|
<br />
<br />
<br />.- "',""
<br />ud:;;~)
<br />
<br />FIRST PERIOD
<br />
<br />t
<br />
<br />
<br />U2
<br />
<br />UPSTREAM
<br />SITE
<br />~
<br />
<br />~
<br />o
<br />-J
<br />U.
<br />::;;
<br /><{
<br />w
<br />ex:
<br />f0-
<br />CI)
<br />
<br />. DOWNSTREAM
<br />,,/
<br />" SITE
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />
<br />".
<br />
<br />UI
<br />
<br />SECOND PERIOD
<br />
<br />THIRD PERIOD
<br />
<br />"
<br />"
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />,
<br />
<br />UNGAGED TRIBUTARY FLOW WAS
<br />/ ASSUMED IN THE REACH, AS
<br />INDICATED BY THIS PEAK.
<br />THEREFORE, THIRD PERIOD WAS
<br />,", NOT USED FOR ANAL YBIS.
<br />.,
<br />. ,
<br />. ,
<br />. ,
<br />. ,
<br />. ,
<br />. ,
<br />,
<br />.
<br />.
<br />.
<br />.
<br />. '.
<br />
<br />"
<br />
<br />
<br />'..
<br />
<br />TIME, IN DAYS ~
<br />Figure 9. Example of selection of streamflow-event periods for analysis of daily streamflow gain and loss.
<br />
<br />The statistical summary of daily estimates for
<br />reach 2 (table 3) indicated that streamflow gains and
<br />losses between -4.9 and 6.0 ft3/s were common during
<br />all four seasons. Although large daily gains and losses
<br />were computed (table 3), generally most of the values
<br />were less than :1:20 percent of the daily mean stream-
<br />flow. During winter, gain-and-Ioss computations were
<br />made for 718 days, resulting in a daily median loss
<br />of zero; during spring, gain-and-Ioss computatious
<br />made for 736 days had a daily mediau gain of 2 ft3/s;
<br />during summer, gain-and-Ioss computations made for
<br />752 days had a daily median loss of I ft3/s; and during
<br />fall, gain-and-Ioss computations made for 753 days
<br />had a daily median loss of I ft3/s (table 3). Statistical
<br />analysis, using a two-tailed t-test (Iman and Conover,
<br />1983), was used to assess whether daily streamflow
<br />gains and losses were significantly differeut from
<br />zero during some seasons. A two-tailed hypothesis
<br />test assumes that the difference can be greater
<br />than or less thau zero. Results of the test using the
<br />1984-92 estimated streamflow gaiu-and-Ioss data
<br />indicated that only during the fall were daily gains or
<br />losses in reach 2 significantly different from zero at
<br />the 95-percent confidence level.
<br />In addition to the more than 2,900 daily
<br />computatious of streamflow gains and losses in
<br />reach 2, boxplots are shown in figure 10 that
<br />summarize streamflow gain-and-Ioss calculations
<br />for more than 100 selected streamflow events between
<br />1984 and 1992. Gain-and-Ioss computations for
<br />
<br />
<br />8 winter streamflow events that were analyzed had a
<br />daily median gain of 1 ft3/s; for 26 spring streamflow
<br />events, a daily median gain of 1.5 ft3/s; for 48 summer
<br />streamflow events, a daily median loss of 3.0 ft3/s; and
<br />for 24 fall streamflow events, a daily median loss of
<br />1.5 ft3/s (table 3). The streamflow-event estimates in
<br />reach 2 ranged from a daily gain of 30 ft3/s to a daily
<br />loss of 58 ft3/s (table 3).
<br />
<br />Reach 3
<br />
<br />Analysis of daily mean streamflow gains and
<br />losses for reach 3 was made for more than 3,200 days
<br />between 1984 and 1992. A constant I-day advance
<br />(traveltime adjustment) of the upstream flow, which
<br />was based on a significant correlation between
<br />upstream and downstream streamflow of 0.74
<br />(p=O.OOOI), was used in reach 3 for the computation of
<br />daily streamflow gains and losses. Visual inspection
<br />of many daily streamflow hydrographs for sites Q3
<br />and Q4 also indicated that a I-day advance of the
<br />upstream hydrograph was an appropriate traveltime
<br />adjustment when using daily streamflow data. The
<br />boxplots of these calculated streamflow gains and
<br />losses are shown in figure 11; a negative value
<br />indicated a streamflow gain and a positive value
<br />indicated a streamflow loss. Median values of stream-
<br />flow gains and losses in reach 3 indicated a daily
<br />streamflow gain during every season (fig. 11).
<br />
<br />STREAMFLOW GAINS AND LOSSES 19
<br />
<br />J
<br />
|