Champaign, IL—In a plain office building on the edge of the cornfields in this college town, Xin-Zhong Liang spends his time trying to hash out equations to illustrate monsoons, rivers and nighttime rainfall.
Success is not easy. At the moment, the tools used to predict what climate will be like in the future can only guess at the impact of things like glaciers, oceans and hurricanes. Big things, on a big scale. But if Liang, son of a Chinese farmer, gets it right, he might be able to tell farmers in Illinois what crops will grow well on their land 10, 20, even 100 years from now when the climate may be much warmer.
A difference of 2 degrees in the earth’s average surface temperature could change crop yields the world over – boosting growing seasons in higher latitudes and narrowing them in others. But adding in changes in rainfall levels – not to mention whether it falls in torrents or gentle drops – complicates any projection of where food will come from in the future and what it will look like. And then there’s the question of how plants will do in an atmosphere with higher carbon dioxide.
“All of these [factors] are in one box, so how can we aggregate their effects and feed it back into the atmosphere?” asked Liang, hopping up from his desk to excitedly scrawl some lines and points on a nearby dry-erase board.
It’s a remarkable prospect being studied on macro and micro levels the world over—from Europe to Champaign—as it could dramatically shift the way farming is done and shift the landscape in terms of which crops feed the planet.
Global climate change models currently try to make predictions by averaging things like temperature and moisture from points that may be as much as 200 miles apart.
But Liang would like to see it done differently, so that the models draw on points that are closer—say 20 miles apart—and have a better chance of predicting the impacts of such small things as clouds.
“Some people call it a no man’s land because we still don’t know all the physics to put in [at the local level],” Liang said.
Liang started out as a global climate change modeler – a numbers guru and physics expert who tries to translate the entire earth environment into codes a computer can read, and advice people can follow. Over the last nine years, though, he has switched over to what is called regional climate change modeling, which involves many of the same core skills but looks to a smaller field.
“It’s very much like engineering,” he said. “You’re doing science, and then you translate your knowledge, your physical understanding into building something.”
First, Liang said, you build the basics – such as a foundation and basement of a house – but once you get to a certain point you can start adding in things like sprinkler systems.
By zooming in on North America, for example, regional models can potentially offer a more detailed picture of water systems—rivers, lakes, even streams— or of the impact global warming could have on specific regions. Because they have a higher resolution than global models, with more points to get information from in a given area, they have the capacity to see things that the global models overlook.
“There could be a mountain there — everything, a forest — and the [global] model doesn’t know, so the accuracy [of global models] has quite large uncertainties,” Liang said.
To fill in the climate picture, Liang and other regional modelers dial up the resolution and pour more physics into climate models. And what’s starting, slowly, to come into focus is more localized information about future temperatures, precipitation and weather patterns.
Getting the mountains and rainfall right in these smaller models takes time and money. Yet, until recently, few resources have been directed at regional modeling in the United States.
Print
Email
Bookmark
