Geothermal / Ground-Source — The Earth as Heat Sink

Geothermal (ground-source) HVAC uses the stable temperature of the earth a few meters down — roughly the local annual-average air temperature, far cooler than summer ambient — as the heat-rejection sink (and winter heat source) for a heat pump. In cooling, that means dumping the building’s heat into ground that’s ~20 °C instead of into 38 °C August air, which raises the heat pump’s efficiency exactly when cooling demand and electricity prices peak. It’s not a separate refrigeration cycle so much as a better sink for one.

The collection behind this article is the design-and-practice canon as catalogued in the Zotero registry (metadata only; full texts live in the library): Egg’s Geothermal HVAC: Green Heating and Cooling (2011) and Modern Geothermal HVAC: Engineering and Control Applications (2013), Lloyd’s Geo Power: Stay Warm, Keep Cool and Save Money (2015), and Kavanaugh’s Geothermal Heating and Cooling: Design of Ground-Source Heat Pump Systems (2014). Between them the titles span the owner’s view (comfort and payback), the engineer’s (loop-field design and controls) and the systems view — ground-source integrated with the same hydronic distribution loops (Hydronics) that sorption and conventional chillers feed.

Why it matters in a heat-driven strategy

Geothermal and adsorption attack cooling cost from two different angles that stack:

Adsorption removes electricity from the driving side (heat instead of compressor work).
Geothermal removes electricity from the rejection side (cool ground instead of hot-air condenser) for whatever compressor work remains.

A site can do both: an adsorption or absorption chiller still needs to reject condenser and absorber heat, and a ground loop is a superior place to put it versus a cooling tower (no water consumption, lower approach temperature). So geothermal is also the natural heat-rejection partner for sorption cooling, not just an alternative to vapor compression.

Cost/feasibility caveat

The economics are dominated by ground-loop installation (drilling/excavation) and by available land and geology. The registry’s titles alone signal the split: Geo Power’s “save money” framing is the owner’s payback question, Kavanaugh’s “design of ground-source heat pump systems” the engineer’s loop-sizing one. Whether it pencils is highly site-specific — soil thermal conductivity, water table, and how much land or borehole depth is available.

The cold sink in practice: Chena Hot Springs

The clearest field demonstration of “a better sink beats a hotter source” is not a ground loop at all but a river. At Chena Hot Springs, Alaska, a custom three-pressure double-lift ammonia-water chiller driven by only 74 °C spring water (5.4 L/s) delivers −29 °C calcium-chloride brine and holds a ~1,000-tonne ice museum at −4 °C through 32 °C summers — because it rejects absorber and condenser heat to 4 °C river water (5.0 L/s). The whole system draws ~33 kW of electricity against 110 kW for the diesel-powered vapor-compression backup it replaced, saving about $400/day at ~30 ¢/kWh, with ~95% availability over its first two seasons. The same 4 °C stream is what lets Chena run the world’s lowest-temperature geothermal power plant, a 730 kW ORC on 73 °C water: NREL notes it is the temperature differential, not the source temperature, that yields an acceptable cycle efficiency.

The sink’s limits showed up too. When the creek warmed to ~7 °C on mid-afternoon summer days — coincident with peak load — chiller efficiency dropped, and the owners planned a 3.9 °C shallow cold-water well to replace the surface water: an earth-coupled sink chosen specifically for its stability, which is this article’s thesis restated. Two other practices transfer to any ground-coupled sorption plant: every surface touching spring water, river water or brine is stainless steel, and the 63 °C spring water leaving the chiller cascades on to heat a greenhouse and an outdoor pool rather than being wasted. The manufacturer’s earlier unit made fishing-fleet ice at Kotzebue for ~10 years from 74 °C diesel jacket water, so the design is proven beyond one site. The full absorption-side story — and why Chena is the exception in a thin record — is in Geothermal-Driven Sorption Cooling.

The other end of the spectrum: deep geothermal as a heat source

Everything above treats the ground as a sink at ~20 °C. Drill kilometres into hot rock and the earth becomes a source: that is EGS. Fervo Energy’s vendor page describes the method — oil-and-gas-derived horizontal directional drilling (many wells from one pad, smaller footprint, access to geologies previously out of reach), downhole fiber-optic sensing to map flow and temperature in real time, and binary-cycle conversion of the heat into 24/7 carbon-free electricity with “extremely low water use.” The page names no projects, power figures or costs; it cites a 2016 NREL estimate that the resource within 5 km could meet all US electricity demand (7 km: five times over), and lists research on in-reservoir energy storage — using the fractured rock as a pressure and thermal buffer so the plant can dispatch against wind and solar, a thermal-reservoir analogue to the storage questions elsewhere in this wiki. The project numbers (Project Red, Cape Station) are compiled in the EGS article.

Between the two ends sit hydrothermal district-heating resources at 70–110 °C, and that middle band is where geothermal heat has actually driven chillers — a thin, mixed record compiled in the sorption article. The ends meet on the rejection side: Chena’s chiller and ORC both work because of a cold sink, not a hot source, and for any heat-driven chiller a colder condenser is worth more than a few extra degrees of drive heat.

Sources

Zotero cooling registry — the four-book geothermal HVAC design canon (Egg 2011/2013, Lloyd 2015, Kavanaugh 2014) this article was compiled from; metadata only
Fervo Energy — Technology — deep EGS as the hot-source counterpart: horizontal drilling, fiber-optic sensing, binary cycle, NREL resource claim, in-reservoir storage
Chena Hot Springs GHC Bulletin 2006 — the cold-sink lesson: 74 °C drive works with a 4 °C river; creek warming, planned cold-water well, cascade, 730 kW ORC