Depth, not years: what a 'failing' geothermal plant actually missed

Decline is the wrong word for most of what gets called a "failing" energy asset. The plant did not run out of resource; it ran out of well.

Lightning Dock, a New Mexico geothermal site that came online in 2013, was written off this decade as marginal: inlet water had fallen to 250°F, far below the 310°F design minimum, and the plant was bleeding money. MIT Technology Review reports that Zanskar bought it in June 2024 on a different theory. The original production wells were only about 2,500 feet deep and sat on the very top of the reservoir, leaving the productive heat below them untouched. A new well drilled to 8,000 feet began operating in May 2025, and a year later it is still flowing at more than 4,000 gallons per minute. The plant is back at full capacity.

Lightning Dock illustrates a potential pattern for sites where wells were similarly shallow and subsurface modeling was cruder. "Decline" may be a reading error about where the heat actually sits rather than a verdict on the resource. The method travels for such sites: re-map with modern tools, drill deeper or off the original pad, then read the data for a year before calling the reservoir exhausted.

The honest limit lives inside the case. Cooling at Lightning Dock ran at roughly 10°F per year against a typical 1 to 2°F per year, so this is not a representative decline curve. The redevelop question only opens for sites where the wells were similarly shallow, not for fields that cooled for ordinary reasons.