Thermal Output Stability in Closed-Loop EGS Systems: Field Observations

The question we hear most frequently from industrial buyers evaluating geothermal supply is: how stable is the output? It is a reasonable question. Industrial manufacturing processes are built around reliable, predictable energy input. An energy source that varies by 15 to 20 percent from hour to hour or day to day creates production planning problems that a manufacturing engineer will rightly resist. This article addresses the thermal output stability question directly, using data from our early-pilot monitoring program rather than theoretical arguments about geothermal characteristics.

What "Stable" Means in This Context

Before presenting the data, we need to define the relevant stability metric. There are three different timescales at which thermal output variation is meaningful, and the causes and magnitudes of variation differ across timescales.

Intraday variation refers to output changes within a single 24-hour period. For a closed-loop geothermal system, intraday variation is driven primarily by the thermal extraction rate (how hard the system is being pushed at any given hour), the working fluid circulation rate, and any scheduled maintenance events. At constant extraction rate, a well-designed closed-loop system should show very low intraday variation in thermal output. Formation temperatures at 2,000 to 3,000 meters are not sensitive to surface temperature changes, precipitation, solar radiation, or any of the weather inputs that drive intraday variation in solar and wind generation.

Seasonal variation refers to output changes over the course of a year. The relevant mechanism here is the thermal recovery of the formation. If a closed-loop system is extracting heat continuously, the near-wellbore zone will cool slightly over time as heat is removed faster than it can be replenished by conduction from the broader formation. In winter, ambient surface temperatures are lower, which improves the efficiency of the ORC rejection cycle but does not directly affect formation temperature at depth. Seasonal variation in output for a well-designed closed-loop system should be small, typically below 5 to 8 percent of rated output across a year.

Long-term decline refers to the gradual reduction in formation temperature at the wellbore over years to decades of operation, as the near-wellbore thermal reservoir is depleted faster than the surrounding formation can recharge it. This is the most important stability consideration for long-term energy planning and the one most commonly cited as a concern by buyers who have read about geothermal resource depletion in conventional systems.

Early-Pilot Data: What We Observed

Our early-pilot monitoring program covers a limited dataset from our initial closed-loop test wellbore in the Wasatch Front region, instrumented with formation temperature sensors at three depth intervals and working fluid temperature sensors at the inlet and outlet of the closed-loop circuit. We want to be explicit: this is early-stage data from a pilot program, not production operational data from a commercial fleet. We are sharing it because it is real data that speaks to the stability question, not because it is a complete or statistically definitive dataset.

Over the monitoring period from initial operation to the present, the inlet temperature of the working fluid returning from the formation depth has shown variability of approximately 2 to 3 percent around the mean operating temperature. The variability is not random noise. It correlates with extraction rate: when we pushed the working fluid circulation rate above the nominal design point in our extraction rate tests, the return temperature dropped by 4 to 6 percent, recovering over a period of 8 to 14 hours as the near-wellbore zone thermally recovered after the extraction rate was returned to nominal. This is expected behavior for a closed-loop system and is consistent with analytical predictions from wellbore heat exchanger models.

The implication for operational planning is clear: the system should not be run at maximum extraction rate continuously. It should be run at a nominal extraction rate that maintains near-wellbore thermal recovery, with headroom reserved for short-duration demand response events. This is how we have designed the dispatch system's operating setpoints.

The Long-Term Decline Question

We cannot answer the long-term decline question from our early-pilot data. We have months of data, not years. What we can say is that our analytical models, calibrated to the formation thermal conductivity and heat capacity values derived from the formation logs on our pilot well, predict long-term thermal decline rates of 0.5 to 1.2 percent per year at our nominal extraction rate. These predictions are consistent with published results from closed-loop wellbore heat exchanger field studies in comparable geological settings in Europe and from North American research programs.

A 0.5 to 1.2 percent per year decline rate means that after 20 years of operation, the system is producing at 76 to 90 percent of its initial rated output. For a system designed with a 25-year operational life, this should be factored into the initial sizing so that the system still meets the site's minimum load requirement at end of life. We build that design margin into our specifications.

We are not saying the decline rate prediction is certain. Formation thermal conductivity, the primary driver of long-term replenishment rate, has uncertainty that propagates into the decline rate estimate. Our model predicts 0.5 to 1.2 percent per year; the actual value for any specific wellbore will fall somewhere in or near that range based on the actual formation properties. Getting formation thermal conductivity right requires either core samples and laboratory measurement, or careful calibration of the model against early operational data. We are doing both.

How This Compares to the Intermittent Alternatives

The stability question is most usefully answered not in absolute terms but in comparison to the alternatives an industrial buyer is evaluating. The relevant comparison is not "is geothermal perfectly stable" but "is geothermal more stable than solar or wind, and by how much."

Solar photovoltaic output in the Utah Basin varies from near-zero at night to rated capacity at peak solar hours, with intraday cloud cover producing output fluctuations of 20 to 60 percent of rated capacity in under 10 minutes during overcast or variable cloud conditions. On a monthly basis, solar output varies from roughly 25 to 30 percent of monthly maximum in December to near-maximum during June and July in this region. On an hourly basis, solar output is zero for roughly 12 to 14 hours per day depending on season.

Wind output in Basin and Range locations varies from near-zero during calm periods to rated capacity during wind events, with variability profiles that are less predictable than solar because wind patterns are governed by weather systems on time horizons of hours to days. Wind capacity factors at good inland sites run 30 to 40 percent annually.

Closed-loop geothermal output, operated at our nominal extraction rate with appropriate thermal headroom management, has an expected capacity factor of 87 to 93 percent annually, with intraday variability of 2 to 3 percent and no dependence on weather. The comparison for a manufacturer that needs 24/7 reliable baseload is not close.

What to Ask Any Geothermal Developer About Stability

If you are evaluating a geothermal supply proposal from any developer, there are three questions about output stability that any serious developer should be able to answer with data rather than assertions.

First, what is the nominal extraction rate as a percentage of the formation's thermal capacity, and how was that capacity estimated? A system operated at a high fraction of its estimated thermal capacity will show faster near-wellbore cooling and more sensitivity to extraction rate changes. A developer who cannot quantify this is not managing the resource carefully.

Second, what long-term decline rate is built into the system sizing and financial model, and what is the source of the thermal conductivity estimate used to predict it? Published formation values or a developer's assumptions should be replaceable with measured values once early operational data exists. Ask to see the calibration plan.

Third, what monitoring and reporting will be provided so that you can verify output stability over time? A developer confident in their system should welcome the transparency of a real-time telemetry feed that shows the site operator the formation temperature, output rate, and operating status. That data should be in the contract, not something you have to request after the fact.