1、NASACONTRACTOR NASA CR-2289 REPORT o* 00 N N = U I VARIATION OF THE LOW LEVEL WINDS DURING THE PASSAGE OF A THUNDERSTORM GUST FRONT by R. W. Siizchir, R. A, Atzthes, utzd He A. Pmofsky Prepared by THE PENNSYLVANIA STATE UNIVERSITY University Park, Pa. 16802 for George C. Alnrsbnll Space Flight Celrt
2、er NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. JULY 1973 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2. GOVERNMNT ACCESSION NO. . REPORT NO. NASA CR-2289 I. TITLE AND SUBTITLE Variation of the Low Level Winds During the Passag
3、e of a Thunderstorm Gust Front I 1. AUTHOR(S) 8. PERFORMING ORGAN1 ZATlON REPOR r j R. W. Sinclair, R. A. Anthes, and H. A. Panofsky 3. RECIPIENTS CATALOG NO. 5. REPORT DATE July 1973 6. PERFORMING ORGANIZATION CCOE Mi13 3. PERFORMING ORGANIZATION NAME AND ADDRESS The Pennsylvania State University U
4、niversity Park, Pennsylvania 16802 10. WORK UNIT NO. 11. CONTRACT OR GRANT NO. NAS8-27334 2. SPONSORING AGENCY NAME AND ADDRESS L 13. TYPE OF REPORT the regression equations for these variables are given. The coherence between microscale wind speed variations at the different levels has the same pro
5、portions as in non-thunderstorm cases. 17. KEt WORDS thunderstorm front gusts aircraft response wind shear 19. SECURITY CLASSIF. (of thlm repart) 20. SECURITY CLA 18. DISTRIBUTION STATEMENT 20 it F. (of thlm pie) 21. NO, OF PAGES 22. PRICE * For sale by the National Technical Information Service, Sp
6、ringfield, Virginia 22151 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-FOREWORD The research reported herein was supported by NASA Contract NAS8- 27334. Dr. George H. Fichtl of the Aerospace Environment Division, Aero-Astrodynamics Laboratory, Mar
7、shall Space Flight Center, was the scientific monitor, and support was provided by Mr. John Enders of the Aeronautical Operating Systems Division, Office of Advanced Research and Technology, NASA Headquarters. The research re2orted in this document is concerned with the results of a study of the emp
8、irical characteristics of thunderstorm gust fronts for the design and safe operation of aeronautical systems. of an investigation of the effect of severe atmospheric forcing on aeronautical systems. Portions of the study relating to vehicle response, numerical modeling of the gust front, and statist
9、ical analysis of evolutionary processes will be made available as they are completed. It is part The data used herein was supplied by NASA through Dr. Fichtl and the National Severe Storms Laboratory through Dr. Edwin Kessler. Mr. Dennis Deaven of the Department of Meteorology at Penn State provided
10、 invaluable aid with the computer analyses used herein. iii Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TABLE OF CONTENTS Page LISTOFTABLE V LIST OF FIGURES vi 1.0 INTRODUCTION 1 1.1 Statement of Problem 1 1.2 Previous Reasearch 1 . . 1.3 Specifi
11、c Goal of This Research 6 2.0 THE OBSERVATIONS 7 2.1 Kennedy Tower Data 7 2.2 NSSL-WKY Tower Data 9 2.3 Tampa Data 9 3.0 ANALYSIS OF TOWER DATA . 11 3.1 Relevant Time Scales . 11 3.2 Smoothing the Wind Data 18 3.3 Summary of the Six Gust Front Cases 18 4.0 MODELING THE THUNDERSTORM GUST FRONT 24 4.1
12、 Scaled Time-height Cross Sections . 24 4.3 4.2 Test of Logarithmic Wind Law Hypothesis 28 Comparison of Roughness Length Under Thunderstorm Conditions with Those Under Non-thunderstorm Conditions 36 Implications of the Logarithmic Wind Law for the 4.4 Prediction of the Gust Front Wind Speeds . 38 5
13、.0 STATISTICAL RELATIONSHIP OF GUST FRONT PARAMETERS TO RAWINSODE AND WAR VARIABLES . 50 5.1 Choice of Independent Variables 50 5.2 The Variation of At Based on a Simple Kinematic Model . 51 5.4 Discussion of Regression Equations 59 5.3 Development of Regression Equations 56 REFERENCES . 64 iv Provi
14、ded by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-LIST OF TABLES Tab1 e Page 1 Values of descriptive parameters for the six gust fronts . 2 Average wind directions, average ratio of 90 m winds to 18 m winds and corresponding roughness lengths for the six gu
15、st fronts. Also shown are previously determined roughness lengths for the listed wind directions 3 Means and standard deviations for each variable and correlation coefficients for each possible pair of variables for the 81 thunderstorm cases from Tampa, Florida 23 40 58 V Provided by IHSNot for Resa
16、leNo reproduction or networking permitted without license from IHS-,-,-Fipure LIST OF FIGURES Page 1 Time-height cross section of the one-second average wind speeds for the Florida gust front of July 27,1967 8 2a Time-height cross section of the 130-second running mean of wind speeds for the Florida
17、 gust front of July 27, 1967. The isotachs in this figure are expressed in mps 12 2b Time-height cross section of the 130-second running mean of wind directions for the Florida gust front of July 27, 1967. The isogons in this figure are expressed in degrees 13 3a Time-height cross section of the 130
18、-second running mean of wind speeds for the Florida gust front of April 29, 1968. The isotachs in this figure are expressed in mps 14 3b Time-height cross section of the 130isecond running mean of wind directions for the Florida gust front of April 29, 1968. The isogons in this figure are expressed
19、in degrees 15 4a Time-height cross section of the 130-second running mean of wind speeds for the Florida gust front of June 25, 1968. The isotachs in this figure are expressed in mps 16 4b Time-height cross section of the 130-second running mean of wind directions for the Florida gust front of June
20、25, 1968. The isogons in this figure are expressed in degrees. . 17 5 Time-height cross section of the hand-smoothed wind speeds for the Oklahoma gust front of April 16, 1969. The isotachs in this figure are expressed in mps . 19 6 Time-height cross section of the manually-smoothed wind speeds for t
21、he Oklahoma gust front of April 29, 1969. The isotachs in this figure are expressed in mps . 20 - 7 Time-height cross section of the manually-smoothed wind speeds for the Oklahoma gust front of May 31, 1969. The isotachs in this figure are expressed in mps . 21 8 Scaled time-height cross section of
22、wind speed for the Florida gust front of July 27, 1967 . 25 vi Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-LIST OF FIGURES (Continued) Figure Page 9 Scaled time-height cross section of wind speed for 26 the Florida gust front of April 29, 1968 .
23、10 Scaled time-height cross section of wind speed for the Florida gust front of June 25, 1968 . 27 11 Ratio of wind speed to speed at 18 m vs. height and scaled time for the Florida gust front of July 27,1967 30 12 Ratio of wind speed to speed at 18 m vs. height and scaled time for the Florida gust
24、front of 31 April 29, 1968. . 13 Ratio of wind speed to speed at 18 m vs. height and scaled time for the Florida gust front of 32 June 25,1968 14 Ratio of wind speed to speed at 18 m vs. height and scaled time for the Oklahoma gust front of April 16, 1969 33 15 Ratio of wind speed to speed at 18 m v
25、s. height and scaled time for the Oklahoma gust front of April 26,1969. 34 . 16 Ratio of wind speed to speed at 18 mvs. height and scaled time for the Oklahoma gust front of 35 May 31, 1969. 17 Roughness length, zo, as a function of direction at the Kennedy tower (Blackadar et al., 1972) 37 18 Rough
26、ness length, zo, as a function of direction at the WKY Oklahoma tower 38 19 Comparison of observed 90 m wind with the 30 m wind multiplied by the average rat.i.0, Vgo/V30, for the Florida gust front of July 27, 1967 . 42 20 Comparison of obeerved 90 m wind with the 30 m wind multiplied by the averag
27、e ratio, VgO/V for the Florida gust front of April 29, 1963: 43 vii Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-LIST OF FIGURES (Continued) Figure Page 21 Comparison of observed 90 m wind with the 30 m wind multiplied by the average ratio, VgO/V3
28、o, for the Florida gust front of June 25, 1968 . . . . . . . . . . . 44 22 Comparison of observed 90 m wind with the 7 m wind multiplied by the average ratio, Vgo/V7, for the Oklahoma gust frnnt of April 16, 1969 . . . . . . - . . . 45 ?3 Comparison of observed 90 m wind with the 7 llli wind multipl
29、ied by the average ratio, Vg0/V, for the Oklahoma gust front of April 26, 1969 . . . . . . . . . . 46 24 Comparison of observed 90 m wind with the 7 m wind multiplied by the average ratio, Vg /v7, for the Oklahoma gust front of May 31, 1869 . . . . . . . . . . . 47 25 Schematic diagram of gust front
30、 . . . . . . . . . . . . . . . . . 52 26 At vs. gust front age . . . . . . . . . . . . . . . . . . . . . . 54 27 At vs. station distance from center of storm track(R). . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 28 Scatter diagram with line of best fit of the speed maximum vs. the avera
31、ge speed before the gustfront . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 29 Scatter diagram with line of best fit of the speed maximum vs. the instability index . . . . . . . . . . . . . 62 viii Provided by IHSNot for ResaleNo reproduction or networking permitted without license from I
32、HS-,-,-1.0 INTRODUCTION 1.1 Statement of Problem One atmospheric phenomenon that may adversely affect the operation of aeronautical systems is the cold surge or gust front accompanying thunderstorms. As the cold downdraft associated with the mature or decaying thunderstorm cell reaches the ground an
33、d spreads out horizontally, this cool, relatively dense air pushes under the warmer, lighter ambient % air. The interface between the two air masses of different densities is called the pseudo cold front or gust front. This front may precede the main storm cell by as much as ten miles. The strong ve
34、rtical and horizontal wind shears associated with the gusty winds behind the front are potential dangers to the safe operation of aircraft and launch vehicles. In this report we investigate the structural properties of six gust fronts. Of particular interest is the variation of wind speed with time
35、and height. In an effort to develop a statistical prediction scheme for the wind speed from the surface to about 150 m during the passage of a gust front, we have scaled the winds so that certain similarities among storms are present. We then attempt to relate the scaling parameters to measurable sy
36、noptic scale and radar variables, based on data from 81 thunderstorm cases in Tampa, Florida. 1.2 Previous Research The low-level variation of wind under high wind conditions has been the subject of numerous empirical studies. As early as 1937, Sherlock and Provided by IHSNot for ResaleNo reproducti
37、on or networking permitted without license from IHS-,-,-Stout (1937) investigated the effect of wind loading of electric power lines during high wind situations. They analyzed the vertical variation (up to 250 feet) of wind during two winter storms, and noted that gust maxima occurred in the upper l
38、evels first: a time lag of several seconds existed between the top and the 50 foot level. Considerable attention has been directed toward the prediction of peak surface wind gusts during thunderstorms. Fawbush and Miller (1954) developed an empirical technique for forecasting maximum surface wind sp
39、eed. 8 Their method, called the downrush temperature technique, was based on Brancatos (1942) discovery that temperatures of downdrafts reaching the surface in thunderstorms are approximately equal to the surface tempera- ture of the moist adiabat through the intersection of the environmental wet-bu
40、lb curve and the zero degree isotherm. Foster (1958) devised a method based on the buoyancy equation to compute the vertical velocity in a downdraft. This calculated downdraft velocity was then correlated with peak wind gusts at the surface. The correlation coefficient of 0.5 was disappointingly sma
41、ll, but it was statistically significant at the one percent level. To test the hypothesis that the downdraft transported upper-level momentum to the surface, the average wind speeds at 500 and 700 millibars were added to the computed downdraft velocity. This addition yielded an improvement in the co
42、rrelation coefficient of only 0.01. Therefore, little useful information had been added. Feteris (1965) extended Fosters and Fawbush and Millers techniques by considering combinations of parameters rather than single parameters as predictors. An important parameter was the instability index, AT, 2 P
43、rovided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-defined as the difference between Fawbush and Millers downrush temperature and the ambient air temperature at the surface. When AT was less than zero, a strong relation existed between maximum wind speed
44、 at the surface and mean vertical wind shear in the troposphere. This relationship supported the hypothesis that transfer of momentum from the middle and upper troposphere occurred under convectively unstable conditions. The results of the multiple linear regression analysts indicated that the combi
45、nation of AT and the surface geostrophic wind speed explained the largest percentage of the variance. On the other hand, the predictors suggested by Fawbush and Miller explained very little variance of the peak wind gust. Miller (1967) stated that the downrush temperature technique was not dgsigned
46、for tropical-type thunderstorms. Because the wet-bulb zero degree isotherm intersection in the subtropics during the summer is well above 10,500 feet, air from this level seldom reaches the surface. Consequently, this technique gives forecasts of wind speed which are too high. also indicated that th
47、e dry instability index showed promise in predicting maximum gust speeds in regions where thunderstorm development is primarily associated with surface heating. Miller In 1965, data from a micrometeorological network of 14 towers equipped with wind sensors at various levels were used to test the dow
48、n- rush temperature technique at Cape Kennedy, Florida (Aerospace Review, 1965). Although this technique tended to overpredict peak wind gusts, it did prove to be useful as a forecast aid. also pointed to the complexity of the problem of predicting peak winds The results from this study 3 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-associated with individual thunderstorms. The regions of high winds were extremely localized and sharply edged. Thus a very dense observational network is necessary in order to accura
