Ocean warming drove 10% rise in 'extreme' rainfall from Atlantic hurricanes in 2020

The 2020 Atlantic hurricane season was the most active on record, featuring an unprecedented 30 named storms. The attribution study, published in Nature Communications, investigates the influence of climate change on extreme rainfall in the north Atlantic over the season.

Between 1850 and 2020, climate change caused a 0.6C rise in sea surface temperatures in the north Atlantic Ocean, the study finds. The authors conclude that this extra energy “intensified” storms in the region, driving an increase in “extreme” rainfall of around 10%.

“This isn’t an end of the century thing that we have to worry about. We actually have to prepare our coastlines for the changes in these storms now,” the lead author of the study tells Carbon Brief.

Separately, a rapid attribution study has analysed the succession of tropical storms and cyclones that hit Madagascar, Malawi and Mozambique over the period of six weeks in early 2022. These storms caused widespread flooding which impacted more than a million people.

The researchers find that climate change made rainfall during these events more extreme. However, the authors emphasise that they cannot quantify the exact influence of climate change – in part due to sparse historical rainfall data in the affected countries.

The Atlantic hurricane season – which runs from 1 June to 30 November – was the most active on record in 2020. The season saw 30 storms that were strong enough to be named – almost half of which intensified into “hurricanes” with sustained wind speeds above 74 miles per hour.

Newly formed cyclones typically move westwards across the Atlantic, growing in size and intensity as they pick up energy from the ocean. Many make landfall in the US and Caribbean, driving storm surges that can wreak damage on coastlines, before they divert northwards and dissipate.

Overall, tropical cyclones that formed during the 2020 Atlantic hurricane season caused at least 417 deaths and more than $51bn in damages – making it the sixth costliest season on record.

The plot below shows the observed storm tracks of hurricanes in the north Atlantic during the 2020 hurricane season. Each series of dots connected by a line indicates the path of a single hurricane. The plot also shows observed sea surface temperatures averaged across the hurricane season, where darker reds indicate higher temperatures.

“As expected, many of these storms occurred over the warm ocean temperatures of the tropical north Atlantic, where the average sea surface temperature during the hurricane season was above 27C,” the study says.

Attribution is a fast-growing field of climate science that aims to identify the “fingerprint” of global warming on the world’s weather – particularly extreme events, such as heatwaves and droughts and storms.

In recent years, scientists have conducted many studies assessing the influence of climate change on individual hurricanes. However, Prof Suzana Camargo from New York’s Columbia University – who was not involved in the study – tells Carbon Brief that this is the first attribution study to investigate a full Atlantic hurricane season.

To run attribution studies, scientists use models to compare the world as it is to a “counterfactual” world without human-caused climate change.

The authors of this study ran a series of “hindcasts” using the Community Earth System Model (CESM) to find the impact of climate change on Atlantic ocean temperatures during the 2020 hurricane season.

A hindcast is “essentially a forecast, but instead of projecting into the future, you’re going back in time and [starting from] a previous state”, Dr Kevin Reed – an associate professor at New York’s Stony Brook University and lead author on the study – tells Carbon Brief.

By comparing the counterfactual and real-world hindcasts, the authors found that the north Atlantic has warmed by 0.6C since 1850. The plot below shows the north Atlantic (dashed line) and global average (solid line) changes in temperature over 1920-2020, according to the CESM model.

Across the north Atlantic basin, 2020 temperatures ranged from 0.4-0.9C above pre-industrial temperatures, the study finds. The plot below shows the variation in global ocean temperature in 2020 compared to 1850. The red shading indicates areas that were warmer in 2020 than 1850.

For every degree that the atmosphere warms, the air is able to hold 7% more moisture – a relationship described by the Clausius-Clapeyron equation. Reed tells Carbon Brief that when the atmosphere can hold more moisture, it can also “rain out” more during storms – resulting in more intense rainfall.

Running their simulations for every three days of the 2020 hurricane season, the authors isolated the rainfall caused by storms and ignored any other rainfall during the hurricane season.

The results show that climate change caused a “subtle” increase in average storm rainfall during the 2020 north Atlantic hurricane season, the paper says, but drove more noticeable changes in “extreme” rainfall.

The study focuses on two measures of extreme rainfall. Extreme three-hourly rainfall looks at the most intense 1% of rainfall over a three hour period. Meanwhile, extreme three-day accumulated rainfall looks at the most intense 1% of rainfall measured over three days. Both are measured in inches.

The study concludes that for tropical storms – which reach minimum speeds of 40 miles per hour – climate change increased the extreme three-hourly storm rainfall rates and extreme three-day accumulated rainfall by 10% and 5%, respectively. For hurricane-strength storms – which exceed speeds of 74 miles per hour – these increases rise to 11% and 8%, respectively.

This shows that “the climate change signal is larger under more intense storms”, Reed tells Carbon Brief. He adds:

“This isn’t an end of the century thing that we have to worry about. We actually have to prepare our coastlines for the changes in these storms now.”

Dr Tom Knutson is a senior scientist at the Geophysical Fluid Dynamics Laboratory in New Jersey and was not involved in the study. He tells Carbon Brief that the study “adds to the body of scientific evidence that this change is likely already underway”.

It is “interesting” that hurricane rainfall is intensifying more rapidly than the 7% per degree rate suggested by the Clausius Clapeyron relationship, Knutson says. He adds:

“Some previous studies have suggested that such a higher rate is partly due to increased wind intensity of hurricanes with climate warming.”

Prof Haiyan Jiang – a professor of Earth and environment at Florida International University, who was also not involved in the study – tells Carbon Brief that the findings “are consistent with the current understanding of the theory that a warming climate will cause heavier rainfall in hurricanes”.

The findings line up well with her own research, which found that over the past 20 years, tropical cyclone rainfall increased by 1.3% per year, she adds.

However, she urges readers to “be cautious”, because the study relies solely on models to determine 2020 rainfall levels and does not use any observational data as a “ground truth” to verify.

Elsewhere, a separate study (pdf) finds that “extremely active tropical cyclone seasons” in the Atlantic are now twice as likely as they were in the 1980s. The authors attribute this increase to ocean warming.

Meanwhile, a separate team of scientists have run a rapid attribution study on a series of storms that hit south-east Africa earlier this year.

(The findings are yet to be published in a peer-reviewed journal. However, the methods used by the World Weather Attribution team in the analysis have been published in previous attribution studies.)

The storms, which included three cyclones, hit Madagascar, Malawi and Mozambique over the period of just six weeks in early 2022. The plot below shows the three-day accumulated rainfall for two of the storms – tropical storm Ana (left) and tropical cyclone Batsirai (right). Darker blue indicates higher levels of rainfall.

Storm Ana was “extreme”, and would happen roughly once every 50 years in today’s climate, the study finds. According to the authors, storm Ana destroyed more than 7,700 homes and flooded more than 37,00 hectares of crops in Mozambique alone. Meanwhile, in Madagascar, an estimated 500,000 people were impacted by the storm.

Storm Batsirai was less rare, ranking as a one-in-two year event, the authors find. Nevertheless, in Madagascar, the death toll from Batsirai was much higher than for storm Ana. The authors emphasised in a press conference that the consecutive storms gave communities little respite, making them increasingly vulnerable to each new storm.

Climate change likely made rainfall during these tropical storms heavier, the study concludes. However, the authors were unable to pin down the exact fingerprint of climate change.

Dr Izidine Pinto is a climate scientist from Mozambique, who worked on the rapid attribution study. He told journalists at a press conference that scarce observational data in the affected countries had limited the scope of their analysis.

For example, he explained that of the 23 weather stations in Mozambique, only four had data sufficient for use in this study – in part because many stations in the country were lost during a lengthy civil war. Meanwhile, in Madagascar and Malawi, there were no weather stations with suitable data for the study, the study notes.

“If we had enough station data, we would be confident enough to say ‘this is how much climate change has contributed to this particular event’,” Pinto told reporters.