Note: the last paragraph in the conclusions was updated on 2/9/2022
The Climate Leadership and Community Protection Act (Climate Act) establishes a “Net Zero” target by 2050 and the Draft Scoping Plan defines how to “achieve the State’s bold clean energy and climate agenda”. However, there hasn’t been a feasibility plan that fully addresses the cost and technology necessary to provide reliable energy in the future all-electric net-zero New York energy system. This is the fourth post in a series of posts describing the problem and the Scoping Plan’s failure to provide a proposal that adequately addresses the problem. This post shows why extremely high prices are a feature and not a bug for any electric system that relies on intermittent wind and solar generation for the majority of its power.
I have written extensively on implementation of the Climate Act because I believe the ambitions for a zero-emissions economy outstrip available technology such that it will adversely affect reliability and affordability, risk safety, affect lifestyles, will have worse impacts on the environment than the purported effects of climate change in New York, and cannot measurably affect global warming when implemented. The opinions expressed in this post do not reflect the position of any of my previous employers or any other company I have been associated with, these comments are mine alone.
The Climate Action Council is responsible for preparing the Scoping Plan that will “achieve the State’s bold clean energy and climate agenda”. The Climate Act requires the Climate Action Council to “[e]valuate, using the best available economic models, emission estimation techniques and other scientific methods, the total potential costs and potential economic and non-economic benefits of the plan for reducing greenhouse gases, and make such evaluation publicly available” in the Scoping Plan. Starting in the fall of 2020 seven advisory panels developed recommended strategies to meet the targets that were presented to the Climate Action Council in the spring of 2021. Those recommendations were translated into specific policy options in an integration analysis by the New York State Energy Research and Development Authority (NYSERDA) and its consultants. The integration analysis was used to develop the Draft Scoping Plan that was released for public comment on December 30, 2021. This draft includes results from the integration analysis on the benefits and costs to achieve the Climate Act goals. The public comment period extends through at least the end of April 2022, and will also include a minimum of six public hearings. The Council will consider the feedback received as it continues to discuss and deliberate on the topics in the Draft as it works towards a final Scoping Plan for release by January 1, 2023.
The Climate Action Council claims that the integration analysis was developed to estimate the economy-wide benefits, costs, and GHG emissions reductions associated with pathways that achieve the Climate Act greenhouse gas emission limits and carbon neutrality goal. It incorporates and builds from Advisory Panel and Working Group recommendations, as well as inputs and insights from complementary analyses, to model and assess multiple mitigation scenarios. In addition, there is historical/archived information is available through the Support Studies section of the Climate Resources webpage, and can found as part of the Pathways to Deep Decarbonization in New York State – Final Report.
This is one of a series of posts describing the reliability problem and the Scoping Plan’s failure to provide a plan that adequately addresses the problem. In the first post I described how the Texas blackouts of February 2021 are the inevitable outcome if the Scoping Plan does not address renewable variability correctly. The second post explained why renewable variability requires massive amounts of over-building to replace existing firm capacity. The New York Independent System Operator’s Comprehensive Reliability Plan documents reliability issues that one of the state’s organizations responsible for reliability is worried about relative to the transition to an emissions-free generating system. This post shows why extremely high prices are an inevitable feature and not a bug for any electric system that relies on intermittent wind and solar generation for the majority of its power.
Renewable Variability and High Electricity Prices
A recent article at Timera Energy, Wind intermittency driving requirement for UK flexibility, includes the following figure. The graph plots the daily wind generation in the United Kingdom and the electricity market day ahead (DA) price. According to the article:
The UK requires significant wind capacity growth in order to reach challenging 2030 & 2035 decarbonisation targets. This renewable rollout comes with challenges however, with the grid already required to cope with periods of wind generation above 14 GW, and below 1 GW. Low wind events (which are often synchronised across NW Europe) have been driving prices upwards in an already high price environment, with only 3 GW of average wind generation coinciding with a National Grid electricity capacity market notice called on 24th Jan (the shaded grey area on the chart) and a £110/MWh day on day increase in DA prices.
The graph highlights why intermittent wind and solar will always lead to high prices. First point is that there is tremendous variation in the wind energy resource. Adding significant solar to the mix changes things a bit but the point is that there will always be times when renewable energy resources are very low. When those resources are low, the price of electricity goes up a lot because the replacement resources are so expensive. Importantly, also note that when the renewable resources are high, the price of electricity goes down a lot. The remainder of this article will consider each of these issues in more detail.
As a meteorologist I am particularly concerned about the variability of wind and solar resources. My over-riding concern is the need to develop a robust estimate of the minimum amount available. If we don’t know that, then it is impossible to develop adequate resources that can provide electricity when it is needed most. Unfortunately, it turns out that the highest loads (very cold and very hot weather) correlate very well to the lowest wind energy resources.
I plan to eventually write an article about this issue in the context of New York. For this post I will describe the issue by referring to a recent article: Storage requirements in a 100% renewable electricity system: Extreme events and inter-annual variability reference that uses a German example. The authors found that this issue is getting increased academic and political attention. Their analysis “explores how such scarcity periods relate to energy storage requirements”. The authors hypothesized a German 100% renewable electricity system and estimated how large the wind, solar, and energy storage components would have to be to provide reliable electricity at all times. Their analysis used 35 years of hourly time series data. Interestingly, “While our time series analysis supports previous findings that periods with persistently scarce supply last no longer than two weeks, we find that the maximum energy deficit occurs over a much longer period of nine weeks”. This occurred because they found that multiple scarce periods can closely follow each other. This means that the energy storage systems have problems re-charging between energy deficit periods. They conclude that focusing on short-duration extreme events or single years can lead to an underestimation of storage requirements and costs of a 100 % renewable system.
Low Renewable Resource Availability Impacts
The ultimate problem with low renewable resource periods is that backup resources are needed to maintain reliability. The German example highlights a couple of points: an extensive time period is needed to find the worst case and the worst case is rare. In my opinion, the Integration Analysis that supports the Draft Scoping Plan document did not evaluate a long enough time period to determine the worst case and I am sure that it did not address the re-charging problem identified in the German example. In the current New York State electric system there are generating units that only operate to provide power during the several times a year when the load peaks. In the future this kind of resource will be needed not only for peak load periods but also low renewable resource availability periods.
In an emissions-free electric system there are two possible solutions. Most straight-forward is to use energy storage during low wind and solar resources periods because the storage could be built in New York such that it is integrated into the existing network. Renewable advocates often argue that if there was a more extensive transmission system that power from locations where the wind and solar resources aren’t low could be used.
I believe that energy storage is the more likely solution for New York. New York has unique reliability constraints because New York City and Long Island are essentially a load pocket. Experience based on historical blackouts has led to the nation’s strictest set of reliability standards designed to promote reliability for New York consumers, including specific reliability rules for the New York City metropolitan area. I believe that those rules will ultimately mean that energy storage resources will have to be used.
Unfortunately, there are issues with energy storage. In the first place, long-duration energy storage batteries that would be a solution for wind and solar deficits are not commercially available. It is difficult to size currently available batteries properly and they are expensive. One alternative is to overbuild wind and solar resources using the rationale that if the resources only can produce half as much expected then build twice as much to avoid energy storage costs. However, as the German example shows the worst case is essentially no wind and solar so over-building is not a complete solution. This is the first inevitable reason that an electric system dependent upon renewable resources is going to be expensive.
The advocates who claim that the wind is always blowing someplace overlook the logistics of such a solution. The worst-case meteorological situation is a large high-pressure system that causes light winds over large areas. Weather patterns are governed by atmospheric longwaves (Rossby waves) that affect the jet stream and pressure systems. The point for this discussion is that “The length of longwaves vary from around 3,700 mi (6,000 km) to 5,000 mi (8,000 km) or more”. In the worst case it is possible that a high-pressure system could cover a significant fraction of the longwave. In other words, to get energy from where the wind is blowing for a wind lull on the East Coast you would have to go to the Rocky Mountains. Imagine all the wind turbines and transmission lines needed to provide all the power needed for the eastern seaboard and all points in between for the worst case. Worse, understand that these resources have to be there but won’t be used except for the, for example, one in ten-year wind lull. A US graph similar to the UK graph example above would show prices much, much higher.
High Renewable Resource Availability Impacts
There is another aspect of over-building as a solution. Current electric pricing schemes pay generators for their cost to operate. Fuel costs are a main driver of their fees and wind has no fuel cost. Consequently, wind generators can offer their power generated at little to no cost. The UK graph shows that effect when the prices are low during periods of high wind generation availability.
This affects overall costs in multiple ways. Because wind and solar displace dispatchable generation sources like nuclear and hydro, it affects their cost recovery. Those resources are needed for reliability so subsidies are needed to keep them solvent. Those subsidies increase overall costs.
Donn Dears explained that there are other hidden costs. For example, the latest natural gas combined cycle (NGCC) power plants can have efficiencies as high as 63%, but with wind and solar on the grid these plants would have to operate in following mode where they respond to wind and solar output ups and downs and adjust output to match intermittent variability. This decreases their efficiency. At some point this inefficiency will lead to higher consumer costs.
In my opinion, the biggest electricity planning problem is ensuring that resources are available for the worst case. In the past this planning has focused on the annual peak load. Decades of experience has led to a resilient system based on dispatchable resources that can meet load during these periods in New York but the experience in Texas in February 2021 indicates that problems can still arise.
In the future, this planning will become more difficult because many of the generating resources are not dispatchable. The Draft Scoping Plan does not adequately address the feasibility of meeting this challenge. The New York Independent System Operator 2021-2030 Comprehensive Reliability Plan notes: “While there are hundreds of projects in the NYISO interconnection queue, there are none that would be capable of providing dispatchable emission-free resources that could perform on a multi-day period to maintain bulk power system reliability. Such resources are not yet widely commercially available.” This concern is basically ignored.
The United Kingdom graph of electricity prices and wind generation output illustrates the inevitable problem with renewable resource prices. There are further examples that costs are increasing markedly there (here and here) and no sign that the Climate Action Council is aware of the problem.
I conclude that maintaining current reliability standards with an electric system that relies heavily on renewable energy sources must increase prices significantly. This is because providing power during peak loads relies on a small set of resources that are rarely used. In the current system when new resources are needed for this application combustion turbines were used because they are the cheapest dispatchable generating technology. In any future zero-emissions system, the required resources will probably be energy storage which is much more expensive and has not been deployed for this application at scale. Proponents that claim that because the wind is always blowing somewhere that all we need to do is develop more transmission underestimate the scale of the resources needed or the distances to locations where the wind is “always” blowing. As a result, this “solution” would be even more expensive..