The most comprehensive review of changes to extreme rainfall yet has confirmed the link between extreme rainfall intensity and global atmospheric temperature, with annual rainfall highs intensifying by around 7% per 1°C warming. University of Adelaide’s Seth Westra and his teammates called on 8326 weather stations across the Earth’s surface with over 30 years’ data from the period between 1900 and 2009. “The real innovation in this study is the strength of the data set we used,” Seth told me.
As a flood engineer, Seth has conducted numerous flood risk studies using highest daily rainfall amounts over a year period for planning. But when trying to understand how these rainfall amounts are changing, he found that many studies did not present their information so engineers could use it easily. He therefore wanted to do a statistical study that could produce more precise estimates. And when he found out that Lisa Alexander at the University of New South Wales, Australia was bringing together measurements of temperature and rainfall extremes, he saw his opportunity.
Lisa is part of a team that has collected data from 11391 weather stations worldwide, a much larger record than previous studies had used. That’s partly to contribute to the Intergovernmental Panel on Climate Change’s fifth assessment report on global warming, due out in stages in late 2013 and 2014. But it meant the massive task of collecting the measurements needed to do the kind of study Seth planned was already being done. “You can imagine the sheer amount of work in getting data from dozens of meteorological agencies from around the world,” he said. “Trying to feed them all into one single repository is a lot of effort.”
With this data in hand, Seth, Lisa and Francis Zwiers from the University of Victoria, Canada, first selected only the 8326 stations with over 30 years’ worth of data. They then used two statistical tests to look at the annual extreme rainfall highs, which they published online last year in a paper in the Journal of Climate. In the first, they simply looked at whether rainfall extremes increase or decrease at each station over time, Seth explained.
Rain not dropping
“We found that two-thirds were increasing, and one-third was decreasing, with a net overall increase in extremes,” he said. A larger proportion of those that were increasing passed the important scientific test of “statistical significance”, he added. “We showed that there being more stations where extreme rainfall is increasing than stations where it’s decreasing is not random chance. You can still have one region increasing while another decreases, but the overall pattern is highly significant.”
The second test in that paper looked at the link between extreme rainfall and atmospheric temperature, which are connected through moisture levels in the air. In the 19th century European physicists Benoît Clapeyron and Rudolf Clausius discovered an important physical rule that shows the atmosphere can hold more water as it gets warmer. Their rule says air can contain around 7% more water vapour for every 1°C warming. In the 21st century, scientists have suggested that extreme rainfall, because it’s linked to the moisture in the atmosphere, should rise at the same rate. And though more recent studies suggest that other factors may complicate the picture, Seth’s team still came up with figures between 5.9 and 7.7% for every 1°C warming. “We treated the data in a variety of ways, but always found it went up about 7% per degree,” he said.
Tropical areas and those nearest the poles saw the strongest increases in their annual rainfall highs. And perhaps surprisingly, the maps the researchers produced show weather stations in some areas prone to drought also seeing an increase in their annual highest rainfall. Seth pointed out that this is because this ‘extreme’ rainfall measure only considers one day, while droughts occur over longer periods. “Droughts are largely because of deficits in seasonal and annual rainfall totals, rather than the intensity of the most extreme rainfall events,” he said. “They are therefore more likely to be controlled by large-scale atmospheric circulation changes.”
And while reinforcing what’s known about extreme rainfall across the world is important, he hopes to be able to look closer. “We need to explore some of the regional variation,” he said. “It was reassuring that the results are consistent with what climate models suggest, but we don’t understand all the physical processes. Getting country scale or local scale data would be valuable in helping with that.”
Westra, S., Alexander, L., & Zwiers, F. (2012). Global increasing trends in annual maximum daily precipitation Journal of Climate DOI: 10.1175/JCLI-D-12-00502.1