In our work on modeling and projections of electric sector greenhouse gas emissions, GPI often sees analysts using federally published electricity generation and emissions data defined by large geographic electricity regions rather than by an individual state or a specific electric utility. While using these large regions is convenient in terms of data availability, it can miss key differences in electric sector emissions that are available at more granular geographic regions.
This post explores electric sector emissions analysis and modeling, using the Midwest as an example, and the implications for how data choices inform decision-making on the future of our electric system.
Key takeaways:
- Detailed data is available at the electric grid or utility level that can help improve the accuracy of modeling electric sector emissions, which can better inform decision making.
- It is important to be intentional about how emissions data is used, especially when it may be used to inform decision makers who are shaping the long-term trajectory of the energy system.
- Future work on emissions assessments and electricity emissions can be made more accurate and useful by learning about what options exist for calculating electricity emissions and when each is the best fit.
Illustrating fuel mix differences across geographies, using the Midwest as an example
Many Midwestern states, for instance, are most often aggregated into a region known as MRO—the Midwest Reliability Organization territory—as defined by the US Environmental Protection Agency (US EPA) and the North American Reliability Corporation (NERC).
Figure 1. Map of various geographies used to calculate electricity sector emissions
Source: Jessi Wyatt, Great Plains Institute, 2019.
In the day-to-day operation of the electric grid, however, it’s the electric utilities that are responsible for delivering power from their power plants to customers, and regional balancing authorities like MISO (discussed below)—the Midwest Independent System Operator—that are responsible for the operation of the grid. It might also be important to consider individual states’ renewable portfolio standards, climate goals, or other policies that set targets for the use of specific energy sources in electric generation.
Each geographic region has a unique electric generation fuel mix
When calculating electricity generation emissions or analytical projections, each geographic definition of the modeled region has a unique generation fuel mix, whether it’s a federally defined region like MRO, a regional transmission organization’s territory like MISO, a specific utility, or a state with a defined electric portfolio (we show this using Minnesota in figure 2).
As electrification and grid decarbonization scenarios are increasingly explored by local governments, states, and regions, GPI has explored the consequences of assuming one electric generation mix over another.
Figure 2. Electric generation fuel mix for MRO, MISO, and the state of Minnesota in 2016, in megawatt-hours (MWh)
Source: Adapted from EPA eGRID Power Generation Data, 2016.
Impact of disparate electricity generation geographies in the Midwest
In practice, electric sector emissions projections for Minnesota and the Midwest are often based on data from the Midwest Reliability Organization (MRO) footprint. This is convenient because most data published by the federal government are already categorized into a North American Electric Reliability Corporation (NERC) region. This includes information such as total power generation, breakdown of production by fuel type, and estimated annual carbon-dioxide equivalent (CO2e) emissions. Using NERC regions as the common geography across datasets can mean that there is consistency across studies. However, the convenience of using NERC regions like MRO to determine electric carbon intensity miss crucial granularity that other electricity generation geographies provide.
Additional options for defining electricity generation geographies include state-specific, regional, or utility-specific data. Each will make more sense in some instances than others, but as a rule of thumb, the more granular you can define your carbon intensity, the more accurate the resulting research will be. This fluctuating degree of accuracy has implications for many applications, such as electricity emission reductions or carbon emission estimations. Conversely, the implications of defaulting to a carbon intensity metric at too broad a scale for the application (e.g., the tradition of defaulting to MRO) can have unintended consequences given the varying assumptions for these electricity regions—an estimation may be too high or too low, depending on what geography is chosen, which has real implications for informing policy decisions and defining thresholds for emissions reductions or achieving reduction compliance.
Looking at the distribution of carbon intensity values shown in Figure 1, MRO is much lower than MISO and significantly higher than the state-specific example given for Minnesota. All the utility-specific examples provided (here, an investor-owned utility, a coop, and a municipal utility) are much lower than non-utility carbon intensities.
Figure 3. GHG Emissions Intensity across Electricity Regions for 2016 (in metric tons CO2e per megawatt-hour (MWh))
Source: Adapted from EPA eGRID Power Generation Data, 2016
In the Midwest, defaulting to MRO to determine carbon intensity or emissions factor risks inaccurate information for state or local levels of application. For example, as states adopt renewable portfolio standards, or communities adopt renewable energy generation goals, the nuance of both their starting point and outcomes risk being lost to inaccurate, overarching electricity market information. To ensure more accuracy a state could use state-specific electricity generation data or a community could use utility-specific data.
Looking out to projections made for each electricity region in 2030, the need for a situation-relevant electricity sector geography—and subsequent emissions factor—remains. While some utilities remain similar between 2016 and 2030, the change for each is not constant across the board. This translates to inaccurate projections when considering a geography unfit to the application.
Figure 4. GHG emissions intensity across electricity regions for 2016 and 2030 (in metric tons CO2e per megawatt-hour (MWh))
Source: Great Plains Institute, January 2020, based on data from US Energy Information Administration, MISO, Xcel Energy, Great River Energy, and Southern Minnesota Municipal Power Agency.
Electric generation data should be chosen carefully given the role of analysis and modeling in decisions
Given how much weight that analysis and modeling of greenhouse gas emissions can have in decision-making, there is value is thinking intentionally about what electricity region should be used to represent an emissions ratio. The decision to employ one electricity generation mix or estimated carbon intensity over another does have consequences and should not be motivated out of default or convenience.
With increasing insight into various electricity region generation carbon intensities, there is an opportunity to ensure that decisions on electricity generation geography are made more thoughtfully. The result will be more accurate research and a better ability for government, communities, and industry sectors to benchmark and set goals along with a more precise understanding of energy generation.
With questions or inquires on the data specifics provided in this post, please contact:
Jessi Wyatt, Energy Planner and Analyst, Great Plains Institute
Email: [email protected] or Phone: 612-400-6292
Keep up-to-date with all GPI news by signing up for our monthly newsletter.
