Location matters: The new science of siting clean energy to push more carbon from the grid

What’s a watt of solar power worth in terms of avoided carbon emissions? The value varies dramatically from place to place and hour to hour — and that makes it challenging to know where you can get the most climate bang for the buck when you’re investing in clean energy.

This is a big deal for corporations looking to cut their carbon footprints, from companies like Microsoft and Google with ambitious plans to power their entire operations from carbon-free energy, to smaller businesses seeking products that allow them to clean up their energy purchases.

Data on the carbon-reduction value of clean power projects in different locations has started to emerge in recent years, giving companies new options to consider when deciding where to invest. The structures that would allow corporate investments to account for these values haven’t yet been fully fleshed out, but a growing number of firms are starting to take them into account anyway — and in some cases, take the next step of incorporating them into clean energy financing.

A new way to finance solar: Selling ​“emissionality” offsets 

Earlier this month, cleantech startup Clearloop broke ground on a 1-megawatt solar project outside Jackson, Tennessee that stakes out a bold new definition of solar’s carbon-reduction value. It’s the first utility-scale solar project in the country to be partially financed by selling the carbon emissions it will displace over its lifetime. These transactions will not take the customary form of renewable energy credits that average out that value over time, but rather of carbon offsets that are directly related to the power grid the project is connected to.

In more specific terms, about $400,000, or roughly one-third of the project’s cost, was raised via the sale of offsets for nearly 60 million pounds of carbon emissions. Buyers range from well-heeled corporations (Intuit) and asset managers (Vista Equity Partners) with clean energy targets to meet, to individuals, including children’s book author Stacy Clark, who is offsetting the carbon associated with the manufacture and shipping of her latest book.

To verify the carbon-displacement value of its future solar production, Clearloop is relying on WattTime, a startup that’s spent the past seven years developing technology to pull emissions data from power plants and energy markets and correlate it to times and places across the power grid. WattTime calls this intersection of variables ​“emissionality,” a play on the renewables industry concept of ​“additionality,” which differentiates between spending to build new wind and solar farms and spending to buy credits from existing ones.

The concept of additionality has had a sizable impact on corporate clean energy investments over the past half-decade or so, spurring companies to seek out projects that deploy additional renewable capacity. In the view of Clearloop CEO Laura Zapata, emissionality could make a similar difference, encouraging companies to make investments that have a clear carbon-reduction value.

Central to the concept of emissionality is data that allows projects like Clearloop’s to measure their ​“marginal emissions” impact — how their additional contributions of clean energy displace what would otherwise be dirty energy on the grids they’re connected to. One of the key findings of Clearloop’s efforts and other similar undertakings has been that a watt of new solar displaces more carbon on a dirtier grid than it would on a cleaner one. That won’t surprise insiders in the clean energy space, but until recently it’s been difficult to quantify.

In California, where solar power already supplies a large and growing share of the state’s electricity, ​“a metric ton of carbon is avoided by building something between 50 to 80 watts of solar,” Zapata explained. ​“In Tennessee or other more fossil-fuel-heavy regions, it’s 30 to 40 watts of solar.”

This estimate depends on a number of assumptions about how the difference between California’s and Tennessee’s emissionality ought to be calculated. The granular data used for such estimates has only recently become available, and companies involved in collecting and analyzing this data are still seeking broader industry consensus on the methodologies that should underpin the calculations involved.

But from Zapata’s perspective, measuring and valuing emissionality is a critical next step in capturing the benefits that aren’t yet being accounted for in most solar financing structures. These typically use average annual emissions reductions as measured by the offset credits exchanged between clean energy developers and buyers.

In the case of Clearloop’s 1-megawatt solar farm, the carbon-offset revenue is ​“fully additional” because the funding was necessary to build it, Zapata said. Clearloop and WattTime have developed a tool that allows carbon-offset buyers to track the emissions reductions from their investments over time in relation to changing grid conditions.

And because this Clearloop project is sited in a lower-income region, said Zapata, ​“it’s not just good for the grid cleanliness but [also] a good way to boost communities with more tax dollars and other benefits,” such as jobs and the economic development that can come from providing carbon-free power to new businesses.

The big question going forward, she said, is whether these emissions reductions can be made a part of standard clean energy transactions. While the Jackson project has followed the methodologies set out in the Greenhouse Gas Protocol, the industry standard for carbon credit transactions, the protocol and other voluntary carbon registries today ​“do not recognize carbon offsets from renewable energy projects in the United States,” she said.

Zapata contends that this needs to change. ​“Renewables are the biggest opportunity we have today to decarbonize our economy. We just need more capital and more creative ways to bring that capital to clean up the grid.”

Emissionality 101

Gavin McCormick, WattTime’s CEO, sees Clearloop’s new project as an important step forward for a concept he’s been promoting for years. (WattTime and Canary Media are both independent subsidiaries of RMI.)

“We need to count the carbon impact of what we’re doing, whether or not we get credit for it,” he said. ​“It’s not who you bought your power from or what [credits] you purchased, but what did you really do for climate change?”

McCormick isn’t alone in making this commonsense observation — or in seeking out the data to prove it. Meredith Fowlie, his doctoral adviser at the University of California at Berkeley, is the co-author of a 2013 paper that found that a unit of solar power generated in Trenton, New Jersey displaced more emissions than a unit of solar in Tucson, Arizona, even though the Arizona panel receives a lot more sunlight over the course of a year and thus produces more energy. 

The difference stems from both the time and the location of that solar production. Arizona, like California, has a lot of solar capacity due to its sunny and temperate climate — and it sometimes generates more solar power at midday than its grid has demand for. That leads to the risk of solar generation being deliberately restricted (what the industry terms ​“curtailment”) unless demand can be shifted to make use of the available solar resources.

So far, WattTime’s data has mostly been used to help behind-the-meter batteries, electric vehicles and other electricity users adjust when they consume electricity in order to fully utilize plentiful solar or wind power and avoid using power derived from fossils.

But according to McCormick, ​“There’s more impact from shifting in place than shifting in time.” WattTime estimates that choosing the locations of new wind and solar projects on the basis of their emissions-reduction impact could increase their reduction potential by up to one-third compared to siting that fails to take this factor into account.

The problem is that the economic considerations that determine where renewable projects actually get built today don’t align well with the goal of maximizing carbon-abatement impact. 

“We’re building all the renewable energy in the same few regions of the country” — states with renewable mandates or locations with the sunniest sun or windiest wind. ​“The solar panels go where it’s sunny, and pretty soon the price is dirt-cheap and we’re throwing away solar.” This is playing out more and more in places like California, which frequently has to curtail solar that exceeds midday grid demand, and West Texas, which is forced to do the same with excess wind power.

Location, location and…congestion

This leads us to another key factor tied to location: a site’s relationship to the transmission grid. If a new renewable project is built in an area with transmission bottlenecks, then some of the power it produces might not be consumed and its value in reducing carbon emissions is lessened.

That’s one of the chief findings from a new tool from REsurety, an energy analytics company providing customers including Microsoft, Hannon Armstrong, Quinbrook Infrastructure Partners and other major clean energy investors with a way to measure emissionality, which it refers to as ​“locational marginal emissions.”

Launched in July in partnership with Microsoft, REsurety’s tool crunches data collected by federal government agencies and from the system operators and transmission organizations that manage transmission grids and energy markets for about two-thirds of U.S. electricity customers. This historical emissions data is tied to the electricity consumption or generation at each node in the grid, and it can shed light on ​“emissions trends that can be expected to persist over time,” said Adam Reeve, the company’s senior vice president of software products.

One driver of the longevity of emissions patterns is the slow-to-change nature of transmission infrastructure, which can take up to a decade to construct. One commonly cited example of this phenomenon is wind power in West Texas, the most heavily developed wind market in the country, where a wind farm’s location on one side or the other of a transmission congestion point can make a huge difference in how much carbon it abates over the course of a year. 

Electricity typically flows out of the windy Panhandle region of Texas and toward load centers such as Dallas and Houston. Once flows reach a certain point, however, ​“the grid gets congested and no more clean electricity can be exported to displace gas or coal generation elsewhere,” Reeve said. At that point, ​“any incremental megawatt of wind in the Panhandle would simply displace another megawatt of wind, which means it doesn’t actually displace any carbon.”

That’s an issue for investors looking to value projects based on their carbon-abatement potential, rather than just their dollar value alone. Transmission congestion is reflected in the locational marginal prices on the Texas grid, but no such market mechanisms reflect the carbon impact.

Having the data on hand can help REsurety’s customers ​“make smarter decisions about how to invest wisely,” Reeve said, whether that’s in terms of buying existing clean energy projects or siting new projects to maximize their carbon impact.

The disconnect between marginal emissions impact and money 

For a company such as Microsoft that’s investing in a range of carbon-mitigation strategies, finding tools to measure the climate impact of these investments is increasingly important, said Brian Janous, the software giant’s general manager of energy and renewables.

“We score projects today based on the megawatt-hours, the prices and the cost of that production,” he said. ​“How do we create this other score, this other dial, that measures the marginal emissions?” Registries such as the Greenhouse Gas Protocol need to catch up with this line of thinking, in Janous’ view.

Reeve agreed that renewable energy credits, the chief store of value for corporate clean energy purchases, ​“are not necessarily indicative of the carbon value of that energy.” One way to address that ​“is essentially attaching a carbon attribute to the renewable energy credit,” he said. That concept is being pursued by the EnergyTag consortium, a group of more than 100 companies including Google and Microsoft that are working with energy companies and grid operators to account for the emissions-intensity of energy on an hourly basis, rather than the annual accounting now applied to these credits.

Kathleen Spees, a principal with The Brattle Group who co-authored a report on REsurety’s approach, highlighted the potential for emissions data to be bundled into the locational marginal price data that forms the backbone of state and regional energy markets across the country.

Mid-Atlantic grid operator PJM has begun publishing marginal emissions rates for informational purposes, although they’re calculated using annual averages for the generators involved rather than more granular data. Spees has proposed an ​“integrated clean capacity market,” one of many concepts being studied by New England’s grid operator as part of its exploration of ways to integrate state decarbonization policy goals into its market structures.

This kind of approach will become increasingly important as renewable energy displaces baseload fossil-fuel-fired power plants in greater proportions, she said. That’s already the case in Ontario, Canada, where hydropower and wind power supply more than 90 percent of annual energy needs but often must be curtailed when generation exceeds demand.

“At some point, the more economical approach is to shift to a technology that’s more expensive on a megawatt-hour basis but is more cost-effective if you’re talking about dollars per ton of carbon,” Spees said.

That might apply to batteries and other energy storage assets, which today are not eligible to produce renewable energy credits but which can store excess clean energy for use at a later time in ways that have emissions-reduction value, Spees said.

“You have to give them an incentive to do the flip — to charge on clean and discharge against fossil,” she said. ​“And until we have data, you have no information basis you can use to provide that operating incentive.”

Catching regulations and markets up to the data

WattTime is just one of a host of groups seeking to bring marginal emissions data into the realm of corporate procurement and greenhouse gas accounting.

Another is Tomorrow, a Danish company that’s developed an ​“electricity map” — an open-source rendering of the carbon-intensity of power grids across the world. The company is working with Google to provide hourly forecasts of emissions-intensity on the grid as part of the tech giant’s efforts to shift its data-center electricity usage to clean power when it’s available.

That’s a shift away from the ​“average” emissions data that now informs most corporate energy purchasing decisions and toward a ​“physical” representation of energy tied to place and time, Tomorrow CEO Olivier Corradi explained in an interview.

So far, the tools and standards used to assess this accounting, like those produced by the global consortium that sets the Greenhouse Gas Protocol and the United Nations’ Clean Development Mechanism, have yet to incorporate these data sources.

It’s difficult to quantify and certify these values because it’s hard to ​“make a calculation that’s uncontroversial,” Corradi said. In simple terms, measuring what would have happened to grid emissions if a solar or wind farm seeking to be credited for reducing them hadn’t been built requires proving that ​“the thing would not happen without you.”

Establishing that proof will demand a lot of data, as well as a convening of the stakeholder process that approves changes to the Greenhouse Gas Protocol, before renewable energy projects like Clearloop’s will be able to attach their emissionality values to their renewable energy production, as opposed to employing the voluntary carbon credit structure the company is using for its Tennessee solar farm.

The Renewable Energy Buyers Alliance, a group representing corporate clean energy purchasers, is pressing for data transparency and harmonization in advance of plans to ​“put a dollar value to” marginal emissions, said Priya Barua, REBA’s director of zero-carbon innovation. ​“How do you account for the avoided emissions if you don’t even have that baseline?”

Corradi agreed that a standardized baseline and shared understanding are important. ​“If you do not let people look under the hood at how we have built this, they will not trust it.” But if trust is built and institutions buy in, then measurements of marginal emissions could give companies a range of new options for cutting carbon, he said.

That’s why Tomorrow has opened up its electricity map process to collaborate with data scientists around the world. ​“We are trying to build the most trustworthy and unequivocal signal.” 

Read a related article: Should Google and Microsoft focus on sourcing their own 100% clean power or cleaning up the dirtiest grids?