Editing IPCC:AR6/SR15/Chapter-2 (section)

==== 2.3.4.1 CDR technologies and deployment levels in 1.5°C pathways ====

<div id="section-2-3-4-1-block-1"></div>

A number of approaches to actively remove carbon-dioxide from the atmosphere are increasingly discussed in the literature (Minx et al., 2018) <sup>[[#fn:r253|253]]</sup> (see also Chapter 4, Section 4.3.7). Approaches under consideration include the enhancement of terrestrial and coastal carbon storage in plants and soils such as afforestation and reforestation (Canadell and Raupach, 2008) <sup>[[#fn:r254|254]]</sup> , soil carbon enhancement (Paustian et al., 2016; Frank et al., 2017; Zomer et al., 2017) <sup>[[#fn:r255|255]]</sup> , and other conservation, restoration, and management options for natural and managed land (Griscom et al., 2017) <sup>[[#fn:r256|256]]</sup> and coastal ecosystems (McLeod et al., 2011) <sup>[[#fn:r257|257]]</sup> . Biochar sequestration (Woolf et al., 2010; Smith, 2016; Werner et al., 2018) <sup>[[#fn:r258|258]]</sup> provides an additional route for terrestrial carbon storage. Other approaches are concerned with storing atmospheric carbon dioxide in geological formations. They include the combination of biomass use for energy production with carbon capture and storage (BECCS) (Obersteiner et al., 2001; Keith and Rhodes, 2002; Gough and Upham, 2011) <sup>[[#fn:r259|259]]</sup> and direct air capture with storage (DACCS) using chemical solvents and sorbents (Zeman and Lackner, 2004; Keith et al., 2006; Socolow et al., 2011) <sup>[[#fn:r260|260]]</sup> . Further approaches investigate the mineralization of atmospheric carbon dioxide (Mazzotti et al., 2005; Matter et al., 2016) <sup>[[#fn:r261|261]]</sup> , including enhanced weathering of rocks (Schuiling and Krijgsman, 2006; Hartmann et al., 2013; Strefler et al., 2018a) <sup>[[#fn:r262|262]]</sup> . A fourth group of approaches is concerned with the sequestration of carbon dioxide in the oceans, for example by means of ocean alkalinization (Kheshgi, 1995; Rau, 2011; Ilyina et al., 2013; Lenton et al., 2018) <sup>[[#fn:r263|263]]</sup> . The costs, CDR potential and environmental side effects of several of these measures are increasingly investigated and compared in the literature, but large uncertainties remain, in particular concerning the feasibility and impact of large-scale deployment of CDR measures (The Royal Society, 2009; Smith et al., 2015; Psarras et al., 2017; Fuss et al., 2018) <sup>[[#fn:r264|264]]</sup> (see Chapter 4.3.7). There are also proposals to remove methane, nitrous oxide and halocarbons via photocatalysis from the atmosphere (Boucher and Folberth, 2010; de Richter et al., 2017) <sup>[[#fn:r265|265]]</sup> , but a broader assessment of their effectiveness, cost and sustainability impacts is lacking to date.

Only some of these approaches have so far been considered in IAMs (see Section 2.3.1.2). The mitigation scenario literature up to AR5 mostly included BECCS and, to a more limited extent, afforestation and reforestation (Clarke et al., 2014) <sup>[[#fn:r266|266]]</sup> . Since then, some 2°C- and 1.5°C-consistent pathways including additional CDR measures such as DACCS (Chen and Tavoni, 2013; Marcucci et al., 2017; Lehtilä and Koljonen, 2018; Strefler et al., 2018b) <sup>[[#fn:r267|267]]</sup> and soil carbon sequestration (Frank et al., 2017) <sup>[[#fn:r268|268]]</sup> have become available. Other, more speculative approaches, in particular ocean-based CDR and removal of non-CO <sub>2</sub> gases, have not yet been taken up by the literature on mitigation pathways. See Supplementary Material 2.SM.1.2 for an overview on the coverage of CDR measures in models which contributed pathways to this assessment. Chapter 4.3.7 assesses the potential, costs, and sustainability implications of the full range of CDR measures.

Integrated assessment modelling has not yet explored land conservation, restoration and management options to remove carbon dioxide from the atmosphere in sufficient depth, despite land management having a potentially considerable impact on the terrestrial carbon stock (Erb et al., 2018) <sup>[[#fn:r269|269]]</sup> . Moreover, associated CDR measures have low technological requirements, and come with potential environmental and social co-benefits (Griscom et al., 2017) <sup>[[#fn:r270|270]]</sup> . Despite the evolving capabilities of IAMs in accounting for a wider range of CDR measures, 1.5°C-consistent pathways assessed here continue to predominantly rely on BECCS and afforestation/reforestation (see Supplementary Material 2.SM.1.2). However, IAMs with spatially explicit land-use modelling include a full accounting of land-use change emissions comprising carbon stored in the terrestrial biosphere and soils. Net CDR in the AFOLU sector, including but not restricted to afforestation and reforestation, can thus in principle be inferred by comparing AFOLU CO <sub>2</sub> emissions between a baseline scenario and a 1.5°C-consistent pathway from the same model and study. However, baseline AFOLU CO <sub>2</sub> emissions can not only be reduced by CDR in the AFOLU sector but also by measures to reduce deforestation and preserve land carbon stocks. The pathway literature and pathway data available to this assessment do not yet allow separating the two contributions. As a conservative approximation, the additional net negative AFOLU CO <sub>2</sub> emissions below the baseline are taken as a proxy for AFOLU CDR in this assessment. Because this does not include CDR that was deployed before reaching net zero AFOLU CO <sub>2</sub> emissions, this approximation is a lower-bound for terrestrial CDR in the AFOLU sector (including all mitigation-policy-related factors that lead to net negative AFOLU CO <sub>2</sub> emissions).

The scale and type of CDR deployment in 1.5°C-consistent pathways varies widely (Figure 2.9 and 2.10). Overall CDR deployment over the 21st century is substantial in most of the pathways, and deployment levels cover a wide range, on the order of 100–1000 Gt CO <sub>2</sub> in 1.5°C pathways with no or limited overshoot (730 [260–1030] GtCO <sub>2</sub> , for median and 5th–95th percentile range). Both BECCS (480 [0–1000] GtCO <sub>2</sub> in 1.5°C pathways with no or limited overshoot) and AFOLU CDR measures including afforestation and reforestation (210 [10-540] GtCO <sub>2</sub> in1.5°C pathways with no or limited overshoot) can play a major role, <sup>[[#fn:4|4]]</sup> but for both cases pathways exist where they play no role at all. This shows the flexibility in substituting between individual CDR measures, once a portfolio of options becomes available. The high end of the CDR deployment range is populated by high overshoot pathways, as illustrated by pathway archetype ''S5'' based on SSP5 (fossil-fuelled development, see Section 2.3.1.1) and characterized by very large BECCS deployment to return warming to 1.5°C by 2100 (Kriegler et al., 2017) <sup>[[#fn:r271|271]]</sup> . In contrast, the low end is populated by a few pathways with no or limited overshoot that limit CDR to on the order of 100–200 GtCO <sub>2</sub> over the 21st century, coming entirely from terrestrial CDR measures with no or small use of BECCS. These are pathways with very low energy demand facilitating the rapid phase-out of fossil fuels and process emissions that exclude BECCS and CCS use (Grubler et al., 2018) <sup>[[#fn:r272|272]]</sup> and/or pathways with rapid shifts to sustainable food consumption freeing up sufficient land areas for afforestation and reforestation (Haberl et al., 2011; van Vuuren et al., 2018) <sup>[[#fn:r273|273]]</sup> . Some pathways use neither BECCS nor afforestation but still rely on CDR through considerable net negative CO <sub>2</sub> emissions in the AFOLU sector around mid-century (Holz et al., 2018b) <sup>[[#fn:r274|274]]</sup> . We conclude that the role of BECCS as a dominant CDR measure in deep mitigation pathways has been reduced since the time of the AR5. This is related to three factors: a larger variation of underlying assumptions about socio-economic drivers (Riahi et al., 2017; Rogelj et al., 2018) <sup>[[#fn:r275|275]]</sup> and associated energy (Grubler et al., 2018) <sup>[[#fn:r276|276]]</sup> and food demand (van Vuuren et al., 2018) <sup>[[#fn:r277|277]]</sup> ; the incorporation of a larger portfolio of mitigation and CDR options (Marcucci et al., 2017; Grubler et al., 2018; Lehtilä and Koljonen, 2018; Liu et al., 2018; van Vuuren et al., 2018) <sup>[[#fn:r278|278]]</sup> ; and targeted analysis of deployment limits for (specific) CDR measures (Holz et al., 2018b; Kriegler et al., 2018a; Strefler et al., 2018b) <sup>[[#fn:r279|279]]</sup> , including the availability of bioenergy (Bauer et al., 2018) <sup>[[#fn:r280|280]]</sup> , CCS (Krey et al., 2014a; Grubler et al., 2018) <sup>[[#fn:r281|281]]</sup> and afforestation (Popp et al., 2014b, 2017) <sup>[[#fn:r282|282]]</sup> . As additional CDR measures are being built into IAMs, the prevalence of BECCS is expected to be further reduced.

<div id="section-2-3-4-1-block-2"></div>

<span id="figure-2.9"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 2.9'''

<span id="section-6"></span>
<!-- IMG CAPTION -->
Cumulative CDR deployment in 1.5°C-consistent pathways in the literature as reported in the database collected for this assessment until 2050 (panel a) and until 2100 (panel b).
<!-- IMG FILE -->
[[File:8cc70c499928e076bd802a762902bc5a Figure-2.9-1024x729.jpg]]

Total CDR comprises all forms of CDR, including AFOLU CDR and BECCS, and, in a few pathways, other CDR measures like DACCS. It does not include CCS combined with fossil fuels (which is not a CDR technology as it does not result in active removal of CO <sub>2</sub> from the atmosphere). AFOLU CDR has not been reported directly and is hence represented by means of a proxy: the additional amount of net negative CO <sub>2</sub> emissions in the AFOLU sector compared to a baseline scenario (see text for a discussion). ‘Compensatory CO <sub>2</sub> ’ depicts the cumulative amount of CDR that is used to neutralize concurrent residual CO <sub>2</sub> emissions. ‘Net negative CO <sub>2</sub> ’ describes the additional amount of CDR that is used to produce net negative CO <sub>2</sub> emissions, once residual CO <sub>2</sub> emissions are neutralized. The two quantities add up to total CDR for individual pathways (not for percentiles and medians, see Footnote 4).

Original Creation for this Report using IAMC 1.5°C Scenario Data hosted by IIASA

<!-- END IMG -->
<div id="section-2-3-4-1-block-3"></div>

As discussed in Section 2.3.2, CDR can be used in two ways in mitigation pathways: (i) to move more rapidly towards the point of carbon neutrality and maintain it afterwards in order to stabilize global mean temperature rise, and (ii) to produce net negative CO <sub>2</sub> emissions, drawing down anthropogenic CO <sub>2</sub> in the atmosphere in order to decline global mean temperature after an overshoot peak (Kriegler et al., 2018b; Obersteiner et al., 2018) <sup>[[#fn:r283|283]]</sup> . Both uses are important in 1.5°C-consistent pathways (Figure 2.9 and 2.10). Because of the tighter remaining 1.5°C carbon budget, and because many pathways in the literature do not restrict exceeding this budget prior to 2100, the relative weight of the net negative emissions component of CDR increases compared to 2°C-consistent pathways. The amount of compensatory CDR remains roughly the same over the century. This is the net effect of stronger deployment of compensatory CDR until mid-century to accelerate the approach to carbon neutrality and less compensatory CDR in the second half of the century due to deeper mitigation of end-use sectors in 1.5°C-consistent pathways (Luderer et al., 2018) <sup>[[#fn:r284|284]]</sup> . Comparing median levels, end-of-century net cumulative CO <sub>2</sub> emissions are roughly 600 GtCO <sub>2</sub> smaller in 1.5°C compared to 2°C-consistent pathways, with approximately two thirds coming from further reductions of gross CO <sub>2</sub> emissions and the remaining third from increased CDR deployment. As a result, median levels of total CDR deployment in 1.5°C-consistent pathways are larger than in 2°C-consistent pathways (Figure 2.9), but with marked variations in each pathway class.

<div id="section-2-3-4-1-block-4"></div>

<span id="figure-2.10"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 2.10'''

<span id="section-7"></span>
<!-- IMG CAPTION -->
Accounting of cumulative CO <sub>2</sub>  emissions for the four 1.5°C-consistent pathway archetypes.
<!-- IMG FILE -->
[[File:23aa679f187a7ec52e7db9c53ef48d4d figure-2.10-1024x993.jpg]]

See top panel for explanation of the bar plots. Total CDR is the difference between gross (red horizontal bar) and net (purple horizontal bar) cumulative CO <sub>2</sub> emissions over the period 2018–2100, and it is equal to the sum of the BECCS (grey) and AFOLU CDR (green) contributions. Cumulative net negative emissions are the difference between peak (orange horizontal bar) and net (purple) cumulative CO <sub>2</sub> emissions. The blue shaded area depicts the estimated range of the remaining carbon budget for a two-in-three to one-in-two chance of staying below1.5°C. The grey shaded area depicts the range when accounting for additional Earth system feedbacks.

Original Creation for this Report using IAMC 1.5°C Scenario Data hosted by IIASA

<!-- END IMG -->
<div id="section-2-3-4-1-block-5"></div>

Ramp-up rates of individual CDR measures in 1.5°C-consistent pathways are provided in Table 2.4. BECCS deployment is still limited in 2030, but ramps up to median levels of 3 (Below-1.5°C), 5 (1.5°C-low-OS) and 7 GtCO <sub>2</sub> yr <sup>−1</sup> (1.5°C-high-OS) in 2050, and to 6 (Below-1.5°C), 12 (1.5°C-low-OS) and 15 GtCO <sub>2</sub> yr <sup>−1</sup> (1.5°C-high-OS) in 2100, respectively. In 1.5°C pathways with no or limited overshoot, this amounts to 0–1, 0–8, and 0–16 GtCO <sub>2</sub> yr <sup>−</sup> <sup>1</sup> in 2030, 2050, and 2100, respectively (ranges refer to the union of the min-max range of the Below-1.5°C and the interquartile range of the 1.5°C-low-OS class; see Table 2.4). Net CDR in the AFOLU sector reaches slightly lower levels in 2050, and stays more constant until 2100. In 1.5°C pathways with no or limited overshoot, AFOLU CDR amounts to 0–5, 1–11, and 1–5 GtCO2 yr <sup>−</sup> <sup>1</sup> (see above for the definition of the ranges) in 2030, 2050, and 2100, respectively. In contrast to BECCS, AFOLU CDR is more strongly deployed in non-overshoot than overshoot pathways. This indicates differences in the timing of the two CDR approaches. Afforestation is scaled up until around mid-century, when the time of carbon neutrality is reached in 1.5°C-consistent pathways, while BECCS is projected to be used predominantly in the 2nd half of the century (Figure 2.5). This reflects the fact that afforestation is a readily available CDR technology, while BECCS is more costly and much less mature a technology. As a result, the two options contribute differently to compensating concurrent CO <sub>2</sub> emissions (until 2050) and to producing net negative CO <sub>2</sub> emissions (post-2050). BECCS deployment is particularly strong in pathways with high overshoots but can also feature in pathways with low overshoot (see Figure 2.5 and 2.10). Annual deployment levels until mid-century are not found to be significantly different between 2°C-consistent pathways and 1.5°C-consistent pathways with no or low overshoot. This suggests similar implementation challenges for ramping up BECCS deployment at the rates projected in the pathways (Honegger and Reiner, 2018; Nemet et al., 2018) <sup>[[#fn:r285|285]]</sup> . The feasibility and sustainability of upscaling CDR at these rates is assessed in Chapter 4.3.7.

Concerns have been raised that building expectations about large-scale CDR deployment in the future can lead to an actual reduction of near-term mitigation efforts (Geden, 2015; Anderson and Peters, 2016; Dooley and Kartha, 2018) <sup>[[#fn:r286|286]]</sup> . The pathway literature confirms that CDR availability influences the shape of mitigation pathways critically (Krey et al., 2014a; Holz et al., 2018b; Kriegler et al., 2018a; Strefler et al., 2018b) <sup>[[#fn:r287|287]]</sup> . Deeper near-term emissions reductions are required to reach the 1.5°C–2°C target range if CDR availability is constrained. As a result, the least-cost benchmark pathways to derive GHG emissions gap estimates (UNEP, 2017) <sup>[[#fn:r288|288]]</sup> are dependent on assumptions about CDR availability. Using GHG benchmarks in climate policy makes implicit assumptions about CDR availability (Fuss et al., 2014; van Vuuren et al., 2017a) <sup>[[#fn:r289|289]]</sup> . At the same time, the literature also shows that rapid and stringent mitigation as well as large-scale CDR deployment occur simultaneously in 1.5°C pathways due to the tight remaining carbon budget (Luderer et al., 2018) <sup>[[#fn:r290|290]]</sup> . Thus, an emissions gap is identified even for high CDR availability (Strefler et al., 2018b) <sup>[[#fn:r291|291]]</sup> , contradicting a wait-and-see approach. There are significant trade-offs between near-term action, overshoot and reliance on CDR deployment in the long-term which are assessed in Section 2.3.5.

<div id="section-2-3-4-1-block-6" class="box"></div>

<span id="box-2.1-bioenergy-and-beccs-deployment-in-integrated-assessment-modelling"></span>