Flooding in Bad Neuenahr, Germany, 15 July FERDINAND MERZBACH/NEWS5/AFP via Getty Images
New research suggests that record-breaking climate events such as Canada's record temperature of almost 5C, which was exceeded last month by Canada, will become more common in the future. This comes after a string of extreme weather events across the globe have questioned the ability of climate models predict such extremes.
Erich Fischer, ETH Zurich, Switzerland and his colleagues used computer models to simulate the maximum temperature for the hottest week in North America and Europe. This was to determine if temperatures could break records. Under certain emissions scenarios, they could. Records were broken by more than one degree by 2030, and not the 0.1C and 0.2C normally predicted.
Researchers conclude that such record-breaking events are largely due to Earth's rapid warming. This is not just because of the increase in temperature, which has been 1.1C thus far. Fischer says it is the pace of change.
Geert Jan Van Oldenborgh, Royal Netherlands Meteorological Institute's meteorologist, said it was worrying that some statistical models suggested that records for North America were impossible. These models use a theoretical distribution for extreme values that gives an upper limit on temperatures within an area. This limit is usually a smooth increase in line with climate changes.
This heatwave then came, and it was well above the regional upper limit. Van Oldenborgh says it is quite surprising and disturbing that our theory of heatwave behavior was so distorted.
Climate scientists have been rattled by the heatwave but it is not the only thing that has happened recently. Germany was hit by deadly floods, while Henan in China saw its most severe rainfall in a millennium. People were killed in flooded subways. It shocked me," Tim Palmer from the University of Oxford says.
Continue reading: Climate change makes extreme flooding more likely in Germany and Belgium
What about future events? According to Peter Stott, UK Met Office, climate models have been able to predict large-scale changes from climate change at a broad scale. The global average temperature increase, as well as the rise in extreme temperatures. This has been clearly indicated and it is actually what's happening.
Stott says that older models didn't capture the intensity of extremes such as those in Canada. Stott says that new climate models provide more spatial detail than weather models. They can predict local extremes better, with a grid of boxes measuring 2 km across. Modelers are also improving their understanding of the mechanisms behind intense, short-term rainfalls like those seen in China and Germany.
Higher resolution models require more computing power. But, resolution is not the only problem for projecting extremes. Another important issue is timescales. Many climate models work on centennial timescales. However, some scientists now use decadal forecasts to predict what the next few years will look like. They have been proven to be able to predict Atlantic hurricanes.
These decadal predictions are definitely on the rise. These predictions are not meant to predict what climate change will do but rather what climate change is doing right now, according to Ted Shepherd, University of Reading, UK.
Continue reading: Climate change makes North American heatwaves 150 times more probable
Although many modellers agree that greater computing power is not a panacea for projecting extremes accurately, it can help. For example, computing power is required to calculate complex Navier-Stokes equations numbers. This can be used for modeling motion in the atmosphere.
Palmer says that more processing power would yield more precise figures. It all boils down to computing. Palmer has called for a CERN on climate change. This supercomputing project could run for 200 million dollars per year. Although it has not yet been realized, there are several initiatives that could help climate models. One example is a EU-funded project to create a digital twin for Earth.
It is important to remember that climate models are constantly improving, according to Tim Osborn, University of East Anglia, UK. It is possible that models cannot simulate records such as those for North America's heat, because they fail to pick up complex combinations of processes like the interaction between clear sky, low soil moisture, and wind direction. But, we don't know this yet.
Improved climate models are essential for adaptation to climate change, and early warning systems that can prevent deaths. It is not as though they are required to mitigate the effects of climate change. Shepherd says that I don't think it is the models. People are taking action not only because of climate change, but also for other reasons. They are not putting their heads in the sand. It is hard to imagine things that have not happened.
Register for the Countdown to CO26 newsletter, which covers this important year in climate policy