Western Electric Grid Emissions

August 15, 2024

“59% of California’s electricity is fossil-free”

On February 22, 2022, the California Energy Commission (CEC) issued a press release New Data Indicates California Remains Ahead of Clean Electricity Goals, claiming that in 2020, 59% of California’s electricity is fossil-free.

Using the CEC Total System Electric Generation data, we indeed find this result, if we take Coal, Natural Gas, Oil, and Other (Waste Heat / Petroleum Coke) as the only fossil sources:

However, the CEC data labels some imported energy as unspecified and that needs to be treated as fossil-free to reach 60%. As there is a strong incentive to classify energy as fossil-free, one would expect that truly fossil-free energy is well accounted for, and therefore one would expect that energy from unspecified sources is actually of fossil origin. Indeed the 2022 data lumps thermal + unspecified vs. non-GHG + renewables. For 2020, treating unspecified as fossil brings the percentage down to 54.6%.

The August 18, 2023 CEC press release Data Show Clean Power Increasing, Fossil Fuel Decreasing in California emphasizes the increase in clean power over the period 2012-2022.

What matters for the climate is GHG emissions, i.e, we want to see a decrease in fossil energy. The increase (or decrease) of clean power, in and of itself, has no impact on the climate. Here is the evolution of fossil generation, still from the same CEC data:

The change year-over-year is very variable, and is even sometime an increase (e.g. 2014).

It is also apparent that while there has been progress since 2012, there has been much less progress since 2017, which warrants a closer look. However, because unspecified lumps together coal and natural gas, the CEC data is not optimal.

The West

The data from the US Energy Information Administration (EIA) allows us to get a more complete picture.

There is no electricity of unspecified origin. Even if that is achieved by estimation, at least it is somehow informed. The small scale production is estimated. We can also look at the 11 Western states individually and as a whole; this is a good approximation of “the” grid.

West AZ CA CO ID MT NM NV UT OR WA
This setting affects both the chart above and the values below.

Because we no longer have unspecified fossil energy, we can better observe the evolution in the more recent years, say since , and quantify the changes by linear regression over those years:

• coal is decreasing at the rate of GWh/year
• natural gas is increasing, although not quite as fast as coal is decreasing, at the rate of GWh/year
• we can ignore petroleum and other fossil
• non-fossil, including hydro, is increasing at the rate of GWh/year; excluding hydro, GWh/year
• overall consumption is increasing at the rate of GWh/year

Assuming the pattern holds, and we indeed want to get rid of coal and natural gas, then:

• in a first phase, coal is going to continue to decrease and reach 0 in
• at that point, the natural gas generation will have reached GWh
• from then on, the excess of clean growth over consumption growth will reduce the natural gas generation, which will reach 0 in (including the progression of hydro) or (excluding the progression of hydro)

Of course, the rates and dates are only predictions, based on the period 2014-2023 being representative of the future. We already know of likely headwinds: the decomissioning of Diablo Canyon will absorb at least two years of non-fossil growth; California’s NEM is likely to slow the growth of small scale solar. The complete decarbonation society requires a much faster growth of consumption than we have experience. Then we will reach the point where original PVs will reach their end of life, and so production will no longer fully contribute to growth of non-fossil production.

Also, the rates and dates vary if we change the period for the prediction, but they do not change qualitatively: 0 coal is pretty much stable at 2032, with 0 natural gas is at least 20 years past the California target.

What about Green House Gas emissions?

We can also look at the GHG emissions directly. For electricity generation, the EIA provides only the CO₂ emissions, and not the other GHG. Values are in t: metric ton or kt: thousand metric tons, etc.

West AZ CA CO ID MT NM NV UT OR WA
This setting affects both the chart above and the values below.

Regression on the emissions from coal gives us zero coal in , consistent with the prediction based on generation. At that point, we will emit kt of CO₂ per year.

Let's assume that we continue on the current track until 0 coal in 2032, emit 110 Mt CO₂/year at that point, and somehow manage to reach 0 fossil in 2045 (the goal for California). Past 2032, we would emit on the order of 110 * (2045-2032) / 2 = 715 Mt CO₂ (of course, that assumes a linear decline).

If, on the other hand, we reach 0 fossil in 2063 (as the data suggests), past 2032 we would emit 110 * (2063-2032) / 2 = 1,705 Mt CO₂, so an excess of about 1 Gt.

In addition to CO₂, a natural gas plant produces about 1 t of NOx for 1000 t of CO₂. The 100-year global warming potential of NOx is 298. So our plant also emits about 300 t of CO₂e in the form of NOx, for a total of 1.3 Gt of CO₂e.

To put this 1.3 Gt in perspective, the IGCC estimates that the worlwide, all sources (not just electricity generation) CO₂e budget for 50/50 chance of a 1.5C=2.7F world is 200 Gt.