Water-Resources Investigations Report 01-4234
Estimates of Evapotranspiration from the Ruby Lake National Wildlife Refuge Area, Ruby Valley, Northeastern Nevada, May 1999-October 2000
Seasonal estimates of ET and the corresponding extent of habitat areas were used to compute annual ET (October 1999-September 2000) in the Ruby Lake NWR (table 4). Because the area of wetland typically decreases during the summer, seasonal estimates of ET were applied to compute an annual estimate.
Major habitats on the Ruby Lake NWR are regularly inventoried by USFWS as part of the refuge's annual water-management plan. Areas of major habitats were determined from field mapping and observation by USFWS personnel in March 2000, when the maximum extent of the wetland occurred during this study. Total wetland area was estimated by determining the percentage of open water and bulrush marsh in each management unit (fig. 2). Habitat areas were determined once more in September 2000 during the minimum extent of the wetland. As the wetland area diminishes during the summer, additional playa areas become exposed. Between March and September 2000, the wetland area decreased by about 4,500 acres and the area of playa increased (table 4). During prolonged dry periods, these playas often become vegetated. Generally, habitat areas other than open water, bulrush marsh, and playas do not change, except during years of above-normal precipitation when flooding occurs in the lower-lying areas.
Summation of monthly ET totals for October 1999 through April 2000 at representative sites was used to compute winter ET for habitats in the wetland, areas of phreatophytic shrubs, and desert-shrub uplands (table 4). Similarly, summation of monthly totals for May 2000 through September 2000 was used to compute summer ET rates. To determine total ET for the remaining major habitats (meadow, grassland, and playa and bare soil), annual ground-water ET rates were derived on the basis of satellite data, adjusted to reflect total ET by adding precipitation, and seasonally proportioned to correspond with changes in habitat area.
Satellite data recently has been used in east- central Nevada to estimate regional ground-water ET based on relations between vegetation indices derived from Landsat data and measured plant cover (Nichols, 2000). Nichols (2000, p. A6, eq. 3) determined that ground-water ET could be estimated as a function of plant cover. Plant cover, in turn, can be determined on a regional scale from Landsat data using easily calculated vegetation indices (Nichols, 2000, p. B6, eqs. 9 and 10). The relation between vegetation indices and plant cover was used together with the relation between plant cover and ground-water ET to determine annual estimates of ground-water ET for meadow, grassland, and areas of playa and bare soil in Ruby Lake NWR. Satellite data used to derive plant cover and compute annual ET rates were obtained on June 10, 1985, and June 29, 1989. Based on this analysis, annual ground-water ET in Ruby Lake NWR is 2.54 ft/yr for meadow, 1.71 ft/yr for grassland, and 0.15 ft/yr for areas of playa and bare soil.
Nichols (2000, p. C12) suggests that winter ground-water ET by vegetation in east-central Nevada accounts for about 26 percent of the annual ground-water ET. Applying this percentage to the annual estimates derived from the Landsat data produces a winter ground-water ET of about 0.66 ft/yr for meadow, about 0.45 ft/yr for grassland, and about 0.04 ft/yr for areas of playa and bare soil. Similarly, estimates of summer ground-water ET are about 1.88 ft/yr for meadow, about 1.26 ft/yr for grassland, and about 0.11 ft/yr for areas of playa and bare soil. Finally, the seasonal rates of ground-water ET were adjusted to account for the volume of precipitation that fell during the data-collection period to arrive at total annual ET.
Total annual ET for meadow, grassland, and playa and bare soil were computed by adding the seasonal amount of precipitation measured at the refuge headquarters to the estimated rate of seasonal ground-water ET. Limited bulk-precipitation gage data, which was collected at the Bowen-ratio sites, suggest some spatial variability in precipitation on the valley floor, particularly during the summer. However, the data set is incomplete due to problems with vandalism and with evaporation of precipitation at the bulk-precipitation gage. Consequently, precipitation measured at the refuge headquarters (fig. 2) was used in the computation of total annual ET. Annual precipitation for the 2000 water year measured at the refuge headquarters was 0.65 ft (7.74 in.), which represents about 58 percent of the long-term annual average (13.3 in.) based on a 30-year record (1961-90). Of the total precipitation, 0.42 ft occurred during the winter (October 1999-April 2000) and 0.23 ft occurred during the summer (May-September 2000). Adjusting the seasonal ground-water ET rates to account for seasonal precipitation results, for meadow, in total winter ET during the data-collection period of about 1.08 ft/yr and total summer ET of about 2.11 ft/yr (table 4). Total winter ET for grassland is about 0.87 ft/yr and in summer about 1.49 ft/yr. Total winter ET for areas of playa and bare soil is about 0.46 ft/yr and in summer about 0.34 ft/yr (table 4). The product of the seasonal habitat areas and ET rates for each habitat was summed to compute an annual ET.
Based on the seasonal distribution of habitats and computed winter and summer ET rates, an estimated 89,000 acre-ft of water was consumed by ET on the refuge during the 2000 water year (table 4). Of this total, more than 55 percent (49,800 acre-ft) is accounted for by ET in the wetland areas. Assuming that the precipitation measured at the refuge headquarters equals the average over the wetland area, about 7,960 acre-ft (16 percent) of the annual ET from the wetland was derived from precipitation and about 41,800 acre-ft was derived from sources other than precipitation during the 2000 water year.
The amount of annual inflow water required to maintain the refuge can be determined, in part, from estimates of annual ET from the wetland. Results of this study suggest that about 49,800 acre-ft were consumed by ET from the wetland based on the seasonal extent of open water and bulrush marsh during the 2000 water year (table 4). The extent of the refuge that is flooded is directly related to the amount of annual precipitation that falls on Ruby Lake, and more importantly, to the annual amount of snow accumulation in the southern Ruby Mountains that ultimately discharges to Ruby Lake. Several years of below-average precipitation can decrease the wetland area. Conversely, several years of above-average precipitation can increase the wetland area. In 1989, after 3 years of near- to below-average precipitation, the open-water area of the wetland covered about 1,030 acres in the South Marsh (fig. 2). For 1989 conditions, Nichols (2000, C17) estimated about 26,800 acre-ft of ET from open water, marsh vegetation, and bare soil that previously was flooded. Although seasonal ET rates were not considered in this estimate, it does provide a probable lower limit of water consumed by ET in the wetland. Comparison of Nichols' preliminary estimate with results of this study suggest that annual variations of ET in the wetland area resulting from climatic variation could be on the order of 20,000 acre-ft.
Previous Section: Evapotranspiration from Habitats
Next Section: Summary
Table of Contents | Home Page