Indicators of Global Climate Change 2023: annual update of key indicators of the state of the climate system and human influence

ERFs were principally assessed in Chap. 7 of AR6 WGI (Forster et al., 2021), which focussed on assessing ERF from changes in atmospheric concentrations; it also supported estimates of ERF in Chap. 6 that attributed forcing to specific precursor emissions (Szopa et al., 2021) and also generated the time history of ERF shown in AR6 WGI Fig. 2.10 and discussed in Chap. 2 (Gulev et al., 2021). Only the concentration-based estimates are updated herein.

The ERF calculation follows the methodology used in AR6 WGI (Smith et al., 2021) as updated by Forster et al. (2023). For each category of forcing, a 100 000-member probabilistic Monte Carlo ensemble is sampled to span the assessed uncertainty range in each forcing. All uncertainties are reported as 5%–95% ranges and provided in square brackets. The methods are all detailed in Sect. S4.

The summary results for the anthropogenic constituents of ERF and solar irradiance in 2023 relative to 1750 are shown in Fig. 3a. In Table 3 these are summarised alongside the equivalent ERFs from AR6 (1750–2019) and last year's update of climate indicators (1750–2022). Figure 3b shows the time evolution of ERF from 1750 to 2023.

Total anthropogenic ERF increased to 2.79 [1.78 to 3.61] W m-2 in 2023 relative to 1750, compared to 2.72 [1.96 to 3.48] W m-2 for 2019 relative to 1750 in AR6. The estimate of ERF for 2023 is lower than the 2.91 [2.19 to 3.63] W m-2 in 2022 evaluated in last year's indicators. The main reason for the decline in 2023 relative to 2022 is a very strong contribution from biomass burning aerosol in 2023, particularly organic carbon emissions which strengthened the negative aerosol ERF (see also Sect. 2.3). Sulfur emissions from shipping have declined since 2020, weakening the aerosol ERF and adding around +0.1 W m-2 over 2020 to 2023 (Gettelman et al., 2024; see Sect. S4.2.2). However, the strengthened negative ERF from increased biomass burning likely dominated the effect of reduced shipping emissions. As discussed in Sect. 2, it is not easy to determine how much of the biomass burning contribution is from natural wildfires in response to 2023's anomalously warm year, which would be a climate feedback rather than a forcing. We follow the convention of CMIP and count all biomass burning emissions as anthropogenic, though this assumption may need revision in future. The approach of including all biomass burning aerosols is consistent with reporting ERF based onpage2634 the concentration increase in GHGs independent of whether CO2 and CH4 are caused by anthropogenic emissions or a smaller part is caused by any feedbacks such as from biomass burning fires or wetlands. However, changes in mineral dust and sea salt are not included in the ERF of aerosols, and any changes are interpreted as being yearly variations or related to feedbacks.

The relative uncertainty in the total ERF was at the lowest reported in 2022 (see Table 3), but with the strengthening of the aerosol ERF due to biomass additional burning, the relative uncertainty in total ERF for 2023 is higher than in 2019 reported in AR6 (Forster et al., 2021). Despite the strong aerosol forcing in 2023, decadal trends in anthropogenic ERF remain high and are over 0.6 W m-2 per decade. These are discussed further in Sect. 7.3.

The ERF from well-mixed GHGs is 3.48 [3.18 to 3.79] W m-2 for 1750–2023, of which 2.28 W m-2 is from CO2, 0.56 W m-2 from CH4, 0.22 W m-2 from N2O and 0.41 W m-2 from halogenated gases (contributions do not sum to total due to rounding). This is an increase from 3.32 [3.03 to 3.61] W m-2 for 1750–2019 in AR6. ERFs from CO2, CH4 and N2O have all increased since the AR6 WG1 assessment for 1750–2019, owing to increases in atmospheric concentrations.

The total aerosol ERF (sum of the ERF from aerosol–radiation interactions (ERFari) and aerosol–cloud interactions (ERFaci)) for 1750–2023 is -1.18 [-2.10 to -0.49] W m-2 compared to -0.98 [-1.58 to -0.40] W m-2 in Forster et al. (2023) and -1.06 [-1.71 to -0.41] W m-2 assessed for 1750–2019 in AR6 WG1. This counters a recent trend of reductions in aerosol forcing and is related in most part to 2023 being an extremely active biomass burning season. Most of this reduction is from ERFaci, which is determined to be -0.91 [-1.80 to -0.27] W m-2 in 2023 compared to -0.77 [-1.33 to -0.23] W m-2 for 1750–2022 (Forster et al., 2023) and -0.84 [-1.45 to -0.25] W m-2 in AR6 for 1750–2019. ERFari for 1750–2023 is -0.26 [-0.50 to -0.03] W m-2, stronger than the -0.21 [-0.42 to 0.00] W m-2 for 1750–2022 and the -0.22 [-0.47 to 0.04] W m-2 assessed for 1750–2019 in AR6 WG1 (Forster et al., 2021). The largest contributions to ERFari are from SO2 (primary source of sulfate aerosol; -0.24 W m-2), BC (+0.16 W m-2), OC (-0.11 W m-2) and NH3 (primary source of nitrate aerosol; -0.04 W m-2). ERFari also includes terms from CH4, N2O, VOCs and NOx which are small.

Ozone ERF is determined to be 0.51 [0.25 to 0.76] W m-2 for 1750–2023, slightly higher than the AR6 assessment of 0.47 [0.24 to 0.71] W m-2 for 1750–2019. This is due to the increase in emissions of some of its precursors (CO, VOC, CH4), but this result is highly uncertain since the preliminary OMI/MLS satellite data indicate tropospheric ozone burden is stable from 2020 to 2023 (meaning that the 2023 level does not reach the 2019 level), which could be partly due to the 2020–2023 levels of tropospheric NO2 being lower than the pre-COVID levels (OMI data from Krotkov et al., 2019). Land-use forcing and stratospheric water vapour from methane oxidation are unchanged (to two decimal places) since AR6. BC emissions increased between 2022 and 2023 and were similar to 2019 levels in 2023, resulting in ERF from light-absorbing particles on snow and ice being 0.08 [0.00 to 0.17] W m-2 for 1750–2023, similar to AR6. We determine from provisional data that aviation activity in 2023 had not yet returned to pre-COVID levels (IATA, 2024). Therefore, ERF from contrails and contrail-induced cirrus remains lower than AR6, at 0.05 [0.02 to 0.09] W m-2 in 2023 compared to 0.06 [0.02 to 0.10] W m-2 in 2019.

The headline assessment of solar ERF is unchanged, at 0.01 [-0.06 to +0.08] W m-2 from pre-industrial to the 2009–2019 solar cycle mean. Separate to the assessment of solar forcing over complete solar cycles, we provide a single-year solar ERF for 2023 of 0.08 [0.00 to +0.16] W m-2. This is higher than the single-year estimate of solar ERF for 2019 (a solar minimum) of -0.02 [-0.08 to 0.06] W m-2.

Volcanic ERF is included in the overall time series (Fig. 3b), but following IPCC convention, we do not provide a single-year estimate for 2023 given the sporadic nature of volcanoes. Alongside the time series of stratospheric aerosol optical depth derived from proxies and satellite products, for 2022 and 2023 we include the stratospheric water vapour contribution from the Hunga Tonga–Hunga Ha'apai (HTHH) eruption derived from Microwave Limb Sounder (MLS) data.

Stratospheric water vapour forcing is estimated to be +0.14 W m-2 in 2022 and +0.18 W m-2 in 2023 and in 2023 almost totally offsets the negative forcing from stratospheric aerosol.