Editor’s note: In July, the Bureau of Reclamation pledged $1.2 million to fund new scientific initiatives that will inform the agency’s management of the Klamath Project. This is the first in a series of articles explaining the past, present and future of that scientific research.
It’s the big question in the Upper Klamath Basin: To what extent does the water level of Upper Klamath Lake affect the health of sucker populations?
After nearly three decades of data collection and modeling, scientists still aren’t quite sure.
C’waam and Koptu (Lost River and shortnose suckers) are endemic to the Klamath watershed and culturally important to the Klamath Tribes. Listed as endangered throughout their entire habitat range under the Endangered Species Act in 1988, sucker populations have been in peril for decades.
Biological opinions informing Reclamation’s operations have recommended that the surface elevation of Upper Klamath Lake, one of the last significant remaining habitats for suckers, remain at least 4,138 feet above sea level at all times in order to maintain that habitat.
With a mean depth of 8 feet, Upper Klamath Lake is remarkably shallow. And thanks to its volcanic sediments, its environment is eutrophic — or high in nutrients like phosphorus. These factors make it an ideal breeding ground for blue-green algae, which grow extensive colonies during the summer and eventually experience a mass die-off, negatively impacting the lake’s water quality and lowering the amount of available oxygen.
These conditions put stress on young suckers, leaving them more vulnerable to parasites, diseases and predators.
The U.S. Geological Survey produces much of the science that informs the biological opinions that concern Upper Klamath Lake.
USGS scientists in the Klamath Basin fall into either the ecosystems mission area, which focuses on species, or the water mission area, which focuses on hydrology. The sucker problem incorporates both areas, merging research on water quality with research on fish health.
Adult suckers aren’t the primary concern for researchers — about 10 percent of them die each year, which is normal.
“What you don’t see is enough individuals coming into the populations to replace them,” said Eric Janney, director of USGS’s Western Fisheries Research Center.
This lack of “recruitment,” as scientists call it, means that not enough juvenile suckers are surviving long enough to reach sexual maturity and ultimately support the longevity of the species.
In general, Janney said, adult fish spawn every spring along the shoreline of Upper Klamath Lake and in the Williamson River. Larvae hatch from eggs and travel into the lake in mid-June, where USGS researchers are able to catch them for study throughout the summer.
But every August, new juveniles almost entirely disappear. Janney said that’s not because they suddenly become adept at avoiding the nets — it’s because they’re dying.
“Over the last 20 years, we’ve seen almost complete mortality every year,” he said. “We don’t have a successful year where fish did really well.”
That makes it difficult to determine whether Reclamation’s maintained lake levels are actually helping the fish, because there’s not one year where they can correlate Upper Klamath Lake’s elevation with increased survival of juvenile suckers.
“That doesn’t mean that if you drew the water down, it wouldn’t have an effect on the suckers,” said Summer Burdick, a fish biologist with USGS who studies juvenile sucker mortality.
Lake levels are just one piece of a complex puzzle of environmental factors that all interact to impact water quality. How big of a piece they are is still up in the air.
Throughout the 1980s leading up to the listing of the suckers under the Endangered Species Act, several hypotheses about their predicament began to float around.
“One of the obvious ones was that the footprint of the irrigation project is within the known distribution of the sucker species,” said Torrey Tyler, supervisory fish biologist for the Bureau of Reclamation’s Klamath Area Office.
In 1987, the Klamath Tribes began a monitoring program that took water quality measurements every two weeks from May to October at a consistent group of sites throughout the lake. They started to see patterns in lake level data and water quality and sent memos to Reclamation recommending that Upper Klamath Lake be kept at pre-dam levels to ensure suckers could continue to live in the environment they had evolved in.
In 1996, USGS released their first study on the relationship between lake level and water quality in Upper Klamath Lake. Using five years of data from 1990-1994 provided by the Klamath Tribes, it analyzed whether different measures of water quality (dissolved oxygen, chlorophyll-a concentrations, pH and water temperature) fluctuated in relation to lake levels each year.
The study didn’t find a significant relationship for the May-October season, but it did find that, in June of each year, the level of the lake did correlate with how early the algal bloom began. That bloom, in turn, determined how frequently pH levels spiked throughout the lake.
This research informed a 2001 biological opinion that required the project to shut off water deliveries, decimating the basin’s farmers and ranchers that year. The next year the National Research Council, part of the National Academy of Sciences, reviewed the opinion in 2002 and found “no clear evidence of a connection between the lake levels and the welfare of the two sucker species.”
Five years of data isn’t enough to establish a trend, but it’s not enough to disprove it, either.
In 2007, USGS released an updated report that included 17 years of water quality and lake level data. The report used linear models to determine whether relationships between those data were statistically significant, and found that, “at present, no single causal factor can be clearly identified” as to why Upper Klamath Lake’s water quality declines so intensely during each summer.
While the 2007 paper didn’t necessarily refute the 1996 paper’s findings, Tyler said it told a slightly different story — that meteorological conditions like air temperature and wind speed had more of a significant effect on algal blooms (and therefore water quality) than lake levels. But even the 2007 paper said that the Klamath Tribes’ water quality dataset “was not designed to address the questions being asked in this report” — data collection on water quality would need to be more continuous than once every two weeks in order to truly understand how conditions varied.
Earlier this year, a paper in the journal Hydrobiologia looked at 27 years of that data with a different approach than the previous two USGS papers. It determined that lake levels don’t directly affect water quality but instead work alongside weather conditions to influence it. The study identified a long-term median lake level that “generally provided the overall lowest risk of poor water quality,” and found that the progression of algal blooms meant that both extreme low and extreme high levels could negatively impact water quality at different times in the summer.
The various environmental factors had non-linear relationships, which made it easy for a typical model to ignore their combined influence on the system. The paper concluded that “although maintenance of the target lake levels over the long-term would be expected to reduce the frequency of poor water quality ... lake level management alone may not necessarily prevent poor water quality.”
A year and a half ago, Reclamation approached USGS and asked for another lake level study as part of the new scientific initiative. By that point, USGS had been collecting its own water quality data on an hourly basis since 2005, building a data set with far more detail than previous ones. Hydrologist Liam Schenk is leading the new effort to analyze that data, and he said the new approach is inspired by the Hydrobiologia paper’s non-linear approach.
Instead of using linear models like in the previous two USGS studies, Schenk’s project employs a cutting-edge machine-learning model that will use thousands of hourly data points to paint a picture of all the interactions between the different systems at play in the lake.
“You don’t come in with any preconceived ideas of what might be affecting water quality,” Schenk said.
If the last two reports were any indication, lake level is part of a suite of factors playing into Upper Klamath Lake’s water quality. That’s why the model provides a list of “relative influences,” which rank each variable (from lake level to air temperature to nutrients entering the lake) based on the extent to which it impacts water quality. The results will be complex, but Schenk hopes it will paint a fuller picture of what goes into water quality in Upper Klamath Lake.
“It’s a dynamic system, so it doesn’t behave in a linear fashion,” he said.
Machine learning models have often been used to study water quality in similar lakes around the world, but Schenk said this is the first time one will be used in Upper Klamath Lake. He said it’s too early to tell whether or not it will improve the science.
While water quality’s relationship to lake levels remains a point of contention in the basin, numerous studies have shown that low lake levels cut off access to sucker spawning grounds on the shores of Upper Klamath Lake. A 2015 USGS study tracked individual tagged adult suckers to determine where they went to spawn between 2006 and 2012.
In 2010, Upper Klamath Lake’s elevation was the lowest it had been in 38 years — so low that it made some sucker spawning spots too difficult for the fish to reach. That led to about 14 percent fewer female and 8 percent fewer male suckers participating in the spawning season that year. Though biologists’ main concern is not having enough juveniles surviving to adulthood, having fewer juveniles to begin with doesn’t help.
Reclamation said the new USGS report should be finished sometime next summer, after which scientists in the basin will initially review it. Then, Reclamation will conduct a review of the report along with the two other USGS studies and the Hydrobiologia paper to complete a synthesis report by early 2022.
Who will do that final summary is still up in the air, though Tyler hopes it will be an unbiased third party.
No one knows what the “new science” will reveal about Upper Klamath Lake, but one thing is certain: Below its carefully managed surface, the lake holds many more secrets than you’d expect.