SANTA ROSA, Calif. - At the northern edge of the Geysers, the world's largest geothermal complex, which sprawls over nearly 40 square miles north of Santa Rosa, Calif., Houston-based power company Calpine is conducting an experiment.
On the surface, not much sets the project apart from the 18 ridge-top power plants and dozens of other drilling platforms here, most of which Calpine owns. The site, high on a sun-baked hillside, is home to a 110-foot drill rig surrounded by a couple of trailers and a mass of machinery.
But two miles underground, something big is happening.
Traditional geothermal systems draw steam from naturally occurring underground reservoirs - typically found only along continental plate boundaries and in areas of subsurface volcanic activity - and pass it through turbines to convert it to electricity.
Calpine's new project, however, takes advantage of the ubiquitous heat of the Earth's crust, drilling into hot, dry rock, injecting cold water under low pressure to expand existing cracks and gaps, and drilling a second well at the other end of the system. Once a link is established, water is pumped into the first well, heated as it passes through the network of fractures, then drawn back up as steam.
These enhanced geothermal systems, or EGS, could theoretically work in all 50 states, a prospect that appeals to the U.S. government.
Over the past four years, the Energy Department has spent $60 million on six EGS demonstration projects in Alaska, Oregon, Idaho, Nevada and California, including Calpine's. This summer, the agency nearly doubled its annual geothermal budget, allotting two-thirds of it - about $44 million - for EGS research.
The technology's profitability remains uncertain, but its potential is huge. The U.S. Geological Survey believes that the West alone - where the geothermal heat EGS requires tends to be shallower, about one to two miles down - may harbor enough energy to supply about 75 percent of the nation's needs. The country's total current production capacity from all power sources is 1,100 gigawatts.
The Energy Department conservatively pegs the entire country's EGS potential at 100 to 500 gigawatts, but a national assessment funded in 2008 by Google's philanthropic arm - which has invested more than $1 billion in various forms of clean energy, ostensibly to feed its server farms - estimates closer to 3,000 gigawatts.
"EGS could be the 'killer app' of the energy world," said Dan Reicher, Google.org's director of Climate and Energy Initiatives.
Tapping even a fraction of that power could be an enormous boon for efforts to rid our energy supply of carbon emissions. Unlike intermittent solar and wind, geothermal can provide always-on power to displace fossil fuels without storage or back up natural gas power plants.
Environmental impacts appear to be relatively few. The EGS fracturing process can cause small earthquakes, but seismic monitoring of Calpine's project has shown the risk to be considerably smaller than with hydraulic fracturing for oil and gas, in part because water is injected at low instead of high pressure. EGS fracturing also uses clean water, and is thus unlikely to contaminate drinking supplies.
Still, geothermal energy supplies only 0.4 percent of the nation's total capacity, mostly due to its upfront cost. A single well can cost $10 million or more, while the infrastructure needed to convert steam to energy runs into the hundreds of millions.
And there's no guarantee of success, even at a known reservoir. In recent years, price competition from natural gas, biomass, solar and wind has further constrained growth.
EGS is even more expensive than traditional geothermal.
Mineral content, temperature and permeability have to be just right, but without reliable remote sensing technologies, now being developed with Energy Department money, developers can't know what's underground until they get there.
"We're all looking at this huge resource and the question is, how can we economically use it?" says Karl Gawell, executive director of the Geothermal Energy Association. "You're going to need sustained government investment to make progress."
Calpine's project is the furthest along, with one well in Oregon and two in Nevada close behind. The Idaho project won't be complete until 2015, and the Alaska experiment is on hold after the first well failed.
Calpine recently proved that its $13 million Geysers experiment can generate enough steam to produce up to 5 MW, enough for about 1,250 homes. There's only one problem: It can't find a power buyer. Until it does, it won't be able to build a new plant nearby or a planned network of wells to feed it.
The Energy Department hopes its investments will help EGS plants produce power at 9 cents per kilowatt-hour by 2020 and 6 cents by 2030 - prices competitive with natural gas and other renewables.
Currently, however, Calpine is trying to find a buyer at 11 or 12 cents in order to afford the plant. But as long as California utilities can get power elsewhere for less, they aren't biting.
"We're going to keep trying," vows Calpine Senior Geologist Mark Walters.
Doug Hollet, manager of the Energy Department's Geothermal Technologies Program, has a more modest aim for now: "Our goal is that when people discuss renewables like wind and solar, they mention geothermal in the same sentence."