Winsor Lowe, interim director of the University of Montana's Wildlife Biology Program, co-wrote a research paper published April 21 in the Proceedings of the National Academy of Sciences on how streamwater chemistry varies across a headwater stream network.

Lowe and co-authors from Virginia Tech, the U.S. Geological Survey, the University of Washington, the Cary Institute of Ecosystem Studies, the University of Connecticut and the U.S. Forest Service Northern Research Station examined 664 water samples collected every 10 meters along 32 tributaries of a stream network in the Hubbard Brook Valley of New Hampshire.

Lowe says this would be like looking at any of the Bitterroot Valley creeks by starting at its headwaters high in the mountains and sampling all the small streams that feed into the system as the stream makes its way down to the Bitterroot River.

Headwater streams perform critical functions for downstream ecosystems, but the complexity of their streamwater chemistry is not well understood. Lowe and his co-authors' findings suggest that in headwater stream networks many factors influence the streamwater chemistry in different locations along the stream's course, and in complex relationships with the surrounding landscape.

Previous studies of streamwater chemistry generally looked at stream characteristics along one branch, focusing on variation from upstream to downstream. This study's looks at an entire stream network and uses more detailed, high-resolution data to show that important stream characteristics also vary across the entire headwater stream system.

For example, concentrations of biologically important nutrients such as nitrogen and carbon may vary significantly along individual tributaries, but also among tributaries or on east- versus west-facing slopes in the same network.

"These other large scale patterns of variation could be important for water quality monitoring and the management of aquatic species that rely on healthy streams and rivers," Lowe said. "Until now, we have not had the high-resolution data or statistical methods needed to detect these broader patterns of variation."

He notes that this research will give other scientists and land managers a new way to study aquatic ecosystems in headwater stream systems.

The paper, "Network analysis reveals multiscale controls on streamwater chemistry," was written by Kevin J. McGuire, Christian E. Torgersen, Gene E. Likens, Donald C. Buso, Winsor H. Lowe and Scott W. Bailey.