Townend and Zoback, 2000]—that rocks underground are always close to failure by fracturing. This scenario can keep certain fractures open as permeable pathways, but it also means that the rocks are highly susceptible to small stress perturbations. Having a kilometers-deep borehole that provides vertically continuous data about stress magnitudes and orientations, pore fluid pressures, and temperatures in the context of lithologic and fracture properties would allow scientists to probe this hypothesis.

Valuable knowledge could come from determining the vertical profile of the subsurface properties at a pristine location and monitoring it as the geothermal project progresses.

Workshop attendees were very interested in the possibility of gaining a better understanding of poromechanical behavior and the processes affecting stress and strain near the interface between the Paleozoic sedimentary rocks and the Precambrian crystalline basement. Fiber-optic sensing along the length of the borehole, along with measurements of the physical properties of rock samples, could provide insights into the mechanical behavior of the subsurface at multiple scales. These insights are important for understanding induced seismicity and mitigating hazards.

Attendees were also enthusiastic about the opportunity to retrieve core samples that could spur a new generation of geological investigation into the thermal history of the rocks beneath upstate New York. This investigation could fill knowledge gaps about what happened to the rocks from the end of deposition in the Devonian (420–360 million years ago); through the burial, fluid flow, diagenesis, and fracturing of the Alleghanian orogeny; to the mid-Mesozoic emplacement of kimberlites and the topographic readjustments and denudation of the Cenozoic (66 million years ago to the present).

Although no geobiologists participated, the workshop participants agreed that sampling could enable study of life-forms found amid the lithologies and conditions that a borehole would traverse. Discussions at the workshop revealed that the Cornell site offers even higher research value if the vertical profile of the thermal, hydrological, mechanical, and chemical properties of the subsurface could be determined at a pristine location not yet subjected to subsurface manipulation, and then monitored through time as the geothermal project progresses.

Where Science and Engineering Challenges Intersect

A major question about a Cornell borehole, one where science and engineering considerations intersect, is whether it should have a narrow or wide diameter. Regardless of diameter, each borehole to 3 kilometers or greater depth will cost several million dollars. Because a wide-diameter borehole down to an exploratory target depth of 4–5 kilometers could serve later as an operational well accommodating a higher flow than a narrow hole, a case can be made for such a hole, which could also facilitate extensive testing and sampling. An alternative strategy is to drill an initial narrow borehole that would allow studies of geologic context and in situ characterization. Later, this borehole could serve as an observatory for downhole monitoring of such things as temperature, pressure, seismic activity, and fluid chemistry.

The risks of these two alternatives are markedly different. A wide-diameter borehole intended to be used later for energy production would require much of its length to be cased with cement. However, some critical rock properties and conditions should be measured or sampled in an uncased borehole, and much of the equipment for permanent monitoring would need to be installed prior to casing. Moreover, several borehole experts foresaw this scenario leading to high costs and high technical risks of losing the operational well if wall rock collapsed and refilled the hole.

Drilling a narrow hole to 4–5 kilometers also comes with significant practical, costly challenges. One such challenge is the limited availability of companies and equipment capable of drilling and coring such a hole at a cost-effective rate. Discussions about the back and forth between science goals and borehole operational goals revealed scenarios that may translate to significantly higher borehole costs and to compromised plans that reduce the scientific scope.

Another alternative emerged from these discussions: A narrow borehole that reaches only 100 meters into the basement would have lower costs and risks than a deep hole. This narrow-hole design would still allow seismological and hydrological experiments to document subsurface conditions before and after artificial permeability construction in a neighboring well, and after a geothermal field begins production. With this approach, the reduced risk to the integrity of a wide-diameter borehole, which could then be cased immediately to protect against collapse, could plausibly lower expenses enough to enable a three-borehole setup. This setup, incorporating one narrow, intermediate-depth hole and two wide (and cased) full-depth boreholes, might not cost much more than two wide boreholes dug with numerous workarounds and compromises.

A Promising Start to the Work Ahead

Cornell sought funds to drill a narrow borehole through the sedimentary rocks and about 100 meters into the crystalline basement. And at the end of July, the U.S. Department of Energy announced selection of the project for funding.

Following the ICDP workshop, Cornell University sought funds from the U.S. Department of Energy (DOE) to drill a narrow borehole through the sedimentary rocks and about 100 meters into the crystalline basement. And at the end of July, DOE announced selection of Cornell’s project for funding. The funds awarded should be sufficient to bore the hole; to obtain several hundred meters of continuous core and spot cores; to analyze pressure, stress, strain, and hydrological and rock properties; and to install permanent monitoring systems.

With this foundation, Cornell and experts like those who participated in the workshop can address many of the scientific and engineering challenges laid out above. Yet some of the fascinating questions articulated at the workshop will require companion projects. We will need to obtain additional samples, install other types of sensors, extract and analyze data shortly after drilling and during years of observation, and conduct complementary experiments and geophysical surveys. Those added-value projects will require that experts lead efforts to seek, in collaboration with Cornell, funds for the worthy, complementary projects. The university is also committed to broad and rapid public dissemination of all data extracted from the borehole.

The ICDP-supported borehole planning workshop, which assembled interested parties with diverse backgrounds and broad expertise, has helped jump-start progress toward using geothermal energy to meet the university’s carbon neutrality goals. The initial idea was to use a geothermal energy borehole as a chance to piggyback some fascinating scientific research onto a practical project. The ICDP workshop revealed that it is also an opportunity to increase the probability that the ESH project itself will be successful because of the “boring” companion science.


Camp, E. R., et al. (2018), A probabilistic application of oil and gas data for exploration stage geothermal reservoir assessment in the Appalachian Basin, Geothermics, 71, 187–199,

Lopez, S., et al. (2010), 40 years of Dogger aquifer management in Ile-de-France, Paris Basin, France, Geothermics, 39, 339–356,

Smith, J. D. (2019), Exploratory spatial data analysis and uncertainty propagation for geothermal resource assessment and reservoir models, Ph.D. thesis, 255 pp., Cornell Univ., Ithaca, N.Y.

Townend, J., and M. D. Zoback (2000), How faulting keeps the crust strong, Geology, 28(5), 399–402,<399:HFKTCS>2.0.CO;2.

Author Information

Teresa Jordan ( and Patrick Fulton, Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, N.Y.; Jefferson Tester, Smith School of Chemical and Biomolecular Engineering, Cornell University, Ithaca, N.Y.; Hiroshi Asanuma, Fukushima Renewable Energy Institute, National Institute of Advanced Industrial Science and Technology, Koriyama, Japan; and David Bruhn, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, Netherlands


Jordan, T.,Fulton, P.,Tester, J.,Asanuma, H., and Bruhn, D. (2020), Exploring by boring: Geothermal wells as research tools, Eos, 101, Published on 10 September 2020.

Text © 2020. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.