Scientific Investigations Report 2006–5212
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006–5212
In December 2004, the USGS provided a hydrologic assessment of the Klamath Basin Water Bank Program for the Bureau of Reclamation (Reclamation) Klamath Basin Area Office (U.S. Geological Survey, written commun., 2005). In the assessment, inconsistencies were found in the river and canal flow records from sites along the Klamath River between Klamath Falls and Keno, Oregon, for 1961–2004. A simple water balance estimated from the records indicated that the Klamath Falls to Keno reach was a losing reach in the 1960s and 1970s and a gaining reach in the 1980s and 1990s. However, ascertaining the cause of these flow inconsistencies was not possible because of the limited scope and time available for the water-bank assessment. The inconsistencies prompted the following questions:
As a result of these unanswered questions, Reclamation requested that the USGS perform a more in-depth analysis of historic and current Klamath Project diversion and return-flow data in order to accurately quantify uncertainty in the data and the benefits of the water-bank operations.
The scope of this assessment was limited to the streamflow-measurement sites directly related to diversion and return flows to and from the Klamath River reach between the Link River at Klamath Falls and Klamath River at Keno sites. The “A” Canal is not in this reach of the Klamath River and therefore was not included in the water balance.
The assessment included:
The assessment study area, also termed “water-balance reach” in this report, is located near Klamath Falls in south-central Oregon (fig. 1). The study area extends from the USGS streamflow-gaging station on the Link River at Klamath Falls (11507500) to the USGS gaging station on the Klamath River at Keno (11509500). The Link River is a short river linking Upper Klamath Lake with Lake Ewauna. Lake Ewauna extends from Klamath Falls about 2 mi downstream before it becomes the Klamath River. The flow-measurement site at the lower end of the study area (RM 231.9) is about 1.5 mi downstream of the Keno Dam.
The Klamath Project provides irrigation water to about 240,000 acres and to the Lower Klamath and Tule Lake National Wildlife Refuges (Bureau of Reclamation, 2000; 2006). Streamflow-measurement sites assessed in this report are a subset of the sites in the Klamath Project. However, these sites were selected because the water-bank assessment (U.S. Geological Survey, written commun., 2005) identified concerns about their accuracy and therefore the accuracy of measured flows to and from the Klamath River in the water-balance reach. The reach also includes non-Klamath Project diversion and return flows associated with agriculture, municipal wastewater, and the wood-products industry in the study reach. The possible influence of these flows on our understanding of Klamath River discharge in the reach also was assessed in this study.
The sum of the inflows and outflows measured at eight sites was used to compute a surface-water balance for the reach of the Klamath River between Link River Dam and Keno Dam (table 1, fig. 2). The resulting calculated net flow was then subtracted from flow at the USGS measurement site at Keno (11509500) to determine whether the reach was losing or gaining. Flow at the four diversion and four return flow-measurement sites were negative and positive terms in the water balance, respectively. If all inflows to and outflows from the reach could be accurately documented and measured, the right side of the equation shown below (Qresidual) would be zero (table 2). However, if that condition is not met and the right side of the equation is nonzero, a gaining reach will have a positive value and a losing reach will have a negative value. Flow-measurement errors can result in a false characterization of a reach as gaining or losing.
Qdownstream_outflow – (Qupstream_inflows – Qupstream_outflows) = Qresidual (1)
where:
| Qdownstream_outflow | is the flow at the downstream Klamath River streamflow-gaging station at Keno, |
| Qupstream_inflows | is the sum of upstream inflows, |
| Qupstream_outflows | is the sum of upstream outflows, and |
| Qresidual | is the unaccounted for gain or loss in flow for the reach. |
The Klamath Project flow-measurement sites are operated by Reclamation and irrigation and drainage districts. The Link River flow data used in the analysis were compiled by PacifiCorp and Reclamation. The Klamath River at Keno streamflow-gaging station (11509500) is operated by the USGS. Monthly flow records from all sites were used to compute the monthly net sum water balance (table 3).
All flow data used in the water-balance analysis came from Reclamation’s hydrologic database with the exception of the USGS Klamath River at Keno (11509500) flow record. All flow records spanned 44 years (1961–2004), except for the Miller Hill Pumping Plant return flow record, which began in water year 1987. (A water year is defined as the 12-month period from October 1 to September 30.)
For most months of the year, water in Lost River Diversion Channel flows toward the Klamath River. During the summer, the flow can be reversed when water is diverted from the Klamath River. Until recently, there has not been a gaging station on the Lost River Diversion Channel close to the Klamath River. Four streamflow-measurement sites (Station 48, Miller Hill Pumping Plant diversions, Miller Hill Pumping Plant return spills, and Lost River Diversion Dam), all located on the Lost River Diversion Channel, were included in the water-balance equation to compute the net flow of Lost River Diversion Channel.
As shown in table 3, there is a trend in net Klamath River flows from negative to positive values. Positive values indicate a gain to the Klamath River in the water-balance reach and negative values indicate a loss of water in the reach. In the 1960s and 1970s, most of the values were negative for October through April, indicating a net loss from the river. In the 1980s and 1990s, almost all values for these months are positive, indicating a net gain to the river. Likewise, many values for May–September changed from double digit negative to single digit negative. The inconsistency is also evident in figure 3, which shows a relation between the sum of April–September net canal flows and April–September net river flows used in the water balance. Net canal flows were calculated as the sum of the diversion flows minus the sum of the return canal flows. Specifically, these include the North Canal, Ady Canal, Miller Hill Pumping Plant, and Station 48 diversion flows minus the Lost River Diversion Dam, Miller Hill Pumping Plant spills, and Klamath Straits Drain F-FF Pumping Plant return flows. Net river flows were computed as Klamath River at Keno flows minus Link River flows.
In figure 3, data points for the first and second 22-year periods (1961–82 and 1983–2004) are separated by a regression line that was computed from data for the entire period. Such a sharp separation or shift for the two time periods is an indication of flow data error or changes in undocumented flows that were not included in the water balance. If there were no water-balance inconsistency over the 44-year period, data points from the first and second 22-year periods would be more evenly distributed on both sides of the regression line. Mean monthly and mean annual net water-balance flow volumes in thousand acre-feet are shown for the two 22-year periods in table 4. The mean annual net water balance for the first and second 22-year periods (1961–82 and 1983–2004) are about –68,000 and 35,000 acre-ft, respectively. The absolute difference between the two periods is about 103,000 acre-ft/yr.
U.S.
Department of the Interior | U.S. Geological
Survey
Persistent URL: https://pubs.water.usgs.gov/sir20065212
Page Contact Information: Publications Team
Page Last Modified: Thursday, 01-Dec-2016 19:16:42 EST
