When it Rains, It Pours: Global Warming and the Rising Frequency of Extreme Precipitation in the United States
Scientists expect that global warming will cause a variety of changes to precipitation patterns in the United States. Many areas will receive increased amounts of rain and snow over the course of a year; some areas will receive less. But scientists expect that, all across the country, the rainstorms and snowstorms that do occur will be more intense – increasing the risk of flooding and other impacts.
In this report, we evaluate trends in the frequency of storms with extreme levels of rainfall or snowfall across the contiguous United States over the last 60 years. We analyze daily precipitation records spanning from 1948 through 2006 at more than 3,000 weather stations in 48 states. We then examine patterns in the timing of heavy precipitation relative to the local climate at each weather station.
We find that storms with extreme amounts of rain or snowfall are happening more often across most of America, consistent with the predicted impact of global warming.
Scientists expect global warming to increase the frequency of heavy precipitation.
As the earth warms, temperate regions of North America will face a growing risk of storms with extreme levels of rain or snowfall.
Global warming increases the intensity of precipitation in two key ways. First, by increasing the temperature of the land and the oceans, global warming causes water to evaporate faster. Second, by increasing air temperature, global warming enables the atmosphere to hold more water vapor. These factors combine to make clouds richer with moisture, making heavy downpours or snowstorms more likely.
The consequences of increasingly intense rainstorms may include flooding, crop damage, pollution of waterways with runoff, erosion, and other environmental and economic damage. During the 20th century, floods caused more property damage and loss of life than any other natural disaster in the United States.
An increase in the number of downpours does not necessarily mean more water will be available.
Scientists expect that extreme downpours will punctuate longer periods of relative dryness, increasing the risk of drought. In the Southwest, for example, total annual precipitation is projected to decline – amplifying the impact of periods of little rainfall between heavy storms.
Even in the rest of the country, where total annual precipitation is expected to increase, more of that precipitation will fall in heavy rainstorms or snowstorms, paradoxically increasing the potential for drought.
As temperatures rise, precipitation will become increasingly likely to fall as rain rather than snow, increasing runoff and likely reducing water supplies in areas dependent on snowpack.
Weather records show that storms with extreme precipitation have become more frequent over the last 60 years.
Consistent with the predicted impacts of global warming, we found that storms with extreme precipitation have increased in frequency by 24 percent across the continental United States since 1948. (According to a statistical analysis of the data, with 95 percent confidence, the increase has been between 22 and 26 percent.)
New England and the Mid-Atlantic experienced the largest increase in extreme precipitation frequency.
New England and the Mid-Atlantic saw storms with extreme precipitation levels increase in frequency by 61 percent and 42 percent, respectively. At the state level, Rhode Island, New Hampshire, Massachusetts, Vermont, New York and Louisiana all saw extreme precipitation events increase in frequency by more than 50 percent.
In the contiguous United States, 40 states experienced a statistically significant trend toward increasingly frequent storms with extreme precipitation. Only one state (Oregon) showed a statistically significant decline in frequency of storms with extreme precipitation. (See Figure ES-1.)
See the report appendices on page 35 for a full list of results by region, state and metropolitan area.
These findings are consistent with previous studies of extreme precipitation patterns, both in the United States and across the globe.
Climate divisions covering more than half of the land area of the United States show a statistically significant trend toward more frequent storms with extreme precipitation.
We also looked at the trend in frequency of storms with extreme precipitation within climate divisions, which are boundaries used by climatologists since the 1950s to aggregate weather observations. Figure ES-2 presents these trends, showing that the largest increases occurred across New England, New York, much of the Great Lakes area, the upper Midwest, plus Louisiana, New Mexico, northern Washington and southern California.
Climate regions covering more than half of the surface area of the contiguous United States show a statistically significant increase in the frequency of storms with extreme precipitation levels.
In contrast, the data show statistically significant decreases in extreme precipitation frequency for climate regions covering only 4 percent of the area of the United States. (Oregon, the northwestern corner of North Dakota, central Arkansas, the southern tip of Lake Michigan, and northern Florida.)
These findings are consistent with previous studies of extreme precipitation patterns, both in the United States and across the globe. For example:
Scientists have observed warmer weather, higher atmospheric moisture content, increased formation of storm clouds, and an increase in thunderstorm activity over the contiguous United States in recent decades.
In 1999, researchers at the Illinois State Water Survey and the National Climatic Data Center (NCDC) found that storms with extreme precipitation became more frequent by about 3 percent per decade from 1931 to 1996. Our findings are consistent with this result.
In 2004, scientists at NCDC concluded that most of the observed increase in storms with heavy and very heavy precipitation levels since the early 1900s had occurred in the last three decades. In other words, they found that the change in extreme precipitation frequency is unusual and relatively recent.
Moreover, NCDC found that extremely heavy storms are increasing in frequency more rapidly than very heavy storms – which in turn are increasing in frequency more rapidly than heavy storms.
The severity of the trend toward more intense downpours in the future depends upon our emissions of the pollution that drives global warming.
Climate models predict that the trend toward increasingly frequent storms with heavy precipitation will intensify in the future. Some amount of change is inevitable given the global warming emissions humans have already created. However, we still have the ability to prevent the worst-case scenarios.
By halting the increase in total U.S. global warming emissions now and reducing emissions by at least 80 percent by mid-century, we can limit the increase in major storm frequency — and thus reduce future risks of flooding and other serious consequences of extreme rainstorms.
To address global warming, America should limit emissions of global warming pollution, while improving energy efficiency and increasing the use of renewable energy.
To protect future generations, the United States should adopt a mandatory cap on global warming pollution that reduces total U.S. emissions by at least 15 to 20 percent by 2020 and by at least 80 percent by 2050.
If policymakers choose a cap-and-trade program to achieve this goal, it should include auctioning 100 percent of emission allowances, rather than giving allowances away to polluters. By auctioning allowances, we can reduce the cost of achieving emission reduction goals, making it more likely that America will succeed.
The United States should also adopt complementary policies to improve energy efficiency and increase the use of clean, renewable energy.
How We Obtained Our Results
In this report, we examine trends in the frequency of extreme precipitation across the contiguous United States from 1948 through 2006. We analyze daily precipitation records obtained from the National Climatic Data Center for more than 3,000 weather stations, identifying storms with extreme 24-hour precipitation totals. We define extreme precipitation relative to the local climate, selecting storms with an average recurrence interval of 1 year or more. In practical terms, this means that we selected the 59 largest storms in terms of total precipitation at each weather station during the 59-year period of analysis, and labeled these “extreme.” We then examined trends in the frequency of these storms over time. For a more detailed explanation, see the “Methodology” section on page 32.