Geoscience Reference
In-Depth Information
40
30
%
20
10
0
0
2
4
6
8
10
Loss rate (mm h 1 )
Fig. 9.21 Frequency (%) of the constant loss rates derived from Pilgrim's (1966) data collection for use in design
flood estimation. The circles represent all loss rates combined as weighted averages from 101
watersheds in the US, 24 watersheds in Australia and 8 watersheds in New Zealand. The triangles
represent the smallest loss rates observed in 60 watersheds in the United States.
infiltration. Various methods have been in use to determine the initial loss, but all of
them have drawbacks. In the most obvious method the initial loss is taken as all the
rain prior to the start of the rise in the stream flow; but this is not always applicable,
because often the rain may be finished before the stream flow hydrograph shows any
rise. This difficulty can be avoided by considering as initial loss the maximal isolated
burst of rainfall observed in the record, which was not reflected in an obvious rise in
the streamflow hydrograph. Another way is to make use of a “typical delay period,”
which can be derived from the record as the delay between short, intense storms and the
subsequent start of the rise in the streamflow hydrograph; when the storm is sufficiently
intense, the initial loss can be assumed to be negligible. This period can then serve to
determine the start of the rainfall excess, i.e. the end of the initial loss period, in storms of
longer duration. Examples of the application of the loss rate methodology can be found
in the papers by Cook (1946) and Laurenson and Pilgrim (1963). Cordery (1970) has
shown how the initial loss can be related to an antecedent precipitation index, which he
used as a measure of the wetness of the catchment.
To give an idea of the values that can be expected for the constant loss rate,
Figure 9.21 presents a summary of the data collection of Pilgrim (1966). The circles
represent the loss rate frequency as a weighted average of 460 values from 101 water-
sheds in the United States, 150 values from 24 Australian watersheds and 116 values
from 8 watersheds in New Zealand; the results for the three data sets were sufficiently
similar so that they could be combined in one single curve. The triangles in Figure 9.21
represent the frequency of the smallest loss rates observed in 60 watersheds in the United
States.
Loss rate proportional to rainfall intensity: the runoff coefficient
In the Rational Method, the peak runoff rate ( Q p /
A ) (expressed as volume rate of flow
per unit catchment area) at the outlet of a catchment is assumed to be a fraction of the
Search WWH ::




Custom Search