Historical greenhouse gas concentrations for climate modelling (CMIP6)

Abstract

Abstract. Atmospheric greenhouse gas (GHG) concentrations are at unprecedented, record-high levels compared to the last 800 000 years. Those elevated GHG concentrations warm the planet and – partially offset by net cooling effects by aerosols – are largely responsible for the observed warming over the past 150 years. An accurate representation of GHG concentrations is hence important to understand and model recent climate change. So far, community efforts to create composite datasets of GHG concentrations with seasonal and latitudinal information have focused on marine boundary layer conditions and recent trends since the 1980s. Here, we provide consolidated datasets of historical atmospheric concentrations (mole fractions) of 43 GHGs to be used in the Climate Model Intercomparison Project – Phase 6 (CMIP6) experiments. The presented datasets are based on AGAGE and NOAA networks, firn and ice core data, and archived air data, and a large set of published studies. In contrast to previous intercomparisons, the new datasets are latitudinally resolved and include seasonality. We focus on the period 1850–2014 for historical CMIP6 runs, but data are also provided for the last 2000 years. We provide consolidated datasets in various spatiotemporal resolutions for carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O), as well as 40 other GHGs, namely 17 ozone-depleting substances, 11 hydrofluorocarbons (HFCs), 9 perfluorocarbons (PFCs), sulfur hexafluoride (SF6), nitrogen trifluoride (NF3) and sulfuryl fluoride (SO2F2). In addition, we provide three equivalence species that aggregate concentrations of GHGs other than CO2, CH4 and N2O, weighted by their radiative forcing efficiencies. For the year 1850, which is used for pre-industrial control runs, we estimate annual global-mean surface concentrations of CO2 at 284.3 ppm, CH4 at 808.2 ppb and N2O at 273.0 ppb. The data are available at https://esgf-node.llnl.gov/search/input4mips/ and http://www.climatecollege.unimelb.edu.au/cmip6. While the minimum CMIP6 recommendation is to use the global- and annual-mean time series, modelling groups can also choose our monthly and latitudinally resolved concentrations, which imply a stronger radiative forcing in the Northern Hemisphere winter (due to the latitudinal gradient and seasonality).

The PMIP4 contribution to CMIP6 – Part 3: The last millennium, scientific objective, and experimental design for the PMIP4 <i>past1000</i> simulations

“…for global mean surface mixing rations, see Fig. 9 in Meinshausen et al, 2017). The CMIP6 reconstruction offers a better representation of latitudinal and seasonal variations and we recommend using this data set for consistency throughout the CE.…”

“…1). The data compilations for surface concentrations of CO 2 , CH 4 , and N 2 O are based on updated instrumental data and ice-core records (Meinshausen et al, 2017). Differences between the new CMIP6 data set and previous estimates for CMIP5 are rather small (e.g.…”

“…GHG (CO 2 , CH 4 , N 2 O) concentrations are provided by Meinshausen et al (2017) for the CMIP6 historical experiments. This data set has been extended to cover the entire CE (1 to 2014 CE).…”

Compiled records of carbon isotopes in atmospheric CO<sub>2</sub> for historical simulations in CMIP6

“…The Coupled Climate-Carbon Cycle MIP (C4MIP) encompasses historical, future, and idealized biogeochemical simulations in both the ocean and the terrestrial biosphere, using climatic forcing from coupled Earth system models (ESMs) as opposed to observations (Jones et al, 2016). For any historical simulations in CMIP6 that use observed atmospheric greenhouse gas concentrations to drive the ESMs, compiled records of atmospheric CO 2 and other greenhouse gas concentrations are provided by Meinshausen et al (2017). Here we describe a compilation of historical data for carbon isotopes in atmospheric CO 2 to support the inclusion of carbon isotope modelling in CMIP6.…”

Model constraints on the anthropogenic carbon budget of the Arctic Ocean

“…In the air-sea flux equation, the atmospheric xCO 2 is corrected for humidity and atmospheric pressure to convert to pCO 2atm . The atmospheric xCO 2 history for 1765-1869 is from Meinshausen et al (2017), while the history for 1870 and beyond is the same as used in the NEMO-PISCES simulations.…”