The first time that the residents of the Cumbrian village of Glenridding were flooded in 2015, they reassured themselves that this was an extreme event – one of those freak occurrences. But when storm Desmond, swiftly followed by storm Eva, resulted in the village being awash for the third time in less than four weeks, it seemed ludicrous to still be labeling these raging floods as rare events. So what is going on? Are extreme climate events becoming more frequent, or were the residents of Glenridding suffering a series of unlucky rolls of the dice?
Rising levels of greenhouse gases result in a warmer atmosphere. Warmer air can hold more water vapour, causing heavier rainfall events. At the same time we’d expect to see an increase in extreme flood events, but this isn’t always the case because floods also depend on other factors, such as whether the soil is saturated before rains fall. The floods we are most interested in are the extreme ones because they cause the most damage and disruption. Such floods can be identified from the large peaks they produce in flow rate in rivers. But because they are rare events it is difficult to find long enough historical records to demonstrate any trends.
To get around this problem, Wouter Berghuijs from ETH Zurich, Switzerland, and colleagues took a regional approach, aggregating data over many locations, in order to provide robust information on the changing nature of extremes across a large number of catchments.
The team analysed daily streamflow observations for the period 1980 to 2009, taken from 309 catchments in eastern Australia, 671 catchments in the continental US, 244 catchments in Brazil and 520 catchments located across Europe. For each catchment, the scientists searched for the largest daily flow rate events during this thirty-year period, in other words, a flood that has a 3.3% chance of occurring in any given year. They then split the data into two time periods – 1980 to 1994 and 1995 to 2009 – and compared the total number of these extreme events in each period.
The results show that both the frequency and magnitude of these extreme flood events has increased, with the total number of extreme floods increasing by an average of 26.6% for the latter time period. The increases were greatest in the northern hemisphere, with European catchments experiencing a 44.4% increase in extreme floods and 21.4% for the US. The changes have been less dramatic in the southern hemisphere, with an increase of 14% for Brazil and 11.6% for Australia.
Understanding the reasons for these changes is still difficult. It’s likely to be a combination of long-term effects such as climate change combined with short-term weather variability. “For example, Australia has suffered a big drought, which reduced the number of floods observed during the early 2000s,” said Berghuijs. “The result was that floods did not increase so strongly in Australia, but that does not mean it is sheltered from climate change.”
By taking this regional approach, the scientists have been able to quantify changes in these rare flood events and show that they have indeed increased as our climate has warmed. Currently flood protection tends to be based on local historical records, but this can be misleading, as the residents of Glenridding found to their cost. The new approach, reported in Environmental Research Letters (ERL) helps scientists and policymakers to understand if flood protection based on standards set in the past is sufficient for present-day conditions.