Interim Report on Flash Floods, Area 5 - Nevada Test Site

Page 1

ft DESERT RESEARCH INSTITUTE <11 UNIVERSITY OF NEVADA SYSTEM ON FLASH FLOODS, AREA 5- NEVADA TEST SITE WATER RESOURCES CENTER

Page 2

This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Department of Energy, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product of process disclosed, or represents that its use would not infringe privately-owned rights. Reference herein to any specific commercial product, process, or service by trade name, mark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. AVAILABLE FROM THE NATIONAL TECHNICAL INFORMATION SERVICE Price: Printed Copy

Page 3

INTERIM REPORT ON FLASH FLOODS, AREA - 5 NEVADA TEST SITE Water Resources Center Desert Research Institute University of Nevada System 1 Associate Research Professor, Water Resources Center The work upon which this report is based was supported by the U.S. Department of Energy under Contract No. DE-AC08-DPO1253.

Page 4

The potential for flash flooding in the vicinity of waste management sites is an important consideration for siting of such facilities. This report describes the research conducted during the first year of effort on potential for flash flooding in Area 5 of the Nevada Test Site. This effort is one aspect of a larger research project directed toward an assessment of the possible migration of radionuclides at NTS away from proposed and currently used disposal locations. This research was supported by the Nevada Operations Office of the U.S. Department of Energy under Contract No. DE-AC08-

Page 5

FLASH FLOOD MODELING PHILOSOPHY - METHODOLOGY ........................... 2 Previous Flash Flood Analyses ........................................................ 2 Flash Flood Modeling ............................................................................... 3 PRECIPITATION DATA ANALYSIS........................................................................... 7 Analysis of Set Hour Precipitation - McCarran Evaluation of Precipitation as aFunction of Time . . 17 SELECTION AND INSTRUMENTATION OF ASTUDYWATERSHED .... 19

Page 6

1. Summary of USGS Flash Flood Equations ........................................... 3 2. Summary of NTS Rain Gage System........................................................... 8 3. Greatest Daily Precipitation at Station Well 5B 4. Weibull Analysis of Data in Table 3................................................ 12 5. Maximum Daily Precipitation Events Corresponding to Given Recurrence Intervals for Selected Stations . . 15 6. Maximum Set Hour Precipitation Events Corresponding to Given Recurrence Intervals, McCarran International 7. Summary of Storm Precipitation in Study Watershed ... 21

Page 7

1 . Flash Flood Methodology................................................................................ 4 4. Sample of Record Produced by Weighing Bucket Gage ... 17 5. Hyetograph of an Unusual Summer Storm ............................................. 18 6. Typical Channel Cross Section ................................................................ 20

Page 8

In this interim report the primary emphasis is on documenta tion of the methods of analysis being used and the progress made to date. Thus, very few conclusions are reached in this report. Although verification of the data sets presently available is a task which must be completed before absolute conclusions can be reached, an examination of the presently available data indicates that consideration must be given to the possibility of flash floods when siting waste management facilities in Area 5 of the Nevada Test Site. The following specific points should be noted. 1. A watershed with washes originating in it which flow past the radioactive waste management site has been selected as the study watershed. 2. In the upper reaches of the study watershed twenty crest stage gages have been installed at nine locations to monitor any flow event which may occur during the term of the study. 3. At each location where crest stage gages have been installed, the existing channel cross section has 4. Within the study watershed, five recording bucket rain gages have been installed. In addition, a recording

Page 9

rain gage has been placed at the waste management site. Three precipitation events have already been measured by this gage system. 5. All of the available daily precipitation data for all rain gages at the Nevada Test Site (NTS) has been obtained and digitized. 6. A number of probability distributions: Gumbel, log normal, gamma and log Pearson Type III, have been fit to the daily precipitation records. It is now possible to estimate the maximum daily precipitation event at any NTS station for a given recurrence interval. 7. The daily and set hour precipitation records for McCarran International Airport have been analyzed. 8. The chart records for the recording rain gage at Well 5B have been obtained and analyzed. It is noted that these records apparently are available back to 1963 with the year 1965 missing. At the present time the records for the station at Yucca are being analyzed. 9. A rough, first cut, permeability map of the study water shed has been made. 10. A literature review of both watershed and flash flood models is in progress.

Page 10

The determination of flood flow frequency and the prediction of peak flood flows is a subject area which has received wide attention in the past. In many areas of the country, the proce dures for establishing flood flow frequencies and estimating the peak flood flow are well defined (U.S. Water Resources Council, 1977). However, in the arid Southwest the prediction of flash flood frequencies and peak flash flood flows is a serious problem which has received very little attention. The primary problem in studying flash floods is the lack of a reliable data base; and although some work in this area has been done, (Magura and Wood, 1980; Christensen and Spahr, 1980; Williams, 1957a,b) in general the prediction of flash flood frequency and peak flow must be considered a developing area of technology. In this report, the flash flood modeling philosophy-methodol ogy is first briefly described and then the progress the investi gators have made studying the historical precipitation records for the Nevada Test Site (NTS) is described in detail. In addition progress made in areas other than precipitation analysis is

Page 11

FLASH FLOOD MODELING PHILOSOPHY-METHODOLOGY During the winter months, storms associated with broad low pressure systems that have developed over the Pacific Ocean move eastward over the southern Nevada area. The precipitation which results from these storms is generally widespread and only rarely intense. Christensen and Spahr, (1980), have assumed that these storms do not constitute a flash flood hazard; however, this may not be a valid conclusion. During the summer months local convective thunderstorms associated with moisture from the Gulf of California and the Southern Pacific Ocean move northeastward over the southern Nevada area. Precipitation from these storms is generally localized and can be intense; thus, this latter type storm is the type which is generally assumed to present a flash PREVIOUS FLASH FLOOD ANALYSES The only known previous analysis of flash floods at NTS was performed by the United States Geological Survey for Topopah Wash (Christensen and Spahr, 1980). In this investigation, regression equations were used to predict the peak flash flood flows for various return periods. The equations used in this analysis are summarized in Table 1. In general, this type of empirical analysis should be avoided unless there exists a specific analysis for the

Page 12

TABLE 1. SUMMARY OF USGS FLASH FLOOD EQUATIONS A * watershed area, square allies 2 a mean basin altitude, thousands of feet L 3 latitude of basin minus 35° area under study. In the case of NTS, a specific analysis does not exist. The primary problems with using such empirical analyses outside of the area for which they were developed are: 1. There is no method for determining if the analysis is conservative for the area under study, and 2. The regression equations do not account for local considerations such as basin permeability, watershed slope and the degree of channelization which could be critical parameters in a specific case. FLASH FLOOD MODELING Figure 1 summarizes in schematic fashion the flash flood modeling methodology presently being followed. The following specific comments regarding this figure should be noted; 1. Hydraulic Field Work: The hydraulic field work has consisted of emplacing a nominal number of crest stage

Page 13

FLASH FLOOD METHODOLOGY FIGURE 1. FLASH FLOOD METHODOLOGY gages throughout the watershed selected for study, Figure 2. This work is being done in the expectation that a flood event may occur sometime during the term of the study. If such an event occurred, it would more than justify the very nominal amount of time and effort which has been spent on this task. In addition to the crest stage gages, channel slopes and cross sections are also being measured. These data are necessary to the modeling effort regardless of whether a flow event occurs or not.

Page 15

2. Precipitation Field Work: The precipitation field work has consisted of emplacing five weighing bucket rain gages in the study watershed to measure the intensity and duration of precipitation events. A recording rain gage was also installed at the waste management site. It should be noted that three precipitation events have already been measured by this rain gage network; these events are described in detail in a subsequent section. 3. Analyses of Historical Precipitation Records; This is the primary subject of this report and is described in detail in the subsequent sections. 4. Unsaturated Flow: This aspect of the total project is essential to the flash flood component in that the percentage of the precipitation which infiltrates and does not appear as runoff must be known for the modeling . 5. Paleohydrology: The work in paleohydrology serves as a basic check on the rest of the flash flood work; i.e., by studying the paleohydrology of the area estimates of previous extreme events can be made on the basis of the debris flow mechanics and geomorphology. 6. Watershed Model; All of the above work areas are intended to serve as an input to the selection or development of an appropriate watershed model. The watershed model is essentially a rational method of estimating the quantity of water which a given precipi tation event would produce at the head of the alluvial fan above the waste management site. 7. Hydraulic Models: The hydraulic models are rational methods of estimating the flow and erosion as functions of time which would result in the vicinity of the waste management site as a result of a given precipitation In general, all of the tasks described by Figure 1 are interlock ing and essential to the development of an adequate method of risk analysis, from the perspective of flash floods, for the waste

Page 16

PRECIPITATION DATA ANALYSIS The historical precipitation records available as a data base for this project consist of forty stations located on the Nevada Test Site and the weather station at McCarran International Airport in Las Vegas, Nevada. The essential information regarding the NTS rain gage system is summarized in Table 2 and the rain gage locations are plotted on an NTS base map in Figure 3. With regard to the information contained in Table 2 and Figure 3, the following should be noted: 1 . The station with the longest period of record is Station 4JA where twenty-one years of daily precipita tion records are available. 2. At many stations the period of record is too short to be of use in any kind of an extreme event analysis. 3. The stations in closest proximity to the study area are Yucca, Well 5B, GMX and Cane Springs. Of these stations, GMX has a very short period of record while the other three stations have comparatively long periods of record. It should be noted that the U.S. Water Resources Council (1977), dealing with the analyses of flood flow frequency from the viewpoint of flow records recommend that the minimum period of record required for analysis is ten years. This recommendation also undoubtedly holds true for precipitation event analyses. 4. Although many of the gages enumerated in Table 2 were or are recording rain gages, no attempt to analyze these records had been made previous to this study.

Page 17

TABLE 2: SUMMARY OF NTS RAIN GAGE SYSTEM Full Years of Daily Area 12 Mesa ME 635,900 894,000 103 3

Page 19

At the McCarran International Airport Station there are essentially forty-three years of daily precipitation records. Although this station was relocated from what is now Nellis Air Force Base to its present location in December, 1948, there is no reason to suspect that this move of 12.6 miles had any significant impact on the daily precipitation record. This station also has twenty-eight years of set hour precipitation records. The McCarran Station represents the longest and probably most accurate precipitation data set available in southern Nevada. Since this project was initiated, all of the daily precipita tion records listed in Table 2 have been digitized to form a co herent data base. In addition, the recording rain gage records for Well 5B and Yucca are being analyzed. It should be noted that for Well 5B only nine years of weighing bucket rain gages records have been located. Efforts are continuing to locate additional records. From the viewpoint of flash floods, precipitation as a function of time is an essential parameter and every effort is being made to locate records of this type for the four stations in the vicinity of the waste management site. ANALYSIS OF DAILY PRECIPITATION RECORDS As noted above, the daily precipitation records for all the NTS precipitation stations have been digitized. From these records, the maximum event for each year has been identified and for each of these events a recurrence interval has been computed by the Weibull formula. It is noted that this procedure is a standard engineering method and a complete description of the method can be found in any standard hydrology text; e.g., Viessman

Page 20

As an example of the analysis used, in Table 3 the- maximum daily precipitation event which occurred each year at Station Well 5B is tabulated. In Table 4 these data have been arranged from largest to smallest and the Weibull probability calculated by: where Pw(r>R) = the Weibull probability of an event R being equalled or exceeded by an event r, N = total number of points comprising the data set and m = rank of event R. The Weibull recurrence interval is then computed as: where Tw = Weibull recurrence interval in years. For example, in Table 4 the probability that a precipitation event of 1.04 inches per day will be equalled or exceeded in any given year is 0.18 and thus we can expect an event of this size to occur about TABLE 3. GREATEST DAILY PRECIPITATION AT STATION WELL 5B FOR YEARS 1964-1979

Page 21

TABLE 4. WEIBULL ANALYSIS OF DATA IN TABLE 3 Once the plotting positions have been established, a probability distribution which satisfactorily describes the data must be found so that predictions of rainfall for a given recurrence interval may be made. For the daily precipitation data considered here, four types of probability distributions were 1. the Gumbel extremal distribution, 2. a shifted lognormal distribution, 3. a shifted Gamma distribution, …