2014-05-15_REVISION - M1987064

• Permit No: M1987064
• IBM Index Class Name: REVISION
• Doc Date: 5/15/2014
• Doc Name: Response to adequacy review comments TR04
• From: Elam Construction, Inc.
• To: DRMS
• Type & Sequence: TR4
• Email Name: KAP
• Media Type: D
• Archive: No

D.C. Goodrich et at /Agricultural and Forest Meteorology 105 (2000) 281 -309
the ground and remote sensing measurements. Sec-
tion 4 describes the theory and modeling approach
to scale in situ measurements as well as the water
balance used to assess the validity of the scaling and
modeling approach. Results are presented and dis-
cussed in Section 5. A brief summary is contained
in Section 6. This is followed by conclusions and
recommendations contained in Section 7.
2. Background
The direct and indirect influence of riparian ET
on basin surface and groundwater resources has been
noted in numerous prior studies (Croft, 1948; Lines,
1996; Bowie and Kam, 1968; Federer, 1973; Singh,
1968; Culler et al., 1982). Both Bowie and Kam
(1968) and Federer (1973) noted significant impacts
on streamflow by ET which were further confirmed
by examination of streamflows after herbiciding or
eliminating riparian vegetation (Culler et al., 1982). A
variety of prior methods have utilized these observed
impacts on adjacent surface water and groundwater
to indirectly estimate riparian ET.
Early methods analyzed baseflow recession curves
at various times of the year to estimate riparian ET at
monthly and seasonal time scales (Langbein, 1942;
Riggs, 1953; Whelan, 1950; Hall, 1968). On a shorter
time scale Tschinkel (1963) and Reigner (1966) uti-
lize diurnal hydrograph changes to compute daily
riparian ET. Daniel (1976), in a similar manner to
Tschinkel (1963), presents a method to compute ET
over several months based on the difference between
observed streamflow and the theoretical "potential
streamflow" determined from a recession index. Lines
(1996) utilized this approach to estimate riparian ET
on the Mojave River for selected years in which sur-
face runoff was minimal by integrating the estimated
streamflow depletion throughout the riparian growing
season. Corell et al. (1996) and Goodrich et al. (1998)
carried out a similar analysis for the San Pedro River
in southeastern Arizona. While periods without ET
and runoff recharge can typically be found by ex-
amining various periods of the discharge record, the
requirement for absence of pumping in this approach
is typically more restrictive in any developed basin.
When storm runoff does occur, the method also as-
sumes baseflows can be easily separated from storm
283
runoff. As noted by Maddock et al. (1998), this is not
a simple task. Another difficulty with this method is
determining the areal extent over which the riparian
ET affects streamflow. If a reach becomes intermit-
tent, this type of analysis is not applicable even though
substantial riparian vegetation may be present.
Gatewood et al. (1950) provide a good description
of six methods other than hydrograph analysis that
were employed to estimate riparian water use in the
Lower Safford Valley in Arizona. The six methods
were: (1) tank or lysimeter, (2) transpiration -well, (3)
seepage run, (4) water balance (inflow — outflow), (5)
chloride- increase, and (6) slope- seepage. All of these
methods estimate riparian ET indirectly through its
observed effects on related stream or aquifer charac-
teristics. As in the case of integrating streamflow de-
pletion, the transpiration -well, chloride- increase, and
slope- seepage methods all exhibit the same difficul-
ties of estimating the areal extent of riparian vegeta-
tion responsible for the indirectly determined riparian
water use. In a well - controlled tank or lysimeter, ac-
curate ET estimates can be obtained via a water bal-
ance calculation but scaling -up these measurements to
the riparian corridor level has been problematic. Lines
(1996) noted that lysimeter or tank ET measurements
(e.g. Gatewood et al., 1950; van Hylckama, 1980) did
not agree well with annual ET estimates derived from
streamflow depletion along the Mojave River. Weeks
et al. (1987) noted that the high rates measured by
lysimeters or tanks are likely due to oasis and plant
age effects and are likely to represent maximum ET
rates occurring in a non -water limited situation. The
seepage run method can only be applied in reaches
with perennial flow and with flows large enough to be
accurately measured by current meter methods. Both
of these conditions are violated in the San Pedro Basin
as well as numerous semi -arid riparian systems.
Other meteorologically based methods such as the
often employed Blaney — Criddle (Shuttleworth, 1993)
approach have also been employed to estimate ripar-
ian ET. In this empirical, temperature -based method,
ET rates are scaled by vegetation area to derive
monthly ET estimates. However, temperature -based
ET- estimation methods are not recommended unless
this is the only available data source (Shuttleworth,
1993). Shuttleworth (1993) also notes that while the
complexity of the Blaney — Criddle equation "is a
tribute to the loyalty of its proponents, it precludes