Natural Gas Initiative is a cross-campus effort of the Precourt Institute for Energy.

Natural Gas Initiative

Integration of Gas and Renewables

Project Description: 
This project consists of three lines of research that explore the role of natural gas as an input fuel in the transition to a large share of intermittent renewable generation in the electricity supply industry. The first is an empirical analysis of the impact of a growing share of wind and solar energy on the hourly operation of natural gas-fired power plants in California and natural gas and coal- fired plants in the rest of the Western Electricity Coordinating Council (WECC). The United States Environmental Protection Agency (EPA) Continuous Emissions Monitoring System (CEMS) data provides information on hourly output levels of all sizeable fossil-fired generation units, allowing assessment of how the within-day operation of these generation facilities is changing in response to the growth in wind and solar generation, particularly in California, where grid-scale wind and solar generation capacity currently provides roughly 25 percent of instate grid-scale generation annually.
The second line is an analysis of how natural gas markets are likely to need to evolve to accommodate large shares of intermittent wind and solar energy. The amount of wind and solar generation capacity in the WECC increases natural gas-fired generators will need to start-up and shutdown more frequently and once operating ramp up and down quickly in response to changes in the production of wind and solar resources. This means that substantial amounts natural gas will need to be moved quickly throughout the gas grid and quickly reduced when these wind and solar resources produce energy. In California, as much as 17,000 MW of natural gas generation capacity and electricity imports from neighboring states must increase their production during the three-hour period at the end of the daylight hours to replace the solar energy from California generation capacity that can no longer produce. An increasing share of this energy is supplied by natural gas-fired generation units in the WECC as more coal-fired generation capacity in the WECC is retired. Natural gas markets in the United States are not designed to deal cost-effectively with the volatility in natural gas deliveries to manage these large and uncertain daily ramps of natural gas- fired generation units. This has created a pressing need for greater spatial differentiation in pricing, and much more frequent settlements in the natural gas market. The major question studied is what economic and reliability benefits from introduction of more granularity (both temporal and spatial) into the Western United States natural gas market.

The third line of research will use an enhanced version of PESD’s Energy Market Game (www.energymarketgame.org) to prototype the impact policies for greater natural gas and intermittent renewables integration developed in the second line of research described above. Different market mechanisms that compensate gas-fired generators for the reliability services they provide will be prototyped. One such mechanism we are developing and testing is Standardized Fixed-Price Forward
Contracts (SFPFCs), which require retailers to purchase their expected energy demand several years ahead. This mechanism provides a revenue stream for gas-fired assets (and other dispatchable units) that is not dependent on how much they run and incentivizes them to be available when they are needed to compensate for low renewable output. In addition to prototyping innovative mechanisms to facilitate gas- renewables integration, game-based simulations can be used to educate policymakers and market participants about how such mechanisms would operate, reducing barriers to implementation.

Major Findings: A major finding from our empirical research using the EPA CEMS data is that total natural gas-fired generation in California has remained roughly constant over the past 20 years despite the addition of more than 16,000 MWs of grid scale solar generation units, approximately 10,000 MW of distributed solar capacity and 6,000 MW of grid scale wind generation capacity in California. Annual greenhouse gas emissions from electricity generation in California have remained roughly constant over the past 20 years, with year-to-year increases and decreases due to hydroelectric energy availability both in and outside of California. Natural gas generation in the remainder of the WECC besides California has also increased over the past 20 years because of the retirement of coal-fired generation capacity outside of California and the construction of new combined cycle gas turbine (CCGT) generation capacity and combustion turbine capacity in the WECC region excluding California, which has led to a slight reduction in GHG emissions from fossil-fuel electricity WECC region excluding California. The Electricity Reliability Council of Texas (ERCOT) has almost 37,000 MW of grid scale wind generation capacity and almost 12,000 MW of grid scale solar capacity. However, because of the retirement of a coal-fired generation capacity and construction of CCGT capacity in the ERCOT region, the CEMS data shows that total GHG emissions in ERCOT has declined by 20 percent between 2010 and 2022.

A major finding from the second line of research using CEMS data on the hourly natural gas consumption units is a dramatic increase in the hour-to-hour volatility natural gas consumption by units in California and the rest of the WECC. There are several factors driving this outcome. First, there is a tremendous amount of contemporaneous correlation in the hourly output of grid scale and distributed solar generation units in California as shown in Wolak (2016).1 Consequently, the demand for natural gas in California can dramatically increase when a significant fraction of California’s solar generation capacity is no longer producing energy and the demand for natural gas can dramatically drop when this solar capacity produces energy. The decline in the amount of coal-fired generation capacity in the WECC over the past ten years implies that virtually all declines in wind and solar generation in California and rest of the WECC must met with an increase in natural gas generation, which further increases the volatility in the hourly demand for natural gas in many balancing areas in the WECC. This line of research also finds an increase in daily natural gas price volatility at major delivery points on the WECC in response to the increased volatility in the consumption of natural gas in the WECC.
The major finding for the third line of research with the Energy Market Game is that a standardized energy contracting based long-term resource adequacy mechanism in wholesale electricity market with a significant share of intermittent renewables is superior to a capacity-based long-term resource adequacy mechanism. Different from fossil fuel-fired generation units, where the assumption of independence of availability to produce energy from individual generation units is a reasonable assumption, the non- dispatchability and high degree of contemporaneous correlation in the hourly output of wind and solar generation units in California documented in Wolak (2016) makes it difficult, if not impossible, to define the firm capacity of these intermittent generation resources. Consequently, the effective load carrying capacity (ELCC) of an incremental one MW solar generation capacity in California given the significant amount of existing in-state solar generation capacity is extremely small.

Students involved with this project (undergraduate/graduate/postdocs)

  • Trevor Davis--Post-Doctoral Scholar at Program on Energy Sustainable Development--Worked on cleaning and preparing the EPA CEMS data for empirical analysis. Primarily worked on software development for the Energy Market Game.
  • Ryan Triolo—Graduate Student in Civil and Environmental Engineering--Worked on empirical analysis of CEMS hourly generation unit-level output data for California and ERCOT.
  • Kurt Sweat—Graduate Student in Economics--Worked on empirical analysis of CEMS data for natural gas consumption in by generation units in California and ERCOT
  • Brennan Bower—Undergraduate in Economics--Worked on analysis of the CEMS data for generation natural gas and coal-fired units located in WECC.
  • Helen Hill—Undergraduate Student in Economics-- Worked on analysis of the CEMS data for generation natural gas and coal-fired units located in WECC.

Academic and non-academic publications (Op-Eds, news magazines, etc.) and conference presentations that came about as a result of this project. Please include publications in-progress.

  • Wolak, Frank A. “What Does of Megawatt-hour of California Solar Energy and Wind Energy Displace” work in progress.
  • Wolak, Frank A. “The Greenhouse Gas Emissions Implications of Large-Scale Wind and Solar Generation Deployment in California versus Texas,” work in progress
  • Jha, Akshaya, and Frank A. Wolak. "Can forward commodity markets improve spot market performance? Evidence from wholesale electricity." American Economic Journal: Economic Policy 15, no. 2 (2023): 292-330.
  • Thurber, Mark C., Fletcher H. Passow, Trevor L. Davis, and Frank A. Wolak. "Game-based investigation of standardized forward contracting for long-term resource adequacy." The Electricity Journal 35, no. 7 (2022): 107156.
  • Wolak, Frank A. "Long-term resource adequacy in wholesale electricity markets with significant intermittent renewables." Environmental and Energy Policy and the Economy 3, no. 1 (2022): 155-220.
  • Wolak, Frank A. "Market design in an intermittent renewable future: Cost recovery with zero-marginal cost resources." IEEE Power and Energy Magazine 19, no. 1 (2021): 29-40.
  • Graf, Christoph, Federico Quaglia, and Frank A. Wolak. "(Machine) learning from the COVID-19 lockdown about electricity market performance with a large share of renewables." Journal of Environmental Economics and Management 105 (2021): 102398.

Related papers are available at: https://web.stanford.edu/group/fwolak/cgi-bin/ 

Presentations:

  • “Long-Term Resource Adequacy with Significant Intermittent Renewables,” Presentation to CREG (Colombia Electricity Market Regulator) March 23, 2020
  • “Long-Term Resource Adequacy with Significant Intermittent Renewables,” Presentation to New England Power Pool Stakeholders on November 5, 2020.
  • “Why federal regulation is not the answer” (March 9, 2021) Opinion piece in Austin American-Statesman in response to February 2021 electricity shortfall in Texas.
  • “Long-Term Resource Adequacy with Significant Intermittent Renewables: The California and Texas Experience,” Presentation to OSINERMIN, Energy Market Regulator in Peru, October 21, 2021.
  • “Australia and Texas: Analysis and Solution to Energy Crises,” Public Lecture Sponsored by Monash University, Melbourne, Australia, September 14, 2022.
  • “Long-Term Resource Adequacy with Significant Intermittent Renewables,” Presentation to Australian Energy Market Commission, September 16, 2022.