The following is a contributed article by Robert Borlick, senior energy advisor at Borlick Energy Consultancy.
This is a response to the Utility Dive opinion editorial authored by Harshit Chatur.
Although it is unclear whether achieving a zero-emissions electric grid by 2050 is physically feasible by 2050, the arguments advanced by Chatur are unpersuasive. Let's examine each of those arguments:
1. Solar and wind renewable energy resources are not controllable sources of power.
This is not quite true. By overbuilding a renewable resource it is possible to get reasonably close to convert it into a dispatchable resource by simply curtailing its output when the energy is not needed. Marc Perez developed this concept at Columbia University as a cheaper alternative to using storage resources to make a renewables dispatchable.
2. Solar and wind installed capacity factors are low relative to fossil fuels
Of course, this is true — so what? Renewable resources can be scaled up to produce equal amounts of energy. The more relevant issue, which Chatur did not address, is the cost of scaling up.
While renewables are more expensive to build than fossil fueled generators, they offer the overwhelming advantages of having no fuel costs and emitting no greenhouse gases. Furthermore, their costs will continue to rapidly decline. When comparing two different options one needs to include all associated costs, which Chatur failed to do.
3. Storage technology must continue to evolve
I agree — and they will. Chatur points out that batteries are expensive and have limited storage capacity. While that's true, there are many services that batteries can provide today at lower costs than fossil fuel plants, e.g., peaking and frequency regulation.
4. Initial capital cost ($/kWh) of renewables is high
Indeed they are, but a more relevant metric is their levelized costs of energy (LCOEs) compared to fossil fueled generation. Today LCOEs for large scale solar PV and onshore wind are already competitive with gas-fired generation. And this is before crediting the renewables for their added value of being emissions-free.
5. Grid transformation will be expensive
Now this is unequivocally true. But the cost of transforming the grid does not equate to the inability to achieve a zero-emission electric grid by 2050.
Chatur could have questioned whether the U.S. can afford this cost but he didn't. One estimate developed by Wood Mackenzie puts the cost at around $15 trillion spread over 30 years.
6. Illustrative costs
The numerical example Chatur presented contains errors, mostly small, but some that are large.
First, he compares the initial cost of a gas-fired plant (presumably a combined cycle plant) with a utility-scale solar plant, but ignores the lifetime fuel and O&M costs of the gas plant and the lifetime societal cost of its emissions. He uses an aggressive estimate of $900/kW for the gas plant and a pessimistic estimate of $1550/kW for the solar plant (utility scale solar PV is currently being installed for less than 1000/kW).
Lastly, he assumes $400/kWh for battery storage (lithium-ion batteries currently cost less than $200/kWh). He also erroneously assumes that the nighttime demand is exactly the same as the daytime demand. A cursory examination of the typical utility's hourly demand over the diurnal cycle reveals that nighttime demand is far less than that for daytime hours. So he has overstated the size of the solar plant and the associated storage.
Granted these numerical errors, though sloppy, are trivial compared to Chatur's unconditional assertion that we cannot achieve zero-emission by 2050.
He then states it would cost trillions of dollars. Indeed it would — and will. But even if it costs twice what Wood Mackenzie recently estimated, the alternative of continuing to rely on fossil fuels will be even more expensive.
And spending $1 trillion to $2 trillion per year will not impose an undue hardship on the US economy. The US already spends about $650 billion per year on fossil fuel subsidies. Eliminating those subsidies would pay for much of the transition to zero emissions and the balance could be paid for through higher taxes on the top 1%.
A more serious concern is whether we can develop the supply chains needed to construct the vast fleet of renewables, nuclear plants and energy storage resources that will be needed.
7. The right mix
Chatur partially redeems himself in this last section by acknowledging the need for a mix of resources that includes more than just renewables and storage. There is a need for constructing a fleet of nuclear plants based on the latest technology to provide emissions-free base load energy, while relying on renewables and storage to supply the rest of our energy needs.
Also, I am not convinced that zero-emissions is the optimal solution. If we satisfy 90%-95% of our energy needs with emissions-free generation and 5%-10% with natural gas, the total cost will be substantially lower. Wind generation typically falls to near zero for as long as several days each year; consequently, the full nameplate capacity of the wind fleet needs to be backed up with other generation and demand response.
Maintaining a fleet of gas-fire plants that only run during the hours of low wind output would be far cheaper than relying on energy storage while producing relatively little greenhouse gases.
Achieving the right resource mix is critical and not enough effort is being dedicated to examining this issue.