Summary
- Further refined data sets for the Oil & Gas Methane Emissions Reduction Project to come up with the use case for the Hyperledger Global Challenge 2022.
Use Case
We define methane and flaring emissions fro Oil & Gas producers based in U.S. oil and gas basins (see image above):
- Permian (P) - Texas, Nevada
- Bakken (B) - North Dakota, Montana
- Niobrara (N) - Wyoming, Colorado, ....
Each basin is assinged 2 Audited Emissions Certificates from the NET network, one for methane emissions (fugitive, venting, incomplete flaring), Bm, Nm, and Pm and one for total methane flaring Bf, Nf, and Pf. We estimate the total methane and flaring emissions, denominated in million tones of CO2e total emissions as compiled in this spreadsheet with sources.
Each basin is also assigned fuel production tokens for oil and natural gas (e.g., Bo and Bg) in metric tons of oil equivalent (mtoe) and billion cubic meters (bcm), respectively. We use 0.86 mtoe/bcm to convert gas to oil equivalent. Note that we have selected mtoe and bcm in our basin examples for the year 2020 detailed in Table 1, however, oil and gas could be converted in any unit base on user preference.
Table 1: Emission and fuel production token data for select oil and gas basins in the U.S. in 2020.
C-NFT component Basin | Methane, mt CO2e input | Flaring, mt CO2e input | Oil prod., mtoe output | Gas prod., bcm output | O&G prod., mtoe (total output) | EF methane kgCO2e/toe partial | EF flaring kgCO2e/toe partial | EF total, kgCO2e/toe total |
|---|---|---|---|---|---|---|---|---|
| Bakken, B | Bm = 35.30 | Bf = 3.17 | Bo = 60.43 | Bg = 27.94 | BT = 84.46 | 418.0 | 37.6 | Be = 455.5 |
| Niobrara, N | Nm = 4.07 | Nf = 0.16 | No = 237.25 | Ng = 56.62 | NT = 81.05 | 50.2 | 2.0 | Ne = 52.2 |
| Permian, P | Pm = 27.53 | Pf = 7.19 | Po = 1597.58 | Pg = 172.83 | PT = 366.54 | 75.1 | 19.6 | Pe = 94.7 |
U.S. Average | 298.5 | 15.2 | 313.7 | |||||
| Global Average | 229.9 | 38.7 | 268.5 |
The data in Table 1 are used to construct Carbon Trakers Non Fungible Tokens (C-NFTs) for each basin (producer), as introduced in this working paper. The C-NFT links the methane and flaring emission tokens to the corresponding fuel production tokens, which, are assigned to each C-NFT at the closure of a audit, verification and certification processes. The C-NFT provides an anchor for the embodied flaring and venting emissions, measured as the carbon intensity or emission factors (see right rolumns in Table 1) of the oil and gas fuel tokens.
- Permian C-NFT carbon intensity anchor point, Pe = (Pm + Pf)/(Po + Pg *0.86 mtoe/bcm) = 94.7 kgCO2e/toe
The emission factors assigned to the C-NFT for each basin, and national and global averages reveal important information about U.S. production at the basin level. First U.S. flaring is well below global averages, and is lowest in the Niobara basin in Wyoming/Colorado. However, the majority of oil and sector waste emissions come from methane, which is more the 25 times more potent than the CO2 produced during methane flaring. U.S. methane emission factor is estimated to be 30% greater than the global average in 2020. As a result the Bakken was the most polluting basin on a per unit basis, with Niobrara achieving the lowest levels.
Note: In the original C-NFT contract fuel tokens are denominated in kgCO2e, by applying the standard scope 1 emission factor of each fuel: oil = 2.966 ton CO2e/toe; gas = 1.875 ton Co2e/kcm. As dsicussed in the call with Si Chen we consider assigning custom units to fuel tokens used in a C-NFT. However, for consistency these should conform to the same unit, e.g., mtoe used in Table 1.
Now any fuel tokens assigned to the C-NFT, can be transferred to an energy trader or fuel consumer (e.g., via a public utility company).
Local Gas Utility buys 1 bcm natural gas with 20% from oil company in Bakken, 20% from Niobrara and 60% from Permian. Now we can calculate the average methane emission factor of the gas supplied by the public utility.
- Local Gas methane emissions per thousand cubic-meter natural gas: (Be*0.2 + Ne*0.2 + Pe*0.6) = 158.36kgCO2e/toe.
More interestingly, the Local Gas Utility can acquire the following natural gas fuel tokens from each C-NFT, 0.2 bcm form Bg and Ng , and 0.6 bcm of Pg. These can be re-sold to consumers at a premium on the emission factor anchor. Clearly, Niabrara could seek a significant premium compared to the average emission factor of the Local Gas Utility: 52.2 vs. 158.36 kgCO2e/toe. Assuming consumers are exposed to a carbon price of 100 USD/tonCO2e, the fuel production linked to the Niabrara contract could seek a premium of up to 9.1 USD/kcm or 0.25 USD/mmtbu, roughly 10% of the Henry Hub Natural gas prices in 2020 (~2.5 USD/mmbtu).
Video 2
Time:
- Wednesday April 15, 2022 at 09 AM Pacific
- Add Climate Action and Accounting SIG calls to your calendar
Dial-In Information: [ZOOM]
For this call, the call-in information is different:
Join Zoom Meeting
https://zoom.us/j/93426785264?pwd=bmJOL2UxTzNHME5HTjg3c3d0QjZVdz09
Meeting ID: 934 2678 5264
Passcode: rQCq8x
|
Hyperledger is committed to creating a safe and welcoming community for all. For more information please visit the Hyperledger Code of Conduct. |
|---|
11 Comments
Bertrand WILLIAMSRIOUX
Arezki Dji Si Chen
Please see the revisions to the meeting notes above. I have collected O&G basin estimates of total methane emissions from the following map of the Permian https://data.permianmap.org/pages/operators (EDF project), and state level aggregates of venting and faring data from the EIA (https://www.eia.gov/dnav/ng/ng_prod_sum_a_EPG0_VGV_mmcf_a.htm) for the Bakken and Niobrara (Wyoming/Colorado). Without methane the total emission factors are not represented of total performance. While U.S. is advanced related to the global average on flaring, its methane emissions were above global estimates in 2020.
I have added a calculation of a premium that could be paid for avoided methane and flaring emiossions by sourcing Niobrara fuel token. The premium is about 10% of the Henry Hub natural gas prices in 2020 (~2.5 USD/mmtbu) assuming a carbon price/value of 100 USD/ton CO2e.
Si Chen
Bertrand WILLIAMSRIOUXThanks for putting this together.
So going through this as a use case to be implemented, can we think of it as 3 oil/gas producers each with audited emissions from the sources of methane:
For demo purposes we could assume each one produces 10% of the total of each basin, so their total audited emissions are 1/10 of the totals for each basin.
We would need some way to transfer of natural gas between oil/gas producer and the utility and then the utility and the consumer, to trigger the transfer of the embedded emissions. How do you see that happening? What are the exact steps to link the Carbon Tracker C-NFT to the audited emissions of the producers, and then transfer them to the customer?
Bertrand WILLIAMSRIOUX
Si Chen
Si Chen
Bertrand WILLIAMSRIOUXSorry this part is still not too clear to me yet. Could you explain a little more? Exactly what is on the C-NFT token? What is a "fuel token"? Does it actually have quantity of fuel, ie tons of oil equivalent or something?
Bertrand WILLIAMSRIOUX
Si Chen. See the modification to Table 1 to define the component of the C-NFT.
Audited emissions are inputs in KG CO2e
Fuel (or products) tokens are outputs. I am revising the contracts so that fuel token units can be user defined:
However, distributing emissions based on downstream scope 3 (instead of energy) will attribute less embodied emissions (e.g., venting, flaring) to fuels with lower emission factor (e.g. natural gas vs. oil). See below for an exmale.
This is something for user to consider - if a fuel (e.g., natural gas) has a lower downstream scope 3 impact per unit energy, should it be assigned less embodied emissions than its co-products (e.g., oil)???
Taking the original Table 1 (above) we express fuel tokens (outputs) from Bakken using downstream scope 3 impact as mt CO2e (instead of mtoe).
oil = 2.966 ton CO2e/toe; gas = 1.875 ton Co2e/kcm = 2.180 ton Co2e/toe
C-NFT component
Basin
Methane, mt CO2e
input
Flaring, mt CO2e
input
Oil prod., mt CO2e
output
Gas prod., mt CO2e
output
O&G prod., mt CO2e
(total output)
EF total, CO2e in / out (%)
total
C-NFT emission factor (EF) defines the embodied (venting/flaring) emissions of each fuel unit: inputs /(total outputs). In this case a simple percentage (Bm+Bf)/(Bo+Bg) = 16%.
What are the embodied emission from buying 1 mtoe of gas from Bakken?
Emission factor fo natural gas (downstream scope 3) is 2.180 ton CO2e/toe.
2.180 ton CO2e/toe * 1 mtoe of gas * 16% = 0.349 mtCO2e embodied emissions.
If we use energy content from the original Table, Be = 0.4555 tCO2e/toe.
For 1 mtoe of gas we get 0.456 mtCO2e > 0.349 mtCO2e.
Si Chen
So the C-NFT for Bakken would be B_e = (B_f + B_m)/(B_T) = (35.3 + 3.17) / 241.12 = 38.47 mtCO2e / 84.46 mtoe = 455.5 kgCO2e/to
So if a utility bought 1 million tons oil equivalent of natural gas from the Bakken oil producer, it would get 455.5 kgCO2e X 1 million in CO2e in its carbon account, or 0.455 mtCO2e.
If you're thinking that C-NFT is just a retired "data point" of carbon intensity, then we would need some token which represents the total "embedded emissions" of a Bakken oil producer, so that it is transferred down its customers. In this case, if the hypothetical Bakken producer is 10% of the Bakken total, it would have 3.847 mtCO2e of embedded emissions against 8.446 mtoe to transfer.
What do you think of this? Or just use the C-NFT as the inventory of embedded emissions?
Bertrand WILLIAMSRIOUX
Si Chen I have considered the option of transferring all emission tokens.
I define two approaches
Original (current): a trail of emission data is tracked up supply chain accounts by referencing (the C-NFT data points)
New (proposed): a cascade of emission data that pools into final accounts
They achieve the same accounting results. The benefit of transferring embedded emissions (proposed) is explicit ownership of emission tokens → real consumption based emission accounting.
In the original C-NFT, embedded emission (input) tokens were defined as retired (non-transferable). Transferring them would require
In the new approach, as a fuel is processed in a supply chain a cascade of embedded emission tokens follow. In addition to transferring product tokens the entire inventory of embedded emission data is also transferred (1) or issued (2) to the buyer's account. This would increase the amount of tx data, compared to referencing a C-NFT data point in the original approach.
Arezki Dji what do you think? There may be good reason to enable both approaches. E.g., give the buyer the option to cascade down token data from the C-NFT trail, creating a new (perhaps more convenient?) data point with all the token data
Si Chen
Generally I'd prefer the new(proposed) approach so the tokens could be transferred.
Is the C-NFT just the embedded emissions of the fuel? Would it need to be linked to a purchase of fuel somehow?
Bertrand WILLIAMSRIOUX
Si Chen No the C-NFT is not just the embedded emission. It contains the embedded emissions and the fuel tokens. It is referred to as an NFT because it holds two unique attributes:
Combined they tell a buyer the amount embedded emissions for a fuel/product token purchase.
The owner of the C-NFT calls it to transfer the fuel tokens. In the new approach the emission tokens could be:
Si Chen
Ok, so the C-NFT has the fuel purchase embedded for example 1 million toe of natural gas?
So it could be used not only to transfer emissions but also transfer fuel as well?
Bertrand WILLIAMSRIOUX
yes