Abstract. Fast emulators of comprehensive climate models are often used to explore the impact of anthropogenic emissions on future climate. A new approach to emulators is introduced that generates means and covariances of monthly averaged climate variables as a function of global mean surface temperature. The emulator is trained with output from a state-of-the-art climate model and serves as a good first-order representation for the evolution of spatially resolved climate variables and their variability. To train the emulator, data is first projected into a reduced-dimensional space; the emulator then learns the dependence of climate variables on global mean surface temperature in the projected space. To recover climate variables in physical space, an inverse transformation is applied. The resulting emulator can cheaply generate means and variances of climate fields averaged over arbitrarily defined regions and in previously unseen warming scenarios. For illustrative purposes, the emulator is applied to predict changes in the mean and variability of monthly values of both surface temperature and relative humidity as a function of global mean surface temperature changes. However, the approach can be applied to any other variable of interest on yearly, monthly or daily timescales.
Geogdzhayev et al. (Thu,) studied this question.