Open-Water Evaporation Results


Average open-water evaporation rates during the energy budget periods ranged from 2.8 mm d^-1 in early October 2008 at the MDL site (MDN was not usable in 2008) to as much as 7.0 mm d^-1 during mid-July 2010 at the MDL site (table 15). Evaporation totals from June 12, 2009, to October 2, 2009, were 615 and 601 mm at the MDL and MDN sites, respectively. The difference between totals at the sites was about 2.3 percent. Evaporation during the same period in 2010 was slightly larger, totaling 636 and 611 mm at MDL and MDN, respectively. Evaporation totals for the entire period of data for 2010, May 29 through October 2, which was two weeks longer than the data collection period in 2009, were 701 and 675 mm at the MDL and MDN sites, respectively. The difference between the 2010 totals at the two sites was about 3.8 percent.


Open-water evaporation measurements generally correspond to the period from June through September. About 61 percent of annual pan evaporation at Klamath Falls occurs during this period, based on mean monthly values from 1949 to 2004 (Western Regional Climate Center, 2012). The surface of Upper Klamath Lake is typically frozen during winter months.


During the study period, water surface temperatures of the lake were highest from late July through the first week of August. In 2008, however, the highest biweekly average surface temperature occurred during the period that extends through mid-August (table 15). Peaks in average surface temperatures tended to lag peaks in net radiation (fig. 20; table 15). Energy-budget evaporation rates are generally at their highest when net radiation is highest (fig. 21), but rates are also affected by other factors, chiefly net changes in energy stored in the lake and the dominant type of heat transfer occurring in the lake as described by Bowen ratios (table 15).


Where data were available to calculate evaporation at both the MDL and MDN sites, differences in evaporation rates averaged about 0.2 mm/d, and the median difference for all energy budget periods was 0.1 mm/d. The maximum difference observed was 0.7 mm/d (table 15). The similarity of evaporation rates determined at the geographically separate MDL and MDN sites indicates that, at the time scales of the 2-week energy-budget periods, the physical conditions controlling the Bowen ratio were similar at both sites. Given that water depth and fetch conditions at these two sites are representative of most of the lake, and that the lake is reasonably well mixed, results from the MDL and MDN sites should be reasonably representative of the lake as a whole.


Comparison of Open-Water Evaporation with Previous Studies and With Wetland Evapotranspiration


Biweekly energy-budget evaporation rates calculated for budget periods in 2008, 2009, and 2010 are similar to evaporation rates determined in previous studies of Upper Klamath Lake. Janssen (2005) determined that 2003 energy budget evaporation rates from June 7 to September 30 averaged 4.2 mm/d, somewhat less than the 5.5 mm/d average measured during a similar period in 2009 and 2010 in this study in spite of slightly warmer monthly average temperatures in 2003. Janssen’s average, however, excludes data from 2 days in late June along with a period of 13 days between June 26 and July 8. Because near-peak evaporation occurs during these missing periods, Janssen’s average probably would have been greater if the missing data had been available.


Using a 1-D surface energy balance model to produce simulations of Upper Klamath Lake daily evaporation for 1950 through 2005, Hostetler (2009) found average May to September evaporation totaled 707 mm. This compares favorably with the late-May to September 2010 totals of 698 and 680 mm measured at the MDL and MDN sites, respectively, during this study.


Seasonal trends in open-water evaporation and wetland ET were similar (figs. 15, 21). In general, open-water evaporation exceeded wetland ET during the periods when both were measured. As expected, E–ET was greatest during the late summer periods (late August to October), due to the greater release of stored heat in the lake compared to the land surface. Differences were smaller during midsummer (late June to early August), a time when vegetation was at full height and stored heat was not yet a factor. A notable exception occurred in late June through mid-July 2010, when E–ET was unusually large. The low lake levels in 2010 probably contributed to this difference through reduced evaporation of standing water from the wetland, as discussed in Daily Evapotranspiration and Crop Coefficients. Early‑season (late May to mid-June) data are spotty (only three periods), but consistently show substantially greater open-water evaporation, as the green vegetation canopy has not fully emerged from the dead stalk mat, and replaced it as the primary exchange surface. Overall, during the periods of open-water data collection, open-water evaporation (the mean of the MDL and MDN sites) was 20 percent greater than wetland ET (a 70 percent-30 percent weighted mean of the bulrush and mixed sites, respectively).


First posted March 4, 2013