Rainfall extremes are increasing around the world, and the increase is linked to the warming of the atmosphere which has taken place since pre-industrial times. This is the conclusion of a recent study which investigated extreme rainfall trends using data from 8326 weather-recording stations globally, some of which have records spanning more than a hundred years.
Of all the stations analysed, we found that two-thirds showed increasing trends over the course of the 20th and early 21st centuries. When we looked at the association between the intensity of rainfall extremes and a record of global mean near-surface atmospheric temperature, rainfall intensity was found to increase at a rate of between 5.9% and 7.7% for each degree, depending on the method of analysis.
This kind of change is precisely what can be expected if one assumes that the intensity of the most extreme rainfall events will scale with the capacity of the atmosphere to hold moisture. This is well known to increase with temperature at a rate of about 7% per degree.
Looking beyond globally averaged numbers, however, we also found distinct regional differences. The greatest increases occurred in the tropical belt; the smallest in the drier mid-latitudes where you will find most of the world’s deserts. In the higher latitudes, particularly in the northern hemisphere, the rate of change was close to the global average. Again, such changes seemed to be in quite close agreement with what global climate models say should happen as a result of global warming: a reassuring case of observations confirming theory.
The implications of this are likely to be significant for flood risk around the world. It’s true that 7% per degree doesn’t sound like much. But if we continue to follow the current trajectory of greenhouse gas emissions, we can probably expect in the order of three to five degrees of warming by the end of the 21st century. If the relationship between extreme rainfall and atmospheric temperature continues to hold, then this could mean as much as a 35% increase in extreme rainfall intensity on average globally.
What does this mean for the capacity of our infrastructure to handle current and future floods? Most flood-defence infrastructure, such as dams, levees, stormwater systems and coastal flood defences, has been designed to handle historical flood risk. If the risk of flooding increases, then such infrastructure will have increasing difficulty managing floods in the future. This would either lead to increased damage costs due to the flooding, or necessitate expensive infrastructure upgrades or resettlement of low-lying communities. Even the increase in extreme rainfall intensity observed thus far is likely to lead to substantial challenges for some existing infrastructure.
Nevertheless our analysis should not be interpreted as suggesting that the rapid increase in flood damage which has occurred over the past few decades is entirely attributable to climate change.
In fact, other changes such as deforestation, rapid urbanisation and an increase in the number of people living in flood plains are likely to account for the bulk of these changes. Furthermore, not all floods are caused by extreme rainfall events. Snow melt and storm surge also contribute to overall flood risk. Antecedent moisture – the wetness of the catchment prior to the flood-producing rainfall event – can also have a substantial influence on flood risk. In some parts of Australia this might even cause flood risk to decrease because of an expected increase in the number or severity of future droughts.
Despite all these caveats, our recent study contributes to the debate on how climate change will affect flood risk, by showing that the intensification of rainfall extremes is not just a projection made by climate models. Rather, it can already can be detected in our observational record.