Executive Summary:
291.9 MW of power from vented carbon negative sources such as landfill gas would make the entire Bitcoin network carbon neutral.
To put in context how achievable this is: if Bitcoin mining companies focused on using previously vented sources of methane emissions (such as: vented landfill gas) it would take only 35 landfills each producing 8 MW power (mid sized landfills) fully combusting their methane for Bitcoin mining) to make the entire Bitcoin network carbon negative.
132.5 MW of flared gas methane mitigation using Bitcoin mining has come online over an 16 month period. This represents growth of 8.3 MW/month. Bitcoin mining using vented methane as power is currently growing at a much slower pace. We anticipate only 3 MW/month being the average for the next 3-4 years. At this growth rate, the Bitcoin network will abate more emissions than it creates by 2027[i].
This puts Bitcoin on track to be the world’s first cryptocurrency, and indeed the first monetary system, to achieve net zero emissions (without purchasing offsets).
Methodology:
Combusting methane from sources such as biogas or landfill gas that would have gone into the atmosphere is one of the four recognized forms of carbon capture technology and is “one of the few carbon-negative fuel sources” in the world.
This counterintuitive fact comes from the fact that methane is a 84x more warming Greenhouse Gas than CO2. So unlike combusting natural gas from a pipeline, when we combust methane that would have otherwise become air-borne (such as biogas and landfill gas), it is extremely carbon negative.
This is important because the United Nations Environment Program as said that reducing methane emissions represents the strongest lever we have to slow climate change over the next 25 years.
Current techniques such as flaring are expensive, place a heavy burden on regulators and worst of all are very inefficient (don’t completely remove the methane, even in perfect conditions). For this reason, UNEP, IPCC and EDG have joined forced to institute a global ban on routine flaring. IEA states that flares are only 92% efficient (8% is still released into the atmosphere).
Using these fact, we:
- Calculated the amount of Bitcoin energy that currently comes from carbon positive sources
- Worked out the amount of emissions this generated (carbon positive quantity)
- Reverse engineered to calculate how much methane would need to be removed from the air via combustion in order to counter-balance these emissions (carbon negative quantity). We factored in that combusting methane produces CO2 which is of course still a greenhouse gas, albeit 84x less warming – to produce a true net emission reduction amount using standard carbon emissions methodologies used by the US EPA and Carbon Credits.
Calculations
1. Calculate the total Bitcoin energy that comes from fossil fuel
Current Estimated Bitcoin annual power consumption: 96.28 TWh
Source: Cambridge University https://ccaf.io/cbeci/index (data is dynamic, and correct as of 29 Aug ’22)
Percentage of fossil fuel based power: 41.5%
59.5% of the Bitcoin network is non-fossil fuel based. My own ground-up model (unpublished) suggests a slightly more conservative figure that is more inline with the Bitcoin Mining Council’s Q4 ’21 estimate of 58.5%, so using the more conservative estimate of the two: fossil fuel sources: = 41.5%
=> 96.28 x 0.415 = 39.96 TWh of fossil-fuel based energy
2. Calculate how many tonnes CO2 emissions this represents
Approach:
Balance of coal:gas (assuming standard global grid-mix proportions): 36.7:23.5 =
NG: 23.5%, Coal 36.7% of global electricity generation Source: Our World in Data https://ourworldindata.org/electricity-mix
36.7/60.2: 23.5/60.2
= 61.0:39.0 Coal:Gas ratio
=> 24.34 TWh from coal, 15.58 TWh from gas
- 0.986 t CO2 /MWh for coal-fired plants
- 0.429 t CO2 /MWh for gas-fired plants
Source: https://www.rte-france.com/en/eco2mix/co2-emissions
% Share of Global Hashrate
- USA : 42%
- Kazakhstan+Iran: 9.7%
- Rest 4.57% (Canada, China, Rest of world)
Source: https://beincrypto.com/bitcoin-mining-renewables-unreported/
We assume a standard 61/39 split between Coal/Gas for fossil fuel component of the rest of the world, whereas Kazakhstan and Iran’s fossil fuel use is effectively 100% coal based.
So limiting the percentage to these 3 regions, we get the following % breakdown
- USA: 42/56.27 = 74.64%
- Kazakhstan+Iran: 42/56.27 = 17.24%
- Rest 4.57/56.27 = 8.12%
- USA Gas:Coal ratio = 38.3%: 21.8% = 0.637:0.363 gas:coal split
- Kazakhstan/Iran Gas:Coal ratio is 0.0:1.0
- Global grid Gas:Coal ratio is 0.39:0.61.0
From 1. Above, there is 39.96 TWh fossil fuel electricity.
So in US that’s 29.83 TWh
- .7464×39.96×0.637 = 19.00 TWh from Gas
- .7464×39.96×0.363 = 10.83 TWh from Coal
Kazakhstan/Iran FF component is 6.89 TWh
- 100% coal => 0.1724×39.96= 6.89 TWh from Coal
Rest of World, standard mix = 61.0:39.0 Coal:Gas ratio
- So that’s 39.96×0.0457 = 1.82 TWh
- 1.82×0.39 = 0.71 TWh from Gas
- 1.82×0.61 = 1.11 from Coal
=>
Total from Gas = 19.71 TWh
Total from Coal = 18.83 TWh
=>
Total emissions from Gas = 19.71 x 429,000
Total emissions from Coal = 18.83 x 986,000
=> Total Bitcoin network emissions = (429,000 x 19.71) + (986,000 x 18.83) =
= 8.456 Mt+ 18.566 Mt = 27.022 Mt CO2-eq per year. (Mt stands for Megatons – one million metric tonnes)
*CO2-eq = CO2 equivalent. Because methane is 84x more warming than CO2 over a 20-year period (86x if feedbacks are included), 1 t methane emissions = 84 t CO2-eq
May 2023 Update: Current network emissions are now 34.8 M t (See BEEST model for raw data).
3. Reverse Engineer to work out how much Landfill gas Power would need to be generated to offset these 34.8 Tonnes CO2-eq
- Methane has a 20-year GWP (Global Warming Potential) of 84 => 34.8 Million Tonnes CO2-eq represents 414,286 t Methane (per year).
- When vented landfill gas, or biogas from wastewater or farms, is combusted – it reduces net CO2-eq emissions by 95.8%
So to offset 414,286 t methane, 414,286/0.958 = 432,449 t methane is required - 1 t Methane when combusted produces 4.79 MWh*
*Methane generates 9.81 Kwh/m3. Microturbines and generators have efficiency of up to 35%. This means that to generate 1 MWh, you would need 1000/(9.81×0.35) = 291 m3. Methane’s density is 0.717kg/m3. So 291 m3 = 0.291×0.717 = 0.208 t (tonnes). So 1t generates 4.79 MWh
So 335,798 t methane generates 432,449 x 4.79 = 2071 GWh (per year)
There are 8760 hours in a year, so this = 2071 GWh/8760 h = 236.5 MW (theoretical)
However, 10% of methane according to UNFCCC is oxides anyway, and uptime of LFG generators can be assumed to be only 90%, therefore the actual megawatts required will be 236.5/(0.9)2 = 291.9 MW
MWh/8760 h = 291.9 MW[ii]
It’s worth noting that if the methane source includes the average ratio of flared:vented landfill gas which is 30% flared/ 70% vented, it would take considerably more power to achieve the same reduction. The calculation is also more complex, as it involves looking at the flared and vented portions separately. I’ve included the calculation at the end of the document for those interested. For those who want the summary: the answer is that it would take 399.6 MW[iii] of energy.
If we attempted to make the Bitcoin network net zero emission only using previously flared gas, the amount of carbon-negative Bitcoin mining power required goes up by an order of magnitude because the flaring process was already eliminating 92% of the methane. In this case, 3225.6 MW[iv] of energy is required.
4. Future projections
Use of flared gas for Bitcoin mining has grown at a rate of 8.3 new MW power every month since May ’21.
We forecast that use of vented gas to power Bitcoin mining will grow at only ~45% the rate due to the additional capital requirements of mining using vented gas (need for Gas Capture and Collection Systems for all venting landfills). This adds around 15% to the project procurement-to installation timeframe and 20% to the capital requirements for a bitcoin mining operator using landfill gas for power (source: Adam Wright, Vespene).
Behind Vespene, there are another 4 companies that we’re aware of, specialising in Landfill gas, who are currently capital raising, or who have recently completed cap-raises. 5 of 10 renewable-based mining operations I have had direct contact with have also said that they are actively evaluating the feasibility of using landfill gas for their next project. So the landscape is similar to the flared gas landscape 2 years ago, albeit with considerably more evidence of interest in the pipeline, perhaps because of the 13x higher level of carbon mitigation.
[i] While hashrate and therefore energy consumption will have increased, network renewable usage and miner efficiency is also increasing – providing a counterbalance to hashrate increase. Using trend maps, we’ll do a deeper dive into the likely Bitcoin net zero emissions data in a separate paper.
[ii] There is academic debate as to whether methane’s 20-year GWP of 84-86, or 100-year GWP of 28-36 is used to calculate CO2-eq emissions. Lately more scholars have been calling for the use of the 20-year GWP, based on the urgent time horizon required to address methane emissions. The UNEP and IPCC are increasingly using this metric as they pivot their focus to prioritizing urgent methane mitigation. Further, there is strong evidence that even the current GWP20 of 84 may now be severely underestimated. This is because methane is now lingering in the atmosphere much longer due to a number of factors. These factors include wildfires releasing CO, which atmospheric hydroxyl now targets in preference to breaking down methane. Even if we were to use the unadjusted and increasingly less favored GWP100 for methane of 30, it would still take only 523.9 MW power to make the Bitcoin network carbon neutral.
[iii] Further Calculations:
First calculate the amount of CO2-eq reduced through Bitcoin mining using 1MW on 30% flared, 70% vented sources
From Calculations 3.3 above, we know that 335,798 t methane combustion is enough to make the Bitcoin network zero emissions, if it came from 100% vented sources. Let’s start with that number, and see how many what the CO2-eq value of this will be when it comes from 30:70 flared:vented sources
CO2-eq reduction through combustion = Flared CO2-eq reduction + Vented CO2-eq reduction
Flared CO2-eq reduction:
Start with 0.3x 100,000 t
= 30,000×84 = 2,520,000 CO2-eq (because methane has 84x the GWP of CO2 over a 20-year period
CO2-eq reduction from combusting flared gas =
(Baseline CO2-eq reduction achieved by flaring) – (Combusted CO2-eq reduction)
= [(0.08x84x30,000) + (0.92×2.75×30,000)] – (2.75×30,000)
= (6.72+2.53-2.75)x30,000
= 6.5×30,000 = 195,000 t CO2-eq
Vented CO2-eq reduction:
Start with 0.3x 100,000 t
= 70,000×84 = 5,880,000 CO2-eq
CO2-eq reduction from combusting vented gas =
Vented total – combusted total
= (84×70,000) – (2.75×70,000)
= 81.25×70,000 = 5,697,500 t CO2-eq
=> CO2-eq reduction through combustion = 195,000 t + 5,697,500 t = 5,882,500 t CO2-eq
From Calculations 2 above, we would need 27.02 t CO2-eq to make the Bitcoin network net zero emissions. So we would therefore need 100,000 x 27.02/5.88 = 459,524 t Methane. Each t generates 4.79 MWh, so this = 2,202 GWh in a year, or 2241,711/8760 = 251.3 MW (theoretical), or 310.2 MW actual when Bitcoin was at 27.02 t emissions. At 34.8 t emissions, this would require 399.6 MW
* Note: the logic for each formula is described in the footnotes of Quantifying the Potential Impact of Bitcoin Mining on Global Methane Emissions
[iv] Using 100% previously flared gas, let’s calculate the amount of energy required to make the Bitcoin network net zero emissions.
From our equation above, we know that 30,000 t previously flared methane, when combusted has a net carbon negative impact of reducing 195,000 t CO2-eq emissions which produces 30,000×4.75/8760 = 16.27MW energy. We need 34.8 Mt CO2-eq emissions reduction however. So this would require 34.8/0.195×16.27 = 2903 MW energy. 10% would have been oxidized in air anyway (UNFCCC), so actual MW required: 2903/0.9 = 3225.6 MW
Document Revisions
28 May: Updated to reflect dynamic emission chart calculation, with new more nuanced predictor of zero-emission date for the Bitcoin network. Updated to reflect current estimated # landfills required to take the network carbon negative based on latest emission data.