“While globally averaged annual mean surface temperatures (GAMST) provide valuable information about the overall warming trend of our planet, the most devastating consequences of climate change for humans and the environment are often associated with climate extremes happening at regional scales in populated areas,” Jana Sillmann of the Center for International Climate and Environmental Research – Oslo (CICERO) told environmentalresearchweb. “We can see that overall the warmest day of the year became warmer in the recent hiatus period, in the observations as well as in the model simulations. The coldest night in a year also became warmer, except in the mid-latitudes, and even exceeds the warming observed in the hot extremes in the high northern latitudes.”

To come up with these results, Sillmann and colleagues from the University of New South Wales, Australia, Canadian Centre for Climate Modelling and Analysis, and Pacific Climate Impacts Consortium, Canada, used the HadEX2 observation-based gridded dataset of climate extreme indices, along with simulations from 27 Coupled Model Intercomparison Project Phase 5 (CMIP5) climate models. They examined the periods 1996–2010 and 1971–2010.


One of the reasons the hiatus attracted so much attention was that it led to the argument that current climate models are not able to reproduce the observed slowdown in mean temperature trends of the last 15 years.

“Many studies concentrated on explaining the potential reasons for this discrepancy,” said Sillmann. “In our study we were interested if this recent discrepancy between observed and simulated GAMST trends would also be reflected in temperature extremes.”

While climate change models have not been able to simulate the hiatus in average temperatures well, the fit between observations and models for extreme temperatures is better, the team found, except for a recent cooling in Northern mid-latitude (20°–45°N) minimum temperature extremes, particularly in South Asia.

“So far, there hasn’t been a study looking at the ability of the current climate model generation to simulate the observed trends in temperature extremes in recent decades on a global scale,” said Sillmann. “Overall we can show that temperature extremes continue to increase over most regions of the world, with no indication of a slow-down or pause in the hiatus period.”

The inability of climate models to simulate the observed cooling pattern in cold extremes across Northern Hemisphere mid-latitudes could actually help scientists understand the hiatus in global mean temperatures. Current candidates for the explanation include a decrease in stratospheric water vapour, an increase in atmospheric aerosol concentrations, and internal climate variability manifested via La-Niña-like decadal cooling together with a vertical redistribution of heat in the ocean.

The researchers believe their findings are “consistent with the suggestion that the recent 15-year period largely represents a highly unusual (extreme) realization of climate as part of internal variability”, as they write in Environmental Research Letters (ERL).

But the team stressed that 15 years is not a long time when it comes to analysis. “We found that, from a statistical point of view, the recent hiatus period is too short to assess the significance of trend patterns, particularly when looking at extreme events,” said Sillmann. “The analysis of…the last 40 years clearly indicates significant increasing trends in the temperature extremes in large parts of the world in both the observations and the model simulations.”

Now the researchers plan to investigate the regional cooling in cold extremes to find out more about underlying mechanisms and how to improve climate models. “With continuous international efforts and research being spent to further improve the observational data basis for climate extremes in regions that are not yet well covered (e.g. Africa, South America), we will be able to extend the evaluation of models to a truly global scale,” said Sillmann.

Related links

Related stories