FLOOD07101 - Laserfiche WebLink

Home Browse Search

J X;~t-IAJStJJJ fMM 0AJ:/<?FTl 111 ~-\~!J!!RESOURCES RESEARCH, VOL, 35, NO, 3, PAGES 785-796, MARCH 1999 A numerical modeling study to investigate the assumptions used in the calculation of probable maximum precipitation Deborah J. Abbs CSIRO Atmospheric Research, Aspendale, Victoria, Australia Abstract. A numerical model of the atmosphere has been employed to evaluate the assumptions used in the simple two-parameter model that is utilized for many probable maximum precipitation (PMP) calculations. These assumptions are (1) the precipitation is linearly related to the precipitable water; (2) the precipitation efficiency of the storm does not change as the moisture available to the storm increases; and (3) terrain modulates the distribution of the precipitation but does not affect the synoptic-scale dynamics of the storm. A single case study is used to illustrate the techniques employed and to describe the results that were common to four case studies. We show that long-lived, moderate rainfall processes are important contributors to the total precipitation produced by the storm. Increases in the moisture availability result in the heavy rainfall beginning earlier, lasting longer, and being more continuDus. As the moisture availability changes, the spatial distribution of the area over which more than 50% of the total rainfall falls as heavy rainfall changes. As the moisture availability is increased, the precipitation efficiency of the storm does not change significantly. Terrain effects are shown to have an effect on the amount of rainfall that occurs over the higher terrain as well as on the distribution of the rainfall due to the "convergence component" of the stonrr. Despite these deficiencies in the assumptions used to estimate PMP, improvements in the estimation of PMP may soon be possible if.increased effort is placed on (amongst other things) the numerical modeling of extreme rainfall events. These improvements are only possible if the results of these effDrts are communicated to the hydrological community. 1. Introduction In this paper, we investigate whether numerical models of the atmosphere may be used as tools for quantitative precipi- tation fon::c;n"tins, oyer catchment lJb~~ r"glunlJ, for t.1xtr4:3mo precipitation events. The concept of probable maximum pre. cipitation (PMP) is used by hydrologists and meteorologists involved in the design of structures, such as dams or bridges, where the need is to compute extreme rainfall (and hence flODd) events, World Meteorological Organization (WMO) (1986, summary] defines the PMP as "the greatest depth of precipitation for a given duration meteorologically possible over a given size storm area at a particular location at a par- ticular time of the year, with no allowance made for long-term climatic trends." It is thus seen as a single deterministic num- ber (governed by ph)'Nif:ol primljpl"u,) U1Mt wuuld nuvur bo 91P ceeded. However, there have been several occasions on which Dbserved rainfalls have exceeded the PMP estimates valid at the time; the occurrence of such events led to a reevaluation of the PMP methodologies and resulted in the development of the generalized techniques that are now used throughout Aus- tralia. A more practical problem associated with an increase in PMP estimates is that if the original PMP estimates are sub- sequently revised upward, this will involve expensive remedial works to upgrade underdesigned spillways. In recent times hydrolosists hove called for n rQ"pprul.ul of the phlloawphJc:.1 underlying dam safety practices [e.g" Laurenson and Pearse, Copyright 1999 by the American Geophysical Union, Paper number 1998WR900013. 00-13-1397/99/1998WR9000 t3$09,oo 1991]. In ffieUnited States a study [National Research Council, 1994] noted the importance of develDping improved hydrom- eteorological analysis procedures for assessing extreme precip- itation. B. Stewart (private communication, 1997) notes the need to "develop a consistent mClhodology for altlmolins PMP that accounts for the natural variability of climate and our relatively short historical records." Consequently, there has been pressure to validate independently, and to impruve, the current PMP methodology. The major rationale fqr the present study was to provide the first steps toward using nu- merical modeling techniques to enhance current PMP proce- dures. Because of the potential hazards associated with overtop- ping of large dams, a high degree of security in the calculation of the spillway design flood is required. The Australian Na- tional Commlnee On Lllrg. O.rno (^NCQ!'P) [l9R6j II.fin.. the annual exceedance probability (AEP) to be the probability that a particular flood value will be exceeded in any year. In Australia the AEP of the design flood is estimated from the frequency analysis of rain/flood records, all of which are typi- cally <100 years in duration. This method of analysis can provide reliable AEP estimates only down to _10-2, certainly insufficient security for a large dam. The design of large dams is based on the much more conservative estimate of the prob- able maximum flood (PMF), which in turn depends on the estimate of the PMP. PMF Is estimated using a design storm m~thud b.."d un Ihv PMP. [n turn, PMP has been historically estimated in a variety of ways using techniques that maximize recorded storms. When the storms maximized are only those that occur on the catchment under consideration, the method is called the "in 70.