Teaching old data new tricks

Fig.1. Rock core samples, pictured, stored at the U.S. Geological Survey's Core Research Center. Data derived from core samples, among other types of samples, are useful in testing climate models. (Credit: USGS, 2012)

There’s an old saying that history repeats itself. But in the case of Carrie Morrill’s research, she’s looking to history to prepare for the future.

“History doesn’t necessarily repeat itself, (but) may be some events could be relevant for the future,” she said.

Morrill, who’s a research scientist at National Oceanic and Atmospheric Administration’s (NOAA) World Data Center for Paleoclimatology, tests climate models against paleoclimatology data to improve predictions of future climate change.

“Climate models are mathematical representations of the climate system built from the laws of physics, which describe for example how fluids move and energy gets transferred. We use paleo data as a sort of ‘ground-truthing’ to make sure that the climate models work,” Morrill explained in a follow-up comment.

These models that researchers construct are “very complex,” she said. They must replicate events in the atmosphere and represent wind patterns, track cycling of water, how heat and energy get transferred and more.

“It’s important to make sure that models are giving the right information,” said Morrill, who also serves as a research associate at the University of Colorado’s Cooperative Institute for Research in Environmental Sciences.

But why?

Scientists use models to project how climate will change. For example, Morrill said models are used to see the effects of greenhouse gases that we, as a society, continue to put in and take out of the atmosphere. Getting a sense of how climate is affected by increases in greenhouse gases allows individuals to understand how the climate will change and to what degree.

But in order for Morrill and her NOAA collaborators - made up of students, post-doctorate researchers, scientists, archive specialists and others - to predict future climate actions, it has to start somewhere. And that’s where paleoclimate data come into play.

The data are derived from various avenues, such as ice and sediment cores, tree rings, corals and more. Morrill said the work she does couldn’t be done through individual samples alone. What she strives for is constructing the bigger picture. Generally, Morrill and her team often turn to scientific data that other people and teams generate. Turning to collections of cores and the accompanying data is where the answers lie.

“It’s really a task to put together all these little pieces,” she said.

But, that task is worth it. Putting together the different pieces of data from various periods - like the Holocene, which Morrill is currently researching - allow scientists to get a more global view of what climate looked like years ago while focusing on individual points in time. This, Morrill says, can unveil patterns.

“Patterns are really important for figuring out past climate change,” she said.

But while the data from these samples can reveal patterns over time, Morrill said she and her team run in to challenges while sifting through the collections’ data to build climate models. The main issue comes in the lack of data across the world.

“It’s not an even spatial coverage of sites,” Morrill said.

While some areas of the world have more documented and archived information about them, Morrill said that generally there is more data on land than oceans. Also, documented datasets of North America and Europe are longer than other areas of the world, posing issues when Morrill and her team are testing models outside of those areas. And Morrill noted, while methods are improving, there are limitations in the dating of samplings. But even with the variance in information sets, Morrill said scientists have done a great job organizing research outings and initiatives.

Morrill will continue to work on climate model and data comparison of different time periods. And she is particularly interested in not just looking at the direction of change but the quantitative differences over time - to test if models can reflect the correct magnitude of change.

“There’s a lot we can potentially learn … in terms of using paleoclimate information,” Morrill said. “...We know in the future it’s going to get warmer; the question is how much warmer?”


References
Li, Yu, Morrill, Carrie (2014). A Holocene East Asian winter monsoon record at the southern edge of the Gobi Desert and its comparison with a transient simulation. Retrieved from http://link.springer.com/article/10.1007%2Fs00382-014-2372-5#page-1.

National Oceanic and Atmospheric Administration Paleoclimatology Program. Retrieved from http://www.ncdc.noaa.gov/data-access/paleoclimatology-data/about-the-paleoclimatology-program.

National Oceanic and Atmospheric Administration Applied Research Center for Paleoclimatology. Retrieved from http://www.ncdc.noaa.gov/climate-information/research-programs/applied-research-center-arc.

Glossary

paleoclimatology
the study of past climates
Holocene Epoch
a time period between beginning 11,500 years ago, at the close of the Paleolithic Ice Age, to the present day