Climate change pushes precipitation — and ability to predict it — to its limits

A satellite image captures Hurricane Laura’s landfall. (Image source: NOAA)

By Sarah Anderson
Medill Reports

Primed by a drought that has lasted longer than the 1930s Dust Bowl, wildfires scorched over 5 million acres of land in the western United States this year. The water level of Lake Powell is dipping dangerously low amid the severe dry spell, threatening the hydroelectric power it generates.

On the other side of the country, a record-breaking 2020 hurricane season produced so many tropical storms that the list of 21 names for them had to be supplemented with Greek alphabet letters for just the second time in history. Hurricane Laura alone caused 42 deaths and almost $20 billion in flood damage.

Climate change is pushing precipitation to both extremes. The elevated annual mean temperature in the United States is accompanied by higher levels of rain and snow in the eastern, southeastern and midwestern regions of the country and decreases in precipitation in the western and southwestern areas, according to the National Oceanic and Atmospheric Administration (NOAA).

“The basic result is that we (will) see more floods and more droughts,” said Richard Alley, a climate scientist and professor of geosciences at Pennsylvania State University, during the annual 2021 Comer Climate Conference, held virtually this year. “You get more intense events in a warmer world.”

As the annual mean temperature increases across the United States, precipitation concentrates in certain regions. (Image source: NOAA)

Climate change has heightened the need to predict and prepare for extreme precipitation events, but it has also impaired the ability to do so. According to NOAA, the United States’ average seasonal precipitation skill — the accuracy in forecasting the amount of precipitation over an upcoming season in a specific region — has declined in recent years.

Precipitation skill varies for different areas, in part because naturally occurring cycles in ocean temperature have a well-documented effect on precipitation trends. El Niño, a climate phenomenon resulting from a warming of the central and eastern Pacific Ocean that occurs every three to seven years on average, has a consistent impact on rainfall in the places along its path each time it strikes.

Seasonal precipitation skill is “actually pretty good in places that are affected by El Niño because we can predict El Niño, and El Niño statistically affects weather patterns,” said David Battisti, a professor of atmospheric sciences at the University of Washington.

The U.S. seasonal precipitation skill has deteriorated since 2013. (Image source: NOAA)

Elsewhere, however, technology has struggled to keep up with new variables as the climate evolves. The impact of climate change on precipitation — and precipitation skill — can be largely traced to the ocean, which stores approximately 93% of excess heat driven by emissions from fossil fuel use, according to the Intergovernmental Panel on Climate Change.

“If you heat the planet, you get more evaporation from the ocean, just like if you heat a pan on your stove, you start to get steam coming off of it,” Battisti said. The additional water vapor in the air means when conditions are right for rain, it rains harder.

As the relationship between rising ocean temperatures and precipitation became clear, the U.S. National Weather Service realized it needed to incorporate ocean processes into its forecasts, said Joellen Russell, a climate scientist and oceanographer at the University of Arizona, at the Comer Climate Conference. However, rather than developing a new climate model that integrated ocean from the start, the ocean element was tacked onto the existing atmosphere component.

“They coupled it, but it wasn’t really designed to be coupled,” Russell said, resulting in a weather simulation that cannot account for the full network of interactions between ocean and atmosphere. It more closely resembles a layer of red painted over a layer of blue than a pointillist blending of red and blue dots that appears purple.

The accuracy of U.S. precipitation prediction tools is also limited by the lack of information on ocean conditions due to shortcomings in operational oceanography, Russell said. For example, orbital satellites often sample from confined and redundant areas, collecting spotty real-time data to feed into the model.

While some scientists focus on how to gather and use current data to further improve seasonal precipitation skill, other researchers are looking to the past to predict the long-term impacts of climate change on precipitation in specific regions. Dylan Parmenter, a Ph.D. student in the Department of Earth and Environmental Sciences at the University of Minnesota, presented his work studying the chemical composition of stalagmites to reconstruct rainfall patterns in the Amazon at the Comer Climate Conference.

Dylan Parmenter collects a cross section of stalagmite from a cave, which shows tree ring-like features corresponding to different time points. (Images courtesy of Dylan Parmenter)

Stalagmites form when water vapor carried from the ocean combines with the soil at a cave site and drips onto the floor of the cave. In this way, “stalagmites are kind of like fossilized precipitation,” Parmenter said. The oxygen atoms in the water vapor exist as two versions — one heavier and one lighter. If it rains as the water vapor cloud travels, the heavier oxygen atoms fall to the ground with the rain before reaching the cave. By measuring the ratio of the two types of oxygen in a sample of stalagmite, Parmenter can estimate the precipitation conditions when the stalagmite was created. To determine the age of the sample, he measures to what extent uranium atoms in the stalagmite have decayed, expelling some of their subatomic particles.

These ancient rainfall records can help scientists understand the effect of past climate patterns on precipitation, providing a useful reference point when predicting the impact of modern climate change.

“If we want to know how precipitation in a certain region is going to change from climate change, there’s nothing to compare that to,” Parmenter said. “Our work is going back and saying: With these natural climate change processes in the past, how did it react?”

While Russell acknowledges there is no time in history perfectly representative of the current climate conditions, the fundamental insight gained from research like Parmenter’s will be critical in informing responses to climate change.

“We are in the undiscovered country of the future, as Shakespeare put it. So, no, it’s not going to be exactly the same as any previous (time),” she said. “But being able to look at how these mechanisms have worked in the past can help us accelerate the learning that is going to be required to do the preventing, the mitigating and the adapting.”

Sarah Anderson is a health, environment and science reporter at Medill and a Ph.D. chemist.  Follow her on Twitter @seanderson63.