NRC: More Data Needed for Long-term Climate Forecasting

by




SAN FRANCISCO — While incremental progress has been made in the ability of researchers to predict climate variations, further improvements will require additional ground- and space-based observations, better modeling capabilities and advancements in the way various data sets are combined, according to a new report from the U.S. National Research Council (NRC).

In the report, “Assessment of Intraseasonal to Interannual Climate Prediction and Predictability,” released Sept. 8, the NRC team cautioned that there was “no silver bullet” that would lead to sudden and significant advances in this type of climate forecasting. Instead, it said, progress will be made through the combined efforts of scientists around the world working to address the shortcomings of current prediction techniques.

The ability to predict climate conditions weeks, months or years in advance has important implications for managing agricultural lands, water resources and energy supplies, said Duane Waliser, chief Earth scientist at NASA’s Jet Propulsion Laboratory in Pasadena, Calif. In addition, these forecasts can alert communities to the potential risks of natural disasters such as floods or hurricanes, he said.

This field of research is relatively new. Unlike weather forecasts, which rely on ongoing observations of atmospheric conditions, climate forecasts are based on models representing the complex environmental processes operating in and among the Earth’s oceans, atmosphere and land surfaces. Those processes “are often incompletely understood and difficult to measure with available observational platforms,” according to the NRC report.

As recently as five years ago, for example, researchers had no capability to forecast intraseasonal climate conditions, or predictions of climate two to four weeks in advance, Waliser said. Only in the last year have scientists begun to demonstrate that capability, he said.

That progress has been the result of concerted efforts by scientists to study climate patterns and continuously improve the computer models that offer clues to future climate conditions. The increasing availability of “satellite data ingested by those models” also is assisting researchers in improving the accuracy of climate forecasts, Waliser said.

Today, the European Centre for Medium-Range Weather Forecasts (ECMWF) relies on data gathered by approximately 60 satellite-based instruments to inform its operational forecasts, compared with fewer than 30 instruments in 2005, according to data provided by ECMWF and included in the NRC report. Still, additional satellite instruments would help to improve climate forecasts, the report said.

If scientists had access to more information on the amount of water trapped in the soil, for instance, they could improve seasonal and sub-seasonal climate forecasts because soil moisture can affect air temperature, said Randal Koster, senior scientist at NASA’s Goddard Space Flight Center in Greenbelt, Md. “If we had better soil moisture measurements, we could take better advantage of that source of predictability,” he said.

While soil moisture can be measured with ground-based instruments, it is difficult to deploy enough of those instruments because moisture levels can vary dramatically even over short distances, Koster said. As a result, scientists are looking to satellite-based instruments to provide information on moisture levels over large regions, he said.

The European Space Agency’s Soil Moisture and Ocean Salinity mission launched in November 2009 completed its on-orbit checkout phase in May and uses an L-band interferometric radiometer to gather data worldwide every three days. In 2014, NASA plans to launch a complementary mission, Soil Moisture Active and Passive, with a radiometer and synthetic aperture radar to measure water levels in the soil and determine whether that water is frozen or thawed.

Data drawn from satellite instruments offer information on soil moisture levels near the ground’s surface. When the data are combined with other sources of information such as precipitation records, they can offer valuable insight into soil moisture levels deeper underground, Koster said.

Still, soil moisture is only one of the many variables that must be better understood, measured and represented in climate models to enable scientists to produce accurate forecasts. “Models can and need to be improved,” Waliser said.

To improve climate forecasts, the NRC panel called for greater collaboration between research scientists working in universities or government agencies and the organizations that issue regular forecasts, such as the U.S. National Weather Service’s National Centers for Environmental Prediction, ECMWF and the International Research Institute for Climate and Society at Columbia University. “Programs that can foster collaboration between these two types of institutions have been successful in advancing [intraseasonal and interannual] forecast quality,” according to the report.

In addition, the NRC panel suggested that organizations producing climate forecasts establish public archives showing all the data used to derive those forecasts. One of the techniques researchers employ to test climate models is hindcasting, or feeding data from a previous season or year into a model and comparing the resulting prediction with what occurred, Koster said. While this technique is extremely useful, it is only possible when researchers have access to accurate historical records of the data used to produce forecasts, Waliser said.