Extreme Summer Weather Threatens Gas Power Plants. Here’s How. 

September 13, 2023 | 7:30 am
Gas plant in California with pipeline in foregroundJohn Ciccarelli/BLM
Paul Arbaje
Energy Analyst

Despite the clean energy transition that is well underway in the United States, methane gas, or natural gas, remains the largest source of U.S. electricity generation. By a lot. 

Renewables are continuing to increase their share of the power sector, rising to almost 23 percent of the nation’s electricity supply last year, which was higher than the generation from both coal-fired and nuclear plants. This is made possible in large part by the steep cost declines of wind and solar. But despite the more favorable economics of solar and wind compared to gas, that figure is only a little more than half of gas’s share of about 40 percent. 

However, one important attribute of gas not directly related to cost is starting to be put under the microscope: gas plants’ reliability. More specifically, the purported ability of gas plants to be available at all times to generate electricity, particularly when the grid needs it most, is increasingly under scrutiny

The grid needs electricity most during extreme temperatures so that we can all stay reasonably cool during sweltering heat waves and reasonably warm during frigid cold winter storms. However, gas-fired power plants–again, the country’s largest source of electricity–are proving to be vulnerable to these extreme temperatures, which is forcing grid planners and watchdogs to re-examine these plants’ contributions to grid reliability.  

But what does extreme weather actually do to cause gas plants to run into trouble? What are the physical limitations of this fossil fuel resource that make it vulnerable during, say, summer droughts, or the types of extreme heat waves the world has been experiencing for months now? 

Those are the questions I’ll be getting into in this blog, using some illustrative examples. And with that preamble, let’s finally dig in. 

Summer brings heat- and water-related problems

Heat waves and droughts are two Danger Season phenomena that can cause gas power plants to underperform or even shut down. This is a particularly dangerous risk given the coincident grid strain and high electricity demand that is typically triggered by these weather conditions. Electricity can be a life necessity during these extremely high-temperature conditions, particularly for more vulnerable populations like those living with disabilities and older people, making it even more pertinent that power shortages don’t ensue. 

(There are other summer weather phenomena related to precipitation and flooding that can cause gas plants to shut down, but for now I’m only looking at heat waves and droughts.) 

Heat waves 

High ambient temperatures make thermal power plants like gas plants less efficient for multiple reasons. The first has to do with the relatively lower mass density of hot air compared to cooler air. Since gas combustion and combined cycle plants rely on burning a mixture of gas and air to spin their turbines, a lower air density leaves less room for gas–the actual fuel–to be burned per unit of air intake volume, thereby reducing the power output. In other words, when the air is hot, gas power plants cannot generate as much electricity. 

Further, gas plants that require cooling, like gas steam and combined cycle plants, are cooled less efficiently in high temperatures. This is true for both “wet” cooling systems that are trying to cool the plants with increasingly warm water, as well as “dry” cooling systems–which are much less common–that are trying to cool the plants using increasingly warm air. 

While these are some of the inherent weaknesses of thermal power, in a world that is currently warming due to climate change, efficiency losses in and of themselves generally don’t become issues for maintaining grid reliability because grid operators expect this to happen. They account for this inefficiency in their planning processes by downwardly adjusting a given power plant’s expected capacity. 

But when heat waves are extreme and persistent enough, gas plants can be “derated”–or reduced in capacity–beyond what’s expected in the grid planning process. Or worse yet, they even can be forced completely offline unexpectedly. Some U.S. grid operators account for these so-called “forced outages” at power plants, but not in a very sophisticated way that reflects the possibility of many plants losing capacity or going offline at the same time.  

In August 2020, California experienced an extreme heat wave that prompted electricity demand to soar to a level that exceeded available supply, leading to the first rolling blackouts of this kind since the state’s energy crisis in the early 2000s. As is generally the case with these large power outage events, many factors were at play, but one of them was the large amount of resources that had committed to being available and were not actually online due to derates and forced outages. The grid operator CAISO, along with state regulators, found that the derates and forced outages of this event mostly hit gas plants and were driven largely by high ambient temperatures.

The rolling blackouts took place two nights in a row on August 14 and August 15, 2020. When CAISO declared “Stage 3” emergencies and started these blackout processes on the 14th, gas plants made up about 79% of the offline capacity. The next night they made up about 73% of the offline capacity. 

Gas plants were the most affected by outages and derates during the 2020 California heat wave. Graph shows net qualifying capacity (NQC) expressed in megawatts (MW) that went offline at generators qualified to ensure resource adequacy (RA). Source: California Independent System Operator 

In the aftermath of those rolling blackouts three years ago, CAISO’s market monitoring unit recommended a series of changes to prevent this from happening again, including a better accounting of temperature-triggered gas plant derates in the planning process aimed at ensuring adequate resources.  

However, the misconception of gas plant reliability unfortunately persisted. Less than a month after the blackouts, California extended the lives of aging gas-fired units at four different power plants, and just recently approved further extensions at three of the plants to 2026. This was done against the wishes of surrounding community members decrying the public health impacts and other environmental injustices of the plants. Though these heavily polluting plants are being kept online to ensure reliability, ironically most of them have underperformed or experienced outages at various points during heat waves in 2020, 2021, and 2022

Continuing to lean heavily on gas plants is a terribly misguided solution to keeping the lights on. This will only become more apparent if we continue down this path as heat waves and other extreme weather events get worse with climate change—the very same climate change being driven in the first place by burning fossil fuels like gas. 


As mentioned above, using warmer water makes gas plants operate less efficiently, but when plants don’t have an adequate water supply due to drought, they can’t operate at all. Water is key to the power-producing mechanisms of both combined cycle and gas steam plants. (Simple-cycle combustion turbines don’t require large amounts of water, so droughts are a comparatively lesser concern for plants that are powered by those turbines.) 

From 2010 to 2015, Texas experienced its second-worst drought in the state’s history. The drought led to three gas-fired steam units totaling 403 megawatts (MW) at the R.W. Miller plant in northern Texas getting taken offline for almost a year, until enough rain replenished the supplying reservoir in 2015. The Electric Reliability Council of Texas (ERCOT) said in a report that while this was an extreme case, 11,000 MW of capacity (gas and other generating capacity) was at risk of going offline if the drought conditions persisted for another year. 

After the drought event ended in 2015, ERCOT started periodically reporting how much generating capacity was potentially at risk of drought-related outages 6 to 18 months into the future. In 7 of the last 10 reports released between January 2021 and August 2023, gas was the main resource at risk by far, in terms of generating capacity. 

Gas plants were the main resource at risk of drought-related outages in ERCOT’s last drought risk report for August 2023. Source: ERCOT 

But ERCOT doesn’t appear to have published any retrospective analyses similar to the 2010-2015 report showing whether any of these power plants actually did end up running into trouble due to drought-related issues. The grid operator does release daily data showing which resources are currently suffering unplanned outages or lost capacity due to various reasons, but those data files only stay on its website for one month and contain very vague language describing the nature of the outages. “Hot weather related” is one example, or an apparent favorite of power plant operators, simply “Other”.  

And this leads me to a more general point about grid reliability information: grid operators must do better and disclose more about when, where, and why gas plants and other resources struggle to meet expectations. 

We need more transparency from grid operators 

As extreme weather events become more common and more extreme with climate change, greater and greater threats will be posed to the U.S. energy system.  

This makes the quality and availability of grid reliability information all the more important, particularly as the country transitions to an electricity system that is largely powered by weather-dependent wind and solar energy along with storage. If we don’t know the specific causes of failure at gas plants and other electricity resources, then grid operators, utilities, and policymakers are going to be less likely to plan accordingly and prevent these failures from blacking out the grid in the future. 

While grid operators have more detailed information from power plant owners about what went wrong in recent extreme weather events, the convention and culture of this sector is to not disclose too many specifics until it’s too late and the grid is pushed to the edge (or over the edge) of a widespread outage event. It’s not unusual that even when outages happen, details about the specific power plants that failed during the outage event are kept undisclosed. This practice and lack of transparency allows power companies and utilities to evade accountability and public scrutiny for being inadequately prepared and overly dependent on a fossil fuel we can no longer count on. 

Let’s return to Texas for a second: the extremely intense heat waves in the south this summer have resulted in ERCOT setting 10 all-time records of peak electricity demand. And that’s just for this summer. There were multiple times during the summer, including a very recent grid emergency, when rolling blackouts were only narrowly avoided. A new weekend peak of 85.1 gigawatts (GW) was also set recently, and ERCOT urged customers to conserve electricity, partially citing in vague terms an “unexpected loss of thermal generation.” The grid operator’s data as of that day showed that the most common reason for unplanned outages was simply “Other,” which tells us nothing about what went wrong. 

Texas is far from alone in this practice of withholding detailed information about the cause of plant outages. During the August 2020 heat wave in California, a combined cycle gas plant called the Blythe Energy Center suddenly went offline when electricity customers needed it most. The grid operator’s report published alongside regulators five months later said this happened because of “plant trouble,” without giving further specifics. A separate state enforcement document contains a summary and a set of investigative findings of the Blythe incident, but that text is redacted, and regulators found that no further action was required. 

Scientists expect heat waves and droughts to become worse as climate change continues, making it crucial that policymakers understand the vulnerabilities of the power system and what can be done to mitigate them. The public also deserves to know the extent of these vulnerabilities, since we’re the ones largely funding, via utility bills, the operation and expansion of this energy infrastructure. We ultimately are all affected by its resilience and its ability, or inability, to properly function. 

Bottom line: gas plants can’t always beat the heat 

Gas still supplies about 40 percent of the country’s electricity, but continuing to power our grid by burning this fossil fuel woefully makes climate change even worse, exacerbating the very same extreme weather that can knock gas plants offline. 

Extreme summer weather phenomena have already triggered rolling blackouts in California and have caused some very close calls in Texas. As we look ahead and expect these phenomena to become more extreme, the country should not be relying on gas as the main energy source to meet future electricity demand, given its evident weaknesses we’ve seen in this type of summer weather. 

The United States must diversify away from gas and other fossil fuels in the power sector and transition to clean, renewable electricity. By making this transition, the country will strengthen the reliability and resiliency of the grid, and end the vicious cycle of burning fossil fuels and making weather more extreme, so that we have to burn more fossil fuels, and so on.