FLOOD01413

• County: Elbert
• Title: Use of Rainfall-Simulator Data in Precipitation-Runoff Modeling Studies
• Date: 1/1/1983
• Prepared For: Elbert County
• Prepared By: USGS
• Floodplain - Doc Type: Educational/Technical/Reference Information

Rainfall-simulator runs made during 1982 were conducted to investigate
the influence of variable rainfall intensity on runoff, and to confirm the
trends in fitted values of hydraulic conductivity determined from the 1981
experiments. Prevailing weather conditions during the spring and summer of
1982 presented a substantially different set of antecedent soil-moisture
conditions than those of the summer of 1981. Frequent showers and thunder-
storms were typical of the 1982 summer season, as opposed to the extremely
dry conditions of 1981. Only one rainfall-similation run conducted during
the summer of 1982 could be considered as having a dry antecedent condition.
The results of the sequence of runs conducted in 1982 are depicted in figures
16, 17, and 18. Comparative summer runs on the upland ponderosa area are
shown in figure 16; similar results for plots 3-5 in the lowland prairie area
are shown in figure 17; and comparative runs on plots 1 and 3 made during the
fall are shown in figure 18.

'I

Results of fitting parameters for plot 1 data (fig. l6A) and the second
run on plot 2 (fig. l6C) show close correspondence between observed and com-
puted runoff. A large discrepancy between observed and computed runoff is
evident for the first run on plot 2 (fig. l6B). The computed response rises
too rapidly and overpredicts the slowly rising limb of the observed hydrograph.
Computed results do not start to converge to the observed runoff rate until
about 30 minutes into the run. Soil temperature at 0.5 in depth was 1020F
at the start of this run. Near-surface soil temperature dropped to about
870F after 15 minutes of 2 in/h rain application and then stabilized at 770F
at 35 minutes into the run; (rainfall rate increased from 2 to 4 in/h at 27
minutes, then dropped back to 2 in/h at 39 minutes). Soil temperature at
0.5 in depth also was high, 960F, at the start of the second run on plot 2
(fig. l6C). However, the temperature dropped very rapidly to 840F after 6
minutes of 4 in/h rain application and changed little for the duration of
the run. Soil temperature at the start of the run on plot 1 (fig. l6A) was
about 850F, dropped to 760F after 12 minutes of 2 in/h' rainfall application,
and changed little thereafter.

The best-fit parameter values for the summer runs on plot 2 (figs. l6B
and l6C) compare reasonably well with those developed from 1981 data. Values
for KSAT in the range of 1.0 to 1.2 apply to all four summer runs. Hydraulic
conductivity is the controlling parameter as noted in the fitting of 1981
simulator results. The fitted result for the run shown in figure l6B could
be improved by an increase in the surface-retention capacity to about 0.2 in;
however, the value of 0.05 in seems more appropriate and consistent with other
results. The fitted value for KSAT of 1.5 in/h shown in figure. l6A (plot 1)
is higher than previously determined values from the summer runs of 1981.

A more intensive before- and after-run soil-sampling effort than that
of the previous summer was undertaken for the June 23 and July 8, 1982 runs
on plots 1 and 2 (figs. l6A and l6B). Results shown in table 5 demonstrate
very pronounced variability in cumulative infiltration from point to point.
Point-to-point variability in water uptake is related to depth of wetting
rather than to large differences in degree of saturation, which is a con-
firmation of previous results.

27

I