EARTH SCIENCES

In an attempt to measure snowmelt in a mountainous region northwest of Boulder, Colo., two University of Wyoming students are determining surface shade through a computer model they’ve created. They hope their supercomputing research eventually can be scaled up for use on Mount Moran, UW’s high-performance computing cluster, and be a useful tool for modeling solar radiation for the entire Upper Colorado River Basin.

Using the Department of Mathematics’ supercomputing lab, Troy Axthelm and Jingyu Li have spent time -- during the past 20 weeks -- creating a surface mapping program that is used to model shading projections over a small mountain range (15 kilometers by 15 kilometers) northwest of Boulder. The model, which currently uses 60 supercomputer nodes (a node is conceptually similar to a desktop computer), shows periods of shade and sun to better determine where and when snowmelt occurs.

In a time-lapse video the two UW students created on their project website, cool colors indicate shade while warm colors denote sunlight. The video starts at midnight and is set at 15-minute increments to show areas of sun and shade over any 24-hour period.  

{media load=media,id=88,width=200,align=left,display=inline}

In an effort to better model watershed in the Colorado River Basin, this program will allow for more accurate solar radiation calculations that can be used to determine snowmelt in various areas of the basin.

“We have divided the model area into different plots and are running different times of day on different nodes,” says Li, a Laramie High School senior who is currently taking three courses at UW.

If an area is completely shaded at a certain time, it receives no direct solar radiation, but some indirect radiation, Axthelm says of the model, which currently uses 4 million elevation points. The points are created down to the size of a 9-square-meter grid.

“We could run it (this model) on that whole area and be fairly accurate over the Colorado River Basin -- if there was no vegetation or clouds,” says Axthelm, a senior from Cody majoring in computer science. “If all of those mountain ranges were bare land, we could do it. But shading is not an ‘on’ or ‘off’ thing. There are degrees of shading. We have a long way to go.”

Summer origins

Their research began during the summer when they assisted Fred Ogden, a UW professor in the Department of Civil and Architectural Engineering, and Craig Douglas, a UW professor of mathematics, with the CI-Water Project, which focuses on the hydrology of the Colorado River Basin. The CI-WATER project will develop a high-resolution, physics-based hydrologic model that is applicable over large areas to help assess long-term impacts of water resources management decisions, natural and man-made land-use changes, and climate variability -- with an emphasis on the Rocky Mountain west region.

After their summer experience, Axthelm and Li saw an undergraduate research opportunity in high-performance computing. They conduct their work as an independent study led by Douglas. Li has taken Douglas’s Computational Methods graduate course. 

“We’re going to incorporate this into the new model we’ve developed for the Upper Colorado River Basin,” Douglas says. “This model will help us with a number of issues we’re dealing with, such as snowmelt and humidity issues for runoff. This is a useful and key component.”

In addition, Douglas says the model the two students created can be used to examine the impact of beetle kill on trees during winter. Tree areas that spend more time in the shade, which are colder, would fare better than those that are exposed to more time in the sun, Douglas says, because cold weather kills the beetles.

“They’re bright and motivated,” Douglas says of Axthelm and Li. “Jingyu is a high school student, and she should be going to graduate school.”

Li was one of five female Wyoming high school students who were named winners of the National Center for Women and Information Technology (NCWIT) Award for Aspirations in Computing at UW’s Women in Science Conference in May.

After Axthelm graduates this semester, he plans to pursue his master’s degree in computer science at UW. During that time, he will work with Tim Kuhfuss, UW’s director, research support, to build an efficient computer extension of Mount Moran to the NCAR-Wyoming Supercomputing Center (NWSC) in Cheyenne.

Scaling up

“We’re working with one test range (now) and we want to get that built up to a scalable model,” Axthelm says. “Once we have the program built, we will model the entire Colorado Range.”

That’s when it will be time to use Mount Moran, nicknamed after a mountain peak in western Wyoming’s portion of the Tetons.

The cluster enables atmospheric and earth sciences faculty -- who will be able to use the NWSC -- to learn what to expect with their software. The cluster provides the opportunity for that group of faculty to work out issues caused by scaling up parallel algorithms from tens or hundreds of processors to thousands of processors, before moving up to tens of thousands of processors on the NWSC.

The cluster also provides a research resource for any UW research faculty member -- such as bioinformaticists, social scientists, pure mathematicians and theoretical physicists – who have a complex problem or whose research doesn’t fall within the scope of the NWSC.

In addition, UW students are welcome to use Mount Moran for their work, which is often in concert with UW faculty.

“We’re starting to max out the nodes in this lab,” Axthelm says. “We’re trying to make sure when we use Mount Moran, we use it efficiently and accurately. We’ve just been testing and testing here. We want to scale it up for Mount Moran.”

“They can now push the code to Mount Moran and get higher parallelism,” Douglas says. “I anticipate they will be doing some of that over Christmas break.”

The work has been challenging and rewarding, Li says.

“It was interesting trying to figure it out for ourselves,” she says. “We really didn’t base our work off of any other model.”

The two students hope to create a poster presentation of their research and attend a few supercomputing conferences.