Assessing the Realism of the TIPMIP ESM idealized Experiment Protocol

The TIPMIP ESM experiment protocol employs a specified CO2-emission rate, unique to each participating model, to achieve a common linear increase in global mean surface air temperature of ~0.2K/decade for integrations started from a pre-industrial (piControl) run. This is referred to as the ramp-up phase of TIPMIP.  At different levels of global warming (GWL) into the ramp-up, models switch to zero CO2 emissions and run in this mode for 500 years. Using CO2 emissions only, it is possible to control the rate of global warming across models and also ensure that models branch into zero-emission runs at the same GWL after the same period and rate of (ramp-up) warming. While the simplicity and commonality of warming across models is a positive feature of the protocol, it is reasonable to ask how representative of real-world global warming the protocol is. We address this question by comparing simulations made with version 1.2 of the UK Earth System Model  (UKESM1.2). We compare a 4-member ensemble of UKESM1.2 following the TIPMIP ramp-up protocol (i.e. CO2 emissions only started from a piControl run) against a 4-member ensemble using full CMIP6 historical forcing started from the same piControl. We compare the two ensembles over a 40-year period that approximately covers global warming of 0.2 to 1.0K above pre-industrial values. This corresponds to the interval 1975-2015 for the historical runs, and to years 10 to 50 of the ramp-up. We compare key metrics across the two ensembles, focusing on radiation and energetics, the cryosphere, carbon cycle, and modes of variability, examining both the mean climate and temporal trends across the 40-year period. Initial results suggest the TIPMIP ramp-up compares well to the full historical runs, as well to observations, providing confidence that conclusions drawn from a multi-model assessment of the TIPMIP protocol will be relevant to the real world.