Building to Resist the Effect of Wind

8/8/2019 Building to Resist the Effect of Wind

1/33

AT MICROFICHEREFERENCELIBRARYA project of Volunteers in Asia~ ~ Q Resist the Effect of Wind, Volume,E:stimatiQl) of Extreme wind Speeds and G\.lide tothe Determination of Wind F o r c e ~ by: Emil Simiu and Richard D. MarshallPublished by:National Bureau of StandardsU.S. Department of CommerceWashington, DC 20234 USAPaper copies are $ 1.30. Ask for stock number003-003-01718-3 when ordering.Available from:superintendent of Documents

US Government DocumentsWashington, DC 20402 USA.Reproduction of this microfiche document in anyform 1s subject to the same res t r ic t ions as thoseof the orig inal document.


2/33

NBS BUILDING SCIENCE SERIES 100Building To Resist The Effect Of WindVOLUME 2. Estimation of Extreme Wind

Speeds and Guide to theDetermination of Wind Forces

u.s. DEPARTMENT OF COMMERCE. NATIONAL BUREAU OF STANDARDS


3/33

........... . . . . , . . , . . . . . . , . . . . " . . . . ~ - - - - - - - - - - - - - - - - - - - - - ~ - - - - - - ~ ~ ~ ~ ~ ~ ~ - -

::, f;' : ,

NBS BUILDING SCIENCE SERIES 100-2

Building To ResistThe Effect Of WIndIn five volumesVOLUME 2: Estimation o( Extreme Wind Speeds and

Guide to the Determination of WindForcesEmil 5imiuRichard D. MarshallCenter for Building TechnologyInstitute for Applied TechnologyNational Bureau of StandardsWashington, D.C. 20234

Sponsored by:The Office of Science and TechnologyAgency for International DevelopmentDepartment of StateWashington, D.C. 20523

u.s. DEPARTMENT OF COMMERCE, Juanita M. Kreps, SecretaryNATIONAL BUREAU OF STANDARDS, Ernest Ambler, Acting DirectorIssued May 1977


4/33


5/33

ABSTRACTThe Agency for International Development sponsored with the National Bureau of Standards, a threeand a half year research project to develop improveddesign criteria for low-rise buidings to better resist theeffects of extreme winds.Project results are presented in five volumes. Volume1 gives a background of the research activities, accomplishments. results, and recommendations. InVolume 3, a guide for improved use of masonryfasteners and timber connectors are discussed.Volume 4 furnishes a methodology to estimate andforecast housing needs at a regional level. Socioeconomic and architectural considerations for thePhilippines. Jamaica, and Bangladesh are presented inVolume 5.Volume 2 consists of two reports. The first reviews thetheoretical and practical considerations that are pertinent to the estimation of probabilistically definedwind speeds. Results of the statisticai analysis of extreme wind data in the Philippines are presented andinterpreted. Recommendations based on these resultsare made with regard to the possible redefinition ofwind zones, and tentative conclusions are drawnregarding the adequacy of design wind speeds currently used in the Philippines . Report two describessome of the more common flow mechanisms whichcreate wind pressureson low-rise buildingsand theeffectsof building geometry on these pressures. It isassumed that the basic wind speeds are known and aprocedure is outlined for calculating design windspeeds which incorporates the expected lifeof thestructure, the mean recurrence interval, and the windspeed averaging time. Pressure coefficients are tabulated for various height-lo-width ratios and roofslopes. The steps required to calculate pressuresandtotaldrag and uplift forces are summarized and an illustrative example is presented.

Key words: Building codes; buildings; codes and standards;housing; hurricanes; pressure coefficients; probability distribution functions; risk; statistical analysis; storms; structural engineering; tropical storms; wind loads; wind speeds.

iii

Cover: Instruments to measure wi"d speed and direction(Ieing installed 0" a 10 meter mast at tlfe project test sitein Quezon City, Philil'pi/les.


6/33

CONTENTS1. ESTIMATION OF EXTREME WIND SPEEDS-APPLICATION TO THE PHILIPPINES

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Wind Speed Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2.1 Type of Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2.2 Averaging Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " . . . . . 31.2.3 Height Above Ground . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.2.4 Distance Inland From the Coastline . . , . . . . . . . . '" . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Probabilistic Models of Extreme -'Vind Speeds . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.4 Assessment of Procedures Based on the Annual Highest Speed . . . . . . . . . . . . . . . . . . . . . . . . . . 41.4.1 Wind C!imates Character ized by Small Values of opt('Y) . . . . . . . . . . . . . . . . . . . . . . . . . . 51.5 Assessment of Procedure Based on the Highest Average Monthly Speed . . . . . . . . .. . . . . . . . 61.6 Statistical Analysis of Extreme Wind Data in the Phil ippines . . . . . . . . . . . . . . . . . . . . . . . . . . . 61.7 Interpretation of Results . . , . . . . . " . . . . . . . . . . . . . . . , . . . . . . . . . . . . , . . . . , . . . . . . . . . . . . . . . . 71.7.1 Zone m .. ,........ ,.......................... ,............................... 71.7.2 Zone II , . . . . . ,. " . . , . '" '" " . . . . . . . . . , . . . . . ' . . . . . . . . . . . . . . . . . . " . . . . . . . . . . . . 81.7.3 Zone I . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.8 Conclusions . . . . . . . . . . . . . . . . . , . . . . . . . ' . . . . . , , .. , . . , . . . . . , . , . . . . . . . . , , , . . . . . . . . . . . . . 9ACKNOWLE()(;MENTS . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. 9

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92. A GUIDE TO THE DETERMINATION OF WIND FORCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

..... '

2.1 Introduction . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 132.2 Aerodynamics of Buildings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132.2.1 Typical Wind Flow Around Buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.2.2 Effect of Roof Slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 142,2.3 Roof Overhangs . . , . . , . , . . . . , . . . . . . . . . . . " . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . 142.3 I>esign Wind Speed '" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.3.1 Mean Recurrence Interval. . . . , . . . . . . . . , , . . . . . . . . . . . . . . . . . . , , . . . ,', . . . . . . . . . . . . . 152.3.2 Risk Factor, . . . . . . . . , . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IS2.3.3 Averaging Time an d Peak Wind Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . 152.4 I>esigh Pressures . . . . . . . . , . . . . . . . . , . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.4.1 Dynamic Pressure . . . . . . . . . . . . . , . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IS2.4.2 Mean and Fluctuating Components of Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.4.3 Pressure Coefficients. . . . . . . . , " . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . : .'. . . . . . . . . . . . . . 162.4.4 Correction Factor for Height of Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.5 Procedure for Calculating Wind Forces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

ACKNOWLE()(;MENTS . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 18APPENDIX A

Illustrative Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22Comment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23FIGURES

Fig. 1 Ratio, r, of Maximum Probable Wind SpeedsAveraged over t seconds to those Averaged over 2 sec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Fig. 2 Quantity B . . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . , .. " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Fig. 3 Probability Plots;(a) Type II Distribution, 'Y =2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

v


7/33

(b) Type I Distribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12Fig. 4 Typical Flow !lattern and Surface Pressures. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 4Fig. 5 Vortices Along Edge of Roof. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15Fig. 6 Areas of Intense Suctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15Fig.7 Typical Record of Wind Speed and Surface PIessure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16TablesTable 1 Suggested Values of Zo for Various Types of Exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Table 2 Maximum Annual Winds (I minute average). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Table 3 Station Descriptions and Estimated Extreme Wind Speeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Table 4 Mean Recurrence Interval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Table 5 Relationships Between Risk of Occurrence, Mean Recurrence Interval and ExpectedLife of Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Table 6 Pressure Coefficients for Walls of Rectangular Buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . , .. 19Table 7 Pressure Coefficients for Roofs of Rectangular Buildings . . . . . . . . . . . . . . . . . . . . . . . . . , .. 20Table 8 Internal Pressure Coefficients for Rectangular Buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21Table 9 Correction Factors (R)for Height of Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , .. 21

Facing Page: A wind sf!IIsor is illstalled 011 the wall of a testhouse ill QUezOIl City, Philippilles. Pressures actillX 011walls alld Oil the roof of the test /luildillg Ilre cOIll'l'rted /'Ythese S('IISOrs illto L'iectriCllI siXlIll/s which are recorded 01/IIIIlXIletic til,,,,.


8/33

1. ESTIMATION OF EXTREMEWIND SPEEDS-APPLICATION TO THEPHILIPPINESbvE. Simiu

1.1 INTRODUCTIONIn modern building COlit'S ilnd ... t , l J 1 d a r d ~ II , b a ~ i ( design wind " ~ ' l l ' l ' d s ar e spl'ciiil'd in explicitlv pro-babilistic terms, At any givcn station a random \'.H1,1-bl e can be defined, which consists of thl' I . H g l ' ~ t \'Carlv

wind speed, I t t h l ' ~ t d t i ( l n l"l'l1l' Illr \\'hllh windrl'cord" ,'\'l'r a numlwr lli (OI1"l'cutl\'l' \'l'.1r" ,11'l'ilvailabl!.', ,hen thl' ClIl1lul,1ti\'l' d i ~ t J Ibutilln tUI1t'tf('11(CDI') p; \ h i ~ ral1dlllll \ dri,1bll' 111.1\', ,It le.l"t II'i tlll'llr\


9/33

be estimated to characterize the probabilistic behaviorof the largest yearly wind fpeeds. The basic designwind speed is then defined as the speed corresponding to a specified value FO of the CDF or, equivalently(in view of th e relation N= 1/ (I Fa> in which N=mean recurrence intervaI), as the speed corresponding to a specified mean recurrence interval. For example, the American National Sta,ndard ASS.l [1]specifies that a basic design wind speed correspondingto a SO-year mean recurrence interval (i .e, to a valueFoofthe CDF equal to 0.9S, or to a probability of exceedance of the basic wind speed in anyone yearequal to 0.02) be used in designing all permanentstructures, except those structures wit h an unusuallyhigh degree of hazard to life and property in case offaHure, for which a lOO-year mean recurrence interval (FO =0.99)must be used, an d structures having nohuman occupantsor where the re is negligible risk tohuman life, for which a 25-year mean recurrence(fO =0.96) may bP. used. A wind speed correspondingto a N-year recurrence interval is commonly referredto as the N-year wind.The mean recurrence intervals specified by buildingcodes, rather than being based on a formal riskanalysis-which is in practice not feasible in the present state of the art--are seiected in such a manner asto yield basic wind speeds which , by professional con~ n s u s , are judged to be adequate from a structuralsafety viewpoint. Nevertheless, it is generallya!"".,umed that adequate probabilistic definitions ofdesign wind speeds offer, at least in theory, the advantage of insuring a certain d ~ e e of consistencywith regard to the effect of the wind loads upon structural safety. This is true in the sen.o;e that, all relevantfactors being equal, if appropriate mean recurrenceintervalsare used in design, the probabilities of failureof buildings in different wind climates will, on theaverage, be the same.In the practical application of the probabilistic approach to the definition of design wind speeds, certain important questions arise. One such question pertains to the type of probability distribl -tion best suitedfor modeling the probabilistic behavior of the extremewinds. The provisions of the National Building Codeof Canada [2] are based upon the assumption that thisbehavior is best modeled by a Type I (Gumbel) distribution. The American National Standard ASS.l [1],on the other hand, assumes that the appropriatemodels are Type II (Frechet) distributions with location parameters equal to zero and with tail lengthparameters dependent only upon type of storm.Finally, Thorn [29J has proposed a model consisting ofa mixed probability distri17ution, the parameters ofwhich are functions of (a) the frequency of occurrenceof tropical cyclones in the 5 longitude-latitude squareunder consideration and (b) the maximum average

2

monthly wind speed recorded at the station investigated. The question of selecting the most appropriatedistribution is one that deserves close attention: indeed, as indicated in References 23 and 22, the magnitude of the basic design wind speed may dependstrongly upon the probabilistic model used.A!>Suming that the type of probability distribution bestsuited for modeling the behavior of the extremewinds is known, a second importan t question arises,viz., that of the errors associated with the probabilisticapproach to the definition of design wind speeds.Such errors d e p ~ n d primarily upon the quality of thedata and upon the length of the record (Le., the sampie size) available for analysis.These questions wi II be dealt with in this work, whichwiII also present results of statistical ana lyses of windspeed data recorded in the Philippines. In the liglit ofthe material presented herein, possible approacheswill be examined to the definition of extreme windspeeds for purposes of structural design in the Philippines.1.2 WIND SPEED OAT AFor the statistical analysis of extremt' wind speeds tobe meaningful, the data used in the analysis must bereliable and must constitute an homogeneous set. Thedata may be considered to be reliable if:

The performance of the instrumentation usedfor obtaining the data (i.e., the sensor and the recording system> can be dete rmined to have been adequate. The sensor was exposed in such a way that itwas not influenced by local flow variations due to theproximity of an obstruction (e.g., building top, ridgeor instrument support).A set of wind speed data is referred to herein ashomogeneous if all the data belonging to the set maybe considered to have been obtained under identicalor c?quivalent conditions. These conditions are determined by the following factors, which will be brieflydiscussed below:

type of instrumentation used averaging time (i.e, whe the r highest gust, fastestmile, one-minute average, five-minute average, etc.

was recorded). height above ground roughness of surrounding terrain (exposure) in the case of tropical cyclone winds, distanceinland from the coastline.

1.2.1 Type of instrumentationIf , during the period of record, mOre than one' type ofinstrument has been employed for obtaining the data,


10/33

the various instrumentcharacteristics (anemometerand recorder) must be carefully taken into accountan d th e data must be adjusted accordingly.1.2.2 Averaging TimeIf various averaging times havebeen used during th eperiod of record, the data must be adjusted to a common averaging time. This ca n be done using graphssuch as those presented in Reference 19 and includedin figure 1 in which Z. is a parameter deJning th eterrain roughness (see, for example, Ref. 10).1.2.3 Height Above GroundIf , during th e period of record, th e elevation of th eanemometer ha d been changed, the data must be adjusted to a cbmmon elevation as follows: Let theroughness length an d the zero plane displacement bedenoted by Z. an d Zd, respectively (Zo' Zd' areparameters which define th e roughness of terrain. seeRef. 10). The relation between th e mean wind speedsU(Z1) an d U(ZJ over horizontal terrain of uniformroughness at elevation Z1 and Z. above ground,respectively, can be written as

U(Z.)---(Z)Suggested. values of th e roughness length Zo are givenin table 1 (see refs. 10,21,7). For example, at Sale,Australia, for terrain described as open grassland withfe w rees, at Cardington, England, for open farmlandbroken bya few trees an d hedge rows, and atHeathrow Airport in London, Z. =0.08 m [10, 21]. AtCranfield, England where the ground upwind of th eanemometer is open for a distance of half a mileacross th e corner of an airfield, and where neighboring land is broken by small hedged fields,.Z. =O.09Sm[91. The values of Z. for built-up terrain should beregarded as tentative. It is noted that Equation 1 is ap plicable to mean winds and should no t be used tor e p r . e - ; ~ n t th e profilesof peak gusts.

Table 1. Sugested Values of Z. for Various Typesof Exposure

Type ofExposureCoastalOpenOutskirts of towns, suburbs


11/33

It is pc ~ n t e d ou t that, just as in the case of Equation 1,errors ar e inherent in Equation 2 that ar e associatedwith th e subjec tive estinl.'!',lon of the roughnessparameters. Also, recent research suggests that in th ecase of tropical cyclone winds Equation 2 underestimates wind speeds over built-up terrain, calculated asfunctions of speeds over open terrain, by amounts ofthe order of 15% or morel17].1.2.4 Distance InI.tnd from the CoastlineThe intensity of hurricane or typhoon winds is adecreasing function of the distance inland from thecoastline. Hurricane wind speeds may be adjusted to acommon distance from the coastline by applyingsuitable reduction factors. Such reduction fadors havebeen proposed by Malkin, according tc -.. ho m theratios of peak gusts at 48, 96 an d 144 km from thecoastline to peak gusts at the coastline are 0.88, 0.82an d 0.78, respectively [8, 141.1.3 PROBABILISTIC MODELS OF EXTREMEWIND SPEEDSThe nature of the variate suggests that an appropriatemodel of extreme wind behavior is provided by probability distributions of the largest values, th e generalexpression for which is 1II]:F(v) = exp j-Htl-#L)/uJI-'Y J #L


12/33

large values of N, such as are of interest in structuralsafety calculations-if the following conditions aresatisfied. First, the value of opt ('Y) for that record islarge, say 'Y .? 40 (opt(y) = value of Y [see eq. 3] forwhich the best distribution fit of the largest values isobtained). Second, meteorological informat ion obtained at the station in question, as well as at nearbystations at which the wind climate is similar, indicates that winds considerably in excess of thosereflected in the record cannot be expected to occur except at intervals many times larger than the recordlength. Wind climates which satisfy these two conditions will be referred to as well-behaved.Assuming that the wind speed data are reliable, lowerbounds for the sampling error in the estimation of theN-year winds in a well -behaved climate may becalculated on the basis of a mathematical result, theCramer-Rao relation, which states that for the type Idistribution (see ref. 11. p. 282)

("') 1.10867 ,va r 1 ? - - - u -nV ("') 0.60793 ,ar u >-u-

n (7)where var (,1), va ; (u ) are th e variances of,1 andu,where,1 and t't are the estimated values of 1 and u,respectively. obtained by using any appropriateestimator consistent with basic statistical theory requirements; u is th e actual value of the scaleparameter and n is the sample size. Using Equations 6and 7, lower bounds for the standard deviation of thesampling error in the estimation of the N-year wind,SD[v(N). can be approximated as follows. Equation 4.in which the parameters1, u are replaced by theirestimates 1, u, is inverted to yield

(8)where

(9)

Then

Equation 10 is based on the assumption that the errorinvolved in neglecting the correlation between 1 andCT is small. The validity of this assumption wasverified by using Monte Carlo simulation techniques.Since th e actual value of CT is no t known, in practicalcalculations the estimated value u is used in Equations6 and 7. For example, the distribution parameterscorrespondin& o the wind speed data at Davao (n =24.

5

see table 2), estimated by using the techniquedescribed in Reference 23. are fl = 38.89 km/hr, fT =9.40 km/hr. It follows from Equation 8 that v (SO) = 75km/hr and from Equations 6. 7, and 10 that SD[ v (SO)]> 5.18 km/hr . Subsidiary calculations not reportedhere have shown that Equation to provides a goodindication of the order of magnitude of the samplingerrors.1.4.1 Wind Climates Characterized by SmallValues of op t (y).Occasionally, a record obtained in well-behavedwind climates may exhibit small values of opt (,.); thiswill occur if that record contains a wind speed thatcorresponds to a large mean recurrence interval.There are regions, however, in which, as a rule, thestatistical analysiS of extreme wind records taken atanyone station yields small values of opt (,.). This isthe case if . in the region considered, winds occur thatare meteorologically distinct from, and considerablystronger than the usual annual extremes. Thus. in theregions where tropical cyclones occur, opt (,.) will ingeneral be small, unless most annual extremes areassociated with tropical cyclone winds. An exampleoi a record for which,. (opt) is small is given in figure3a, which represents the probability plot with 'Y = opt(,.) = 2 for the annual extreme fastest mile-speedsrecorded in 1949-73 at the Corpus Christi , Texas, airport. For purposes of comparison. the same data havebeen fitted to a type I distribution (opt ( ,.) = 00 , or Eq.4); the fit in this case is seen to be exceedingly poor,i.e., the plot dev iates strongly from a st raight line (fig.3b). As shown in Reference 23. a measure of the goodness of fit is given by the extent to which the probability plot correlation coefficient is close to unity;this coefficient is printed out in figures 3a and 3b.To small values of the tail length parameter there frequently correspond implausibly high values of theestimated speeds for large recurrence intervals. In thecase of the 1912-48 record at Corpus Christi , for example, opt (,.) = 2 and the estimated 5-minute average is327 mph (155 m/s) for a 1000-year wind, which ishighly unlikely on meteorological grounds. For 20-year records, the situation may be even worse: thus,for the 1917-36 Corpus Christi record, which containsan exceptionally high wind speed due to the 1919hurricane [3, 25], op t (,.) = 1 and the calculated 1000-year wind is 1952 mph (873 m/s) [23], a ridiculousresult. Also, the situation is not likely to improve significantly if the record length increases. From a 74-year record, a plo t quite similar to figure 3 wouldpresumably be obtained, with twice as many pointssimilarly dispersed, to which there would corresponda similar least squares line on probability paper.It may be stated, consequently, that while in the caseof well-behaved climates it appears reasonable to in-


13/33

fer from a good fit of the probability curve to the datathat the tail of the curve adequately describes the extreme winds, such an inference is no t always justifiedifopt(y) is small.It may be argued that one could avoid obtainingunreasonable extreme values by postulating that theannual largest winds are described by a probabilitydistribution of the type I, i.e., by assigning the value y= 00 to th e tail length parameter. This has been doneby Court [3J an d Kintanar [12]. As can be seen infigure 4, the correspond ing fit may be quite poor.However, the estimated extremes at the distributiontails will be reduced. The drawback of this approachis that unreasonably low estimated extremes may beobtained. For example, at Key West, Florida, if allthree parameters of Equation 3 are estimated as inReference 23, to the 1912-48 record there correspondsv (tOO) == 99 mp h (44.2m/s) and v (1000) = 188 mph (84m / s ~ Reference 23. I f it is postulated that y = 00 ,then v (SO) =70 mp h (31.7 m/s), v ( I00) =77 mph (34.4m/s) an d v (t000) =97 mph (38.8 m) (23), an unlikelyresult in view of the high frequency of occurrence ofhurricanes (about 1 in 7 years) at Key West.It may also be argued that since the estimated extremes resulting from small values of y (say y < 4)may be too large, and those corresponding to y = 00may be too small, a probability distribution that mightyield reasonable results is one in which y has an intermediate value, say 4< y < 9. Such an approachhas been proposed by Thom an d will now be examined.1.5 ASSESSMENT OF PROCEDURE BASED

ON THE HIGHEST AVERAGEMONTHLY SPEEDThe procedure for estimating extreme winds in hurricane-prone regions on the basis of annual highestwinds at a station was seen to have the followingshortcomings. First, because hurricane winds arerelatively rare events, the available data may not contain wind speeds associated with major hurriCane oc-, currences and are therefore not representative of thewind climate at the station considered (see the case ofCalapan in Section 1.7 of this report). Second, inregions subjected to winds that are meteorologicallydistinct from, and considerably stronger than theusual annual extJ'emes, implausible estimates may beobtained.The model proposed by Thom lEq. 5J in Reference 29represents an attempt to eliminate these shortcomings. It can be easily shown by applying the intermediate value theorem, that if this model is assumed,the estimated extreme winds may be obtained by inverting an expression of the form:

6

in which 4.5 < y (v) < 9. If the mean rate of arrival oftropical cyclones in the region considered is high,then y ( til will be closer to 4.5. Othe rwise, y (til will becloser to 9; in re3ions where hurricanescannot he expected to occur, y (t) = 9.1n order that estimates notbe based upon possibly unrepresentat ive annual extreme data, Thom's model does not make use of annual extreme speeds. Rather, the parameter f7 is estimated from the maximum of the average monthlywind speeds on record at the location considered,presumably a quantity for which the variability issmall.While the quasi-universal climatological distributionproposed by Thon is tentative, it will yield resultswhich, for a first approximation, may in certain casesbe regarded as acceptable. This model has recentlybeen used by Evans l6J as a basis for obtaining designwind speeds for Jamaica. Estimates of extreme speedsobtained by Evans are substantially higher than theresults obtained by SheJlard [20J in his 1971 analysisof Commonwealth Caribbean wind data.It was shown in the preceding section that the approach which utilizes the series of annual largestspeeds may fail in regions in which hurricanes occur.For such regions, therefore, it may be that alternativeapproaches need to be deve!oped. Among such approaches is one in which esti mates of extreme windsare based upon the follOWing information:

average number of hurricanes affecting thecoastal sector considered (per year) probability distribution of hurricane intensities radial dimensions of hurricanes dependence of wind speeds upon centralpressure and distance from hurricane center.

This approach appears to provide useful es timates ofextreme winds corresponding to large recurrence intervals-which are of interest in ultimate strengthcalculations--and is currently under study at the National Bureau of Standards.1.6 STATISTICAL ANALYSIS OF EXTREMEWIND DATA IN THE PHILIPPINESThrough the courtesy of the Philippine Atmospheric,Geophysical and Astronomical Services Administration (PAGASA), 16 sets of data were obtained consisting of maximum yearly wind speeds recorded duringat least 14 consecut ive years. The data for each of the16 stations are listed in table 2. Table 3 includes sub-jective station descript ions provided by P AGASApersonnel an d the results of the analysis. In Table 3are listed VN"pt(oy)= N-year wind based on the distribution for which the best fit of the largest values isobtained and VNO = N-year wind based on the type Idistribution, N =mean recurrence interval in years.


14/33

TABLE 2. MAXIMUM ANNUAL WINDS (ONE MINUTE AVERAGES)-- -- . - -NO. Station Period of Record Maximum Annual Winds for Each Year of Record Ckm/hourl1 Davao 1950-73 39,52,40,39,40,37,35,35,32,40,40,40,80,48,48,48,56,46,52,50,46,52,46,462 Cagayande 1950-73 47,24,19,13,19,19,12,12,12,19,16,14,21,6,24,17, 19,37,37,46,37,48,41,41Oro3 Zamboaml:a 1950-73 486440,3948614340484548,72 48,48,50 56 68 67 70,56 617461.784 Pasal'Citv 1950-73 89103899272 72 64 72 97 72 81666':1.74.1306580111837420080111 565 Manila 1949-70 72,105,97 89,101,97,100,105,81,72,97,89, 121,105,100,168.74,89,107,111,96,2006 Manila 1902-40 46,56,65,80,73,55,77,70,41,68,50,69,64,68,42, 41,67,54,70,83,58,53,6(1,51,45,Central 63.50.70.52,55,58,37.100,60,45,52,103,53,567 Mirador 1914-40 79.79.70,79.121,102,94,72,105,107,122,58.89, I 07,93,63,77.92.63,53,93.73.75.49,39,68,838 uaguio 1950-73 9 7 . 1 1 J ~ , 5 6 . ' n , ' : I 7 . I H , 6 4 , M , 6 1 , 4 1 S , 1 f 7 , 41S,53,':I1S, Ill.lU7.ISU, 144.111i.IU:".15/ , I I ~ . I I Y , 1.1.19 Calapan 1959-73 145,185.97.72.40,68,40.96,10'1,41,33,111,102,111,83

10 ~ u r i g a o 1 ' : I ~ 2 - 7 3 1115,43.1 Li,411,40,40,43,48,64,64,1I9,39,74.52.711.104.167,111.96.107, 16,7411 Laoag 1949-73 34,72.118,71,1011,100,64. 111.111.64.81,79.64.105,90,144,144,78,120,137,1110,67,89,(,7ZO12 110110 \949-73 9 7 , 1 U 6 , 6 4 , 6 4 , ' : I 7 , 6 4 , 6 4 , M . 7 2 , 7 4 . M . 7 I S , 7 1 \ , 1 \ ' : I , 7 4 , . 6 1 , ~ : . ! , 6 1 , 1 S ' : I . 5 6 . 7 4 13 L ~ u 1':I::>U-7J ' J ~ .:.0, ,L ,4S,81.04.4K.4K.4K.4:1.4I1,4:1.64.h4.:.o.4!:1.:.o. .'J:I.o:>.:>o. UU.l4.;>'-! ..___14 Legaspi 1955-73 40.97.97.4S. 129.185:)7.B2.89.\04.74.89. 148 74 70 174 1482048115 Tacloban 1949-73 I:l7.58, 106.1 I:l.56.68.60,47,48.69.87,42. \05.93.;4.111.

70.194.I:n.lh7 63 III 1551046716 Infanta 1960-73 60.43.45.50.128.39.133.126.46.46.189.85.104.83

TABLE 3. STATION DESCRIPTIONS AND ESTIMATED EXTREME WIND SPEEDSWias !Zone- Period of No. of AMmomeler

No . SUlion !f tlWf.15 Record Yellrs Elevillion Cmelers)

I Davao III ,q5tI-7J HIZ CaJlily.n III , q 5 t ~ 7 J de Oro3 Zamb""nll" III ,q5tI-7J 24 III.. "V ,'v II I ~ " ' - _4 (Manta II l ~ q - 7 n " Tllp l,f 5 ..t u r ~ buildin'h Manila n 1"'124(1 W h"CentralMlraaor II 1 ~ 1 ~ - 4 f l "7 I I ~ BagUill (I:;UU m .It-Wt.Baguiu II ,q5fI-7J 111

;,9 Cal.pan II , q ~ 7 J 15 iOto I""rill"O I 19. . " 7II I..loaa n 1949-73 2., 1212 I l ioio II 1 9 ~ - 7 3 24 ; ' : ; ' ~ : : ' , . o \3 Cebu n 1950-73 24 1014 1 , ~ , , . : r . 19 -locl_n I 1949-73 Z, ~ . , m o""vo'1m bldg.ll'b n anla '_'-1. 14 1111

a3 CUp anemometer; mean speed averaged over one minute."Mean speed averaged Over one minute.cTrees at East side of anemometer_

Dflcriptionof Terrain opllylTuwn 00

Ol"'n 00l l \ i rpmt -rurt ,\n'.l 7Op,mm"kl

I Muunt.lin hl p 00sea J ~ V l . 1 ) Mllut.lin top 00(l:;UU m .lblwt.,*a level)Top of hill; 00town on on e!iide.L . I ~ C i l y 40Topuf 40rnhUIo.. .n 00I-.-own 2Airport ...Ai!Port 9(1!!OPO .m 14~ ~ ~ ~ I i I O r y

own; "" 'don Hal area

vopdyl' ' fkm/hrlN v oofN fkm/hrlN=50 N=IOO N=IIIOO N=50 N=IOO N=IIIOO

7411JOh

-'3:

----

1.7 INTERPRETATION OF RESULTSThe results will be grouped into three classes, according to the wind zone (as defined in Ref. 15) in whichthe stations considered are located (table 3).1.7.1 Zone II I

It is noted that fO,r all three Zone III stations listed inTable 3, opt (y ) = 00. It is convenient to adjust thespeeds at Davao and Cagayan de Oro to open terrainexposure. On the basis of the ter rain descriptions ofTable 3, i f it is assumed Zo = 0.30 m, Zd =0, ZOI =0.08m, Zdl = 0, it follows from Equation 2 that7


15/33

UOO}== 0.8U (10)where U(10}, U 1(10} are mean speeds above ground intown and open exposure, respectively. Thus, inDavao and Cagayan de Oro the calculated 5 0 - y ~ a r mean speeds at 10 m above ground in open terrain are58 mph and 47 mp h (94 km/hr and 76 km/hr) respectively, versus 55 mph, (88 km/hr) in Zamboanga. If thecorresponding highest gustsare obtained by multiplying.the one-minute meansby a factor of, ~ y , 1.20 (seefig. H, the estimated highest 50-yr gustsat Davao,Cagayan de Oro and Zamboanga are at most 94 x 1.20== 113 km/hr, (70mph), i.e., conSiderably lower thanthe value specified for design purposes by the Nationa1 Structural Code ofthe Philippines [I5] for ZoneflI, viz., 95 mp h (153 km/hr). This suggests tha t the requirements of Reference 15 regarding wind loading inthe Zone III portion of Mindanao are conservativeand might be somewhat reduced. (It can be easilyshown, on the basis of Eq. 2 and figure 1, that thisstatement holds even if it is assumed that the errors inthe estimation of the parameter values Zo == 0.30 mand Z 01 =0.08 m are of the order of as much as 50%.)To validate such a conclusion it would however benecessary to determine, from long-term records oftropical cyclone occurences, that the 1950-73 data atthe three stations analyzed are indeed representativefor southern Mindanao.1.7.2 Zone II.Several difficulties arise in interpreting the results forthe Zone II stations in table 3. It is noted, first, that theresults obtained at stations in and near Manila (stations 4, 5, 6 in table 3) are widely divergent. The discrepancies between the results for Pasay City andManila may be due to the different elevations of therespective anemometers. It may alsobe conjecturedthat the discrepancies between these results and thoseobtained from the 1902-1940 Manila Central recordare due to differences in the averaging times and inthe exposure, elevation and calibration of the instruments, as well as to possibly inaccurate estimates ofthe maximum speed in Manila and Pasay City in 1970(200 km/hr, see table 2).The estimated wind speeds at Baguio based upon the1950-73 record are higher than those obtained fromthe 1914-40 data. No explanation is offered for thesedifferences; an investigation into their causes seemswarranted.The record at Calapan illustrates the limitations of theapproach to the definition o f design wind speedsbased on the statistical analysis of the highest annualwinds. From the data covering the period 1961-72, theestimated 50-yr wind based on a Type I distribution is88 mph (141 km/hr) [12], versus 131 mph (209 km/hd,

8

as obtained i f the data covering the period 1959-1973are used (see table 3). Since wind loads are propor tional to the square of the wind speeds, the ratio between the respective estimated winds loads is(209/141)2 =2.2.Although the record at Pasay City is best fitted by atype II distribution with opt (y ) == 2. it is unlikely, asnoted previous ly, that such a distribution correctlydescribes the behavior of the extreme winds. This isobvious, particularly in the case of the 1000-yr wind,which, on physical grounds, could not possibly atta in509 mph (820 km/hr) (see table 3).The National Structural Code of the Philippinesspecifies, for Zone II and elevations under 9.15 m, adesign wind of 109 mph (175 km/hr). In the light ofthe data shown in table 2, the value appears to bereasonable. It will be noted that tables 2 and 3, andfigure 13 of Reference 15 indicate that the extremespeeds and the frequency of occurrence of tropicalcyclones, are considerably h igher at Laoag than atCebu. This suggests that Zone II could be divided, accordingly, into two subzones. with wind load requirements higher in the northern than in the southernsubzone.1.7.3 Zone I.As indicated previously. i f y(opt) is smal l, i.c., if thedifferences am0ng maximum wind speeds rL' 'vrdedin various years are large, the probability d i ~ " , ,:.tions that best fit the dnta may not describe em .the extreme wind speeds for large recurrence in'.,>;vals. The minimum and the maximum winds for theperiod of record are, at Legaspi, 25 mph (40 km/hr)and 127 mph (204 km/hr), respectively, and, atTac1oban, 26 mph (42 km/hr) and 120 mph (194km/hr), respectively. In the writer's opinion, thereliability of the N-year wind estimates obtained atthese stations for N==50, 100 and 1000 is thereforedoubtful. The same comment applies to the estimatesfor Infanta, where the record length is quite insufficient (14 yearsl. The writer therefore believes that theresults of table 3 should not be used to assess the adequacy of the design wind speed requirement for ZoneI specified in Reference 15. Rather, it is reasonable tobase such an assessment on a comparison betweenwind speeds in Zone I and in areas affected by hur-ricanes in the United States. In the light of U.S. experience, it is the opinion of the writer that from sucha comparison it follows that the 124 mph (200 km/hr)wind speed requirement for Zone I and elevationsunder 30 ft. (9.15 m) is adequate for structural designpurposes.


16/33

1.8 CONCLUSIONSFrom the analysis of available extreme speed data inthe Philippines, the following conclusions may bedrawn:1. The design wind speeds specified by the National

Structural Code of the Philipp ines for the Zone IIIpart of Mindanao appear to be conservative andmightbe somewhat reduced. For thi s conclusion tobe validated, it would be necessary to determine,from long-term records of tropical cyclone occurrences, that the data analyzed herein are representative for southern Mindanao.2. Methodological difficult ies and uncertainties withregard to the reliability of the data preclude, at thistime, the estimation for Zones II and I of N-year extreme winds that could be used, with a sufficientdegree of confidence, as design values within the. framework of an explicitly probabilistic code.3. According to the data included herein, Zone II canbe divided into two subzones, with wind load requirements higher in the northern than in thesouthern subzone.4. The data included herein suggest that the windspeed requi rement specified by the National Structural Code of the Philippines for Zone I is adequatefor purposes of structural design, except as notedbelow.5. Higher wind speed values than those specified bythe National Structural Code of the Philippinesshould be used-except perhaps in the Zone III partof Mindanao-in open, and in coastal exposure.6. Improved design criteria for Zones II and I. including possible redefinitions of these zones, could inthe future be achieved by applying themethodology briefly described at the end of thesection"Assessment of Procedure Based on theHighest Average Monthly Speed." This would require, in addition to data on the frequency of occurrence of tropical cyclones at various locations inthe Philippines, that the following data be available:a. Reliable wind speeds, carefully defined withrespect to terrain roughness, averaging time anddistance from shore line.b. ApprOXimate radial dimensions of tropicalcyclones.c. Approximate dependence of tropical cyclonespeeds upon minimum central pressure and distance from storm center.

9

ACKNOWLEDGMENTSThe writer wishes to express his indebtedness and appreciation to Dr. Roman L. Kintanar, Mr. ManuelBonjoc, Mr. Bayani S. Lomotan, Mr. Jesus E. Calooy,Mr. Leonicio A. Amadore, Mr. Samuel B. Landet, andMr. Daniel Dimagiba, of the Phil ippine Atmospheric,Geophysical Astronomical Services Administration(PAGASA), for kindly permitting him to use thePAGASA records and facilities an d for their effectiveand generous help. He also wishes to thank Dr. R. D.Marshall of the Center for Building Technology, Institute for Applied Technology, National Bureau ofStandards , for useful comments and criticism of thiswork. The computer program used here wasdeveloped by Dr. J.J. Filliben, of the StatisticalEngineering Laboratory, National Bureau of Standards.REFERENCES(I ) Building Code Requirements for Minimum DesignLoads in Buildings and Other Structures,A58.1-1972 (New York: American National

Standards Institute, 1972).(2) Canadian Structural Design Manual (SupplementNo.4 to the National Building Code ofCanada) (National Research Council ofCanada, 1970).

(3) Court, A., "Wind Extremes as Design Factors,"Journal of the Franklin Institute, vol. 256 (July1953), pp. 39-55.(4) Csanady, G. T., "On the Resistance Law of a Turbulent Ekman Layer," Journal of the At-mospheric Sciences, vol. 24 (September 1967),pp.467-471.(5) Davenport, A. G., "The Dependence of WindLoads Upon Meteorological Parameters," Pro-ceedings, Vol. 1 (Internat ional Research Semi

nar on Wind Effects on Buildings and Structures) (Toronto: University of Toronto Press,1968).

(6) Evans, C. J., "Design Valuesof Extreme Winds inJamaica" (Paper presented at CaribbeanRegional Conference, Kingston, Jamaica,November 6-7,1975).(7) Fichtl, G., and McVehil, G., LongitudinalandLateral Spectra of Turbulence in the A tomosphericBoundary Layer, Technical Note D-5584(Washington, D.C.: Nat ional Aeronautics andSpace Administration, 1970).(8) Goldman, J.L., and Ushijima, T., "Decrease inMaxinum Hurricane Winds after Landfall,"Journal of the Structura I Division, vol. 100, no.

STl, proc. paper 10295 (New York: AmericanSociety of Civil Engineers, January 1974), pp.129-141. .


17/33

(9) Harr is, R.I., "Measurementsof Wind Structure AtHeights Up to 585 ft Above Ground Leve\,"Proceedings{Symposium on Wind Effects onBuildings and Structures) (Leicestershire:Loughborough University of Technology,1968).(10) Hell iwel l, N.C., "Wind Over London," Proceed-ings (Third International Cc-,ference of WindEffects on Building and Structures) (Tokyo,1971) .(11) Johnson, N.L., and Kotz, 5., Continous Univl,riateDistributions, vol. 1 (Wiley, 1970)(12) Kintanar, R. L., "Climatology and Wind-RelatedProblems in the Philippines," Development ofImproved Design Criteria to Better Resist theEffects ofExtreme Winds for Low-Rise Buildingsin Developing Countries, BSS 56 (Washington,D.


18/33

r1-0

o-a

0-6

04 I 2

II min.

-.:::::::.......... .......... ::::::::::---......... :----r--

I10 min. -

! - - -

r--r--

If - - - I h o u r -

.r-...

r---- -Zo ;r 0.03-0.08mZo ;rO.l5-0.30m

Zo;r I.OOm4 6 10 20 40 60 100 200 400600 1000 2000 3600

Time, tt secondsFIGURE 1. RATIO, 1, OF MAXIMUM PROBABLE WIND SPEEDS AVERAGED OVER t SECONDSTO THOSE AVERAGED OVER 2 SEC.

1.50 , . , . . . . --- .--r--- . . . , . - . . . , . -- . . . , . - . . . , . ---- . ....

lAO t -+ - -+ - - -+ -"" ' :

1.30 H - - - H C - - + - ~ - + - - _ _ + ' ~ - + - a : - l i o t f E : I

0.2 0.5 1.0 1.5 2.0ROUGHNESS LENGTH ZolN METERSFIGURE 2. QUANTITY f3

11


19/33

1______ ______ ____ ____ _________ 1_ _ _ _ _ _ _ _1_________ 1-________ ____95.0000000=MAXIII x

x -I1I19.0000000

83.0000000

77.0000000

71.0000000

II1

III65.0000000=1410-III59.0000000

53.0000000

1t7.0000000 XX X

111.0000010 XXXXXXXJ XU

35.0000000=MIN- X

XXXXXX

xx

l!

x Xx XXx

1____________ 1___________1____________1___________ 1____________1___________1__ t. S O O ~ l 8 1 2.2021567 3 . 9 0 ~ 1 ! \ ~ 5.60111139 7.!061112SEXTREME VALUE nPE 2 (CAUCHY TYPEI PliO!!. PLOT WITH EXP. PAQ. = 2.000000000' SAII'LE SIZE N :: 37PROBABILITY PLOT CORRELATION COEFFICIEOlT = .97191 ESTIMATED INTERCEPT = :51 .071 110 9 ESTIMATED SLOP = 9 . "757"7FIGURE la. TYPE II DISTRIBUTION, "y = 2.1____ ___ _______ ______ ___ __1_____1'____ ____

95.000COOO=MAX-1II

.19.0000000

13.0000000

77.0000000

71.0000000

IIII1IIII65.0000000:M10-III59.00000dO

53.0000000

" ' .0000000

111.0000010 III35.0000000=MIN- XX XXx X X X

xxxXX XX"

Xxx XX X'

X X XXX X x X

lClC

x X

lC

III

1____________ 1___________ 1____________ 1___________ 1____________ 1 ___________ 1__- - - - . - . .__ 1___________ 1-1.312981t7 - . 0 " 2 ~ 1 0 7 1.29722311 2.637327.. 3.977113111EXTREME VALUE TYPE 1 (EXPONENTIAL TYPEI PRORAIlILITY PLOT THE: SAMPLE SIZE N = 37PROBABILITY PLOT CORIIELATlON COFFICIENT = . M I O ~ ESTltIIlTO INTERCEPT = II 033329 ESTIMATED SLOPE = ,. "21209FIGURE lb. TYPE IDISTRIBUTION. Facing Page: This wind tunnel at the University 0; thePhi11il,ines is used to study wind effects on scale-modelbuildings. Shown is a model of the CARE, Inc., test house.

Thl! rows of Mocks on the floor offlle tunnl!l xenerate turbulence orgustiness similar to that observed in full scale.12


20/33

2. A GUIDE TO THEDETERMINATION OFWIND FORCES

2.1 IN TRODUCTION1hh p . l ~ " t ' r t i t ,l ] ... \ \ Itll tht' 11.1(111"1' ,1\ \\ : : " j ~ j \ l \ \ .IT' '11J\\1bl l l ld l l lg '" t i l l ' p r l ' ~ ' l I r t " gl ' l l l ' r lt l 'd ;1\' ',\ 1 lei ,I lI d t I tdt 'h . ' r !111n. l tHJ ll l i t I \ I T ~ - t . ' ' ' ' ' " h tin..:, ,H i : ' l l l !d l l l : " ' : t ' ! ~ :l\1'llt..., ,1-.\\"1,11,1 \111 t f l t . ' \ I \ t ' r , I I ! ...t r l l l t l ; r t ' It :-"',\..., i l ! l l t " i ti ll ~ l l l j d l n h . d t " " I . ~ J l l d II i . J l l t q - lL ! !h t' .\ I!! ' tl lt, r l! , l \ , 11:,11\ IUt l l f l l ' d 111 tht' t td!\I \ \ '111\:,- .t \ ! l t l l1"" ' ,I ' ld t d 11' \j\ ' I , ' !

, \ t T tl 1 I ]1 ) III i 111 h ' I ",, h \',' r '. + ! l "1 i 'n I 11 \ \ : , \ t11..11!1lt ' l l ,", ill! l .!nd h , l " \ ' , 1 ht'lt..:,f\; '\1 \\ ll.l!il r , I ! I I I J \ i \\ 1111l!t' \\ t' t ',.i I !)L:, II tt l I2.2 AfRODYN AMICS OF HUILDINCSi Ii \' t! ( ... ',1 t 'I'. ! 11 t i {I I \ lllllll J I ! i I L,"'" t ,111 l' t rt ' I l l 1\


21/33

FIGURE 4. TYPICAL FLOW PATTERNANDSURFACE PRESSURES.wake. However, it has been established that the patterns of wind flow around bluff bodies such as thebuilding in figuie4 do not change appreciably with achange in wind speed.This allows dimensionless pressure cQefficients (to beciiscussed later) determined for one wind speed to beapplied to all wind speeds. In general, the windpressure is a Tllaximum near the centerof the windwiud waHand drops off rapidly near the corners.Pressureson the side or endwal ls are also nonuniform; the most intense suctions occurring justdownstream of the windward corners.2.2.2 Effect ofRoofSlopeThe pressures acting on a roof are highly dependentupon the slope oftI1e roof, general ly being positive. over the windward portion for sloPes greate r than 30degrees. For slopes lessthan 30 degrees, the windward slope can be subjected to severe suctions whichreach ama"imum at a slope of approximately 10degrees. Under extreme wind conditions,these suc-. iOJ'lS canbe of sufficient intensity to overcome thede"d w e ~ g h t o f t h e b u i l d i n g ~ thus requiring a positivet.iedownor a n c h o r ~ g e system extending from the roof.to the foundation to prevent lossof the roof system orupliftof the entire building.Intense suctions are likely to occur along the edges ofroofsand along ridge lines due to separation ordetachment of the flow at these points. For certa incombinations of roof slope and wind direction, a conical. vortex can.be developed along the windwardedges'of the roof as shown in figure 5. This is a "roi ling up" of the flow into a helical pattern with veryhigh speeds and, consequently, very intense suctions.If not adequately provided for in the design, thesevortices a long the edges of the roof ca n ca use localfailures ofthe. roofing, often leading to complete lossof the roof. Areaswhere intense suctions can be expected are shown in f i g u r e ~ . 2;2;3R()C}fQverhangsIn caIculatingthe total uplift load on a roof, thepressure acting on the underside of roof overhangs


22/33

must also be included. These pressures are usuallypositive an d the resultant force acts in the same direction as th e uplift force due to suction on the top surface of the roof. Pressures acting on the inside of thebuiding (to be discussed later) can also contribute tothe total uplift force an d must likewise be accountedfor.

VORTICES PRODUCEDALONG EDGE OF ROOFWHEN WIND BLOWSON TO A CORNER

FIGURE S. VORTICES ALONG EDGE OFROOF.

AREAS WHERE HIGHSUCTIONS MUST BEALLOWED FOR ON THECLADDING

FIGURE 6. AREAS OF INTENSE SUCTIONS.2.3 DESIGN W I ~ SPEEDSeveral factors must be considered in selecting a windspeed on which to base the design loads for a buildingor other structure. These include the climatology ofthe geographic area, the general terrain roughness,local topographical features, height of the building,expected life of the building an d acceptable level ofrisk of exceeding the design load. The assessment olclimatological wind data an d the procedure for obtaining basic wind speeds are discussed in section I !J.The selection of the basic wind speed and the det . rmination of modifying factors to obtain the designwind speed are discussed in the following sections.2.3.1 Mean Recunence IntervalThe selection of a mean recurrence interval, withwhich there is associated a certain basic wind speed,depends upon the intended purpose of a building an dthe consequences of failure. The mean tecurrence intervals in 'table 4 are recommended for the variousclasses of structures.2.3.2 Risk FactorThere is always a certain risk that wind speeds in excess of the basic wind speed will occur during the ex-

15

pected life of a building. For example the probabilitythat the basic wind speed associated with a 50-yearmean recurrence interval will be exceeded at leastonce in 50 years is 0.63. The relationship between riskof occurrence during the expected building life andthe mean recurrence interval is given in table 5. Itshould be noted that the risk of exceeding the basicwind speed is, in genera l, not equal to the risk offailure.2.3.3 Averaging Time and Peak Wind SpeedIt is well known that the longer the time interval overwhich the wind speed is averaged, the lower the h1dicated peak wind speed wiII be. The calculated designloads will thus depend upon the averaging time usedto determine the design wind speeds. In this document, it has been assumed that all speeds used inpressure and load calculations are based upon anaveraging time of 2 seconds. Wind speeds for averaging times other than 2 seconds can bt' converted into2-second average speeds using the proceduredescribed in section 1.0.2.4 DESIGN PRESSURES2.4.1 Dynamic PressureWhen a fluid such as ai r is brought to rest by impacting on a body, the kinetic energy of the moving air isconverted to a dynamiC pressure 'I, in accordancewith the formula

q= 1/2 pUwhere q = Nlm 1 , p is the mass density of the air inkglm 1 an d U is the free-stream or undisturbed windspeed in m/s. The mass density of air varies with temperature and barometric pressure, having a value of1.225 kglm 1 at standard atmospheric conditions. Inthe case of tropical storms, the mass density may be 5to 10 percent lower. However, this is offset somewhatby the effect of heavy rainfall, and the value quotedabove should be used for all wind pressure calculations, i.e.,

q= 0.613 Ul2.4.2 Mean and Fluctuating Components of

Pressure

(2)

As in the case of wind speed, pressures acting on abuilding are not steady, but fluctuate in a randommanner about some mean value. A typical recordingof wind speed and pressure at a point on the roof of ahouse is shown in figure 7.A close inspection of figure 7 reveals the followingcharacteristics:


23/33

(a) The average or mean pressure is negative (suction)(b) Pressure fluctuations tl'nd to occur in bursts(d Maximum departures from the mean are in the

negative (suction) direction(d) The peak values far exceed the mean value

2 0 ~ ~ ~ ~ t f ~ ~ ~ ~ ~ ~ ~ ~ ~ : f t i ~ ! i ~ ~ ~ ro .+ 9 r '- . .~ ~ _ ~ ~ - l Q - r - + ~ . _.

tDIE - SECOlIIlS r-

200__00o-100-200

FIGURE 1. TYPICAL RECORD OF WINDSPEED AND SURFACEPRESSLTRE.To quantify these pressures, it is essential that a sufficiently long time interval be used to obtain a stablemean. p. The fluctuations an ' described by thei r standard deviation or root-mean-square. Prms' takenabout the mean. Finally. the peak pressure f1uchlations arl' dl'scribed by a peak factor. g. which indicates the numbt.'r of standard deviat ions that the peakprl'SSllfl' dl'viatl's from thl' ml'an. Thus, the peakPTl'S.


24/33

2.4.4 Correction Factor for Height of BuildingThe pressure coefficil'nts described above are based onbuilding heights of 33 ft 00 m) and peak wind speedsat 33 ft (10 m) above ground, averaged over 2seconds.Overall loads calculated for buildings appreciabl y lessthan 33 ft (10 m) in height (measured to eaves orparapet) will thus be overestimated if these coefficients ar e used without modification. On the otherhand, tributary areas such as doors, windows, cIadding and roofing elements will respond to pressurefluctuations with duration times considerably lessthan 2 seconds. To account for thi s, the pressures mustbe multiplied by the correc tion factors, R, in table 9.Thus the expression for the net pressure acting on abuiiding surface becomes

1'= q(CpR -C,I;R;) (6)and the force acting normal to a surface of area A

where Rand R; are correction factors for external andinternal pressures, respectively.2.5 PROCEDURE FOR CALCULATINGWIND FORCESThe procedure for calculating wind forces on a building is summarized in the following steps.

1. Select the appropriate mean recurrence interval

knots

from table 42. Check the associated factor of risk in table 5 andselect a longer mean recurrence interval i f ap

propriate.3. Determine the basic wind speed for this meanrecurrence interval and the appropriate terrainroughness and type of exposure as outlined in section 1.

4. Convert the resulting basic wind speed to a 2-second mean speed using the procedure described insection 1.

5. Calculate the dynamic pressure qusing the expression'1=0.613 U 2

6. Select the appropriate pressure coefficients fromtables 6,7 an d 8.7 Select the appropriat e correction factors from ta

ble9.8. Calculate the pressures from the expressions

p= qCpRor

p = q(CpR - Cp;R;)9. Multiply these pressures by the respective surface

areas to obtain the wind forces.10. Sum appropriate components of these forces to

obtain net uplift and drag loads.

o 10 20 30 40 SO 60 70 80 90 100 110I " " ' " " !. . " ' , , , r " ' " , " " " , , " ,r , , ! , " ,r , 1 ! , 'I !. " " , , ,,I , " ,r " , , /, , , .r " , , , 1 , .r , " I, " , I " , Imp.h.o 10 20 30 40 50 60 70 80 90 100 110 120 130

VELOCITY V 1""111,,1"''''11.1'''''11''''11.1,,1.1',,11,,,.1'11,1111.1.,,,,,,"1II!,I,!"!,!!,!!",!.,!,!,, !,I,!,,!,!,.!, !"I!I"!." .!., , .1m/sec,0 5 10 15 20 25 30 35 40 45 50 55I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I . . I I I I I I I I I I I I I I I I IIbf/ft2

0 2 3 4 5 6 7 8 910 15 20 25 30 35 40, , , , , , , , , , I I I ' , , J I / ' , J I , , , I ! I ! ! I II I I II /m20 100 200 400 600 800 1000 1200 1400 1600 1800 2000DYNAMIC I ' , I , rI I I I I I I I , , I ' I II I I I I I I I I I IPRESSURE q hkgf/m2 60 70 80 90 100 120 140 160 180 20010 20 30 40 50I I I III lid I I I I I I I I I I II I II I I I I I I I I ICONVERSION CHART FOR WIND SPEED AND DYNAMIC PRESSURE HEAD

17

60I


25/33

ACKNOWLEDGMENTSAcknowledgment is made to the Building ResearchEstablishment (UK) for the illustrations used in thisdocument. The write r also wishes to acknowledgeuseful comments and suggestions provided by members of the Philipp ine Advisory Committee and byDr. Emil Simiu of the Center for Building Technology.

TABLE 4 MEAN RECURRENCE INTERVALClassof structure Mean recurrence interval yearsAll structures other than those set out below. 50Structures which have special post-disaster functions, e.g. hospitals, rommunicat ionsbuild-ing,;, etc. HIOStructures presenting a l o ~ degree o( hazard to life and other property in the case offailure. 20

TABLE S. RELATIONSHIPS BETWEEN RISK OF OCCURRENCE, MEAN RECURRENCE INTERVAL AND EXPECTED LIFE OF BUILDINGDesired Risk of exceeding in N years the wind speed correspondingLifetime to the indicated mean recurrence interval

NYears 0.632 0.50 0.40 0.30 0.20 0.10Mean Recurrence Interval in Years10 10 15 20 29 45 9520 20 29 39 56 90 19050 50 72 98 140 224 475100 100 144 196 280 448 949

Note: From this table it will be seen that there isa IO'}/ risk that the wind speed corresponding toa mean recurrence interval of 475years will be exceeded in a lifetime of 50 years.18


26/33

TABLE 6. PRESSURE COEFFICIENTS FOR WALLS OF RECTANGULAR BUILDINGSCp fo r Face LocalBuilding Building Wind Angle

Height/Width Length/width u CpRatio Ratio tDegrees, A B C DII lUI -U5 -lI.n -lI.n

I


27/33

TABLE 7. PRESSURE COEFFIOENTS FOR ROOFS OF RECTANGULAR BUILDINGSBuDding WinclAngle Area Roof Slope IJHeight/Width a Designation De, reesRatio (Degrees) 0 10 2if 25

EF -1.0 -1.0 -0.4 -03GH -0.6 -0.6 -0.8 -0.60 J -1.6 -1.9 -1.9 -1.6K -1.4 -1.4 -2.0 -1.6

h/w


28/33

TABLE 8. INTERNAL PRESSURE COEFFICIENTS FOR RECTANGULAR BUILDINGSCondition

Two opposite walls equally permeable. other walls impermeable:(a) Wind normal to permeable wall(b) Wind normal to impermeable wall

Four walls equally permeable

Dominant opening on one wall. other walls of equal permeability:(a) Dominant opening on windward wall. having a ratioof permeability of windward wall to total permeability ofotht!r walls and roofs subject to external suction. equalto--

2"or more(b) Dominant opening .10 leeward wall(e) Dominant opening on side wall(d) Dominant opening in a roof segment

Internal pressure coefficient Cpi

+0.3-0.3-0.3 or +0.2 whichever is the more severe forcombined loadings

+0.5+0.6+0.8value of e'l for leeward external wall surfacevalue of e for side external wall surfacevalue of C: for external surface of roof segment

Notes: (I) Internal pressures developed within an enclosed structure may be positiveor negative depending on the position andsize of the openings.(2) In the context of table 8 the permeability of a surface is measured by the total area of openings in the surface under consideration.(3) The value of ~ ' J ; c a n be limited or controlled to advantage by deliberate distribution of permeability in the wall or roof.or by the deliberate provision of a venting device which can serve as a dominant opening at a position having a suitableexter'nal pressure coefficient.An example of such is a ridge ventilator on a low-pitch roof. and this. under all directionsof wind. can reduce the uplift force on the roof.

TABLE 9. CORRECTION FACTOR (R) FOR HEIGHT OF BUILDINGTenain Structural System Area h< 5 5 < h < 10- -Walls Overall 0.85 1.00Elements 1.00 1.20SmoothZo < 0.12 m- Overall 0.85 1.00Roofs Elements 1.05 1.25

Internal Pressure 0.85 1.00Walls Overall 0.75 1.00Elements 0.90 1.20RoughZo > 0.12 mRoofs Overall 0.75 1.00Elements 0.95 1.25Internal Pressure 0.75 1.00

Notes: (l)The term "Overall" refers to the entire area of a given wall or roof slope.(2) The term "Elements" refers to roof and cladding elements. doors. windows. etc.(3) The terrain roughness parameter Z. must be estimated subjectively. The following values are suggested for varioustypes of exposure.

TYPE OF EXPOSURECoastalOpen countryOutskirts of towns. suburbsCenters of towns

Z. (meters)0.0050.010.020.12ol3.;:} '30'0.40

21


29/33

APPENDIX AILLUSTRATIVE EXAMPLEA housing development is to be located in flat, opencountry on th e outskirtsof Zamboanga, Philippines,an d will ultimately consist of several hundred singlefamily dwellingsof quite similar geomei:ry. Theperiod of construction is anticipated to be from 10 to15 years. The basic plan dimensions are 6.2 x 7.5 mand the height to the eaves is 2.7 m. The gable roof hasan overhang of 0.7 m on all sides an d a slope of 10degrees . Openings for doors an d windows are evenlydistributed on th e exterior walls.BecaU!ie the development is to be built over a period ofseveral years, it would not be appropriate to assume abuilt-up area 1n selecting the basic wind speed andflat, open country willbe assumed here.From table 4, a mean recurrence interval of 50 yearsis selected an d it is considered that the associated riskof exceeding th e basic wind speed (0.632) in table 5 isacceptable.From section 1, the I-minute average wind speed(N=50) for Zamboanga is 88 km/hr (Type I distribution). Since this is based on data obtained in opencountry at 10 m above ground, this speed can be converted directly to the design speed. Also, from,section 1 the ratio of the I-minute speed to th e 2-secondpeak speed is 0.82. Thus the design speed is

U = 88/0.82 = 107.3 km/hr = 29.8 m/sThe dynamic presure is calculated from equation 2 ofsection 2.4.1

q= 0.613U2 = (0.613) (29.8)2 = 544 N/m2Wind pressures are next calculated using equations4-6 and the coefficients presented in table 6-9. Notethat

h/ w = 2.7/6.2 = 0.44an d l/w = 7.5/6.2 = 1.21WALLSInspection of tables 6 an d 8 reveals that the worstcases ar e walls A an d C with the wind blowing normal to th e ridge. For wall A, Cl!=0.8 an d for wall C,Cp = -0.6. The local Cp is -1.2. The internal pressurecoefficients can range fr.om 0.2 to -0.3. Table 9 indicates that the reduction factor is 0.6.5 for walls an d internal pressures an d 1.00 for cladding elements, doors,windows, etc.

22

For wall A,

For wall C,

p= (544) 10.8-(-0.3)](0.85)= 509N/m2

p= (544) i 0.6-(0.2) J(O.85)= -370 N/m2For cladding elements, the worst cases are

and

p= (544) [0.8 - (-0.3) (0.85)]= 574 N/m2

p = (544) [-0.6 - (0.2) (0.85)]= -419 N/m2

For local pressures acting on strips of width 0.2 w =1.2 m at each corner,

ROOF

p = (544) (-1.2 - 0.2) (0.85)=-647 N/m2

Inspection of table 7 reveals that the greates t upl iftpressures on extended an'as occur when the wind isblowing along the ridge.For sections E and G,

p= (544) [-1.1 - (O.2)J (0.85)= -601 N/m2For sections F and H,

p = (544) 1-0.6 - (0.2)] (0.85)= -370 N/m2

Pressures acting on roofing elements in sections E andG ar.e ot;tained as follows:

p= (544) [(-1.1)( 1.05) - (0.2)(0.85)]= -727 N/m2

and for sections F and H,p= (544) [(-0.6)(1.05) - (0.2)(0.85)]

= -438 N/m2Localized pressures act on ".trips of width 0.15 w =0.93 m as shown in table 7. The worst case occurs forarea Jwith th e wind bicwmg normal to thE' ridge.Note that th e uplift pressure under the eaves mustalso be included.


30/33

p= 544) L1.9 - (0.8)] (0.85)=-1.2k N/ml

For area K in section F. this negative pressure or suction is slightly less

p= (544) [-1.4 - (0.8)] (0.85)= -l.Ok N/m lAlong the ridge (area K). the localized pressure is

p= 544) [-1.4 - (0.2)] (0.85)=-740N/m2

TOTAL UPLIFf FORCEThe total uplift force on the building is calculated forthe wind blowing normal to the ridge as follows:Area of one roofslope = 7.5 + (2)(0.7)] [6.2/ (2Cos 10)+0.71= 8.9)(3.85)=34.2m2Note that areas E. F. G and H include areas Jand Kwhen calculating overall loads.Uplift = 544) (t.0 +0.6) (34.2) (Cos 100 ) (0.85)+ (544)(6.2)(7.5) (0.2)(0.85)

= 29.2kNTOTAL DRAG FORCEThe total drag force (neglecting the roof) is calculatedas the sum of the loads on the windward and leewardwalls.

Drag = 544)(2.7)(7.5) [0.8 - (-0.5)] (0.85)=12.2kN

23

COMMENTThe loads calculated above are the loads that canreasonably be expected to occur under the conditionsstated in the example. They should be considered asthe minimum suitable loads for use with stresses andload factors appropriate for the type of structuralmaterial used.For geographical areas exhibiting large variations inannual extreme wind speeds. the basic wind speedshould be selected. with caution. The application ofprobabilistic models of extreme wind speeds andsome of their limitations are discussed in section 1.0.


31/33

ANNOUNCEMENT OF NEW PUBLICATIONS INBUILDING SCIENCE SERIESSuperintendent of Documents,Government Printing Office,Washington, D.C., 20402Dear Sir:

Please add my name to the announcement list of new pUblications to beissued in the series: National Bureau of Standards Building Science Series.~ a m e ___________________________________________________----------------COmpany_________________________________________________ ---------------AddreM__________________________________________________ - - - - - - - - - - - - - - -City____________________State__________ . _________Zip Code_____ _________ _(Notification key N-339)


32/33

NBS TECHNICAL PUBLICATIONSPERIODICALS

JOURNAL OF RESEARCH reports National Bureauof Standards research and development in physics,mathematics, and chemistry. I t is published in twosections, a"ailable separately: Physics and Chemist ry (Sedion A)Papers of interest primarily to scientists workingthese fields. This section covers a broad range of physlcal and chemical research, with major emphasis onstandards of physical measurement, fundamental constants, and properties of matter. Issued six times a year.t\nnual subscription: Domestic, $17.00; Foreign, $21.25. Mathematical Scieru:es (Section B)Studies and compilations designed mainly for the mathematician and theoretical physicist. Topics in mathematical statistics, theory of experiment design, numericalanalysis, theoretical physics and chemistry, logical design and programming of computers and computer systems. Short numerical tables. Issued quarterly. Annualsubscription: Domestic, $9.00; Foreign, $11.25.DIMENSIONS/NBS (formerly Technical News Bulletin)-This monthly magazine is published to informscientists, engineers, businessmen, industry, teachers,students, and consumerS of the latest advances inscience and technology, with primary emphasis on thework at NBS. The magazine highlights and reviewssuch issues as energy research, fire protection, buildingtechnology, metric conversion, pollution abatement,health and safety, and consumer product performance.In addition, i t reports the results of Bureau programsin measurement standards and techniques, properties ofmstter and materials, engineering standards and services, instrumentation, and automatic data processing.Annual subSCription: Domestic, $12.50; Foreign,$15.65.1I0NPEIIODICALS

Monographs-Major contributions to the technical liter ature on varioua subjects related to the Bureau's scientific and technical activities.Budbooks-Recommended codes of engineering andindustrial practice (including safety codes) developedin cooperation with interested industries, professionalorganizations, and regulatory bodies.Special Publications-Includeproceedings of conferencessponsored by NBS, NBS annual reports, and otherspecial publications appropriate to this grouping suchas wall charts, pocket cards, and bibliographies.Applied Mathematics Series-Mathematical tables, manuals. and studies of special interest to physicists, engineers, chemists, biologists, mathematicians, computer programmers, and others engaged in scientificand technical work.National Standard Reference Data Series-Providesquantitative data on the physical and chemical properties of materials, compiled from the world's literatureand critically evaluated. Developed under a world-wideprogram coordinated by NBS. Program under authorityof National Standard Data Act (Public Law 90-396).

NOTE: At present the principal publication outlet forthese data is the Journal of Physical and ChemicalReference Data (JPCRD) published quarterly fo r NBSby the American Chemical Society (ACS) and the American Institute of Physics (AlP). Subscriptions, reprints,and supplements available from ACS, 1155 SixteenthSt . N.W., Wash. D. C. 20056 Building Science Series-Disseminates technical information developed at the Bureau on building materials,components, systems, and whole structures. The seriespresents research results, test methods, and performance criteria related to the structural and environmentalfunctions and the durability and safety characteristicsof building elements and systems.Teehnical Notes-Studies or reports which are completein themselves bu t restrictive in their treatment of asubject. Analogous to monographs bu t not so comprehensive in scope or definitive in treatment of the subject area. Often serve as a vehicle fo r final reports ofwork performed at NBS under the sponsorship of othergovernment agencies.Voluntary Product Standards-Developed under procedures published by the Department of Commerce in Part10 Title 15, of the Code of Federal Regulations. Thep u ~ p o s e of the standards is to establish nationally recognized requirements fo r products, and to provide allconcerned interests with a basis for common understanding of the characteristics of the products. NBSadministers this program as a supplement to the activities of the private sector s t a n d a r d i ~ i n g organizations.Consumer Information Series-Practical information,based on NBS research and experience, covering areasof interest to the consumer. Easily understandable language and illustrations provide useful background knowledge for shopping in today's technological marketplace.Order above NBS publications from: Superintendentof Document8, Government Printing Office, Washington,D.C.2040S.Order following NBS publication8-NBSIB'8 and FIPSfTom the National Technical Information Service8,Springfield, Va.2S161.Federal Information Processing Standards Publications(FIPS PUBS)-Publications in this series collectivelyconstitute the Federal Information Processing Standards Register. Register serves as the official source ofinformation in the Federal Government regarding standards issued by NBS pursuant to the Federal Propertyand Administrative Services Act of 1949 as amended,Public Law 89-306 (79 Stat. 1127), and as implementedby Executive Order 11717 (38 FR 12315, dated May 11,1973) and Part 6 of Title 15 CFR (Code of FederalRegulations) .NBS Interagency Reports (NBSIR)-A special series ofinterim or final reports on work performed by NBS foroutside sponsors (both government and non-government). In general, initial distribution is handled by thesponsor; public distribution is by the National T e c h I ~ i . cal Information Services (Springfield, Va. 22161) lDpaper copy or microfiche form.

BIBLIOGRAPHIC SUBSCRIPTION SERVICESThe following current-awareness and literature-surveybibliographies are inued periodically by the Bureau:Cryogenic Data Center Current Awareness Service. Aliterature survey issued biweekly. Annual subscription: Domestic, 125.00 ; Foreign, 13O.00t.Liqlliled Natural Gu. A literature survey issued quarterly. Annual 8ubscription: $20.00.

Superconducting Devices and Materials. A literaturesurvey issued quarterly. Annual subscription: $30.00'.Send subscription orders and remittances for the preceding bibliographic services to National Bureau ofStandards, Cryogenic Data Center (275.02) Boulder,Colorado 80302.


33/33

u.. . . . .AIIIIIIMf OF caM.1IICI...........................o.c. 20234-FFICIAL BUSINESSPenalty fa r PrWe&e Use 300

POSTAGE AND !'tEa PAIDu.s. DEPARTMENT OF COMMEACICOM_211

SPECIAL FOURTHCLASS RATEBOOK