Quantcast Precipitation and Snowmelt

The following examples illustrate the important concepts useful for applying the design storm.
The interpolated 2-year, 24-hour rainfall depth for Raleigh, North Carolina, is 3.6 in. (Hershfield,
1961; SCS, 1986). Substituting the return period of two years and rearranging equation 1.1.3 yields
the following probability.
= 0.5
There is a 50% probability that a storm of 3.6 in or greater will occur in any year. Similarly, the annual
probabilities of the 10-year and 25-year design storms are 0.1 and 0.04.
The maps contained in the rainfall frequency atlas (Hershfield, 1961 and SCS, 1986) were obtained
from smoothing, translating and interpolating rainfall data at point locations. Rainfall depths obtained
from the atlas and other similar sources should only be regarded as representative values and
uncertainties or errors are associated with each point estimate. Tung (1987) quantified the rainfall
estimate uncertainty procedures for estimating the mean and coefficient of variation of design storm
rainfall depth for 1-100 year return periods and 30 minute to 24 hour storm durations.
We suggest the 2-year, 10-year and 25-year design storms as a range of precipitation inputs which
is preferred over a single point estimate for an event-based assessment. The use of a range provides an
indication of the variability in extreme events. The 2-year design storm is the maximum expected
precipitation value for an average year. The 10-year design storm provides an intermediate rainfall and
shares the same return period as critical stream low flows (7Q10). We also recommend the (25-year
design storm as suggested by Young et al., 1982) for evaluating nonpoint source runoff as in the case of
feedlots. Maps providing rainfall depths for the 48 contiguous states for the 2-year, 10-year, and 25-
year design storms are available from (SCS, 1986; Hershfield, 1961). We suggest examining weather
records to determine average storm duration during the season of interest.
Snowmelt estimation is essential to evaluating the water balance and may transport a significant
portion of pollutant runoff in northern temperate regions. Modeling snowmelt is difficult since many
natural processes affect the snowpack energy balance including solar radiation and the changes in
sensible heat flux that vary with wind speed. Factors affecting the energy balance include radiation
exchanges at the snow surface, sensible heat exchanges with the overlaying air, condensation,
evaporation and freezing in addition to the melting phase changes of snow.
Hendrik et al. (1971) developed a snowpack energy balance model that features a daily time-step
and requires a large set of site specific daily weather data. Spatial Variability in snowmelt is
represented by selecting topographically uniform snowmelt zones with respect to slope and directional
aspect (north to south) that are assumed to produce similar snowmelt runoff. Model terms include


Privacy Statement - Copyright Information. - Contact Us

Integrated Publishing, Inc.