What We Publish

Greenhouse gases, climate change and the transition from coal to low-carbon electricity

Nathan P. Myhrvold and Ken Caldeira · Environmental Research Letters 7, 014019 · 2012

Key finding. Replacing 1 terawatt of coal-fired electrical capacity over 40 years avoids half of the warming the coal system would have caused only after 43 to 53 years in a transition to solar photovoltaics, and a transition to natural gas takes a century or longer to reach even a 25 per cent reduction.

Two bar charts comparing energy technologies for a 40-year transition away from 1 terawatt of coal-fired capacity. Panel A plots temperature change over 100 years, from 0 to 0.35 degrees Celsius, as a pair of columns per technology for high and low life-cycle emissions estimates, each column stacked into contributions from coal burned during the phaseout, construction, operation, and waste heat and transmission. Staying with coal gives about 0.32 degrees and conservation about 0.06; natural gas and coal with carbon capture reach 0.24 to 0.27 and are dominated by operating emissions, while wind, solar, and nuclear fall between 0.06 and 0.15 and are dominated by construction. Panel B plots years to reduce warming by 0, 25, and 50 per cent below the coal baseline, on an axis running to 150 years. Conservation reaches the three marks at about 0, 20, and 40 years and wind and solar photovoltaics only a little later, while the natural gas and coal-with-carbon-capture bars run off the axis, labelled greater than 500 years.
The lower panel is the paper's argument in one picture. Technologies are separated far less by how much warming they eventually avoid than by how long they take to start avoiding it, and the partial measures sit decades behind the near-zero-emission ones. Figure 3 from Myhrvold and Caldeira (2012), Environmental Research Letters 7, 014019. Reproduced under CC BY 3.0. Extracted from the published PDF and resized for web display.

What question did this research address?

Comparisons between electricity technologies are usually made on annual emissions per unit of generation, as though a plant's climate effect began the day it opened. Building power plants also emits greenhouse gases, and the coal fleet being displaced goes on emitting throughout the years it takes to phase out.

This paper asked what warming actually follows from an energy system transition once both are counted, and how long a build-out has to run before it delivers a measurable reduction in warming.

What did we find?

The study couples life-cycle greenhouse gas emissions for each generating technology, taken across the published high and low estimates, to a simple climate model with a climate sensitivity of 3.18 degrees Celsius per CO2 doubling and a diffusive ocean. Transitions are simulated from 100 gigawatts to 10 terawatts in scale and 1 to 100 years in duration.

Where the emissions fall differs by technology. For wind, solar, and nuclear plants, construction dominates lifetime emissions; for coal and gas plants, operation does, with typically under 1 per cent of lifetime emissions attributable to construction.

Because construction comes first, a rapid build-out of low-emission plants can raise temperatures above the no-transition case at the outset. A 40-year transition from coal to solar photovoltaics spends 1.4 to 6.9 years warmer than if the transition had not been undertaken.

Idealized conservation — phasing coal out over 40 years and replacing it with nothing — sets the floor at 20 years to a 25 per cent reduction in coal-caused warming and 40 years to a 50 per cent reduction. Wind, solar thermal, solar photovoltaics, and nuclear at their low life-cycle estimates come within a few years of that floor.

Partial measures fare much worse. Natural gas emits roughly half what coal does per unit of capacity yet needs a century or more for a 25 per cent reduction in warming, and coal with carbon capture and storage reaches 25 per cent only after 26 to 110 years. More generally, any technology that cuts emissions relative to coal by a factor of only two to five requires century-long times to accrue substantial temperature reductions.

The delay is not an artifact of the transition schedule. Exponential and logistic build-outs, other build-out scales, and assumed rates of technological improvement all leave the central conclusion intact. Turning 1 terawatt of coal capacity off instantaneously, with no replacement at all, still takes 11 years to halve the warming that capacity would have caused.

Why does it matter?

The result reframes what the delay means. Because most of the benefit of a transition arrives only after it is complete, starting later does not postpone the benefit by the length of the delay alone — it pushes the entire lagged payoff further into a century when the damages are larger.

It also separates technologies that look similar on an annual-emissions ledger. A factor-of- two improvement over coal and a near-zero-emission technology differ modestly in tonnes per kilowatt-hour but differ by many decades in when they start reducing warming.

Judging energy policy on the emissions rate of a finished system misses the emissions of building it and of running the old system meanwhile. On the timescales politics works with, those two are most of the story.

Citation

Nathan P. Myhrvold and Ken Caldeira (2012). Greenhouse gases, climate change and the transition from coal to low-carbon electricity. Environmental Research Letters 7, 014019.

Related