This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


In Southern California Gas Plant Debate, Solar+Storage Could Be the Winner

Rosana Francescato's picture
Communications Director Clean Coalition

Rosana is Director of Communications at the Clean Coalition, a nonprofit organization whose mission is to accelerate the transition to renewable energy and a modern grid through technical...

  • Member since 2018
  • 131 items added with 64,537 views
  • Sep 14, 2017 10:00 am GMT

Your access to Member Features is limited.

Solar power is becoming competitive with natural gas — and  is already cheaper than new fossil fuel capacity in many countries.

Add storage to the mix, and the costs come down even more. That’s what the Clean Coalition has found in a recently released study, which showed that solar+storage would be cheaper than two proposed gas plant projects in Southern California.

The proposal for the gas plants calls for building a new Puente Power Project plant and refurbishing the existing Ellwood Peaker Plant. Both are part of the Moorpark Subarea, which includes the cities of Oxnard, Santa Barbara, and Goleta. They were approved to meet local electricity capacity requirements that are currently served by about 2000 megawatts (MW) from the Mandalay and Ormond Beach power plants in this grid-constrained area of the Southern California Edison (SCE) service territory. Because both the Mandalay and Ormond facilities are out of date and will not conform to new state regulations, they’re expected to be retired at the end of 2020.

According to the Clean Coalition’s models of alternatives to the gas plants, a solar+storage solution would cost $267 million to install, compared to $299 million for the Puente proposal. Solar+storage could replace both Puente and Ellwood for approximately $406 million.

This is in stark contrast to a study conducted by the California Independent System Operator(CAISO), which manages the state’s electric grid. CAISO’s study evaluated only storage, while ignoring far more cost-effective solar+storage, and concluded that replacing the Puente plant with incremental distributed energy resources and storage would cost $805 million, with a cost of up to $1.1 billion to replace both the Puente and the Ellwood plants.

Including 30 years of operations, maintenance, and fuel at current values would add over $550 million to the SCE Puente figure; adding these costs to the Clean Coalition solar+storage and CAISO storage-only cases would add far less. Note that the only “fuel” costs associated with storage are round-trip inefficiencies, which are minor compared to a peaker plant burning natural gas.

“The CAISO study was highly valuable in demonstrating that distributed renewables are technically capable of meeting the reliability needs of the Moorpark Subarea,” said Craig Lewis, Executive Director of the Clean Coalition. “However, it overlooked a number of significant factors, including updated costs for solar+storage and the opportunity to apply the 30% Investment Tax Credit (ITC) to storage. With our experience in staging complex projects, the Clean Coalition was able to develop the most comprehensive model to date.”

The Clean Coalition notes that their model addresses these issues:

  • The model uses a cost-effective solar+storage solution, rather than modeling storage alone as was done by CAISO.
  • The Clean Coalition uses up-to-date component cost estimates for 2018, compared to CAISO’s outdated storage costs from 2014. The cost of storage has fallen by over 40% since then.
  • The Clean Coalition appropriately sizes the storage required, by modeling the real generation and dispatch capabilities of solar+storage. CAISO’s unrealistic profile of solar output and storage dispatch resulted in underestimating the energy generation of solar by nearly half and oversizing of storage.
  • The Clean Coalition includes the impact of the 30% federal ITC, which can substantially lower the cost of solar+storage facilities, provided that 70% of the storage charging comes from co-located renewables. Because CAISO modeled additional storage without renewables, it could not account for reaping ITC benefits that result from an implementation featuring solar+storage.
  • For demand response costs, which CAISO overestimated, the Clean Coalition model uses data from an April 2017 Lawrence Berkeley National Labs analysis, as well as current demand response contract costs as reported by Greentech Media in April 2017.

The CAISO study, as well as subsequent analysis by Greentech Media, also left out the costs of operations, maintenance, and fuel. These are expected to run approximately $19 million per year for Puente based on current costs, making it about twice as expensive as a solar+storage design. Accounting for these costs would raise the total cost of Puente to over $870 million over thirty years. A comparable calculation for a solar+storage facility would run about $462 million. Including the health, mortality, and social costs of carbon from the natural gas plant would increase the cost of Puente and Ellwood dramatically.

“Beyond being more cost-effective, the solar+storage approach provides substantial additional functionality and community benefits,” said Mr. Lewis. “And any quantification of the health and environmental value of solar+storage versus gas plants just adds to the vastly superior value of solar+storage.”

More details can be found in the Clean Coalition filing to the California Energy Commission (CEC) on the Puente Power Project and the Clean Coalition model for alternatives to Puente and Ellwood, both available online. A prior Clean Coalition cost analysis of a solar+storage alternative to the Ellwood Peaker plant is also available online.

Late last week, SCE filed a motion to strike the Clean Coalition’s testimony to the CEC on the Puente project. It’s still unknown whether that attempt will be successful. An evidentiary hearing on the Puente project is scheduled for September 14 – 15 in Oxnard.

Rosana Francescato's picture
Thank Rosana for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Bob Meinetz's picture
Bob Meinetz on Sep 14, 2017

Rosana, apples/oranges.

Solar, as an intermittent source of electricity, is only “competitive” with gas for those who don’t mind using candles at night. Quakers, and a few hobbyists.

Without denying either acknowledgement of their individual merits, a Nissan Sentra is not “competitive” with a Ferrari Testarossa (nor, vice versa).

Rosana Francescato's picture
Rosana Francescato on Sep 14, 2017

The comparison here is to a solar+storage solution, which solves the issue of intermittence.

Bob Meinetz's picture
Bob Meinetz on Sep 14, 2017

Rosana, can you give me an example of anywhere on the planet that has been achieved without backup from fossil fuels?

Joe Deely's picture
Joe Deely on Sep 14, 2017

I heard CEOs of two Bay Area CCAs mention recently that they were looking at PPA proposals for solar+storage that were cheaper than NG alternative.

Engineer- Poet's picture
Engineer- Poet on Sep 14, 2017

Solves?  Really?  Here’s what the actual report says:

A PV Solar and storage system of 120 MW of groundmount solar and 75 MW of energy storage with a total capacity of 220 MWh could meet the entirety of the capacity requirement.

Less than 3 hours of storage.  Even if it started fully-charged in the morning, one cloudy day would empty it… and then you’d have nothing for the evening.  This system is going to be totally dependent on natural-gas backups.

The spreadsheet details that the storage costs are assumed to be $400/kWh.  This is in line with Tesla numbers, but cost increases linearly with hours of storage.  Providing 8 hours of storage to provide overnight operation (600 MWh at 75 MW) plus an additional 120 MW of PV capacity to charge it raises the cost to $583 million.

And you yourself wrote this:

The Clean Coalition includes the impact of the 30% federal ITC, which can substantially lower the cost of solar+storage facilities, provided that 70% of the storage charging comes from co-located renewables.

It doesn’t actually cost less, it’s just someone else paying almost a third of it.  Even with the ITC, the full-overnight version costs over $400 million.

The capacity factor of PV is something like 20%.  At 1.43/W, that’s $7.15 per average watt.  It makes far more sense to cease the war on Diablo Canyon than to cover more ground with black rectangles.

Bob Meinetz's picture
Bob Meinetz on Sep 14, 2017

EP, I just received word that California climate bill SB-100, which would have limited CA climate targets to “eligible renewable energy sources” only, is dead in the water.

Seems there are California industrialists and agriculturalists who depend on reliable electricity, make up a significant proportion of California’s tax base, and subscribe to your line of thinking.

Bob Meinetz's picture
Bob Meinetz on Sep 14, 2017

Joe, did the CEOs of the CCAs mention how they will confirm the electricity they’re buying is exclusively generated by solar panels, or are they taking somebody’s word for it? I’d hate to think tens of thousands of Marin County marijuana growers had put explicit faith in their CCA, only to discover their NG “alternative” is CAISO’s grid mix wrapped up in green packaging.

True, it’s hard to keep grow lights burning 24/7 when you’re at the mercy of the earth’s rotation and weather, even with batteries. Maybe CCAs could purchase Renewable Energy Credits (RECs), or coupons good at Wyoming restaurants, or credits on souvenirs at Intermountain’s Gift Shop in exchange for enough coal to get them through the hard times (?)

Nathan Wilson's picture
Nathan Wilson on Sep 15, 2017

Scanning through both the CAISO and Clean Coalition documents, I’m skeptical that the report by the engineering team and that from the lawyer/biologist really addressed the same issues. There seem to be plenty of holes in the Clean Coalition analysis.

I’m willing to accept that the CAISO report may have used some pessimistic old pricing for PV and storage, but the Clean Coalition appear to ignore some serious problems with their proposal. They reduce the needed storage by claiming PV will support a portion of the load, but neither report shows the maximum cloudy/smoggy day demand. They criticize CAISO for pricing expensive roof-mounted PV, but don’t explain where (in this densely populated Los Angeles suburb) their proposed ground-mounted system will be located (the CAISO report says the limiting condition is failure of nearby transmission lines). It is also highly troubling that Clean Coalition chose to compare the operating cost of the two systems without the applying a time value to future fuel purchases, and apparently ignoring battery replacement entirely.

As the California grid evolves over time, we can expect more and more solar power to be supplied from in-land desert locations. A natural compliment to this is not more PV in cities, but rather a combination of storage and backup thermal generation near the the load (i.e. what CAISO studied).

Bob Meinetz's picture
Bob Meinetz on Sep 15, 2017

Rosana, what evidence do you have “solar+storage” solves the problem of intermittence? There are practical obstacles to your theory which make it a non-starter:

1) No amount of storage could prevent, during an extended period of cloudy weather, the grid running out of electricity – forcing hospitals, law enforcement, fire departments, airports, and other critical facilities to shut down. Not an option.

2) Storage expense – to power California’s grid for a single cloudy day, at today’s prices, would require constructing li-ion battery banks at a cost of ~$700 billion, or roughly four times the entire annual state budget. They would need to be replaced every 7-10 years. Not an option.

3) Expense of solar PV – Topaz Solar Farm, one of the world’s largest, is capable of generating an average of ~2 GWh on a cloudless day. California burns through ~480 GWh / day. California would have to build 240 Topaz-scale farms, occupying 2,160 square miles of land, at a cost of $2 billion each ($480 billion, total) to keep up with consumption. They would need to be replaced every 25 years. Not an option.

These are optimal assumptions, in every instance giving your theory the benefit of the doubt. They show “solar+storage” to be obviously not a viable option.

Joe Deely's picture
Joe Deely on Sep 15, 2017

You guys seem to be worried about storage succeeding in CA.

Nathan has it mostly right:

As the California grid evolves over time, we can expect more and more solar power to be supplied from in-land desert locations. A natural compliment to this is not more PV in cities, but rather a combination of storage and backup thermal generation near the the load (i.e. what CAISO studied).

Solar continues to get cheaper as evidenced by it recently meeting the $1/watt Sunshot goal.

There is no end in sight for continuing improvements on the solar roadmap so more and more utility scale solar power will be added to CA grid. In a few years, Commercial rooftop solar will get close to $1/watt and this will greatly accelerate commercial installations.

Storage is just getting started but based on recent EV announcements it looks like WW production of batteries is gonna accelerate quickly. By 2025, I would expect the WW battery market to be about 500GWh/year and growing quickly.

Within 5 years CAISO might have about 6GWh of Battery storage. As you can see from here this would be 1% of the daily load.
Currently, NG share of electricity is about 180GWh per day. Along with increases in renewables – additional storage will gradually lower this number. NG production easily be down to 100 GWh/day by 2030.

Long story short – CA is gonna have a lot of NG plants with low Capacity factors. Adding any new NG capacity will only exacerbate this issue and so this Puente plant should only be added as a last resort.

Joe Deely's picture
Joe Deely on Sep 15, 2017


Thought you as well as others on this site might like to look at this recently released report from the California Energy Commission.

It’s a regular recap they do tracking progress toward the CA legislated Renewable share goals – 33% by 2020 and 50% by 2030. It’s really well done.

The 33% goal for renewables by 2020 will probably be met by 2017/2018. Hopefully the legislature will pass legislation to move the 50% by 2030 goal up to 2025. The 2030 goal is too easy. Better yet from your point of view maybe they should pass a goal for zero carbon share – including nuclear and Large Hydro. How about an 80% ZC goal by 2030?

Also, here’s a comment from the report regarding CCAs:

At the beginning of 2017, five CCAs were
operating in California and collectively
serving 915,000 customers: MCE, Sonoma
Clean Power, Clean Power SF, Lancaster
Choice Energy, and Peninsula Clean
Energy.18 By July, three more CCAs –
Silicon Valley Clean Energy, Apple Valley
Choice Energy, and Redwood Coast Energy
Authority – began serving customers; the
CPUC anticipates another CCA is soon to be
operational, and an additional 11 CCAs are
prospective or under exploration. Recent
estimates predict that as much as 25 percent
of IOU retail electric load could be unbundled by the end of 2017 by CCAs, self-generation, and electric service providers; this number could reach 85 percent in the next decade, or as many as 15 million-20 million customers

As you can see Bob – not just Marin. Also, here is an example of a recent PPA that was signed by Peninsula Clean Energy. You will see many more of these coming over the next 5-10 years as CCAs establish themselves.

Joe Deely's picture
Joe Deely on Sep 15, 2017

The point made in the study is that CAISO used old costs and these costs are changing rapidly. CAISO’s model used data from a 2016 NREL report. However, there is now a 2017 version available. See figure ES-1 in this report.

The $1.45/W has now dropped to $1.03/W. Boom.

The capacity factor for solar PV as determined by EIA for 2016 was 27.2%. So really about $4 per average watt vs the $7.15 you calculated. Will be even cheaper by 2018/2019 which is when this plant would actually be built.

As for the storage – bid it out and see what price comes back for a 2018/2019 project. Let me give you a hint – the bid amount is gonna be a lot lower than $400/kWh.

I am not familiar with grid config in this area so perhaps it still makes sense to have a small NG plant there, but let’s at least have the numbers correct.

Thorkil Soee's picture
Thorkil Soee on Sep 16, 2017

Or giving up the green dream and go nuclear.

Rosana Francescato's picture
Rosana Francescato on Sep 19, 2017

We’re talking about replacing the Puente plant here, not about powering all of California. Solar+storage facilities like the one proposed are in place and working now. Tesla’s 17 MW solar + 52 MWh storage facility on Kauai can store enough energy to power 4,500 homes through the night. It does this at a cost of 13.9 cents per kWh for the 20-year contract, under half the price the island currently pays for power generated from burning diesel fuel, its prevailing source of energy. And AES’ scalable 28 MW solar + 100 MWh storage facility, also on Kauai, will provide 11% of the island’s electricity needs starting next year, at a game-changing 11 cents per kWh for dispatchable solar:

Rosana Francescato's picture
Rosana Francescato on Sep 19, 2017

The solar+storage alternative to Puente would not have to be sited in just one location. And even a non-ground-mounted system would be significantly more cost-effective than what CAISO modeled. They had many other flaws in their model, including outdated storage costs.

Bob Meinetz's picture
Bob Meinetz on Sep 19, 2017

Rosana, solar+storage will never “replace” the Puente plant. During any extended period of cloudy weather the storage part will run out of electricity and CAISO will be forced to fall back on burning gas.

Last year, customers on Kauai burned through 452 GWh of energy. At the island’s capacity factor of 24%, Tesla’s 13MW facility will be capable of generating 27 GWh, or slightly more than half the energy you say it will (6%). Solar panels could be free, but (for lack of other renewable sources) when Tesla’s batteries come up short Kauai will be forced to fall back on diesel fuel.

Unfortunately, we can’t pay the sun to shine more than it does – that’s what I’m talking about.

Bob Meinetz's picture
Bob Meinetz on Sep 19, 2017

Joe, your assessment for rooftop isn’t backed up by facts:

Seems like just yesterday there was no end in sight for Solar City.

Rex Berglund's picture
Rex Berglund on Sep 19, 2017

Joe was talking utility scale and commercial rooftop, not residential. In fact, the article to which you linked says “The Q2 2017 Solar Market Insight report found that non-residential solar capacity — which includes commercial and industrial, public, and community solar — grew by 30 percent year-over-year.”

Bob Meinetz's picture
Bob Meinetz on Sep 20, 2017

Rex, for the last two years in California solar has been tanking on all fronts:

Solar power led the nation last year among new sources of electricity production, but growth slowed significantly as California homeowners and businesses cooled to the idea of rooftop panels.

Maybe California has cooled to the idea of rooftop panels because they have a little experience with them, and they aren’t delivering as promised:

Overall, solar still provides just 4% of the nation’s electricity capacity…The California Solar Energy Industries Assn. warns that if policymakers fail to pay attention to changing growth patterns, rooftop solar could face significant harm…”You try to do the math with consumers, you can’t even with a straight face look across the table and explain to consumers how they’re going to save,” Del Chiaro [California SEIA Executive Director Bernadette Del Chiaro] said. “Policies need to stop harming this market.”

So even SEIA can’t keep a straight face when trying to explain to consumers how they’re going to save…maybe it isn’t policies which are harming the market, but pretending a fickle, intermittent source of energy…isn’t:

SEIA expects the number of solar installations to triple the current 42.8 gigawatts of capacity over the next five years…Part of the growth will be driven by a growing number of state requirements such as California’s 50% clean energy mandate by 2030. Senate leader Kevin de Leon (D-Los Angeles) proposes to make it 100% clean energy by 2045.

Kevin DeLeon’s 100% renewable energy mandate was pulled two weeks ago after word got around Mark Z. Jacobson’s misty-eyed renewables vision had been debunked, DeLeon had the sense to recognize a false flag raised by natural gas interests, and utilities – who also have a little experience with solar – balked at the idea.

Hopefully Minnesota will learn from California’s expensive lesson.

Helmut Frik's picture
Helmut Frik on Sep 20, 2017

In germany, costs for commercial rooftops is in the area 0f 0,8-0,9 €/ W, and utility scale solar at 0,6-0,7€/W. The prices in the US are ikely to follow this example.

Rex Berglund's picture
Rex Berglund on Sep 20, 2017

tanking on all fronts

except commercial and industrial, public, and community solar, which in Q2 grew by 30 percent year-over-year.

Joe Deely's picture
Joe Deely on Sep 20, 2017


Here’s what I said:

In a few years, Commercial rooftop solar will get close to $1/watt and this will greatly accelerate commercial installations.

Once commercial hits that price, developers will be actively looking for large commercial rooftops and working with owners to build out these solar installations. It’s pretty simple economics. This may occur slightly earlier in CCA communities which will be actively seeking local solar.

As for residential rooftop – I don’t really think that segment will take off till at least the mid 20’s. By then, costs will be low enough that most new houses will have some solar installed and it will make sense to add solar to any home remodel. Storage will also start coming into play around the mid 20s.

CA has about 5GW of rooftop solar now. This number will easily climb to 15GW by 2030.

Joe Deely's picture
Joe Deely on Sep 20, 2017

As for solar tanking in CA over the past two years… let’s take a look at your favorite source of data the EIA.

Here is YTD solar in CA for
2017 (thru June) – 15,628 GWh
2016 (thru June) – 11,709 GWh

up 33% Y-Y

How about if we look back two years?
2015 (thru June) – 8,633Gwh –

CA solar generation is up 81% over the last two years.

That’s tanking?? Hmm… hope we keep tanking then.

So how about your favorite source of electricity? Natural Gas?

2015(thru June): 49,202 GWh
2016(thru June): 42,238 GWh
2017(thru June): 35,003 GWh

Down 29% over the last two years. Isn’t it NG that is tanking in CA?

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »