Abstract
Using a statistical method, we examined the time of emergence (ToE), which was defined as the year when the global mean surface temperature (GMST) trend exceeds the trend explained by internal variability, using 33 Coupled Model Intercomparison Project Phase 5 (CMIP5) climate models for 1861–2100. We found that the GMST trend explained by internal variability showed minimal changes even though the GMST trend rapidly increased under the Representative Climate Pathway 8.5 scenario. However, there was large intermodel diversity for the ToEs in CMIP5 climate models. The intensity of Arctic amplification was closely associated with intermodel diversity of the ToEs. A significant difference in the amount of Arctic sea ice extent during the reference period (1861‐1910) played a role in determining whether there was slow or fast simulation of the ToE in CMIP5 climate models. We argue that a slow or fast ToE is influenced by how the climate models initially simulate the sea ice extent during the reference period.
Plain Language Summary
In our society, it is quite important to know the time of emergence (ToE), which is typically defined as the notion that a new climate state can be said to have emerged when it deviates significantly from a prior reference period, to correctly build up an adaptation and mitigation strategy in a changing climate. Although we did not use a typical definition of ToE, we found that the ToEs simulated in climate models are quite diverse. We found the intensity of Arctic amplification was closely associated with intermodel diversity of the ToEs. Our further analysis indicated that the amount of sea ice extent in the Arctic during the reference period may influence the ToEs in climate model. Therefore, it is quite important to correctly simulate the Arctic sea ice extent to correctly project the ToEs in a changing climate.
Key Points
We define the time of emergence as the year when the GMST trend exceeds the trend explained by IV in climate models
The diversity of GMST in CMIP5 climate models after the mid‐21st century could be mainly attributed to the model's characteristics
The sea ice extent during the reference period plays a key role in determining the ToE in CMIP5 climate models
