FEATURES:      

Responding to Earth's Changing Climate     
A Model of Success     
Pieces in a Global Picture     

A Model of Success

Read any daily paper, listen to any radio news or watch any Weather Channel update and the weather report is likely to cover not just today's forecast, but in most cases, the next five to seven days. The long-range forecast has simplified our lives in many ways, thanks largely to the pioneering efforts of Distinguished University Professor Eugenia Kalnay, whose work has not only improved the accuracy of weather forecasts but has broadened our understanding of climate changes as well.

Story by Nancy Grund
Photography by John T. Consoli

"Today, a four-day forecast is as good as a two-day forecast was 15 years ago. Our skill at weather forecasting seems to double every decade or two," says Kalnay, a professor in and former chair of the Department of Meteorology.

What Kalnay doesn't say is that much of this improvement is due to her skill as a scientist and manager.

Before coming to the university, Kalnay directed the Environmental Modeling Center of the National Centers for Environmental Prediction where all computer models for the National Weather Service and private forecasts are generated. Kalnay, along with National Weather Service colleague Zoltan Toth, conceived a new approach called "ensemble forecasting," which is now the standard mode of forecasting around the world.

Ensemble forecasting combines several forecasts into one via a weighted average to produce a more reliable forecast over an extended period of time. With this approach, an initial analysis determines a particular set of atmospheric conditions, which are placed into a control model. Then the conditions are slightly changed or "perturbed" for each forecast. The most widely used method for generating perturbations is the "breeding of growing modes" developed by Kalnay and Toth. This BGM method uncovers and follows the growing errors that make the largest impact on a forecast.

Kalnay describes how improvements in forecasting have occurred. "Today's computer forecasts are based on atmospheric models that, given the present observations of temperature, winds and moisture, simulate how the atmosphere will evolve. The growing sophistication of atmospheric modeling and enhanced balloon and satellite observations used to estimate the initial state of the model have contributed to a new science known as 'data assimilation' and are responsible for the improvements," she explains.

"The development of these computer models is one of the most striking scientific achievements of this century, contributing enormously to the world's economic well-being," Kalnay says.

While at the National Weather Service, Kalnay also directed the team that carried out the National Centers for Environmental Prediction/National Center for Atmospheric Research 50-Year Reanalysis. According to Kalnay, the paper documenting that project is the most cited research paper in the past decade in the earth sciences field.

"We created 50 years of data showing the state of the atmosphere and the weather for the whole globe," offers Kalnay. "The reanalysis project made this information easily accessible to atmospheric scientists.

"As a result, an enormous number of research projects on weather and climate became feasible for many people throughout the world," she adds. "In a smaller way, it's like making the genome of the atmosphere available to researchers."

Analyzing Climate Change

It was the reanalysis project that raised Kalnay's interest in the factors contributing to global warming. Former Maryland researcher Ming Cai recalls how Kalnay approached him with preliminary data. "She showed small differences in temperatures from surface stations and from the reanalysis through the 1960s and the growing temperature differences more recently," remembers Cai. "We were both interested in studying if these changes could be due to urban impact."

Kalnay and Cai employed computer modeling to analyze climate changes in the United States and their findings have sparked controversy in the scientific community and among special interest groups. Their study, published last spring in Nature, suggests that rising temperatures in the United States over the past 50 years are due, in large part, to changes in how we use land, particularly the growth of cities and the spread of agriculture. Previously, many scientists attributed the global warming trend largely to increased greenhouse gases resulting from power plants, cars and other human activities.

"The work that Cai and I have done has obviously touched on an important problem in the world right now," says Kalnay. Still, she is surprised by the attention her work has garnered, such as its inclusion in Discover magazine's top 100 science stories of 2003.

"Global warming is a real fact, it is happening," says Cai, now an associate professor at Florida State University. "With a computer model, we can investigate what options are available to reduce the local impact. We can simulate the impact and look at what we can do to change or reduce the land use effect."

Kalnay and Cai estimated the impact of urbanization and other land uses on climate change by comparing two sets of temperature measurements for the continental United States. Trends were observed at 1,982 surface weather stations along with data collected from satellites and weather balloons used for the reanalysis project.

Surface stations registered warmer temperatures than the satellite and balloon instruments over the last five decades. The results indicate that land use changes in the United States in the last 40 years have contributed to an increase in the mean surface temperature of more than 0.2 degrees Fahrenheit, twice as high as previous estimates.

The study also showed the daily difference between the daytime maximum temperature and the nightly minimum temperature has shrunk since the 1950s. While part of this is due to greenhouse warming, Cai and Kalnay propose half of the effect is caused by urbanization and other land-use changes.

"Previous methods for estimating the impact of land use change relied on measures--population counts or satellite measures of light at night--that provide an indication of the effects of urbanization but not of other changes in land use," reports Kalnay. "Our method covers all changes in land use."

Records show that the earth has warmed by about one degree Fahrenheit during the past century and scientists expect the average global temperature to increase an additional 2-to-6 degrees over the next century. Many scientists believe that even the smallest changes in average temperature over a period of years can cause dramatic changes in the climate.

Kalnay is clear that their recent research by no means diminishes the importance of the greenhouse effect over the last several decades. The study does, she says, "suggest that we have to pay more attention and do whatever is possible to decrease major modifications in land surfaces."

Kalnay and Cai are now extending their U.S. study of climate change to the rest of the world, beginning with Kalnay's native Argentina and Cai's native China. Already, early results from South America support their conclusions.

"If our interpretation is correct, the impact may be even more prominent in other countries, such as China, where most major land development occurred after the 1980s," notes Cai. "The use of different data sets to achieve the same results would further strengthen our arguments."

An Unlikely Career Path

Given Kalnay's wide-ranging contributions to meteorology, her entry into the field followed an unlikely path. It began during the summer break of her freshman year in college, when, unbeknownst to Kalnay, her widowed mother changed Kalnay's major from physics to meteorology. Meteorology scholarships, available from Argentina's National Weather Service, were easier to come by than physics scholarships. Kalnay acquiesced to the change and the rest is history.

At the University of Buenos Aires, a surprising 40 percent of science students were women and Kalnay expected the same when she came to the United States to study. "Even at that time, most of the meteorologists in Argentina were women. It was a culture shock to find America was really behind in women's liberation," she says.

Kalnay battled the odds to become the first woman to receive a doctoral degree in meteorology from the Massachusetts Institute of Technology. Following faculty appointments at the University of Montevideo, Uruguay, and MIT, she worked at NASA/Goddard Space Flight Center for eight years.

It was at Goddard that she learned to balance the demands of a scientist with those of a manager, when she was selected as branch head, succeeding a manager with a rather authoritarian style. "I knew I couldn't imitate him, so I copied a style employed by a Quaker woman who had managed a human rights group in Boston. She had a talent for listening to everyone and she managed by consensus," relates Kalnay. "I wasn't sure whether the style would ever be successful with strong, opinionated male scientists, but it was amazing how well it worked."

Looking back on all her achievements, Kalnay considers her role as a good mentor and manager among her most significant accomplishments. She compares her management style to the subtle movements on a Ouija board. "Early in my life, my mother forced me to take piano lessons and I really had no talent for it," she recalls. "At one point, I [with several others] posed the question to the Ouija board, 'Should I take piano?' planning to force the answer to 'no,' " admits Kalnay. "I didn't even have to push the marker." Just the slight nudging on my part made the rest of the group move the marker slowly in the direction I wanted."

Kalnay is confident that "if you can manage like that, gently pushing people along and moving them in your direction in a subtle way, you can be successful."

As a professor, Kalnay is doing her part to help young women students who are pursuing the sciences. She has mentored undergraduates through the university's award-winning Research Internships in Science and Engineering, or RISE, program. And she continues to work with meteorology graduate students on a variety of major research challenges.

Kalnay is pursuing further improvements in weather and climate forecasting with another Distinguished University Professor, James Yorke, a professor of math and physics who last year won the Japan Prize for his seminal work in chaos research. Kalnay and Yorke are extending accurate weather forecasts to two weeks and improving the ability of models to predict weather in the shortest scales in space and time, such as predicting that a severe storm will hit a specific location at a specific time. They also are improving monthly and seasonal estimations of climate irregularities such as El Nino.

This summer the university will host a symposium to celebrate the 50th anniversary of the development of "operational numerical weather prediction" and the pioneering developments and continuing improvements in forecast modeling that have occurred in that time. And though she is far too modest to say so herself, it is fitting that Kalnay, who has been so instrumental in many of the recent improvements, is one of the lead organizers.