Potential Two Million Jobs from Rooftop Solar, for 10 Years

By Will Driscoll

The continental U.S. could produce 38 percent of its electricity from rooftop solar installations, according to a report from the National Renewable Energy Laboratory (NREL).  Yes, that’s a lot:

  • It’s 1,118 gigawatts of solar capacity—or almost 80 times the 2016 U.S. solar installations of 14.6 gigawatts.
  • Installing that much rooftop solar would yield about two million jobs per year for ten years—that’s eight times the number of U.S. solar jobs in 2016.

The NREL analysis evaluated the potential for solar on buildings with at least one unshaded roof plane that is either nearly flat, or faces east, southeast, south, southwest, or west.  If any such roof plane could accommodate at least 1.5 kilowatts of solar panels, NREL modeled solar on that roof plane.  Summing across all buildings yielded a technical potential of 1,118 gigawatts of rooftop solar.  NREL found that 66 percent of large building rooftop area is suitable for solar, versus 49 percent for medium buildings and 26 percent for small buildings.

The technical potential is simply what the laws of physics allow, combined with common sense—i.e., no north-facing panels. (NREL did count west-facing panels, which have value for meeting late afternoon electricity demand, and east-facing panels, which are equally productive.)  NREL assumed an average solar panel efficiency of 16 percent, and noted that if panels averaging 20 percent efficiency were used, the solar potential would be 25 percent greater (because 20 is that much greater than 16).  At least three firms make solar panels exceeding 20 percent efficiency.

The technical potential is just a theoretical concept.  Yet the economic potential–that is, the rooftop solar installations that would save building owners money–may not be far behind.  Especially over the next ten years, as solar costs keep falling due to technology improvements and economies of scale, the economic potential will keep rising.  Already, more building owners each year realize they can save money with rooftop solar.

The Solar Energy Industries Association reported 2016 U.S. solar installations of 14.6 gigawatts.  Installing NREL’s potential 1,118 gigawatts of rooftop solar over ten years would mean 112 gigawatts per year, or about eight times the amount installed in 2016.  The Solar Foundation counted 260,077 U.S. solar workers in 2016, so an eightfold increase from that level would be about 2 million jobs—again, for a ten-year period. Utility-scale solar jobs would be additional.  The U.S. construction industry—with 9.6 million jobs in 2015—would far outpace the rooftop solar industry, but rooftop solar would make a solid contribution.

For rooftop installations, the number of jobs per gigawatt installed would arguably be higher than the U.S. average in 2016, because rooftop jobs are smaller and more labor-intensive than utility-scale solar projects.  On the other hand, with a big increase in the size of the rooftop solar industry, economies of scale should also come into play.  So on balance, a potential two million jobs per year for ten years seems like a good ballpark estimate.

Will Driscoll is a writer and analyst.  Previously he conducted environmental analyses for EPA, as a project manager for ICF Consulting.  He earned a master’s degree in economics and policy from Princeton.

Dominion’s Own Model Shows that 15 Gigawatts of Solar Would Save Virginia Customers $1.5 Billion

By Will Driscoll

Dominion Virginia Power has modeled a resource plan with 15 gigawatts of solar power, which it calculated would save Virginia customers $1.5 billion compared to a plan that includes a $19 billion nuclear reactor.  Yet when the company submitted its menu of resource options to regulators at the State Corporation Commission as part of its 2016 Integrated Resource Plan (IRP), it included the North Anna 3 nuclear plant while omitting the high-solar option.

The high-solar option only became public when attorneys Will Cleveland and Peter Stein of the Southern Environmental Law Center (SELC), representing an environmental coalition, asked the right questions during the discovery phase of the IRP proceedings.

Utilities in 33 states must periodically file an IRP.  The IRP is intended to define the least-cost set of resources that can meet forecasted electricity demand plus a reserve margin, while also meeting the state’s policy goals on renewables and efficiency.  Utilities use computer models to develop an IRP.

Dominion’s utility planning model generated the 15,000-megawatt solar option when the utility set no constraint on the amount of solar that could be added.

The high-solar plan would actually save Virginians much more than $1.5 billion, according to an expert witness in the IRP hearing, former Texas Public Utility Commissioner Karl Rabago.  The projected $1.5 billion in savings would be after Dominion’s projected $5.8 billion of solar integration costs (i.e., any costs needed to adapt the grid for a high level of solar).  Yet the $5.8 billion value “is at least 54 to 84 percent higher than the PJM high and low [integration cost] numbers that [Dominion] cites,” Rabago said.  Thus, “the overall savings … [with] a more reasonable approach to the integration costs would be much higher than $1.5 billion.” (PJM is a regional transmission organization that coordinates the movement of electricity through Virginia, Maryland, Delaware, New Jersey, Pennsylvania, Ohio, the District of Columbia, and parts of seven other states.)

To those who have followed the low and still-falling costs of utility-scale solar, it may not be surprising that solar, including any integration costs, would cost less than the proposed North Anna 3 nuclear reactor.  But to learn that Dominion’s own utility planning model presented that result to Dominion is a revelation.

To justify discarding the high-solar option, Dominion executive Robert Thomas said that “15,000 megawatts of solar… was a lot of land.” Yet data from the National Renewable Energy Laboratory show that this amount of solar would need only 0.4 percent of Virginia’s land area (i.e., 15000 MW times 7.9 acres per MW, divided by 27.376 million acres of land).  Mr. Thomas also said that the high-solar option “could create reliability issues,” yet high-renewables utilities in Iowa, South Dakota, California and Europe are highly reliable, thanks to accurate day-ahead weather forecasting and sophisticated utility “unit commitment” models that are also available to Dominion.

The State Corporation Commission, in its final order regarding Dominion’s IRP, did not mention the high-solar option.  The SCC approved the IRP as submitted, noting that “approval of an IRP does not in any way create the slightest presumption that resource options contained in the approved IRP will be approved in a future certificate of public convenience and necessity (“CPCN”), rate adjustment clause (“RAC”), fuel factor, or other type of proceeding governed by different statutes.”

SELC attorney Will Cleveland called on Dominion and the SCC to do better next time: “Citing ‘feasibility concerns,’ Dominion rejected and buried the high solar resource plan without any legitimate analysis of whether the plan was in fact feasible. Virginia ratepayers deserve the lowest-cost, cleanest energy available, and it is increasingly clear that means more solar, not more fossil fuels or nuclear. In the future, Dominion should not be allowed to dismiss the cheaper, cleaner resource plan without a full analysis.”

The environmental coalition represented by SELC consisted of Appalachian Voices, Chesapeake Climate Action Network, and the Natural Resources Defense Council.

 

Making the Case for Energy Storage in Integrated Resource Planning

Utilities are missing the opportunity to procure advanced storage as a cost-effective capacity resource, argues the Energy Storage Association.

by Will Driscoll

With utilities preparing to invest billions of dollars in system capacity over the next several years, “the time is now” for utilities to include energy storage in their integrated resource planning, according to the Energy Storage Association.

To make its case, ESA recently prepared a primer on how to evaluate advanced energy storage as a long-term option for system capacity and flexibility. The primer builds on a 2014 Navigant Consulting report prepared for the trade group.

Utilities in 33 states are currently required to file and gain regulatory approval for their integrated resource plans. IRPs are created to determine the least-cost combination of resources that enables a utility to meet forecasted demand, as well as some established reserve margin, over a specified future period, typically ranging from 10 to 20 years.

Energy storage is becoming a more compelling resource option, ESA argues, given the declining costs of energy storage and a rising number of large-scale deployments. Non-hydro utility-scale storage deployments in the U.S. have grown from about 60 megawatts in 2013 to 192 megawatts in 2015, and single deployments under contract now range up to 100 megawatts. U.S. deployments to date have been concentrated in California and the PJM wholesale electricity market, where storage is primarily used for renewables integration and ancillary benefits.

While some utilities have expressed interest in studying energy storage in the context of resource planning, ESA notes that informational barriers remain. The problem is that many IRP modeling systems are not granular enough to capture the flexibility of storage operations, and use inaccurate and outdated cost information.

“Utilities are thus missing the opportunity to analyze, evaluate and procure advanced storage as a cost-effective capacity resource, putting ratepayers at risk of significant imprudent investments,” the primer states.

 

Access to energy storage data has been a concern for some utilities, said Jason Burwen, policy and advocacy director at ESA. But “now there is data on cost and performance,” he said. Energy storage “is no longer something that can be screened out.”

Today, a utility’s challenge relates more to choosing a resource planning model that can fully represent the benefits of storage and how the technology functions.

Accounting for time intervals in IRP modeling is one of the challenges. “Typical production cost models are relatively simple and calculate economic options by modeling generator operations to meet expected load for each hour chronologically over a period of many years,” according to the ESA paper. “The main shortcoming of this type of model is that advanced storage can provide flexibility services on an intra-hourly basis, and there is no way to capture that service in an hourly model.” Simpler models that extrapolate from a small sample of hours also exclude significant storage services.

ESA says the solution is to use fine-grained planning models that examine sub-hourly intervals. Several commercial models that can examine intra-hourly dynamics already exist.

Another issue is that current IRP modeling systems fail to account for the full range of benefits that energy storage offers. In addition to capacity, advanced energy storage provides high value flexibility services, like frequency regulation or ramping support. But planners don’t always have the right tools to evaluate those flexibility benefits and subtract them from the cost of storage — or what ESA calls a “net cost of capacity” analysis of capacity investment options.

To capture the benefits of storage, the primer highlights a net cost of capacity formula used by Portland General Electric in its 2016 draft integrated resource plan: Net cost of capacity = Total installed cost – Operational benefits (i.e., flexibility of operations and avoided costs). Preliminary findings from the draft IRP found that operational benefits of storage were expected to be approximately double the capacity value ($90 per kilowatt per year versus $40 per kilowatt per year, respectively).

Many operational benefits of energy storage accrue as avoided costs, the primer states. These benefits include reduced operating reserve requirements, reduced curtailment of renewable resources and reduced local emissions for areas with emissions restrictions. The ESA primer includes the following table to illustrate the operational benefits that storage can provide.

Finally, ESA notes that it’s critical that planners use the latest advanced storage cost estimates and forecasts for cost model inputs. Cost data is now available from several sources, including GTM Research. Energy storage has demonstrated a declining cost curve over time, due to increasing scale and manufacturing capability. ESA’s projected net cost of storage is shown by the blue band in the graph below.

As energy storage becomes more competitive, ESA argues that the case for utilities to include storage in their IRPs becomes more compelling. And as resource mixes continue to include more variable generation sources, ESA notes that supply flexibility will become essential. In this context, the trade group urged utility regulators to use their existing authority to ask utilities to appropriately consider storage in resource planning “and ensure they meet their duty to ratepayers.”

Will Driscoll is a writer and analyst. Previously, he served as a project manager at ICF Consulting, where he conducted analyses for the U.S. Environmental Protection Agency.

This article was originally published on GreenTechMedia.com.