A pathway to assess chemistry-climate models' tropospheric ozone radiative forcing from satellites

An instantaneous radiative kernel (IRK) in 9.6-m m ozone band from satellite represents the sensitivity of TOA fluxes to the vertical distributions of the geophysical variables, such as ozone (O₃), water vapor (H₂O), air temperature (T_a), and surface temperature (T_s). It has been used as a powerful tool to quantify the radiative effect due to these key variables. In this presentation, we first show that, applying these kernels in clear sky to the biases in present-day simulations from a suite of models in the Chemistry-Climate Model Initiative (CCMI) project and the Atmospheric Chemistry Climate Model Inter-comparison Project (ACCMIP), as compared to reanalysis assimilating satellite observations, the models have large differences in TOA flux, attributable to different geophysical variables. Model simulations diverge most from observations in the tropics. The primary drivers are tropical mid and upper tropospheric ozone, followed by tropical lower tropospheric H₂O. On the other hand, the T_aand T_s' radiative bias are negligible components in all models. Overall, the multi-model ensemble mean for the total radiative bias is -132.9±98 mWm^-2, indicating that they are too atmospherically opaque, thereby reducing the sensitivity of TOA flux to ozone and potentially underestimate ozone radiative forcing. We also find that the inter-model TOA OLR difference is well anti-correlated with their ozone band flux bias. This suggests that there is a significant radiative compensation in the calculation of model outgoing longwave radiation.

The further study find the satellite observed relative humidity is strongly correlated with the O₃IRK distributions, especially over the ITCZ. This suggests the O₃greenhouse effect is not only controlled by O₃distribution but also is effected by the changes in hydrological cycle due to climate change (Kuai et al., 2017).