Residual emissions are carbon emissions that are difficult or impossible to (fully) eliminate due to technological or other limitations. These emissions are extremely important, because – given that we effectively cannot avoid them – they need to be compensated with various kinds of carbon removal technologies (mainly carbon capture and storage) to be able to reach net zero (or carbon-neutrality). Carbon removal is very energy intensive, however, and there are limits to how much carbon we can remove. Estimates of these limits vary, but the most optimistic ones are typically somewhere in the neighborhood of 25Gt/year, or roughly half our current yearly CO₂-e emissions. (Less optimistic estimates tend to be closer to 5Gt/year.) The energy costs of removing 5Gt of CO₂ are probably close to 2.5PWh, which corresponds to approximately 1.6% of current global energy production or the combined electricity output of 285 “typical” nuclear power plants.1 I have on a number of occasions claimed that approximately 37% of CO₂-e emissions at the current state of technology are residual, which is very likely much more than what carbon removal can compensate.2 But let’s not take that percentage at face value and have a closer look at the data, estimates, and calculations behind it.

To calculate total residual emissions, we need two kinds of data: (1) contributions to yearly global emissions of various kinds of emitters (i.e., steel industry, rice farming, road transport, and so forth); and (2) estimates of how much of the emissions per kind of emitter are residual. The first kind of data is available at and elsewhere. The second can be inferred from a variety of sources, but mostly has to be estimated. The mathematics involved are extremely simple. Total residual emissions (as a percentage of total emissions) (\(RE\)) are the sum of the residual emissions per contributing sector (or kind of emitter):
$$ RE = \sum_i C_i R_i , $$ in which \(C_i\) is the percentage that a sector contributes to total CO₂ emissions, and \(R_i\) is the percentage of that sector’s emissions that are residual.

emissions related to energy use

sector \(i\) emission contribution \(C_i\) remarks residual emissions \(R_i\) product \(C_i R_i\)
iron and steel 7.2% Requires burning large amounts of coal. No prospect for significant reduction. \(R_i\) may be close to 100%. 80% 5.8%
other metals 0.7% Differs between types of metal. 50% 0.35%
chemical industry 3.6% 25% 0.9%
other industry 12.7% Probably very significant emission reduction possible in most industries. 15% 1.9%
agriculture and fishing 1.7% Ships and agricultural machinery need fossil fuels. Very limited possibilities for electrification. 90% 1.5%
road transport 11.9% Assuming almost complete electrification. 5% 1.9%
aviation 1.9% Virtually no reduction possible. 95% 1.8%
shipping 1.7% Virtually no reduction possible. 95% 1.6%
rail transport 0.4% Assuming almost complete electrification. 5% 0.0%
pipelines 0.3% 25% 0.1%
residential buildings 10.9% Assuming almost complete electrification, better insulation, and so forth. 5% 0.5%
commercial buildings 6.6% Assuming almost complete electrification, better insulation, and so forth. 5% 0.3%
unallocated fuel combustion 7.8% Various kinds of fuel burning that don’t fit in other categories (and that are often hard to control). Including, for example, burning wood for cooking or heating. 50% 3.9%
fugitive emissions from energy production 5.8% These would be significantly reduced if energy production would no longer predominantly rely on fossil fuels. 10% 0.6%

emissions related to agriculture and land use

sector \(i\) emission contribution \(C_i\) remarks residual emissions \(R_i\) product \(C_i R_i\)
grassland 0.1% No (significant) reduction possible. 100% 0.1%
cropland 1.4% No (significant) reduction possible. 100% 1.4%
agricultural soils 4.1% Mainly due to exposure of soil carbon through tillage and drainage. No significant reduction possible without giving up industrial agriculture (and killing the billions of people that depend on its products). 90% 3.7%
rice cultivation 1.3% No (significant) reduction possible. 100% 1.3%
crop burning 3.5% Could be significantly reduced in principle if (costly) alternatives are offered, but those are likely to require more agricultural machinery with their own emissions. 20% 0.7%
livestock and manure 5.8% Could be significantly reduced if sufficient people switch to a vegan diet (and/or if lab-grown meat takes off). 40% 2.3%
deforestation 2.2% Hard to say… There are reforestation plans, but there also increasing land requirements for energy production, and deforestation in the tropics seems to be escalating. But let’s be optimistic. 10% 0.2%

other emissions

sector \(i\) emission contribution \(C_i\) remarks residual emissions \(R_i\) product \(C_i R_i\)
cement 3.0 These emissions are technologically unavoidable. 95% 2.9%
chemical industry 2.2% Emissions related to various processes (rather than energy use) in the chemical industry. Significant capture and storage at the source may be possible. 30% 0.7%
landfills (waste) 1.9% The main alternative (burning waste) would be even worse. No significant reduction likely. 90% 1.7%
wastewater 1.3% No significant reduction likely. 90% 1.2%

total residual emissions / closing comments

Adding up the numbers in the column marked “product \(C_i R_i\)” gives the total residual emissions. That total is 36% (and thus, slightly lower than my previous estimate), which corresponds to roughly 18Gt CO₂-e. However, due to economic growth and associated growth in consumption, this latter number will rise, even if total emissions decrease. If we assume a steady-state economy (which is impossible under capitalism, but let’s ignore that), then this number (i.e., 18Gt) would not increase, and it would require roughly 5.6% of current global energy production (or 1027 typical nuclear power plants) to compensate it with carbon removal technology. Maybe that’s possible in principle, but given how slow carbon removal is growing, I am not optimistic. I doubt that we’ll ever be able to even remove a tenth of what is required. Notice also that little effort is made to reduce the part of emissions that aren’t residual. At the current state of technology, we could in theory reduce our emissions to 36%, but in practice, our emissions are still growing and the effect of reduction efforts are still in the single digits at best.

If you found this article and/or other articles in this blog useful or valuable, please consider making a small financial contribution to support this blog, 𝐹=𝑚𝑎, and its author. You can find 𝐹=𝑚𝑎’s Patreon page here.


  1. Obviously, it would completely defeat the purpose to use fossil-fueled power plants for carbon removal. In fact, all fossil-fueled power plants will have to be replaced with (relatively!) carbon-neutral alternatives (solar, wind, nuclear, and so forth) to have a chance at reducing emissions to levels anywhere near what carbon capture could possibly handle.
  2. And certainly much more than what we will be able to compensate in the first half of the current century considering the extremely slow growth of carbon removal technology and – especially – it’s actual employment.

Leave a Reply

Your email address will not be published. Required fields are marked *