Radiative effects of CH4, N2O, halocarbons and the foreign‐broadened H2O continuum: A GCM experiment

Abstract

The simplified exchange approximation (SEA) method for calculation of infrared radiative transfer, used for general circulation model (GCM) climate simulations at the Geophysical Fluid Dynamics Laboratory (GFDL) and other institutions, has been updated to permit inclusion of the effects of methane (CH4), nitrous oxide (N2O), halocarbons, and water‐vapor‐air molecular broadening (foreign broadening). The effects of CH4 and N2O are incorporated by interpolation of line‐by‐line (LBL) transmissivity calculations evaluated at standard species concentrations; halocarbon effects are calculated from transmissivities computed using recently measured frequency‐dependent absorption coefficients. The effects of foreign broadening are included by adoption of the “CKD” formalism for the water vapor continuum [Clough et al., 1989]. For a standard midlatitude summer profile, the change in the net infrared flux at the model tropopause due to the inclusion of present‐day concentrations of CH4 and N2O is evaluated to within ∼5% of corresponding LBL results; the change in net flux at the tropopause upon inclusion of 1 ppbv of CFC‐11, CFC‐12, CFC‐113, and HCFC‐22 is within ∼10% of the LBL results. Tropospheric heating rate changes resulting from the introduction of trace species (CH4, N2O, and halocarbons) are calculated to within ∼0.03 K/d of the LBL results. Introduction of the CKD water vapor continuum causes LBL‐computed heating rates to decrease by up to ∼0.4 K/d in the upper troposphere and to increase by up to ∼0.25 K/d in the midtroposphere; the SEA method gives changes within ∼0.05 K/d of the LBL values. The revised SEA formulation has been incorporated into the GFDL “SKYHI” GCM. Two simulations (using fixed sea surface temperatures and prescribed clouds) have been performed to determine the changes to the model climate from that of a control calculation upon inclusion of (1) the trace species and (2) the foreign‐broadened water vapor continuum. When the trace species are added, statistically significant warming (∼1 K) occurs in the annual‐mean tropical upper troposphere, while cooling (∼1.5 K) is noted in the upper stratosphere and stratopause region. The changes are generally similar to annual‐mean equilibrium calculations made using a radiative‐convective model assuming fixed dynamical heating. The effects of the CKD water vapor continuum include cooling (∼1 K) in the annual‐mean troposphere above ∼6 km, with significant warming in the lower troposphere. When effects of both trace gases and the CKD continuum are included, the annual‐mean temperature increases below ∼5 km and cools between 5 and 10 km, indicating that continuum effects dominate in determining temperature changes in the lower and middle troposphere. Above, trace gas effects dominate, resulting in warming in the tropical upper troposphere and cooling in most of the middle atmosphere. Clear‐sky outgoing longwave irradiances have been computed for observed European Centre for Medium‐Range Weather Forecasting atmospheric profiles using three versions of the SEA formulation, including the effects of (1) water vapor, carbon dioxide, and ozone; (2) the above species plus present‐day concentrations of the new trace species; (3) all of the above species plus the CKD H2O continuum. Results for all three cases are within ∼10 W/m2 of corresponding Earth Radiation Budget Experiment clear‐sky irradiance measurements. The combined effect of trace gases and the CKD continuum result in a decrease of ∼8 W/m2 in the computed irradiances.