Originally published at Climate Etc.
Russel Schussler and Roger Caiazza
Electric generation plans need to be well crafted and carefully considered. Because of concerns around climate change many politicians have become galvanized to hastily enact legislation to target net-zero anthropogenic greenhouse gas emissions by 2050. The authors argue that the more seriously you take climate change, the more important it becomes that you have a good plan for electric generation in the near and midterm planning arena. Taking foolish actions in the near to mid-range time periods will not help with CO2 reductions or climate change and may be far worse than doing nothing. Maybe we all could compromise and find a less grand something that has more likely benefits with far fewer threats to reliability, affordability, and overall environmental impacts.
The authors have both been writing about the proposed net-zero transition by 2050 for years. Schussler (aka the Planning Engineer) has been writing about the challenges of “green energy” since 2014 at the Judith Curry’s Climate Etc. blog. Caiazza has focused on New York energy and environmental issues at Pragmatic Environmentalist of New York blog since 2017. Since the original proposal for New York’s Climate Leadership and Community Protection Act (Climate Act) in 2019, he has written over 280 articles about that plan to transition to net zero by 2050.
Traditional Generation Planning
Utilities used to look at 30-year time periods in developing their generation expansion plans. This was not because they believed anyone could forecast what might happen 30 years into the future, but rather because of the recognition of the futility of such efforts. Decisions were made about the next ten years or so, but the later years tested the flexibility of the plans. Because power plants have a long life, many different scenarios were studied in the additional 20 years or so after the plant addition. Commercial technologies were supported by more dependable cost and performance estimates than what could be obtained for newer technologies, but it was recognized that all parameters could change across any technologies. Scenarios would vary fuel prices and availability, potential environmental requirements, as well as other varying system requirements. Back then, no one had the hubris to say this is what the system would, or should look like 20 or 30 years from now. Planners sought to make decisions that would be flexible enough to work well across a variety of future potential scenarios. The hope was for this plan to work with and adapt to the emerging future.
Some jurisdictions have made commitments to completely transform their electric generating systems in less than 30 years. Rather than intending to be flexible in the mid to long term, these plans are often overly prescriptive. This post addresses the potential consequences and suggests a less risky approach.
New York Climate Act
New York’s Climate Act is a good example of prescriptive net-zero legislation. Implementation to meet the following inflexible targets has begun:
- Reduce GHG emissions to 60 percent of 1990 emissions levels by 2030;
- Zero GHG emissions from electricity production by 2040; and
- Reduce GHG emissions to less than 15 percent of 1990 emissions levels by 2050, with offsets to reduce net emissions to zero.
New York passed the Climate Act in 2019 effective 1/1/2020. The legislation established a Climate Action Council to prepare the Scoping Plan that outlines how to “achieve the State’s bold clean energy and climate agenda.” In brief, that plan is to electrify everything possible and power the electric grid with zero-emissions generating resources by 2040. The Integration Analysis prepared by the New York State Energy Research and Development Authority (NYSERDA) and its consultants quantified the impact of the electrification strategies. The Final Scoping Plan was completed at the end of 2022. In 2023 the New York State Department of Environmental Conservation and the Legislature are supposed to promulgate the necessary regulations and legislation to fulfill the recommendations in the Scoping Plan.
There are deep flaws in the New York implementation process. The Scoping Plan is just an outline list of control strategies that NYSERDA claims will reduce emissions as needed and provide reliable electricity. NYSERDA, New York State Independent System Operator (NYISO), and New York State Reliability Council (NYSRC) have not done a consolidated feasibility analysis that addresses the fundamental question: will it work? There are significant differences between the Final Scoping Plan and NYISO 2021-2040 System & Resource Outlook. The following figure from the Resource Outlook summarizes the key findings that are applicable to any net-zero by 2050 initiative. Our biggest concern is that both resource projections rely on untested technology. The Resource Outlook notes:
By 2040, all existing fossil generators are assumed to be retired to achieve the Climate Act target for a zero-emission grid and are replaced by Dispatchable Emission-Free Resources (DEFRs). These resources represent a proxy technology that will meet the flexibility and emissions-free energy needs of the future system but are not yet mature technologies that are commercially available (some examples include hydrogen, renewable natural gas, and small modular nuclear reactors).
What are the characteristics of Good Plans versus Bad plans
In this section we consider the characteristic and provide commentary in italics relative to the New York Scoping Plan.
Bad plans assume that critical elements of the future are all known. Bad plans are narrowly constructed to a specified future. They risk not allowing the flexibility to adapt when things turn out differently than was planned. Good plans look at their impacts or current decisions across a wide variety of potential futures. Good plans provide flexibility and nimbleness for when future conditions change.
The NY Climate Act electrifies as much as possible to decarbonize and presumes all the elements necessary to accomplish the transition are known. The critical element of future expected load must be well known to determine generation resource requirements. Future net-zero load is a function of increased electricity for heating, cooking, water, and electric vehicles at the same time there is increased emphasis on energy efficiency and conservation. Projections in this instance are anything but well known.
Good plans understand that the power supply system and power grid are very complicated systems requiring careful design, construction, and operation. Great consideration is given to the architecture of the system and how it will work. A poor plan leaves the power system and grid as an unplanned afterthought. It specifies some goals and ingredients but ignores the greater system.
The basis of the Climate Act electric grid transition plan is the wind, water, and solar (WWS) approach championed by Stanford University Professor Mark Jacobson. The approach had outsized influence on the members of the Climate Action Council but there are issues with this work. Advocates of this particular transition approach have overstated its findings, it does not put appropriate emphasis on the high load and low renewable resource problem, and understates the challenges of a quick transition to a zero-emissions electrical grid.
Bad plans are one-size fits all. They employ a presumption of what is best and fail to take in the particular specific considerations that can vary across time and place. Good plans recognize that what works in one area, may be less appropriate in another. Good plans seek to capitalize on differing advantages wherever and whenever they may occur.
The New York electrical grid is pretty much two different grids. There is a traditional grid Upstate but there are unique problems in New York City. Experience has shown that sufficient in-city generation must be available to account for the loss of a transmission line into the New York City load pocket or blackouts can occur. The Scoping Plan does not adequately address these differences in their on-size fits all plan.
Good generation plans recognize how people prefer to use electricity. If behavior needs to be changed, they are sensitive to the capabilities and limits of incentives. Depending on the generation mix the value of electricity will likely vary considerably across hours, days, months, and seasons. Good plans will seek to provide value. Bad plans tend not to differentiate between when and how energy might be supplied. Plans crafted based on just average use and average costs will likely not have good results. Traditionally generation planning recognized baseload, intermediate and peaking needs. While many seem to forget these distinctions when comparing alternatives, their importance has not diminished.
The New York plan presumes that net-zero transition to net-zero required changes to personal energy choice preferences will be universally accepted. The behavioral changes required by the Scoping Plan are massive (e.g., type of vehicles, heating your home, and cooking your food). Furthermore, there may be limits on the timing of electric usage. Modeling assumptions on the effects of these changes to personal habits are important for planning but also very uncertain if people do not make the changes expected. It is highly unlikely that load shifting and energy conservation will prevent a markedly higher electric load peak in winter mornings. The Scoping Plan compounds these issues because it does not adequately address the baseload, intermediate and peaking requirements naively arguing that “smart” planning will mitigate issues associated with them.
Good plans look at major environmental impacts across the production and lifetime of a resource. Bad plans tend to look only at marginal impacts when the facilities are operating. Tremendous resources and costs are incurred just getting a generating resource in place. Generally, the longer that resource can operate, the better its average environmental impact might be. Good plans should consider the realistic lifetime of potential resource. Many “green” resources projected to last 30 years fall far shy of 20 years. Conventional resources typically are capable of lasting many years beyond the thirty-year study life.
The Climate Act takes this concern to a higher level. Many life-cycle environmental impacts of fossil generating resources are considered. None of the life-cycle environmental impacts of wind, solar, and energy storage are considered. The Integration Analysis assumes that all wind, solar, and energy storage resources keep operating from the present until 2050. Furthermore, the Climate Action Council has tried to appease climate justice advocates who fervently believe that the risks of fossil-fired generating resources are so great that existing resources must be shut down as soon as possible. Their concern is at odds with consideration of environmental impacts across the production and lifetime of all resources.
Good plans rely on proven technology that can fulfill the specific requirements. For example, providing power for periods of peak load is required for reliable power when it is needed most. Peak loads are typically associated with the hottest and coldest periods of the year when electricity is used for cooling and heating. Typically, those periods occur less than 5% of the time so a technology should be as low cost as possible to keep the price of electricity down during peak loads. A good plan would make the sensible decision to keep an old fossil fired plant around to help the system meet peak loads. Fossil-fired steam boiler electric generating units are a proven technology that can be used to meet this need.
For many years New York City peak load requirements were met with simple-cycle gas turbines installed in the early 1970’s. However, those units were old, inefficient, and had unacceptably high emission rates so, after a multi-year process of reliability planning the State has instituted a regulation to phase them out. After the regulation was promulgated the Environmental Justice (EJ) community glommed on to the issue of peaking power plants: “Fossil peaker plants in New York City are perhaps the most egregious energy-related example of what environmental injustice means today”. Even though the poorly controlled peaking turbines are being phased out, the issue remains a point of contention. Now the EJ organizations are demanding that all fossil-fired power plants in New York City be shut down including the remaining steam boilers even though they meet all emission limits and do not contribute to the alleged health benefits in disadvantaged communities near the facilities. The proposed solution to use renewable energy and energy storage replaces proven technology with one that has not been proven on the scale necessary to keep the lights on in New York City.
Bad plans presume that a new technology can fulfill specific needs. A necessary component of any future system is dependable emergency capacity. For example, a system might need emergency capacity once every five years due to extreme weather either causing very high loads, an unexpected long-term outage of existing resources, or because of an extended drought of wind and solar resources. A bad plan proposes a new technology for this emergency requirement. In order to provide capacity in a zero-emissions electric system a new category of generating resources called Dispatchable Emissions-Free Resources (DEFR) has been suggested to keep the lights on during periods of extended low wind and solar resource availability. In Wyoming, PacifiCorp’s 2021 integrated resource plan (IRP) includes a resource labelled as “non-emitting peaker plants” that is unexplained but appears to be the same as DEFR. The New York Independent System Operator (NYISO) 2021-2040 System Resource Outlook states:
DEFRs that provide sustained on-demand power and system stability will be essential to meeting policy objectives while maintaining a reliable electric grid. While essential to the grid of the future, such DEFR technologies are not commercially viable today. DEFRs will require committed public and private investment in research and development efforts to identify the most efficient and cost-effective technologies with a view towards the development and eventual adoption of commercially viable resources. The development and construction lead times necessary for these technologies may extend beyond policy target dates.
In both instances no specific technology has been specified. The New York Scoping Plan DEFR placeholder is producing and storing “green” hydrogen for use when needed.
This is the fatal flaw of the New York Scoping Plan. The NYISO 2021-2040 System & Resource Outlook states that “To achieve an emission-free grid, Dispatchable Emission-Free Resources (DEFRs) must be developed and deployed”. This magical resource does not exist! The Scoping Plan uses “Green” hydrogen as a placeholder for the technology and predicts that it will be used on average around 3% of the time. The fantasy of the Scoping Plan is that developing the infrastructure to produce hydrogen, store it, and then produce electricity in hydrogen fuel cells can provide affordable and reliable energy to keep the lights on. The costs will be astronomical for a resource used so little presuming that the technological issues can be overcome.
What are the ingredients of a compromise plan?
As mentioned above good plansrecognize how people prefer to use electricity. Electricity usage across a region rarely drops to zero, but at times demand peaks for limited periods of time. It may make sense to build high fixed cost, low variable cost resources (Nuclear, Coal and Combined Cycle) to meet the baseload needs of system. If the plant can run all the time with low variable cost the higher investment cost can be justified. It does not make sense to put in such facilities to serve load levels that only occur rarely. For this component of the load it makes more sense to put in low cost infrastructure that might have higher marginal costs. Between these two conditions there are loads levels that may be present for a few hours a day. To meet these loads, it is usually better to put in plants with moderate costs and moderate marginal costs. This is the thinking behind traditional utility planning which looked at peaking, intermediate and baseload needs in terms of generation fitted for those specific characteristics. There is one other type of generation, intermittent. Intermittent typically was low-cost generation that although it could not be counted on, it could be used to back off generation using higher priced fuels. In looking at the ingredients below it will helpful to consider where they may be most appropriate.
Wind, Solar and Batteries can work to displace fossil fuel generation. With backup from batteries the energy provided can be made to have more value. Unexpected and innovative changes in the capabilities of batteries could be a game changer, but it is too soon to count on timing in this arena. The narrative that these “zero-emissions” resources have zero downsides is false. The construction of wind and solar takes a lot of resources; their construction has a lot of environmental consequences; and fabrication uses a lot of energy that will be difficult to displace away from fossil fuels (making steel for example).
Nuclear power works well to meet baseload needs. It also works supports the transmission system providing needed electrical characteristics commonly called Essential Reliability Services. Nuclear plants can be planned and operated to provide some ramping and load following capabilities. Nuclear offers the best opportunity to reduce dependence upon fossil fuels for electric generation because it is the only proven technology with no emissions that can be scaled up in the immediate future.
Hydro expansion is very unlikely. Environmental considerations make it unlikely that additional locations for hydro generation could be developed. Similarly, there are limited opportunities for additional pumped storage, but there may be some areas where such might be pursued. Finally, geothermal plants when feasible are a good resource, but opportunities for exploiting this resource are limited.
Natural Gas combustion turbines and combined cycle are best suited to fill in the gaps when reliable and functional generation additions are needed. As more environmentally desirable units become capable of doing the job, eventually new construction should be halted and existing units phased out as they age. Keep in mind the US through fracking reduced CO2 more effectively than Germany did with their massive expenditures on “clean” resources.
Existing Resources such as coal- and oil-fired boilers should not be ignored for future plans. It is extremely unlikely that new plants burning those fuels will be built in the US in the foreseeable future. Clean coal was on the table a few years back, but highly visible failures coupled with environmental concerns have closed this door for a while. The cost differential between oil and natural gas as well as the efficiency relative to a combined cycle combustion turbine precludes construction of oil-fired boilers. However, the existing fleet of these plants could be kept around for limited peaking power needs, emergency power, and long-term temporary system needs.
Other potential ingredients to a future plan include technologies currently on the drawing board. Examples include tidal energy, biofuels, fusion, big HVDC ties and so on. These new technologies will have to prove themselves before they are employed as anchoring technologies in good plans. Most new technologies will not prove themselves in the next 10 to 20 years if history is a guide. But some might. While we can’t dependably plan on unproven technology, we must be ready to jump on anything valuable that works. Such technology will likely be available and workable in niche applications many years before they can be deployed more broadly in long term plans.
Smart Grids have also been touted as a component of future electric systems. This is a favorite approach of visionary academics, to concerns about observed and emerging grid problems. In the New York net-zero transition planning process many issues were dismissed with a call for “Smart Grids” as if that would magically solve everything. Modern grids are “smart” but as with any “smart” technology there are all kinds of applications that could be adopted, so of course it is not a panacea for future grid plans.
Energy Efficiency is another favorite future grid resource for the naïve. When concerns about peak loads and the necessary infrastructure are raised, the response is to double down on energy efficiency and energy conservation programs to flatten the peak loads. Of course, if the goal is to decarbonize by electrifying everything, then the load will have to increase to cover building heating, cooking, and hot water. Add in battery electric vehicles and this approach can only hope to reduce the peak but it will never eliminate the need for a peaking power generation resource.
A Good Enough Plan
Assuming the plan is a compromise between net zero and a working power system, the biggest step would be to commit to getting as much nuclear power as possible into the mix as soon as possible. This best supports the grid and reduces CO2. We need to figure out how to get plants built more efficiently and quickly. Adding nuclear must be the centerpiece and driver for meeting emerging generation needs. Under reasonable regulations, it is the only zero-emissions technology that can be scaled up and provide reliable and dispatchable power.
The continued massive ramp up of wind and solar does not make sense currently. There are major reliability concerns which would emerge with the introduction of high level of intermittent asynchronous wind and solar power. Such programs distract from the needed focus upon nuclear programs. As technology improves and better resource choice emerges, large scale existing wind and solar that requires some sort of dispatchable emissions-free resource are likely to become dinosaurs.
At this time, it appears that plans for the addition of fossil-fired plants would center around the gaps where new nuclear power can not be made available or meeting peak demand levels not met by current resource plans. Natural gas plants will be a good compromise. Lower cost combustion turbines will have long term value to aid with ramping, meeting peaking needs and providing emergency power. Higher cost more efficient combined cycle plant will make sense the longer the delay for nuclear development. They can serve variable load levels that occur regularly but vary considerably day to day.
The potential for additional hydro is low, but any ability to effectively exploit remaining opportunities should be considered. Additionally, some areas may offer the potential for the addition of pumped storage hydro or geothermal power Hopefully battery technology will improve and its ability to support energy needs and the grid can be expanded and amplified.
The authors have recognized for years that the economics, even without all the environmental and regulatory considerations, will not support building a new steam boiler plant in the US. Gas is just too cheap in the US compared to coal or oil. New coal is a non-starter given the need for elaborate and expensive pollution controls. However, this does not mean it makes sense to retire functioning coal, gas, and oil plants. In many cases they will be the best emergency back resource available across the board considering economics, environmental impact, and reliability.
There is another economics aspect of our good enough plan that needs to be stressed. The plan does not require the development and deployment of the magical dispatchable emissions-free resource that is a necessary component in a electric system that relies on wind, solar, and energy storage. Eliminating the cost of a brand-new resource to fulfill a very limited role will make this approach cheaper than any net-zero alternative.
There is a segment of society that is invested in the need to do “something” about climate change by mitigating emissions. A good enough plan would support R&D on clean technologies for future generation, energy storage, and transmission system support. Currently, these clean technologies are simply not ready to provide reliable and affordable energy. The developing world will not use zero-emission technologies until they can provide electricity cheaper than existing resources so this R&D is necessary for a global solution. In addition, if the full life-cycle impacts of those technologies are considered, then they are not nearly as “clean” as commonly portrayed.
The proposed good enough plan provides direction but is not overly constraining. It’s hard to know the future, but it’s a safe bet that any plan will not anticipate some critical twists that will emerge down the road. This plan would lay a strong foundation. A major shift to the nuclear plants that are the obvious best choice for baseload power, supplemented with natural gas units, and retention of on the ground facilities should be the framework of a good enough plan. Good enough plans are also flexible so integration of newer technologies when and as warranted is a reasonable attainable path without major downsides. This good enough plan may get you to net zero before the more ambitious ones. It is likely to have less carbon emissions than the more aggressive plans over time. It certainly will be more reliable and affordable.