Relation Between Selected Water-Quality Constituents and Lake Stage in Upper Klamath and Agency Lakes, Oregon By Tamara M. Wood, Gregory J. Fuhrer, and Jennifer L. Morace SUMMARY Upper Klamath Lake is a large (140 square-mile), shallow (mean depth about 8 ft) lake in south-central Oregon that the historical record indicates has been eutrophic since its discovery by non-Native Americans. In recent decades, however, the lake has had annual occurrences of near-monoculture blooms of the blue-green alga Aphanizomenon flos-aquae. In 1988 two sucker species endemic to the lake, the Lost River sucker (Deltistes luxatus) and the shortnose sucker (Chasmistes brevirostris), were listed as endangered by the U.S. Fish and Wildlife Service, and it has been proposed that the poor water quality conditions associated with extremely long and productive blooms are contributing to the decline of those species. It has also been proposed that the low lake levels made possible by the construction of a dam at the outlet from the lake in 1921 have contributed to worsening water quality through a variety of possible mechanisms (Jacob Kann, Klamath Tribes, written com-mun., 1995). One such mechanism would be an increase in internal phosphorus loading from resuspended sediments (Jacoby and others, 1982), resulting from an increase in bottom shear stresses at lower lake levels (Laenen and LeTourneau, 1996), leading in turn to more intense algal blooms. Another possible mechanism is an earlier triggering of algal blooms. When early spring lake levels are low, greater light intensity at the sediment surface might speed recruitment of algal cells from the sediments. Sediment recruitment has been shown to be an important contributor to water column biomass increases in A. flos aquae (Barbiero and Kann, 1994) and Gloeotrichia echinulata (Barbiero, 1993). An earlier bloom could result in poor water quality conditions occurring earlier in the year, when young-of-the-year fish may be more susceptible to those conditions. Lake level can also influence water quality directly. An increased frequency of sediment resuspension at lower lake levels could increase chemical and biological oxygen demand, resulting in decreased dissolved oxygen concentrations. Sediment oxygen demand also may be enhanced at lower lake levels because it is concentrated over a smaller volume of water. Some compensation for increased oxygen demand at lower lake levels might be provided by increased reaeration, if the water column mixes from top to bottom more frequently. Based on the analysis of data that they have been collecting for several years, the Klamath Tribes recently recommended that the Bureau of Reclamation (Reclamation) modify the operating plan for the dam to make the minimum lake levels for the June-August period more closely resemble pre-dam conditions (Jacob Kann, written commun., 1995). The U.S. Geological Survey (USGS) was asked to analyze the available data for the lake and to assess whether the evidence exists to conclude that year-to-year differences in certain lake water-quality variables are related to year-to-year differences in lake level. The results of the analysis will be used as scientific input in the process of developing an operating plan for the Link River Dam. Datasets Two water-quality datasets were analyzed. The first was a series of hourly records of pH, dissolved oxygen, and water temperature, each of approximately a week's duration. The records were collected at 3 sites over 3 years, 1992 through 1994, with enough consistency to define the seasonal patterns. This dataset provided information about the diel extremes in dissolved oxygen and pH and the seasonal pattern in the diel cycle, but measurements were limited to a depth of 1 m(3.28 ft). The second dataset was a set of depth profiles of pH and dissolved oxygen and concurrent depth-integrated samples for nutrients and chlo-rophyll-a. The profiles were collected at approximately biweekly intervals at nine sites (seven in Upper Klamath and two in Agency Lake) over the 5 years 1990 through 1994. These depth profiles provided information on the depth-dependence of dissolved oxygen and pH, and allowed more extensive year-to-year comparisons than did the hourly records. Because measurements were made at each site only once during the sampling day, however, they did not capture the daily extremes in water quality. Lake level is measured daily by the USGS at three sites around the lake: Rocky Point, Rattlesnake Point, and near the city of Klamath Falls. These daily measurements are then used to compute a spatially weighted average of the lake level that is reported in the USGS annual Water-Data Report for Oregon. The average lake levels were used in this report. Two climatic datasets were used in this report; both were collected at the Klamath Falls airport. Air temperature was recorded as a daily maximum and daily minimum value. Cloud cover was quantized on a daily basis into one of seven levels. Because the focus of this study was primarily to examine possible relations between water quality and lake level, the lake level data provide an important context for the discussions that follow.
Includes bibliographical references (p. 57)