Probing the Depths of Crater Lake

Douglas W Larson. American Scientist. Volume 90, Issue 1. Jan/Feb 2002.

Over the past 100 years more than 25 million people have visited Oregon’s Crater Lake National Park to behold the majestic beauty of what is perhaps the most remarkable lake on earth. At nearly 600 meters deep, Crater Lake is the deepest lake in the U.S. and the seventh deepest in the world. To view the lake, visitors must first journey up the slopes of an ancient volcano, traversing vast pumice fields and scattered mountain meadows, before climbing steep, forested slopes that rise abruptly to the jagged rim of a crater hundreds of meters above the lake’s surface. Their first breathtaking glimpse of the lake is of an enormous volcanic depression, or caldera, filled with incredibly dear, blue water.

Indeed, this experience is just what geologist Clarence Dutton of the U.S. Geological Survey had promised in his 1886 essay on Crater Lake published in Science: “As the visitor reaches the brink of the cliff, he suddenly sees below him an expanse of ultramarine blue of a richness and intensity which he has probably never seen before, and will not likely see again. Lake Tahoe may rival this color, but cannot surpass it … It is difficult to compare this scene with any other in the world, for there is none that sufficiently resembles it.”

Beginning in 1885, Dutton and William Steel, a Portland businessman, lobbied the U.S. Congress relentlessly for legislation to protect and preserve the lake that Dutton had praised as “a little sheet of water which is destined to take high rank among the wonders of the world.” Finally, in 1902, Congress passed legislation designating Crater Lake as a national park. On May 22, 1902, President Theodore Roosevelt signed the legislation, assuring the nation that the lake’s beauty and uniqueness would be protected for “present and future generations.”

Regrettably, the protection that Roosevelt envisioned for Crater Lake has not always been assured. Precious little scientific research characterized the lake’s ecology during its first 80 years of park status. Thus when water clarity appeared to be declining in the late 1970s, there was a slim scientific foundation on which to base judgments. Fortunately, Crater Lake has enjoyed considerable scientific interest since then-investigations in which I have participated. In the following paragraphs, I shall describe much that we have learned about the lake and also present some of the questions that remain. Only continued scrutiny can assure future generations’ enjoyment of this wonderful resource.

“A Little Sheet of Water”

Crater Lake occupies a 10-kilometer-wide caldera that was formed roughly 7,000 years ago by the climactic eruption and collapse of Mount Mazama, a volcano in the southern Oregon Cascade Range that was perhaps 3,500 to 4,000 meters high. The recurrence of smaller eruptions and lava flows produced an emergent cinder cone, known as Wizard Island, a submerged cinder cone called Merriam Cone and a dome on the floor of the basin. The lake is enclosed by steep caldera walls that tower 150 to more than 600 meters above the lake’s surface. The lake’s maximum depth is somewhere between 589 and 596 meters. Interestingly, Dutton and Steel recorded a maximum depth of 609 meters in 1886, using only piano wire to sound the lake at 168 locations.

The lake is essentially a closed basin, meaning that no permanent streams enter or exit. Water enters as precipitation falling directly on the lake (about 80 percent of the annual water input) and as snowmelt or rain running off the caldera walls. Precipitation occurs mostly as snowfall, which averages about 13 meters per year. Lake water is lost through seepage (perhaps 50 to nearly 70 percent of the total loss) and evaporation. Since about 1900, lake surface elevation has fluctuated nearly five meters, reaching its highest recorded elevation (1,883.93 meters above mean sea level) in 1975 and falling to its lowest recorded elevation (1,879.02 meters) in September 1942. The residence time of water in Crater Lake is about 150 years.

Thermal gradients are typical of deep, high-mountain, temperate lakes. During most of the year (October-May) the lake is essentially isothermal at around 4 degrees Celsius. The lake rarely freezes over: Ice cover was reported for 1898 and 1924, and in 1949 an ice layer 5 to 30 centimeters thick covered the lake for three months. The lake usually begins to thermally stratify in June. By late summer, maximum surface temperatures may reach 18 degrees, which is rarely exceeded. Summertime temperatures seldom exceed 5 degrees below 40 meters, and 4 degrees below 100 meters. Near-bottom temperatures fall to 3.5 degrees.

The lake is more chemically concentrated than other oligotrophic, high-elevation lakes in the Oregon Cascade Range. The average concentration of total dissolved solids (80 milligrams per liter) is four to five times greater than most Cascade lakes. Total alkalinity, a measure of the water’s capacity to neutralize acid, or buffering capacity, averages 27 milligrams per liter (as calcium carbonate), which is roughly three times greater than the average alkalinity obtained for 63 other Cascade lakes. Specific conductance, which is proportional to the water’s ionic concentration, or salinity, averages 115 micromhos (the reciprocal of microohm, a unit of resistance) per centimeter; this value is nearly seven times greater than the average conductivity of 63 other Cascade lakes. Silica, a vital nutrient used by diatoms to construct frustules, or cell walls, averages 8.0 milligrams per liter, which is typical of Cascade lakes. But other algal nutrients, especially phosphorus and nitrogen, are exceptionally low. Concentrations of dissolved inorganic phosphorus, for example, generally range between 9.0 and 18 micrograms per liter. Concentrations of dissolved inorganic nitrogen are considerably lower (usually one microgram per liter or less in the upper 200 meters), suggesting that primary production by plant life is nitrogen-limited.

Crater Lake has long been celebrated for its extraordinary water clarity and intensely blue color. During August 1937, Arthur Hasler from the University of Wisconsin lowered a 20-centimeter diameter Secchi disk into the lake on three different days and observed that it disappeared at depths of 36, 39 and 40 meters. (By comparison, Secchi-disk transparency of other Cascade lakes is considerably less, ranging between 1.5 and 20 meters for 72 lakes tested in 1981-82.)

In 1968 and ’69, John Donaldson and graduate students from Oregon State University recorded five Secchi-disk readings under optimal weather and lake-surface conditions; these ranged between 32 and 44 meters (average: 38.1 meters), with the highest reading obtained with a 100-centimeter diameter disk. This exceptional clarity was attributed largely to the paucity of suspended particulate matter, which minimizes the scattering of solar radiation as it penetrates vertically (downwelling light) in the lake. Indeed, light penetration is considerable, with one percent of surface incident radiation still remaining at depths of between 80 and 100 meters. G. Evelyn Hutchinson, the renowned limnologist at Yale University and long-time Marginalia columnist in these pages, described the lake in his 1957 A Treatise on Limnology, Vol. 1, as “almost optically pure,” with the maximum transmission of light (around a wavelength of 450 nanometers) limited almost solely by molecular scattering. The lake’s deep-blue color, for instance, results from the molecular backscatter of downwelling light, predominantly the short wavelengths in the visible light spectrum.

Logistical Challenges

Between 1902 and 1982, only a dozen or so scientific expeditions were mounted to study the unique qualities of Crater Lake. Generally, these investigations were short-term (some lasting for only a day or two), narrow in scope and conducted during summer months. This record of sporadic, short-lived investigations was the result of several factors, although the leading one was probably the lake’s nearly inaccessible location. Researchers reached the lake along a single, tortuous trail that switchbacked down the steep caldera wall for nearly two kilometers. Sensitive instruments and other limnological gear had to be transported up and down the trail on a motorized trail packer. Weather conditions were occasionally unfavorable and even dangerous; winter research was not even attempted until 1971, when oceanographers from Oregon State University risked whiteouts, avalanches and hypothermia to collect wintertime data. Research boats were usually launched and retrieved on steep snow-chutes descending several hundred meters to the edge of the lake.

The scientists who conducted these expeditions provided their own funding and much of the logistical support. Oddly, the National Park Service expressed little or no interest in pursuing scientific research of the lake, even though the agency is solely responsible for the lake’s management and protection. In fact, no park service limnological monitoring and research program existed at Crater Lake until 1983, when the U.S. Congress passed legislation (Public Law 97-250) mandating the agency to investigate possible lake deterioration. Thus the few scientists who studied the lake before 1983 were mostly self-funded university professors and their graduate students, off-duty park naturalists and volunteers. Between 1978 and 1984, while working as visiting scientists in the park service’s Volunteers in Parks Program, Cliff Dahm of the University of New Mexico and I often conducted our research from a two-person rubber dinghy equipped with a three-horsepower outboard motor. Water-sampling gear and various instruments were lowered into the lake with a homemade wooden winch and cable reel containing about 650 meters of rope.

Today, the park service has a 30-foot-long research vessel permanently stationed on Crater Lake. The boat was purchased in 1995 at a cost of about $130,000 and was lowered to the lake by helicopter. Water and biological samples are analyzed at a fully equipped laboratory, located near the park’s headquarters building. During winter, researchers are periodically flown by helicopter to Wizard Island where the park service maintains a field laboratory and sleeping quarters to accommodate wintertime studies of the lake. A full-time park service scientific staff implements the program, assisted by university scientists working on long-term research projects under contracts with the park service.

Ecological Challenges

The plants and animals in Crater Lake are challenged as well by the lake’s extreme environment, which they endure through various adaptations. Consider, for example, the vertical distribution of phytoplankton. The lake’s phytoplankton community consists of around 160 taxa, nearly all of which (140 taxa) were identified between 1978 and 1982 by Stan Geiger, a consulting algologist from Portland. Diatoms comprise up to 70 percent of the total phytoplankton assemblage. Living phytoplankton is found at depths greater than 300 meters, although 95 percent of the phytoplankton is concentrated between the surface and 200 meters.

The lakes summertime phytoplankton assemblage is partitioned into three distinct strata, each featuring fairly low species diversity and dominance by a single species. In the upper 20 meters, where waters are nitrogen-deficient and illuminated by high-intensity light, the pennate diatom Nitzschia gracilis proliferates during July and August, occasionally exceeding 1,000 cells per milliliter. Conversely, in the lowermost stratum (160-200 meters), the centric diatom Stephanodiscus hantzschii thrives in an environment where temperatures are less than 4 degrees and light less than 0.1 percent of surface incident radiation. Between these extremes is a middle zone in which Tribonema affine, a yellow-green alga, generally predominates, with Nitzschia gracilis and Stephanodiscus hantzschii scarcely present in most samples, This three-tier structure is analogous, perhaps, to a tropical rain forest in which communities of organisms aggregate into vertically distinct, environmentally disparate zones down through the forest canopy.

Plant life also exists in near-shore waters, especially around Wizard Island where luxuriant, mixed beds of rooted plants (water buttercup, bitter cress), algae (Cladophora sp., Nostoc sp.) and moss (Drepanocladus aduncus) cover the bottoms of island coves. Near-shore rocks and sediments support substantial growths of epilithic diatoms, including more than a hundred species, some of which are rare. Nitrogen-fixing blue-green algae (Nostoc sp., Tolypothrix sp.) dominate the periphyton community, however, suggesting further that the lake is nitrogen-deficient.

Offshore, between depths of 25 and 140 meters, heavy growths of moss (D. aduncus) cover the bottom. Deep-dwelling benthic moss was first observed in 1937 by Arthur Hasler, who described a collection of moss from 120 meters as his “most startling biological finding in Crater Lake.” The moss also supports a diverse assemblage of epiphytic algae, including diatoms, filamentous green algae (Chlorophyta), blue-green algae (Cyanophyta) and yellow-green algae (Xanthophyceae). Forty-one species and varieties of epiphytic diatoms, some described as rare, were found on a single moss sample collected from 29 meters in 1989.

The lake’s zooplankton community is marked by very low species diversity: Only two cladocerans (Daphnia pulicaria and Bosmina longirostris) and 11 rotifers (Keratella cochlearis, Keratella quadrata, Filinia terminalis, Kellicottia longispina, Polyarthra dolichoptera, Synchaeta oblonga, Synchaeta lakowitziana, Philodina acuticornis, Conochilus unicornis, Collotheca pelagica and Asplanchna sp.) are present. In 1967 and `68, Owen Hoffman of Oregon State University first documented the vertical migration of cladocerans in Crater Lake, noting for example that Daphnia resided at depths of around 60 meters during the day, then migrated to the lake surface at night. Hoffman concluded that vertical migration might confer a reproductive advantage. But current investigators believe that Daphnia congregate in the deeper, low– light waters during the day to avoid predation by zooplanktivorous fish, principally kokanee salmon (Oncorhynchus nerka).

Crater Lake was presumably fishless before 1888. In August of that year, William Steel and two colleagues transported 600 fingerling rainbow trout (Oncorhynchus mykiss) to the lake from a nearby ranch. The 37 trout that survived the trip were planted in the lake on September 1. The first trout-some measuring 76 centimeters in length-were caught in 1901. Between 1910 and 1941, the park service stocked the lake with nearly 2 million fish, including rainbow trout, kokanee, brown trout (Salmo trutta), cutthroat trout (Oncorhynchus clarki) and coho salmon (Oncorhynchus kisutch). Currently, only kokanee and rainbow trout are present, although a brown trout was reportedly caught in 1966. The kokanee are most abundant, as evidenced by large schools of smolt-size salmon often numbering several hundred-thousand individuals each.

The Case of Optical Deterioration

Secchi-disk readings obtained during summer 1978 indicated that the clarity of Crater Lake was possibly diminishing. These readings, the first taken since 1969, were all 30 meters or less. We (Dahm, Geiger and I) reported our findings to the park service in October 1978, emphasizing that the data were very preliminary but warning of possible optical deterioration. Unfortunately, historical Secchi-disk data for Crater Lake were scarce, amounting to a few reliable measurements obtained by Hasler in 1937 and Donaldson in 1968-69. Thus, it was virtually impossible to tell whether the 1978 data represented a long-term downward trend or a short-term downward fluctuation. But 46 additional readings obtained from 1979 through 1984 reinforced the notion that clarity had diminished. These readings, 80 percent of which were 30 meters or less, ranged from 22 to 37 meters (average: 28.9 meters).

In response to park service questions about probable cause, we cited a basic limnological axiom: Reduced Secchi transparency is often caused by increased phytoplankton abundance, which is a precursor of accelerated eutrophication of oligotrophic lakes. We also cited the work of Charles Goldman at Lake Tahoe, a deep (505 meters), oligotrophic lake on the California-Nevada border. Goldman, of the University of California at Davis, found that yearly-average Secchi-disk readings decreased from about 30 meters in 1958 to 20 meters in 1990. He attributed this decline to increased phytoplankton abundance caused by nutrient (nitrogen) enrichment from various human sources.

We speculated that the phytoplankter Nitzschia gracilis had become more abundant in Crater Lake because of an unnatural increase in nutrient concentrations, particularly nitrogen. Because nitrogen concentrations are extremely small, additions of nitrogen from sewage could greatly stimulate the growth of phytoplankton and other lake algae.

The proliferation of Nitzschia gracilis was first observed in 1978. But because of the scarcity of pre-1978 phytoplankton data, we were unable to determine if this was a regular occurrence or a newly developing one. Only two previous studies of phytoplankton abundance had been done, the first in 1913 by George Kemmerer and others of the U.S. Bureau of Fisheries and the second on July 18, 1940, by oceanographers Clinton Utterback, Rex Robinson and Lyman Phifer of the University of Washington. Kemmerer’s group found that maximum phytoplankton densities occurred between 60 and 80 meters on August 1 and between 100 and 150 meters on September 5. They also reported that Asterionella sp. was the only diatom in the lake. Utterback and colleagues observed the following: Phytoplankton were most abundant between 75 and 150 meters; practically no phytoplankton existed in the lake’s upper 20 meters; the bulk of the phytoplankton consisted of the filamentous blue-green alga Anabaena sp. (which was never found again); and diatoms constituted about 15 percent of the total phytoplankton assemblage. Neither study offered us much guidance.

The source of nitrogen enrichment, we suspected, was the park’s antiquated sewage disposal facilities on the caldera rim, specifically a septic tank-drainfield system that processed an estimated 16 million gallons of raw sewage every summer. The system is perched 200-250 meters above the surface of the lake in pumice soils described by the U.S. Geological Survey as “so highly permeable that in places all precipitation infiltrates where it falls.” Installed in the mid-1940s to accommodate about 200,000 summertime visitors, the system had become inadequate by the 1970s when the summer visitation rate had increased nearly threefold. The system was upgraded in 1975 shortly after the park’s main source of drinking water, a spring-fed creek, was grossly contaminated with sewage, causing more than a thousand cases of diarrhea and other waterborne ailments among park tourists and staff.

Concerned that the lake was being degraded, perhaps irreversibly, we set out to determine whether sewage was reaching the lake. In 1983, we discovered that one of the springs emerging along the caldera wall and flowing into the lake contained roughly 10 times more nitrate-nitrogen than any of the other 40 to 50 caldera springs tested. Using maps and sketchy geological information, we were able to determine that this spring, designated Spring 42, was linked to an aquifer flowing directly beneath the septic tank-drainfield system. We surmised that septic wastewater was percolating through drainfield soils into the aquifer. Although we strongly recommended that a dye-injection study be conducted to trace the pathway of wastewater through underlying soils and rock, the work was never attempted. Consequently, the question of sewage contamination was never completely resolved. Faced with other evidence, however, the park service acknowledged in 1987 that sewage was probably entering the lake.

In 1991, 13 years after first being alerted to the possibility of sewage contamination, the park service removed the septic tanks from the rim and diverted the sewage through a new $3 million pipeline. Since then, lake-water clarity has seemingly improved: In 1997, park service limnologists reported maximum average summertime Secchi depths of 41.5 meters with a 20-centimeter (diameter) disk and 49.2 meters with the 100-centimeter disk. But whether this optical improvement is the result of sewage diversion is still unknown.

Nitrogen: An Alternative Hypothesis

During the lengthy debate over sewage contamination and its effects on the lake, alternative hypotheses were sought to explain how other sources of nitrogen might stimulate excess phytoplankton production. One theory argued that nitrogen from remineralized organic matter on the lake bottom is circulated into the euphotic zone by convection. This process of nitrogen upwelling is driven by warm, chemically concentrated waters emanating from active hydrothermal vents on the lake bottom. Hydrothermal fluids warm the lake’s deep waters, causing them to convect, which mixes the lake vertically while transporting nutrients and other dissolved chemicals toward the lake surface.

The presence of active lake-bottom hydrothermal vents was first proposed in 1968 by A. S. Van Denburgh of the U.S. Geological Survey. Active vents, according to Van Denburgh, accounted for the lake’s relatively high sulfate and chloride concentrations, averaging 10.5 and 10.2 milligrams per liter, respectively. In 1983, David Williams of the U.S. Geological Survey and Richard Von Herzen of Woods Hole Oceanographic Institution measured conductive heat flow through lake-bottom sediments, concluding that heat flow into the lake causes deep waters to convect slowly. They also estimated that lake-bottom thermal springs (they discovered two) discharged 6.35 x 10^sup 9^ grams of dissolved solids into the lake annually.

In 1988 and 1989, oceanographers Robert Collier and Jack Dymond of Oregon State University explored the bottom of Crater Lake in the submersible Deep Rover, making a total of about 50 dives. They discovered evidence of hydrothermal venting, including: higher temperature and salinity gradients extending several meters upward from the lake bottom; prolific mats of chemolithotrophic bacteria, whose interstitial water temperatures exceeded the temperature of ambient lake-bottom water by more than 15 degrees; and brine pools on the lake bottom that were about 10 times more chemically concentrated than ambient water. They concluded that up– welling of nitrate-nitrogen from deep waters contributes more than 85 percent of the total new nitrogen entering the lake’s euphotic zone, with the balance derived from allochthonous (outside the lake) sources, such as direct precipitation and watershed runoff. If correct, then perhaps the nitrogen derived from sewage would not have been a significant nutrient source for phytoplankton.

Legacy for “Present and Future Generations”

Now that the septic system has been removed, it will probably never be determined which hypothesis is correct. Indeed, in one sense the question is moot: The pollution of this unique body of water with septic wastewater was clearly not the legacy that Americans had in mind a century ago when the lake became a national park. Fortunately, not only has the sewage threat been removed, but a scientifically based monitoring and research program is also now in place, enabling scientists to make new discoveries and acquire a long-term limnological database for tracking lake conditions.

In 1886, Clarence Dutton marveled at visitors’ responses to Crater Lake: “It was touching to see the worthy but untutored people, who had ridden a hundred miles in freight wagons to behold it, vainly striving to keep back tears as they poured forth their exclamations of wonder and joy akin to pain.” Visitors are still moved by the incredible beauty of Crater Lake. Unfortunately, they may take it all for granted, assuming that once a threatened natural treasure like Crater Lake has been designated a national park, the threat is gone and the park is preserved forever in an unalterable state. But the lake is a fragile environment besieged by over a half-million tourists each summer and pressured from all sides by relentless cultural expansion, including logging operations, ski resorts, highways and real estate development. We can hope that, at the end of the next century, visitors will find that the Crater Lake extolled by Clarence Dutton in 1886 has not lost its incomparable charm.