OHSU

Researchers Study Loss of Salmon Habitat in the Columbia

David Jay

09/24/03

PORTLAND, Ore. — Using a simple, yet sophisticated computer analysis, Oregon researchers have determined that flood-control dikes in the Columbia River floodplain and reduction of peak water flows have reduced shallow water habitat for juvenile salmon by about 62 percent during the crucial May to July period when salmon most need such shallow waters for their transition to the Pacific Ocean. The computer models, developed by environmental scientists at Oregon Health & Science University's OGI School of Science & Engineering, are efficient for the very long simulations needed to understand historic changes and climatic impacts on shallow water habitat.

"This is the first study to separate the effects of flow regulation and diking on salmon habitat loss in the Columbia River, Oregon's largest river," said Tobias Kukulka, M.S., the lead author of the study who is now in doctoral studies at the University of Rhode Island. "Taken individually, diking and flow-cycle alteration would have reduced spring-freshet shallow water habitat by 52 percent and 29 percent respectively."

The two-part study, funded by the United States Army Corps of Engineers Portland District and NOAA Fisheries, was conducted under the direction of David Jay, Ph.D., associate professor of environmental and biomolecular systems in the OGI School of Science & Engineering. The studies were posted online Sept. 22, 2003, on the American Geophysical Union's (AGU) website and published soon in the AGU's JGR-Oceans.

Columbia River hydrology has changed drastically in response to human activities and climate at the same time that salmon populations have greatly decreased. To better understand the impacts of hydrologic alteration on salmon populations in the tidal-fluvial environment (the area between the estuary and the river) Kukulka and Jay studied the estuarine shallow water habitat, which is known to play a major role in the survival of seaward-migrating juvenile salmon.

"There are few studies on the tidal-fluvial environment," said Jay. "Every river has such an area, and methods used by hydrologists for rivers and oceanographers for estuaries don't work that well. This is also a crucial area for juvenile salmon because many young salmon feed and prepare physiologically and behaviorally in the fresh, shallow waters for a rapid transition to ocean saltwater.

"The Columbia was historically the world's largest producer of Chinook salmon and of all salmon species, Chinook are especially dependent on shallow water habitat," added Jay. "The Columbia has only about 30 miles of estuary, but more than 110 miles of tidal river and none of this area has been studied much."

However, predicting tidal-fluvial water levels poses a challenge, in large part because of the unpredictability of the river flow and hydropower operations. The first part of Kukulka and Jay's study describes a new model of the tides with variable river flow. The second part of the study discusses specific changes within 40 kilometers of shallow water habitat.

The computer model that was developed for the study enabled the scientists to run a 25-year historical case in just a few minutes. "We can run a variety of scenarios with ease," Jay said. "While much more complex models could be used, the fact is that floodplain topography is poorly known and very detailed results might not be very accurate on a point-by-point basis."

The AGU-published studies also are unique from a tidal dynamic point of view, noted Jay. "We showed how tides can be predicted in an environment where there is another major factor (fluctuating river flow) in addition to astronomical factors (the orbital mechanics of the earth, sun and moon) affecting the tides. While we know of one other computer system that can do this, ours is more compact, closer to the actual physics of the problem, and works better when river flow changes rapidly, as it sometimes does in the Columbia, as opposed to seasonally."

Kukulka and Jay's work is part of a broader, multi-institutional effort to understand the Columbia River ecosystem. In that effort, another OGI research team is using detailed numerical computer models and an extensive real-time observation network to understand and predict circulation and physical habitat in the lower Columbia, including such factors as water level, velocity, salinity and water temperature.

"Our work nicely complements Tobias and David's research. While we are providing spatial and temporal detail over weeks, months or a few years for a range of parameters, the Kukulka-Jay work is providing an important longer-term perspective at lesser detail and more focus on water levels," said Antonio Baptista, Ph.D., professor and head of environmental and biomolecular systems in the OGI School of Science & Engineering.