Which emissions question are you asking? Inventories, consequences and procurement claims
Put one new load on the Plains Grid and get four different CO₂ answers: an inventory share, a short-run change, a long-run change and a claim
A 300 MW load on the Plains Grid, 7 November 2025: four questions, four answers
| Question | Data center (7,200 MWh) | Electrolyzer (3,600 MWh) |
|---|---|---|
| 1 · Inventory share, at average factors | 1,611 t | 553 t |
| 2 · Short-run change at the plants | +3,039 t | +1,435 t |
| 2 · The same, with a must-run rule | +2,211 t | +607 t |
| 3 · Long run, a new 300 MW gas plant | +2,372 t | +934 t |
| 3 · Long run, new wind for its yearly energy | +133 t | +24 t |
| 4 · Claim, certificates matched over the year | 0 t | 0 t |
| 4 · Claim, new wind matched hour by hour (short hours at the grid average, a proxy) | 441 t | 137 t |
Claims follow illustrative rules after the GHG Protocol's Scope 2 Guidance (2015): certificates are assumed to be the buyer's alone, retired for it and from the same market.
New wind is assumed to be built for the buyer (additional) and to reach this grid (deliverable).
The electrolyzer runs from 00:00 to 12:00, the 12 hours with the lowest average factor (0.10 to 0.17 t CO₂ per MWh). Each of its extra MWh is still served by the combined-cycle plants, at 0.40 t.
With JavaScript on, you pick the load, switch the model's rule, choose what gets built and compare five purchases.
Illustrative grid. The Plains Grid is fictional: a 4,000 MW peak; 3,000 MW of wind, 600 MW of solar, 2,000 MW of gas combined-cycle plants and 2,300 MW of gas peakers.
Real plant data. Gas plants' full-load heat rates are 7,535 and 10,975 Btu per kWh (EIA, 2025), held at every output; gas emits 52.91 kg CO₂ per MMBtu (EIA).
From heat to CO₂. 7,535 Btu per kWh = 7.535 MMBtu per MWh; × 52.91 kg = 399 kg. So 0.40 and 0.58 t per MWh, counted at the stack; t is a metric tonne, 1,000 kg.
Real hours. Demand, wind and sun follow the Southwest Power Pool's hours of 7 November 2025 (EIA-930), wind and sun scaled to EIA's 2025 capacity factors (Table 6.07.B). Each hour runs cheapest first, with no ramp limits.
About 30 guided minutes in seven steps.
What you just did, as diagrams
- Inventory
- Average factors share a total that already happened: the data center's 1,611 t and everyone else's 12,055 t add up to the grid's 13,666 t.
- Short run
- Marginal factors ask what changes: the data center adds 3,039 t, mostly at the combined-cycle plants.
- The model matters
- With a rule that keeps 800 MW of combined-cycle plants running, night wind would be spilled, and the electrolyzer's change falls from 1,435 t to 607 t.
- Long run
- What gets built decides the answer: +2,372 t with a new gas plant, +133 t with new wind, for the data center that day.
- Claim
- A purchase sets what may be reported under a stated rule. In these scenarios the plants change only when we assume it adds clean power that reaches this grid.
Go deeper · AnalystTwo ways to report electricity's CO₂
- Scope 2. A company’s reported emissions from the electricity, heat or steam it buys. Where buyers can choose contracts, the GHG Protocol’s Scope 2 Guidance (2015) asks for two counts.
- Location-based. Each MWh is counted at the grid’s average factor where it is used: question 1.
- Market-based. Each MWh is counted at the factor of the electricity the buyer purchased, through contracts or certificates: question 4.
- Conditions. Among other criteria, a certificate must be the buyer’s alone, retired for it and from the same market (Table 7.1). Unclaimed use takes a residual mix; this lesson uses the grid average instead, as a proxy.
- Rules under review. The GHG Protocol consulted on a revision from 20 October 2025 to 31 January 2026. Hourly matching and deliverability are proposals, not adopted rules; its summary of 29 July 2026 reports low support for them as proposed.
- Neither method measures what the plants do differently: that is questions 2 and 3.
Go deeper · ModelerWhere marginal factors come from
- From a dispatch model. Solve the hour with and without one more MWh; the change in CO₂ is the hour’s marginal factor under that model’s rules (fuel prices, minimum outputs, lines).
- From statistics. Regress the hourly change in a region’s CO₂ on the change in its generation over years of data (Siler-Evans et al., 2012).
- Long run. A planning model adds the load and compares the build-out with and without it (Gagnon and Cole, 2022).
- Next. M6 puts carbon and a matching rule into such a planning model.
Go deeperWhat this lesson leaves out
- A year: one windy November day stands for no other day.
- Life-cycle emissions: building plants, and methane leaks before the gas reaches the plant.
- Imports, exports and the lines inside the grid: its plants serve its load from one point.
- Start-ups, ramp limits and reserves, which change which plant responds.
- Part-load efficiency: each plant burns its full-load fuel per MWh at every output.
Check yourself
Myth or fact? “The hours with the lowest average CO₂ factor are the hours when extra use adds the least CO₂.”
- Myth
- Fact
Answer
Myth, here. From 00:00 to 12:00 the average is 0.10 to 0.17 t per MWh, yet each extra MWh comes from the combined-cycle plants at 0.40 t.
Myth or fact? “When a buyer may claim zero emissions, its use adds no CO₂ at the plants.”
- Myth
- Fact
Answer
Myth. With certificates from an existing wind farm the data center may claim 0 t, while the plants emit 3,039 t more that day: the purchase added nothing to the grid.
Myth or fact? “Inventory shares of every user add up to the grid's total; marginal changes need not.”
- Myth
- Fact
Answer
Fact. Shares divide a total that happened. A marginal change answers a different question: what one user's choice changes.
In one hour the plants emit 2,000 t of CO₂ for 8,000 MWh of use. A load uses 100 MWh in that hour. What is its inventory share?
- 2,000 t
- 25 t
- 40 t
- 0.25 t
Answer
25 t: the average factor is 2,000 ÷ 8,000 = 0.25 t per MWh, times 100 MWh.
40 t would be the 0.40 t of a combined-cycle plant on the margin: question 2, not question 1.
Numbers, sources and what’s next
Numbers to remember
| Number | What it means | Source |
|---|---|---|
| 52.91 kg per MMBtu | CO₂ from burning natural gas, per million Btu of heat; a gas plant's CO₂ per MWh is this times its heat rate. | EIA, Carbon Dioxide Emissions Coefficients · released 18 September 2024 |
| 7,535 and 10,975 Btu per kWh | Average tested heat rates of U.S. natural gas combined-cycle plants and gas turbines in 2025. | EIA, Electric Power Annual, Table 8.2 · 2025 data, released 21 October 2026 |
Next
In the real world
In the reported scenarios, reductions in attributed emissions were on the order of one to two percent of regional totals or less.
Reductions are computed under average-emission-intensity accounting with perfect foresight; they are not estimates of the consequential (marginal) system response.
Case file CF6: how much shifting load to cleaner hours can cut emissions, and when.
When gas-fired backup power became much more expensive, the model built fuel cells and turned stored green hydrogen back into electricity.
Industry then drew more hydrogen from gas reformers, and the round trip from power to hydrogen and back lost energy, so total emissions rose although gas-plant emissions fell.
Inputs reflect 2022–2023 conditions; post-2023 federal hydrogen credit rules are not modeled.
Case file CF3: where hydrogen should go when nobody knows how much wind will be built.
Sources and further reading
- EIA. Electric Power Annual, Table 8.2: Average Tested Heat Rates by Prime Mover and Energy Source, 2025 data (released 21 October 2026); Table 7.4 (natural gas at $3.85 per MMBtu in 2025).
- EIA. Carbon Dioxide Emissions Coefficients, 18 September 2024.
- EIA. Assumptions to the Annual Energy Outlook 2026: Electricity Market Module, Table 3 (a new combined-cycle plant; variable O&M), April 2026.
- EIA. Hourly Electric Grid Monitor (Form EIA-930), Southwest Power Pool, 2025; Electric Power Monthly, Table 6.07.B, issue of 23 October 2026.
- GHG Protocol. Scope 2 Guidance (2015), Table 7.1: Scope 2 Quality Criteria, page 60.
- GHG Protocol. Public consultations on Scope 2 and on consequential electricity-sector accounting, opened 20 October 2025; Scope 2 Public Consultation Feedback Summary: Executive Summary, 29 July 2026.
- Siler-Evans, K., Azevedo, I. L. and Morgan, M. G. (2012). Marginal Emissions Factors for the U.S. Electricity System. Environmental Science & Technology 46(9), 4742–4748.
- Gagnon, P. and Cole, W. (2022). Planning for the evolution of the electric grid with a long-run marginal emission rate. iScience 25(3), 103915.
- Ricks, W., Xu, Q. and Jenkins, J. (2023). Minimizing emissions from grid-based hydrogen production in the United States. Environmental Research Letters 18(1), 014025.
- Li et al. (2025). Real-time greenhouse gas emission intensity informed demand-side load regulation for power grid decarbonization. Cell Reports Sustainability 2, 100367.
- Li et al. (2025). Hydrogen’s role in expanding renewable energy and reducing carbon emissions in Texas. International Journal of Hydrogen Energy 113, 77–89.
- Earlier lessons: lesson 12 (cost against CO₂), lesson 8 (the marginal plant), A7 (attributed CO₂ and the plants' CO₂), A16 (hourly matching for hydrogen).
Spotted an error, or stuck? Write to hlli@ksu.edu. Corrections are made in place and dated here.
Text licensed CC BY 4.0, code MIT: teachers and students may reuse and adapt it, with credit to Honglin Li.