USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2008–5076

Back to Table of Contents

Implications for Water Quality

Because severe water-quality problems in UKL are detrimental to the survival of endangered suckers, the hydrodynamic model is particularly useful insofar as it can provide insight into the processes that control water quality. The simulations of hydrodynamics and heat transport in 2005 and 2006 result in the following observations that have important implications for water quality: first, that water moving northward through the trench to the west of Bare Island is routed both to the north and to the south of the lake, and second, that the water originating from deeper in the trench is routed preferentially toward the north, while the water from nearer the surface is routed preferentially toward the south.

These observations were illustrated and quantified by using model simulations of conservative tracers that were designed to demonstrate how water from the trench is transported to the rest of the lake. First, the polygons in the numerical grid east of Eagle Ridge and south of Bare Island that were deeper than 4.5 m were identified (fig. 25). Within those polygons, and within layers deeper than 4.5 m, the concentration of tracer T1 was set to 1 (arbitrary units) at every time step, and within those layers shallower than 4.5 m the concentration of tracer T1 was set to 0 at every time step. Within the same polygons, and within layers deeper than 4.5 m, the concentration of tracer T2 was set to 0 at every time step, and within layers shallower than 4.5 m the concentration of tracer T2 was set to 1 at every time step. Within polygons outside of those identified, the initial concentrations of T1 and T2 were set to 0. Thus, the concentration of T1 and T2 at any point in the grid is a measure of the fraction of water at that point that originated in the trench at depths deeper than 4.5 m and shallower than 4.5 m, respectively. Assigning the concentrations of two distinct tracers in this way allowed tracking of the surface water of the trench separately from the deeper water. The tracer experiments began on August 1, 2005, and used the calibrated hydrodynamic and heat-transport model described above. The concentrations in the trench were assigned continuously, rather than as an initial condition at a single point in time, in order to observe the continuous transport of water through the trench and out into the lake for several days; as a result the total mass of each tracer increases throughout the simulation. For that reason, the simulations are most interesting over the first few days and were stopped after 10 days.

“Snapshots” of the concentration of tracers T1 and T2 in the surface layer of the grid at the end of day 5 are shown in figure 26. The snapshots show that water that exits the trench at the west of Bare Island (fig. 1) flows both to the east around the island and to the west into the northern part of the lake. The bifurcation of the flow at Eagle Point results in the development of two different circulation patterns, the longer of which carries water around the northern end of the lake and down the eastern shoreline, then turns around north of Buck Island. This circulation pattern is responsible for the direct influence of the water exiting the trench on water quality in the northern part of the lake. A second, shorter circulation pattern carries water around the north and east sides of Bare Island and turns around north of Howard Bay. Previous numerical experiments with this model have shown that the relative strengths of these two circulation patterns varies with the strength of the wind forcing, such that a stronger prevailing wind is more effective at pushing water into the northern part of the lake and enhances the larger pattern at the expense of the smaller (Wood and Cheng, 2006).

Consideration of the sum of the concentration of the two tracers together allows some quantification of the first observation provided above. The sum of concentrations of tracers T1 and T2 shows that water leaving the trench at the beginning of the simulation is detected at site MDN, centrally located in the northern part of the lake, in about 3 days, and the fraction of the water at site MDN that can be traced back to the trench increases rapidly over 7 days to a total of about 60 percent (fig. 27A). The concentrations were simulated in the middle of the water column, but there is little top-to-bottom difference in the simulated tracer concentration at either site MDN or MDL. At site MDL, located in the central part of the lake, the influence of the trench can be detected sooner, within about 1 day, and within 3 days the trench accounts for about 80 percent of the water at site MDL. Thus, it appears that the influence of the trench at site MDL through the shorter circulation pattern is both quicker and quantitatively larger than at site MDN. Note that these fractions of trench water and the travel time from the trench apply specifically to the wind conditions that were measured during the period of the simulation, from August 1 to 10, 2005, when the winds were weak to moderate and with no major reversals in direction; relative fractions of trench water and travel times would be expected to change with different wind conditions (Wood and Cheng, 2006). Although the relative strength of the shorter and longer circulation loops varies with wind conditions, the observation that water moving northward through the trench to the west of Bare Island is routed both to the north and to the south of the lake is robust under conditions of prevailing winds.

The second observation, that the water originating from deeper in the trench is routed preferentially toward the north, while the water from nearer the surface is routed preferentially toward the south, is quantified by considering the difference in the time series of tracers T1 and T2 at the two sites. When the flow through the trench west of Bare Island splits into two different pathways, more of the surface water flows to the east around the north side of Bare Island, and more of the bottom water flows to the west into the northern part of the lake. Thus, while the overall influence of the trench at site MDN is less than that at site MDL, the influence of water from below 4.5 m is less at site MDL (fluctuating between about 0 and 10 percent during days 5–10 of the simulation) than at site MDN (fluctuating between about 10 and 20 percent during days 5–10 of the simulation; fig. 27B). Similarly, the influence of water from above 4.5 m is less at site MDN (fluctuating between about 30 and 40 percent during days 5–10 of the simulation) than at site MDL (fluctuating between about 90 and 100 percent during days 5–10 of the simulation; fig. 27C).

The trench plays a significant role in the overall oxygen budget of UKL, a conclusion supported by at least two separate datasets. First, light and dark bottle experiments have shown that the trench (site MDT) is characterized by net oxygen consumption, whereas much of the rest of the lake is characterized by net oxygen production (fig. 28, Appendix A). The data shown in figure 28 were collected between May and September of 2006; the methods used are discussed in Hoilman and others (2008). Second, oxygen isotopes of the dissolved oxygen in the lake are consistent with respiratory processes dominating in the trench and with domination by photosynthetic production away from the trench (S.R. Silva, U.S. Geological Survey, unpub. data, 2007). In addition, dissolved oxygen measurements from continuous monitors show that the lowest dissolved oxygen events in the northern part of the lake are not restricted to that part of the lake, but consistently are concurrent with very low dissolved oxygen concentrations in the trench as well (Wood and others 2006, Hoilman and others, 2008). Of particular interest is the influence the trench has on dissolved oxygen concentrations in the prime adult sucker habitat in the northern part of the lake, to the north and west of Eagle Point (Wood and others, 2006).

Observations from continuous monitors in 2005 show that the low dissolved oxygen concentrations that accompanied the annual bloom decline at the end of July 2005 were more severe in the northern part of the lake, as measured at site MDN, and were closer to the concentrations (particularly the near-bottom concentrations) detected in the trench at site MDT, than the concentrations in the southern part of the lake, as measured at site MDL (fig. 29). This is consistent with the numerical tracer experiments that demonstrate a greater influence of the near-bottom water in the trench at site MDN than at site MDL and less influence of the surface water. The implications are twofold. First, the near-bottom water has a lower dissolved oxygen concentration than the near-surface water. Dissolved oxygen is not a conservative quantity, however, and given that water at site MDT has several days of travel time to reach either site MDN or site MDL, the differing influence of water from the bottom of the trench on these sites does not fully explain why such different concentrations were measured by the continuous monitors at these sites. Another possibility is that as water passes through the trench, the surface and bottom of the water column develop different capacities for replacement of the oxygen consumed in the trench once the water exits, as discussed below.

Vertical velocities measured by the current profiler at site ADCP1 in 2005 showed that at this site, but not at others, there was a clear distinction between the upper water column, where the AFA population was dominated by rising colonies, and the lower water column, where sinking colonies dominated the population (fig. 30; Gartner and others, 2007). The velocities measured by the ADCP are not water velocities, but rather the velocities of particles suspended in the water column. Vertical water velocities are expected to be one to two orders of magnitude lower than the vertical velocities measured by the ADCP. This suggests that under certain circumstances the effect of the two different circulation patterns is to concentrate rising colonies in the area between Bare Island and Howard Bay, and toward the western shoreline, and to send water relatively depleted of colonies (because they settled out in the trench), or dominated by sinking colonies, into the northern part of the lake. This may imply a greater capacity for photosynthetic production, on a per volume basis, in the central part of the lake between Bare Island and Howard Bay than in the northern part of the lake. Photosynthetic production of oxygen is a dominant term in the oxygen budget and dominates high respiratory and other consumptive demands over much of the lake. In the trench the opposite is true—consumption dominates production. If the water exiting the trench to the west around Eagle Point loses much of its capacity for photosynthetic production for at least a period of several days, while water column oxygen demands remain high, then the dissolved oxygen in the northern part of the lake, in the upper lobe of the larger clockwise circulation pattern, will be replenished more slowly than in the central part of the lake. This would be consistent with the bottle incubation experiments that showed that site RPT (like site MDL, located in the southern part of the lake) was consistently a site of net production of dissolved oxygen, whereas the results at site MDN were variable (fig. 28). Further development and testing of this hypothesis, both with numerical modeling and field work designed to better understand the physiology of the AFA blooms, is needed.

Back to Table of Contents