Additional Perspective About The Texas Power Blackout
“With the early release of generating plant outage data, BTU Analytics was able to study the performance of individual plants by fuel source. Their analysis may suggest that the loss of wind power kicked off the blackout. ”
A recent analysis by BTU Analytics (BTU) of the data for individual power plant performance during the Texas Blackout offers some interesting perspective on how that crisis unfolded, its magnitude and geographic scope. We are utilizing charts from a BTU presentation to help illustrate these points. BTU’s analysis was made possible after the Texas Public Utility Commission (PUC) waived the 60-day waiting period to disclose generator outage data, along with unmasking additional plant data from an earlier voluntary filing the plants made with the Electric Reliability Council of Texas (ERCOT). Despite this additional data, there remains more information necessary to fully understand what happened to the suppliers of power to the grid that resulted in the massive blackouts. That additional data will be necessary before we can begin to assign blame for the outages, but more importantly, to be positioned to prescribe possible remedies to prevent such a crisis from occurring in the future. The humanitarian suffering experienced during the crisis demands that remedies be found to prevent it from ever happening again, but the possible changes needed must undergo scrutiny to ensure that we are not merely creating new risks as we try to address prior system shortcomings.
The PUC data enabled BTU to assemble a chart showing each thermal power plant, organized by fuel, that experienced an outage and/or a capacity derating during the weather event of mid-February. Total thermal capacity of 25.5 gigawatts (GW) was impacted, as shown in the chart. Six coal plants accounted for 5.1 GW of output lost, while the lone nuclear plant outage accounted for 1.4 GW of lost generating capacity. Forty natural gas plants, representing 19.0 GW of capacity, were also impacted, accounting for the bulk of thermal generating capacity that experienced an outage or derating. The chart does not include any impacted thermal plants with under 100 megawatts (MW) of capacity. Interestingly, three major power plant outages – the South Texas Nuclear plant and two coal generators - collectively, represented 21% of the total.
Exhibit 12. Details On The Thermal Power Plants With Outages SOURCE: BTU Analytics
Possibly the more interesting chart is the one showing the progression of capacity outages by fuel type from the time the cold front first arrived until early in March. While the chart begins with February 10th, which was shortly after the polar vortex arrived, we are mostly interested in the changes in power outages by fuel type from then until February 22nd.
Exhibit 13. Tracking Timing Of Power Outages By Fuel Type SOURCE: BTU Analytics
As the chart shows, wind capacity began to fail massively during February 10th, losing nearly half its installed capacity of 28 GW. After reaching that point on February 11th, wind capacity recovered slightly, and then essentially remained stable until the end of the night of February 14th, when wind quickly lost an additional roughly 4.5 GW of capacity. At the same time wind capacity was falling, both coal and natural gas were beginning to experience capacity outages. From February 10th to February 15th, gas lost about 5 GW of capacity. Coal, after experiencing nearly 3 GW of outage on February 12th, recovered nearly half of that loss to only about 1-2 GW of outage.
When February 15th arrived, we see that wind capacity fell slightly, just ahead of declines in capacity for both coal and natural gas, which lost output simultaneously. Shortly after the spike in coal and natural gas outages, the South Texas nuclear plant went down. At this point, every source of power supply had been impacted by the bitter cold and the ice and snow. A key unknown from this analysis is how much of these outages were due to fuel supply problems versus electricity failures.
While we await clarity on the fuel versus electricity question, there was a hearing before the House Science, Space and Technology Committee last Thursday that raised some concerns. Beth Garza, who was director of ERCTO’s Independent Market Monitor from 2014-2019, testified that she had heard that “maybe half the outages at gas plants were due to the lack of fuel.” This is a hearsay observation. A spokesperson for ERCOT stated that their data has not changed, and she could not comment on Ms. Garza’s characterization of the cause of the outages. Based on the data, ERCOT has blamed frozen wellheads and pipeline and power outages at compressor stations for 37% of lost natural gas capacity, or approximately 9,300 of the 25,000 MW of lost generation. The chart on generation outages by fuel source presented by ERCOT at its urgent board meeting on February 24th, also shows the rise in wind power outages that matches with the start of the rise in gas generation capacity outages rising. As the wind outage rose more sharply, so, too, did gas generation capacity outages. That chart is consistent with the BTU chart.
If we look at coal in the BTU chart, it appears its capacity was coming back into operation, while wind, natural gas and nuclear capacity remained sidelined. Suddenly, the coal outage spiked, suggesting that its plants may have been the victim of electricity blackouts, rather than issues with frozen fuel stockpiles.
Exhibit 14. Where Texas Power Plants Suffered Outages SOURCE: BTU Analytics
A map of the plant locations that experienced outages or deratings, divided by whether they were thermal or renewable, shows interesting geographic patterns. All the plant outages in the Panhandle region and extending across the northern edge of Texas were renewables. We also observed that along a rough line extending south from the Texas/Oklahoma border through the Dallas/Ft. Worth metroplex and on south to the Austin and San Antonio regions, the power plants impacted were primarily renewables.
The largest concentration of renewable and thermal plants experiencing severe outages was found in the Permian Basin region of West Texas. A second concentration of severe renewable and thermal plant outages can be seen in the Lower Rio Grande Valley. Given the geographic spread of these plant outages, we suspect that electricity outages, either through rolling blackouts or electric power plant failures, probably had a lot to do with that outcome. It would be interesting to see the geographical location of those power plants that did not suffer outages or significant deratings.
We found these three charts providing additional perspective on the nature of the blackout enlightening. It is still too early to assign blame. We need more detailed data on the reasons for each plant’s performance before knowing how the grid’s operations might be altered to avoid another major weather-related blackout.