A statistical “chum emergence model” that has proved out in recent years' testing is giving hydro and salmon managers another tool for assessing conditions experienced by the threatened species in the often roiled water below Bonneville Dam.
The model assembles data retrieved from water temperature sensors implanted in one of the Columbia River chum salmon’s key spawning grounds, the Ives Island channel that hugs the Washington shore just below the dam. It evaluates how close eggs planted there are to hatching. Knowing with some precision when emergence is likely to occur allows dam managers to manipulate operations, such as spill, to reduce the risk of harm to chum sac fry that may still be in the gravel.
The spill, which in recent years has started April 10, is implemented to provide a passage route for juvenile salmon and steelhead that are headed toward the Pacific Ocean. But the water plunging down from spill gates also serves to boost total dissolved gas in the water below the dam. TDG at heightened levels can be harmful to aquatic life.
The model was developed through research at Ives Island and in the laboratory by the Pacific Northwest National Laboratory at the request of the U.S. Army Corps of Engineers, which operates Bonneville and other dams in the Federal Columbia River Power System. PNNL, located in Richland in eastern Washington, is one of the U.S. Department of Energy's 10 national laboratories, managed by DOE's Office of Science. PNNL also performs research for other DOE offices as well as government agencies, universities and industry.
The model is now the property of the Corps, which will continue to repopulate it with new data through time. The temperature sensors chronicle the mixing of warmer water that upwells at the favored Ives redd-building sites and cooler river water to evaluate the “accumulated thermal units” absorbed at redds’ various sites and depths.
The channel study area varies in width depending on the amount of water being discharged from Bonneville and is about three-quarters of a mile long. The area was outfitted with onset loggers at 31 locations, three temperature sensors that deliver data electronically in real-time and four other locations with temperature sensors that monitor the warm upwellings. The onset logger sensors were set in the gravel by divers. Each is manually retrieved about once a month so that accumulated data can be downloaded. The sensors are then redeployed in the gravel bed at egg pocket depth.
"It's a good model that technical managers can use" to make operational decisions, the Corps' Dennis Schwartz said during a recent presentation to the Technical Management Team. The TMT's federal, state and tribal hydro and fish managers consider day-to-day operations that might be made to benefit fish.
The model is designed to tell system operators the complete distribution of chum salmon emergence (first emergence through 50 percent emergence through final emergence) in order to balance chum salmon redd protection and power system operations.
Laboratory studies by PNNL suggest thermal threshold of 932 ATUs bring chum to a 50 percent emergence. The ATU is calculated as sum of daily average temperature from a given start date. That start in the Ives Island is the date of spawning at each redd site.
Keeping an eye on environmental and other conditions for threatened chum salmon has become an extremely technical endeavor. For instance hydro and salmon managers carefully monitor the appearance of chum in the area below Bonneville and elsewhere, such as near Multnomah Falls, so they can decide when to establish a minimum tailwater elevation that will have to be maintained until after emergence so that redds won’t be dewatered over the winter.
First listed under the Endangered Species Act in 1999 the “evolutionarily significant unit” includes all naturally spawned populations of chum salmon in the Columbia River and its tributaries in Washington and Oregon, as well as three artificial propagation programs: the Chinook River (Sea Resources Hatchery), Grays River, and Washougal River/Duncan Creek chum hatchery programs.
NOAA Fisheries' biological opinion for the federal hydro system calls for a tailwater elevation of 11.5 feet beginning the first week of November (or when chum arrive) and ending by Dec. 31 if the water supply forecast indicates that this elevation can be maintained through emergence. BiOps judge whether federal actions jeopardize the survival of listed fish stocks and can prescribe mitigation actions intended to improve survival.
Chum salmon spawn in the Ives Island area in December, but don’t emerge until the spring of the following year.
In a “normal” or higher water year, keeping those redds covered in springtime is usually not a problem because redds built at an elevation of 11.5 feet or lower are usually covered by a thick layer of water resulting from precipitation and early snowmelt runoff. As an example, the average elevation over the past week has been well over 19 feet and in February it averaged almost 18 feet.
NOAA Fisheries’ Paul Wagner said the model would be particularly helpful in years with lower flows and thus less “depth compensation” to shield the redds from TDG.
“This is the tool that will be used going forward,” Wagner said of the emergence model.
The U.S. Environmental Protection Agency in 1972 established a nationwide water quality standard for TDG saturation at 110 percent that remains in effect, though state waivers to the standard have been approved beginning in the 1990s to aid juvenile salmonid downstream migration. Studies reviewed as part of the 1995 and 2000 biological opinions indicated that TDG saturation levels between 110 percent and 120 percent had minimal effects on aquatic organisms. The waivers allow up to 115 percent TDG saturation in dam forebays and up to 120 percent in dam tailraces.
Recent PNNL laboratory research has shown that various chum salmon life stages “are extremely tolerant” of TDG, more so than other salmon species, Schwartz said.