Using New Marginal Emissions Data to Improve State Renewable Portfolio Standards – Columbia University SIPA Center for Global Energy Policy

This report presents the author’s research and views. This does not necessarily reflect the views of the Center for Global Energy Policy. The work may be subject to further revisions. SIPA contributions to CGEP are general-purpose gifts, giving the Center discretion in how to allocate these funds. Rare cases of sponsored projects are clearly indicated.

A complete list of financial sponsors of the Columbia University SIPA Center for Global Energy Policy can be found on our website at Our partners. Below is a list of members who currently belong to the CGEP Annual Visionary Circle. This list is updated periodically.

Executive Summary

Most U.S. states use a Renewable Portfolio Standard (RPS) to achieve clean energy goals. RPS programs typically set annual levels of clean energy production, but ignore significant differences in the intensity of grid greenhouse gas (GHG) emissions at different times of day and in different locations. Newly available location-based marginal emissions (LME) data, which is collected at thousands of physical locations and updated every five minutes, provides insight into where and when the electricity sector produces the most and least greenhouse gas emissions. Including LME in the RPS would enable countries to identify and reward “high impact” clean energy production: that which replaces the dirtiest generation.

This report examines the impact that incentives for clean energy production during periods and locations where LME occurs may have on emission reductions in RPS programs. In five scenarios based on data from four states on the PJM grid (Illinois, New Jersey, Pennsylvania, and Virginia), the authors examine hypothetical changes in energy production across times and geographic areas where different clean and dirty generation mixes occur. A proof-of-concept conducted by the authors showed that shifting clean energy production to the three dirtiest hours of the day resulted in emissions reductions of approximately 10% compared to the base case. Geographically relocating production to move energy to a dirtier location resulted in emissions reductions of 9-20%, depending on the LME composition in a given location compared to the baseline.

States can leverage the following LME trends to improve the effectiveness of their compliance programs:

• LME grids change significantly throughout the day. Variations in how generating resources constitute a marginal unit (i.e., the last unit needed to support load in a given area) as well as congestion cause the average LME to vary by 200 pounds/megawatt-hour throughout the day.
• LMEs also vary depending on the season. In summer, peak loads typically occur in the afternoon because the cooling load drives consumption during the hottest part of the day. In winter, peak loads typically shift to the evening when consumers return home and start using more electricity.
• LME averages also vary among states, especially in winter. Illinois had the lowest average LME of the four states studied, but also had the largest spread of LME between summer and winter. This can be attributed, at least in part, to the abundance of wind power in Illinois, which typically produces more electricity during the winter months.