In May, the International Energy Agency (IEA) released a report ‘Net zero by 2050 – a roadmap for the global energy sector’ whose impact went beyond the energy sector and made it into general news headlines. Unsurprisingly, the tough regime it prescribes to bring global energy sector carbon emissions down to zero in just 30 years includes some dramatic steps – such as an end to new gas heaters by 2025 – which are highly tangible for the general public.
However, the sensation caused by this report was also somewhat surprising, given that the IEA is no stranger to rapid decarbonisation scenarios. The past four editions of its flagship ‘World Energy Outlook’ have included a Sustainable Development Scenario (SDS), which takes the global energy sector to net zero by 2070; this schedule is considered appropriate to meet the principal Paris Agreement goal of keeping global warming ‘well below 2°C’. Bringing net zero forward to 2050 corresponds to the Agreement’s less binding call on signatories to ‘pursue efforts to limit warming to 1.5°C’. Despite the enormous challenge posed even by the SDS, the tougher target has recently received growing attention, with many – largely high-income – countries setting out their own ambition to reach net zero by 2050.
Ambitious decarbonisation scenarios such as these tend to require carbon capture utilisation and storage technology (CCUS) on a massive scale, and the latest report is no exception. Despite seeing fairly limited deployment to date, CCUS is an indispensable technology for rapid decarbonisation because of the wide range of difficult problems it helps to solve. Firstly, it provides a decarbonisation solution for industries like cement and steel, which have few viable alternatives; secondly, it can help deal with recently built emitters that can’t easily be closed, including power plants which may be crucial for meeting energy demand; thirdly, it can play a significant role in the production of low-carbon hydrogen; and lastly and perhaps most importantly from a net-zero perspective, it can be used to permanently remove CO2 from the atmosphere – offsetting any remaining emissions.
My first thought regarding the IEA’s ‘Net zero by 2050’ report was therefore to look at how the role of CCUS may have changed relative to the slower transition to net zero laid out in the SDS. In the SDS, a significant proportion of the final CCUS capacity adopted is rolled out towards the end of the scenario, post-2050, when all forms of the technology should be well established and the more challenging CO2 cuts must be made. From a starting point of 50 Mt of CO2 captured annually in 2020, this rate increases to 10 Gt by 2070, passing through 5.6 Gt of CO2 in 2050.
In the IEA’s latest scenario – accelerated by 20 years – the total amount of CO2 captured in 2050 is increased to 7.6 Gt, making for a much more rapid initial roll-out, but ultimately managing to reach net zero with significantly less CCUS. Interestingly, the IEA already published a (much less-heralded) ‘net zero by 2050’ scenario as part of last year’s WEO, which still called for over 10 Gt of CO2 to be captured in 2050. So, the recent report does seem to represent a somewhat reduced role for the technology, at least over this time period. Given the scenarios publish fairly sparse detail on CCUS, sometimes using different metrics, it can be tricky to see the detail underlying this change, but some conclusions can be drawn.
By 2050, both scenarios see CCUS mop up around 3.5 Gt of CO2 emissions from all fossil fuel-based processes, including heavy industry, hydrogen production, and power plants. In the power sector, 222 GW of coal power plants and 171 GW of natural gas power plants are equipped with CCS, which is again similar to the levels in the Sustainable Development Scenario in 2050. Fitting fossil power plants with CCS is seen by the IEA as essential for preventing excessively early closure of recently built plants in power-hungry nations, as well as playing a useful role in backing up intermittent wind and solar power. However, interestingly, these decarbonised fossil fleets produce much less electricity in the accelerated net zero scenario: 36% less in the case of coal and nearly 50% less in the case of gas. This probably reflects the fact that more rapid decarbonisation requires a faster scale up of other low-carbon generation. However, the fact that a similar number of CCS-equipped plants are still required is revealing of the role for this technology as a little used but still essential tool for grid balancing.
In the ‘Net zero by 2050’ scenario, there is no electricity generated by coal without CCUS by as early as 2040 – requiring a precipitous fall in output over the next 20 years, given that coal is still the world’s largest source of power. This trajectory contrasts slightly with the SDS, in which a small amount of ‘unabated’ coal sticks around until around 2060. In both scenarios, gas-fired power generation plays a more significant role, with some uncaptured gas power emissions remaining even in the 2050 net zero scenario, presumably offset by CO2 removal from the atmosphere.
Both IEA scenarios rely heavily on low-carbon hydrogen as a heating and transport fuel, as well as using it to fuel dispatchable power plants. IEA scenarios typically make use of fossil-derived ‘blue hydrogen’ with CCS and electrolysis-derived ‘green hydrogen’ in roughly equal proportions. The total of 520 Mt of low-carbon hydrogen required in the net zero 2050 scenario (with over 200 Mt based on CO2 capture) is similar to the amount used by 2070 in the SDS, and accounts for a much bigger proportion of the total fossil CO2 captured in 2050. So, the accelerated decarbonisation has clearly not dented the need for this future fuel in a net-zero world, but the pace of expansion is nearly doubled as a result.
Direct air capture (DAC) – used to remove CO2 from the atmosphere – is scaled up much more rapidly in the new report, with 600 Mt of CO2 captured in this way by 2050, compared to only 100 Mt by the same date in the SDS. Given that this technology will only be demonstrated at large-scale for the first time in around 2025, this represents a highly ambitious increase in deployment rate. However, the SDS uses still more DAC when it finally achieves net zero in 2070.
Bio-energy with CCS (BECCS) is the other main way in which carbon capture can be used to achieve CO2 removal or ‘negative emissions’, and it sees roughly similar levels of expansion by 2050 in the two scenarios. On the other hand, it is again much less relied upon in the final vision of net zero, as the SDS then ramps up its deployment three-fold to reach 300 Mt in 2070.
In short, a major difference between the two reports is the extent to which they rely on these ‘negative emissions’ to balance the books; net zero in 2050 has nearly 2 Gt compared to nearly 3 Gt in the SDS. This likely reflects the constraints of a more rushed transition – carbon removal technologies are already ramped up as fast as possible, so any remaining emissions have to be cut faster in order to match the removals in time for the earlier deadline.
Despite these nuances between the two scenarios, it is clear that CCUS must play a major role in reaching net zero on any reasonable timescale, and will be required at a scale far beyond the levels we see today. It is also worth noting that net zero is not the end of the story. If the energy transition is able to continue into net negative emissions – removing more CO2 from the atmosphere than we emit – even a 2070 target may not be too late to bring warming down to 1.5°C. For this ‘clean-up job’, carbon capture will be essential.