NREL: no differences in greenhouse gas emissions between PV technologies

PV Insider speaks with Garvin Heath, Senior Scientist at NREL’s Strategic Energy Analysis Center

Over 400 hundred life cycle assessments of the greenhouse gas emissions of the various PV technologies have been conducted and published over the last few decades. Researchers at NREL have now harmonised these into one meta-analysis.

Garvin Heath, Senior Scientist at NREL’s Strategic Energy Analysis Center is a co-author of two papers as part of a wide ranging harmonisation of Life Cycle Assessments (LCAs) of greenhouse gases  published in a special supplemental issue of the Journal of Industrial Ecology on Meta-Analysis of LCAs; covering all electricity generation sources. But first, he saw the need to get more conclusive results for stakeholders by harmonising the research on PV, with over 400 studies.

“There were hundreds of LCAs published and no one had taken the time to do any sort of review and no one had done it with any really systematic review and meta analysis on this wealth of previous research so the aim of this harmonisation project was to achieve those goals,” says Heath.

Most of the studies that passed their rigorous screening process were published by researchers at universities and national labs, many of them by similar author groups.

“We did a rigorous screening process to select those that met quality and relevance criteria,” Heath explains. “So a number of studies that were in this possible universe were whittled down to many fewer than 400 and from there we did our meta analysis process.”

There’s been intense interest in PV, because it’s something that holds the promise of a low greenhouse gas emissions profile, especially from the operations perspective, but it does have some embodied greenhouse gas emissions from upstream in the lifecycle; such as from making steel for trackers for example.

Various stakeholders have been interested in studying those figures and how they might change with different designs: there has been a great deal of change in the design of cells and modules, and systems over the course of the last few decades.

Fewer differences overall

Previous to NREL’s harmonised meta-analysis, the inconsistency between the methodologies of the multitude of studies had hampered direct comparison of their results. What NREL did find in the refining process, however, was that the results converged more, with the technologies largely overlapping.

“We did not find there were significant differences between the different types of PV that we evaluated, so really it ought to be viewed in the context of the whole electricity sector, and other options, rather than a myopic focus on one rather than another PV technology,” says Heath.

He emphasised that compared to coal and natural gas, solar has virtually no significant life cycle greenhouse gas emissions. There are no surprise revelations of greenhouse gas emissions in these studies or in their harmonised summary, and that regulating entities are aware of this.

These sentiments have been echoed in another context, by SolarReserve CEO Kevin Smith.


“Permitting agencies - at least in the US - understand the very minute risks from any form of solar, when compared to the alternatives; natural gas or coal,” he has said, speaking generally of the entire environmental permitting considerations, rather than just of greenhouse gas emissions.

He believes the regulatory entities take a balanced view. “I think generally they look at it in perspective with the other risks and the alternatives,”he says. “I mean in the overall analysis, you can look at the pollution that comes off a coal or nat gas plant, versus no pollution off a solar plant.”

Location, location, location

While some of the differentials revolved around the design used and minor differences among researchers’ estimates, the biggest was location.

One of the few differences spring not so much from a particular technology, whether thin film, or traditional crystalline, but from the location of a manufacturing facility, since electricity supplies vary from one region to another.

In the US, for example, solar panels are mostly manufactured in the Southwest: in California, which is currently supplied with more than 20 per cent from renewable sources (and an additional 15 per cent from hydropower) with the remainder from from natural gas.

But Tucson, Arizona, by contrast, still gets 75 per cent of its electricity from coal, so panels manufactured in Tucson would then have a higher emissions profile.

Nations with a higher percentage of renewable energy on the grid, would have an advantage in offering manufacturing with lower greenhouse gases.

So - who wants to know?

While it might seem remarkable for there to have been as many as 400 studies of the relative greenhouse gas impacts of the various PV technologies, no nefarious motivations appear to be behind these. While all of these LCAs have long differed in methods and assumptions, that has often been for legitimate reasons.

Heath says the interest in the results comes from a really broad range of governing stakeholders and regulatory bodies. “This harmonisation study was sponsored by the US DOE, which uses the results to guide some of their R&D regime-making and energy analysis,” he points out.

But there have been many other agencies at federal, state and local levels, including public utility commissions, who use the data. States have been interested in using it in integrated resource planning. And the interest is not just within the US. Researchers globally use the data, with the IPPC probably being the best-known entity that is utilising the results.

“Certainly, for the researchers who are interested for instance in doing a look at the breadth of the electric sector in the US - or other places in the world - and are doing specific research on a new LCA or technology; or want to know what the greenhouse gas emissions profile is, on an LC basis; this is a good place to turn to for them to do climate model projections or any other kind of energy scenario estimations of greenhouse gas emissions, and to have a lot of robustness behind it.”