NBC - 1970 - Climatic Data

 

CLIMATIC INFORMATION for

BUILDING DESIGN IN CANADA

SUPPLEMENT No 1 T O THE NATIONAL BUILDING CODE OF C A N A D A

ssued by the

ASSOCIATE

COMM CO MMITTEE ITTEE ON THE NATIONAL BUILDING NATIO NA TIONA NAL L RESEARCH COUNCIL COUNCIL

CODE

OTTAWA CANADA

rice 25 cents

NRC No 11153

Copyright NRC-CNRC

 

NATIONAL RESEARCH COUNCIL ASSOCIATE COMMITTEE ON THE NATIONAL BUILDING CODE 1968 1970

R. F. Legget (Ch (Chai airm rman an))

G.C. Lount

D.C. Beam

I

J.D. Beaty

D.A. Matheson

R.A. Bird

H.H.G. Moody

S.D.C. Chutter

A.T. Muir

W.G. Connelly R.F. DeGrace

L.P. Picard K.R. Rybka

H.B. H. B. Dickens (Vi (Vice ce Chairman)

S.A. Sasso*

A.F. Duffus

R.A.W. Switzer

3.5. Dussault

I

W.R. Edmonds*

C.D. Carruthers Carruthers (ex o ff ffic icio io))

H. Elder

P. Dobush (ex o ff ffic icio io))

J.L. Jolicoeur

C.G.E. Downing

H.A. Lawless

T. T.R. R. Durley (ex of fi fici cio) o)

Maclennan

Campbel Camp belll (ex off cio)

L. L.A. A. Kay (ex of offic ficio io)) R.S. Ferguson Resear Research ch Ad Advi viso sor) r) Deceased

J. J.M. M. Robe R obertson rtson (Secretary (Sec retary))

Copyright NRC-CNRC  

CLIMATIC INFORMATION FOR BUILDING DESIGN I N C N D SUPPLEMENT N o N

L B U I L D I N G C O DE DE O F C

TION

N

D

TABLE OF CONTENTS

..................... uly Design Temperatures Temperatures . . . . . . . . . . . . . . . . . . . . . . . eating Degree Days . . . . . . . . . . . . . . . . . . . . . . . . . ........................... ainfall Intensity

anuary Design Temperatures

Page

Chart No.

4

1 an d 2

5

3

and 5

6

ne Day Rainfall

7

7

nnual Total Precipitation

7

8

7

9

....................... now Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . in ind d Effe cts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ermafrost

...............................

eismic Zones References

able of D esign D Dat ataa for Selected Locations

10 11

11

11

12

11

4

Copyright NRC-CNRC  

NOT S

Copyright NRC-CNRC  

CLIM

T I C I NF O R M

T IO N

or

B U I L D IN IN G D E S I G N IN C

N

D

by Donald

W.

oyd

D.O.T. Meteorologist Meteorologist with DB RINRC ) A joint contribution from the Meteorological Branch, Department of Transport, and the

Division of Building Research, National Research Council.

The great diversit diversityy of climate in Canada has a consi considerable derable effect on the p erforman ce of buildings, and consequently the design of buildings should reflect this diversity. The purposes of this ha ndbook are: firstly, by means of m aps, to indicate the vari variabil abilit ityy and general distribution of earthquake intensi intensity, ty, permafrost, and those cli climatic matic elements tha t are most frequently considered sid ered in building design; secondly, t o explain briefl brieflyy how the desi design gn weather valu values es are computed; an d, thir thirdly, dly, to present recomm ended des desig ignn data for a number of citi cities es and towns and smaller small er populated places places.. I t is no t practical t o lis listt value valuess for all munici municipaliti palities es b ut recommended design weather data for any location in Canada can be obtained by writing to the Secretary, Associat Assoc iatee C om m ittee on t he National Buildi Building ng Cod e, National Resear Research ch Council, Ottaw a. The choice of climatic elements that are included in this handbook and the form in which they are expres expressed sed has bee n dicta ted largel largelyy by the requirem ent for specif specific ic val values ues in several seve ral sec sections tions of t h e Na tional Buil Building ding Co de of Canada. A few add ition al charts are included. The following notes explain briefly the significance of these particular elements in building design and indicate what observations were used and how they were analysed to yield the required design values. To select design values for other locations in Canada, the observed or computed val values ues of each element for speci specific fic obser observation vation sstations tations were plotted on m aps to th e scal scale e oofthe f one inch t o 100 of miles 1 in 5,000 ,000The . Is Isoli olines nes were drawn on these working to show distributions the or design values. charts in this handbook have been charts reduced from the working charts, but these small copies are not intended as a source of design values. In the Table , des design ign weather d ata a re listed for over 600 loc ations, which have been chosen for a variety of reasons. Incorporated cities and towns with populations of over 5,000 have been included unless they are close to other larger cities. For sparcely populated areas many small smaller er towns and villa villages ges have been listed listed.. Th e de desig signn weather data for weather station s themselves are the most reliable and hence these stations have often been listed in preference to locations with somewhat larger populations. A number of requests for recommended design weather d ata for oth er locations have been rec receive eivedd an d where m ost of th e elem ents were eesti sti--

mated thes thesee wer weree also added t o the list list.. The tabulated valu values es are those recommended b y th e Associate Associ ate Com mittee on t h e National Build Building ing Code and are not neces necessar saril ilyy the same as those used in local byla bylaws. ws. In som e cases the valu values es obtained fro m the large-s large-scale cale charts have not bee n rounded off, for reasons explained later. Some municipalities may wish to round off these values in in th eir bylaws. Neither the charts no r the list list of desig designn vvalu alues es should be expec ted to gi give ve a complete variations ions of these climatic elements. If application is made t o the Secretary as picture of the variat mentioned above then values will be estimated for any location not included in the Table using the list of observed or computed values for weather stations, the large-scale manuscript charts and an y more recent information tha t is avai availabl lable. e. In th e absence ooff we ather observatio observations ns a t any particul part icular ar location, a knowledge of the loca locall topography may be important. F or example, ccol oldd air has a tendency t o collec collectt in depress depressions, ions, precipitation frequ ently incr increases eases with elevation, and winds a re gen generally erally str stronger onger near larg largee bodies of water. These and othe r relati relationships onships affect the correspond corresponding ing design valu values, es, and will be ta ken i nto consideration w here possib possible le in answering inquiries.

Copyright NRC-CNRC  

All the weather records that were used in preparing the charts were, of necessity, observed obse rved at inhabited locations, and hence th e charts apply only t o populated areas. areas. This is is parparticul ticularly significa signi ficant nt int omthe ounmo taino us areas where the lines line s on th e charts apply t o thdifferent e populateddarly late valle valleys ys and not untain slopes and high passes passes, , where, in some cases, casonly es, very conditions conditi ons are known to exist. exist.

JANUARY DESIGN TEMPERATURES (CHARTS 1 A N D 2 A building an d its heating system system should be designe designedd t o maintain the inside inside tem peratu re at some p rede term ined leve level. l. T o do this it is necessa necessary ry t o know the m ost sever severee weather weather conditions under which th e system system will be expected t o fun ctio n satisfa satisfactori ctorily. ly. Failure Failure t o m aintain the insi in side de tem perature a t th e pre determ ined level level will will usually usually not be seriou seriouss if th e temperature drop is not great and if th e duratio n is not long. Th e outside conditions used used for design should, therefore, no t b e th e most severe severe in many years, bu t sh ould be the som ewhat less less severe severe conditions that are occasiona occasionall llyy bu t no t greatly greatly exceeded. Winter design Winter design tem peratu re is based o n an analysis of of winter air temperatures o nly. Wind Wind and solar solar radiation also also affect th e inside tem peratu re of m ost buildings bu t there is no convenient way of com binating their their effects with that of o utside air temperature. Som e qu ite comp lex me thod s of taking account of severa severall weather elements have been devised devised and us used ed in r ecent years years bu t the use o f average average wind an d radiation co nditions is usually usually satisfactory for design design purposes. purposes. The choice of a m ethod t o determine the the winter desig designn temperature depends to some extent on the form of the avail availab able le temperature records. records. I n Canada, hourly temperatures temperatures in degrees degr ees Fahrenheit for ten succes successi sive ve years were availabl availablee on punche d cards for over 10 0 stations, and from these cards it was possibl possiblee t o obta in frequency distributions. Th e winter design design temperature is defined, therefore, as the lowest temperature at or below which only a certain small percentage percent age of t he hou rly outsid e air air tem peratures in Ja nua ry occur. Th e Climatology Divisio Division, n, Meteorolog Meteor ologica icall Branch, Branch, D epartmen t of Transport, prepared tabulations showing showing the num ber of hours in January in the ten years from 195 1 o 19 60 incl inclus usiv ivee in which the tem perature fell fell in each of over over 50 two degr ee inte interva rvals ls.. From this it was possibl possiblee t o select select the two de gre e iinte nterval rval below bel ow which which only a small small number of temperatures fell. fell. T o find find th e requir required ed tem perature to t he nearest degree degree an interpo lation rule was devised devised based o n the n ormal distribution. Using Using this rule it was possible to select the temperature below which 1 per cent or 2 temperatures fell.

per cent of th e January

Tabulations and January design design temperatures for 118 stations were were obtained. T he temperaturess were plotted on m aps and used rature used t o estimate estimate desi design gn temperatures for th e other stations in the Table. Table. Since Since t he pattern of January design design temper ature charts is simil similar ar to that of mean annual minimum minimum temperature charts, charts, th e latter chart in the Atlas Atlas of Canada Canada 1) infl influenced uenced the pattern of these these January desig designn temperature charts in the Far North where hourly temperature

observations a re scarce. Most of th e design design tem peratu res on the 2 per cen t chart in the Prairie Prairie Provinces Provinces and Briti Bri tish sh Columbia Columbia are 5 to 10 degree degreess hig higher her th an th ey are on the corresponding corresponding chart b y T hom as in t he 195 3 edition of th e National Buildi Building ng Code of C anada (2). Each chart is based on tempera ttuu re s fo r o n ly a 1 0 -y -y e ar ar p e rio rio d : th e 1 9 5 3 ch ch a rt o n th e p e riod riod fro m 1 9 4 1 t o 1 9 5 0 a n d th e current chart o n the period from 19 51 to 1960. Th e differenc differences es emphasize emphasize the statisti statistical cal inadequacyy of a 10-year inadequac 10-year period, but unfortunately tabulations of h ourly temperature distribudistributions f or longer periods are no t availab available. le. The earlier period includes includes the unusually cold January of 19 50 when th e aver average age temperature in th e four western provinc provinces es range rangedd from 1 2 to 3 2 degrees degr ees below normal. By om itting this ex ceptional m on th it is is thou ght tha t th e present value valuess willl mo re nearly describe a typical winter. A m ore recent tab ulatio n of hou rly temp eratu re diswil distributions tributions for all months fo r th e 10-year 10-year period period 1957 t o 1966 has has been publishe publishedd for 8 3 weather stations (3). The earlier tabulation for 35 more stations is still the best basis for a consistent set of design temperatures b ut th e mo re recen t tabu lation could provide provide design design tem pera tures for othe r m onth s besides besides January. In most case casess the temperatures temperatures obser observed ved a t airports have had t o be used and no adjustadjustments have been made to allow allow for the city effect effect.. T h e January January w inter desig designn temperatures are not reliable reliable to w ithin ithin on e degree, bu t the differences differences between th e 1 and 2 per cent values values

Copyright NRC-CNRC  

(which average about four degrees) are more reliable. The design temperatures, therefore, are liste listedd to th e nearest nearest degree as an indication of these difference differences. s. T h e 2 per cent January design temperature is the value ordinarily used in the design of heating heat ing systems. systems. In speci special al ca cases ses when the contro l of inside inside tem peratu re is mo re critical, critical, the 1 per cent value may b e used. JULY DESIGN TEMPERATURES (CHARTS

3

AND 4

A building building and its cooling and dehum idifying ssystem ystem should be designed designed t o maintain th e inside temperature and humidity at certain predetermined levels. To do this it is necessary to know th e m ost severe weather conditions under which th e system system wil willl be expected to fun ction satis sat isfac factor toril ily. y. Fai Failur luree to maintai maintainn th e ins inside ide temperature and hu midity a t the prede term ined le level velss will usual usually ly no t be serious if the increases increases in in tem per atu re and hum idity are no t great an d if the duration is not long. The outside coilditions used for design should, therefore, not be the most seve severe re in many years, years, but should be the somew hat lless ess sseve evere re conditions th at a re occa occasionsional ally ly bu t n ot greatly exceeded. Th e summ er desi design gn tem peratur es in in this supp lement are based based on an analysi analysiss of July air temperatures and humidities only. Wind an d solar radiation also also affect the insi inside de temperatu re of most buildings and may in som e cases cases be of m ore im portan ce than th e outside air temperature. I t seems, however, tha t no meth od of all allowing owing for variat variations ions in radiation has yet become generally accepted. When requirements have been standardized it may be possible to provide more complete weather information for summer conditions but in the meantime only dry-bulb and wet-bulb design temperatures can be provided. The frequency distributions of combinations of dry-bulb and wet-bulb temperatures for eachh m onth from June to September hav eac havee been tabulated for 33 Cana Canadia diann weather sstat tations ions by Boughner (4). If th e sum mer dry-bulb and wet-bulb design design temperatures are defined defined as the temperatures that are exceeded 2 per cent of of th e hour s in in July, then design design values values can be obtaine d directly for these 33 stations. As mentioned above, th e pattern of January desi design gn tempe rature is is simi simila larr to th at for the mean annual minimum. In the same way, the pattern of July design temperature is much like that of the mean annual maximum. Crow (5) used these similarities as a basis for computing designn temp eratures f or pplace desig lacess in the U. U.S. S.A. A. for which on ly daily daily maxim um an d minim um temperatures were observed. Th e July dry-bulb dry-bulb design design temperatures for th e 33 Canadian stations were subtracted from the mean annual maximum temperatures (for the same period of years) and t he diff difference erencess p lotted on a m ap. Th e differences are all between 3 and 1 1 degrees degrees.. With

this small range, range, th e 33 stations seem to b e enough t o esta establi blish sh diff differenc erences es (w ithin an accuracy of about one degr degree) ee) for an y location. Mean Mean annual maxi maxima ma base basedd o n the period period 1 92 1 to 19 50 were avai availa lable ble for a bo ut 17 0 llocati ocations. ons. Fo r these, the differences were read or estimated from the map and July dry-bulb dry-bulb design design temperatures obtained. These 1 70 stations were used used to preparee Ch art 3 from wh ich val prepar values ues were estimated for ab ou t 450 additio nal locat locations. ions. A mo re recent tabulation of h ourly temperatu re distributions distributions for all months for t he 1010-yea yearr period period 1957 to 19 66 hhas as been published for 8 3 weather statio ns (3 (3). ). Th e 1 70 stations used for Cha rt 3 are probably still the best basis for a consistent set of design temperatures but the more recent tabulations could provide desig designn tem peratu res for ot her m on ths besides besides July. The July wet-bulb des desig ignn temperatur temperatures es for th e 3 3 stations stations were plotted directl directlyy on a map. The range range from 62 t o 75 (excl (excludi uding ng Yukon and NWT) is a little more than for the dry-bulb differences, but is still small enough to yield reasonably accurate wet-bulb design temperatures. Th e north ern part of th e ch art was no t draw n in because d ata are ver veryy sparse sparse and because cool cooling ing and dehumidifying are seldom seldom needed in tha t area. July 2 per cent wet-bulb wet-bulb design design temp sratures values were read from the map for all locations with dry-bulb design values of 75 F or higher. In many cas cases es the tem peratures observed observed a t airp orts have had t o be used used and no adjustments hav havee been m ade t o allow allow for th e city effect. effect. Th e summ er desi design gn temperatures are no t relia reliable ble to within within on e degree degree although although they have been estimated and tabulated t o th e nea nearest rest deg degree. ree.

Copyright NRC-CNRC  

HEATING HEATI NG DEGREE-DAYS (CHART 5) It has long long bee9 known that th e amount o off fuel or energ energy y requ ed to keep the interior of a building building at abo ut 70 F whenJhe whenJhe outside air temperature if if below 65 F iiss roughly roughly proportional to the difference between 65 F and the outside temperature. Wind speed and solar radiation, and the extent to which a building is exposed to these elements, also affect the heat required, but there is no convenient way of combining these effects. For average wind and radiation conditions, however however,, the proportionality with the temperature difference difference still holds and hence the heating heatin g degreedays are based based on temperature alone. Since the fuel requir required ed is also proportional to the duration of cold weather, weather, a convenient convenient meJhod meJ hod of combining these elements of temper temperature ature and time is to add th e differences between 65 F and ando the mean temperatu res for every day in the year when the t he mean temperature tempe rature is below 65 F. F.dt dt is assumed assumed that no heat h eat is required when the mean outside air temperature temper ature for the day is 65 F or higher. higher. Daily degr eed eedays ays have been computed compu ted for many years at Victoria, Win Winnipeg, nipeg, Tor onto ont o and Halifax. The values given in the Table for these four cities are the average annual totals for the 3030-yea yearr per period iod from 1931 to 1960. Daily degreedays are not available for the full 30-year period for other stations. An approximate but reasonably accurate method of obtaining degreedays from monthly mean temperatures was devised by Thom (6). This method was used by Thomas and Boyd 7 ) in 1956 to compute compute normal monthly and annual degreeday totals base based d on the period 1921 to 1950, which were used as a basis for the map in the 1961 Supplement. In 1964 an electronic computer at the National Research Council Computation Centre was used to compute monthly and annual degree deg ree-da -days ys for over 60 0 stations stations based on th e period 1931 to 1960 (8). The annual totals were were plotted on a map (Chart 5) and use used d to estimate values values for locatio locations ns without weather weather stations. stations. Computed values are shown in the Table to the nearest unit as computed but should not be relied relied on t o within less less than 10 0 t o 15 0 degreedays. Values read read from the manuscript manuscript chart are to the nea neares restt 100 degreeday degreedays. s.

RAINFALL INTENSITY (CHART 6) Roof drainage systems are designed to carry off the rainwater from the most intense rainfall rainfa ll that is likely to occur. A certain amou nt of time is require required d for th e rainwat rainwater er t o flow

across or down the roof before it enters the gutter or drainage system. This results in the smoothing out of the most rapid changes in rainfall intensit intensity. y. The drainage system, therefore, need cope only with the th e flow of rainwater produced by th e average average rainfall intensity inten sity over a period of a few minutes which can be called the concentration time. In Canada, it has been customary to use the 15-minute rainfall that will probably be exceeded on the average once in ten years. The concentration time for small roofs is much less than 15 minutes and hence the design intensity will will be exceeded more frequently freque ntly than once in ten year years. s. The Th e safety factors facto rs included in th the e tables in Part 7 of the National Building Building Code, wi will ll probably reduce the frequency to a reasonable value and, in addition, the occasional failure of a roof drainage system will will no t be particularly serious in mos mostt case cases. s.

Chart 6 is a revision of the corresponding charts by Thomas (2) and by Bruce (9).which show the 15-minute rainfall, in inches, that will probably be exceeded on the average once in ten year years. s. rain As Bruce there the re wereaavai available lable to length him only nine From locations locat ions inhe Canada with recording gauge explained, gauge observations covering reasonable of time. these computed comput ed the 15-minute, ten-year, rainfall. rainfall. Bruce also computed comput ed th e maximum 6-hour rainfall expecte expected d once in ten years for 85 locations locations and used used th e ratios of 15-minute to 6-hour rainfalls rainfalls at six six stations where where both were were avai availab lable le to t o estima estimate te t he 15-minute val value ue for the oth er locations. locations. RainRainfall intensities for some locations in northern Canada estimated by the United States Weather Bureau and the latest rainfall intensity figures for cities in the United States near the Canadian border were also also used. used. Since the publication of his paper Bruce has analysed t he rainfall records from several additional stations and these have been used in redrawing the chart. .

It would be ver very y difficult t o estimate the t he p attern of rainfall rainfall intensity in Britis British h Columbia where precipitation is extremely variable. Along the coast an at atte te mpt mp t ha hass been made, based on reports from Victoria and Vancouver and a few stations in the State of Washington. In the

Copyright NRC-CNRC  

interior of British Columbia, th e value of 0. 0.6 6 inch based based on a seven-year record from Pen Pentict ticton on is the only available available indication of the intensity for all the valleys valleys in the southeast s outheastern ern part of t he Province. In the Fraser Valley and further north, the value may be slightly smaller. ONE-DAY RAINFALL (CHART 7 ) If, for any reason, a roofdrainage system becomes ineffective, the accumulation of rainwaterr may b e great enough in some cases wate cases to cause a significant increase in the load on t he roof. Although the period during which rainwater may accumulate is unknown, it is common practice t o use the maximum one-day rainfall for estimating the addi additional tional load. For most weather stations in Canada the total rainfall for each day is published. The maximum oneday one day rainfall (as (as it is usuall usually y c calle alled) d) for sever several al hundre hundred d stations station s has been determined and published by the Climatology Division (10). Since these values are all for predetermined 24-hour periods, beginning beginning and ending at the th e same time each m morning, orning, it is probable that most of them have been exceeded in periods of 24 hours ho urs including including parts of two consecutive days. The Th e maximum 24-hour 24-hour rainfall (i.e (i.e.. any 24-hour 24-hour period) according t o Hershfie Hers hfield ld and Wilso Wilson n is, on the th e average, average, about abo ut 113 1 13 per cent of the th e maximum maximum oneday rainfall (1 1). Most of the rainfall amounts that were used in preparing Chart 7 were based o n reports covering between 20 and an d 30 year years. s. These maximum values differ greatly within relatively relativ ely small areas where little difference would would be expected. The variable variable length of of records may account ac count for part of this variability which might be reduced b y a n analys analysis is of annual a nnual maxima instead of merely selecting the maximum in the period of record. Whatever the reason, the variability has necessitated necessitated a considerable amount of smoothing in drawing the chart and hence the isolines do not in all cases agree with the observed maximum oneday rainfalls. The tabulated values val ues are intended t o be representativ representative e of t he immediate area, and therefore therefor e include some local local variations which which cannot ca nnot be shown on the small small-s -scal cale e chart. ANNUAL ANN UAL TOTAL PRECIPITATION PRECI PITATION (CHART 8)

The tota l am ount oun t of precipitation that th at normally falls in one year is frequently frequen tly u used sed as as a general indication of t he wetness of of a climate. As such it is thoug ht t o have a place in this handbook. Total precipitation is the sum in inches of the measured depth of rainwater and one tenth of t he meas measured ured depth dep th of snow (since th e average average density of fresh ssnow now is about abo ut one o ne tent h tha t of water). The average average annual to tal precipitations for fo r the 30-ye 30-year ar period period from 1 92 921 1 t o 1950 195 0 were used in preparing Chart 8. Th The e values values were were selected from a list of precipitat pr ecipitation ion normals prepared prepared by the th e Climatology Divisi Division on (12). All stations with with records for fo r the full f ull 30 years were plotted on the map or compared compared with nearby stations that had already already been been p lotted t o ensure consistenc consistency. y. Many Ma ny adjusted values were used in areas where unadjusted 30-year values were not available. Th e corresponding chart in the t he Atlas of Canada (1) was used for reference. SNOW LOADS (CHART 9) The roof of a buildin building g should should be able able to sup por t,th e greatest weight of of snow t ha hatt is likely t o accumulate ac cumulate on it. Some observations observations of snow loads on roofs have have been made in recent years, but they are not sufficiently numerous to form the th e basis basis for a snow load load chart. Similarly, observations of the weight or water equivalent of the snow on the ground are not sufficient for such a chart. Although th e roof load and water equivalent observations are neces necessary sary (a (ass mentioned mentione d below) the chart must be based on the more numerous observations of the depth of snow on the ground. The estimation of the th e design snow load on a roof from fr om snow depth observations involves involves the following steps: steps: 1. The d epth ept h of snow on the ground which will be equalled or exceeded exceeded once onc e in 30 years years,, on the th e average, average, iiss computed. 2. A density density is assu assumed med and used used to t o convert snow depths to loads. 3. An adj adjustment ustment is added t o allow for the increase in th e load caused by rainwater absorbed by the snow.

Copyright NRC-CNRC  

4. Because Because th e accumulation of snow on roofs is often different differe nt from that t hat on th the e ground, certain adjustments adjustme nts should be made to the ground snow load t o provide a design design snow snow load on a roof. These steps are explained in more detail in the th e following paragr paragraphs. aphs. The annual maximum maximum depths of sno snow w on t he ground for periods rangin ranging g from 10 to 18 years were available available for over over 200 stations. These data were assembled assembled and analysed using using Gumbel's Gumbel 's extreme extrem e value method met hod as explained by Boyd (13). The resulting chart ch art showed the distribution in Canada of the snow depth which will probably be equalled or exceeded on the average once in 30 years, or which has a probability of 1 n 30 of being exceeded in any one year. The specific gravity of old snow generally ranges from 0.2 t o 0.4 times that tha t of water. It is usuall usually y assumed in Canada ttha hatt 0.1 is the th e average specific specif ic gravity of new snow. The 30-year maximum maximu m sno snow w de pth wil willl almost certainly occur immediately after af ter an unusually heavy heavy snowfall and hence a large proportion of the snow cover will have a low density. It therefore seemed reasonable t o assume a mean specific gravity under these unusual circumstances of abo ut 0.2 for the whole sno snow w cover. cover. In practice it is convenient t o assume th at one inch of snow cove coverr corresponds to a load of exactly one pound per square foot. This corresponds to a specific gravity gravit y of 0.192, the th e value which w was as used used in preparing th e Chart. Because the heaviest loads in Canada frequently occur when early spring rain adds to an already heavy snow load, it was considered advisable to increase the snow load by the load of rainwater that it might retain. It is convenient to use the maximum oneday rainfall in the period of the t he yea yearr when snow de depth pth s are greatest. Boyd has explained how a 2- or 3-month period was selected (1 (13). 3). The results from several winters of a survey of snow loads on roofs indicated that averag ave rage e roof loads were generally much less less tha n loads on the ground. The conditions under

which th e design design snow load on o n t he roof may be taken as 80 or 60 per cent of th e ground snow load are given in Section 4.1 4.1 of tthe he National Building Code 1 965. 965 . Th e Code also permits further furt her decreases decrea ses in desig design n snow loads for steeply sloping roofs, bur requires substantial substanti al increase increasess for roofs where snow accumulation may be more rapid. Recommended adjustment adjus tmentss are give given n in Supplement No. No. 4 to t he National Buil Building ding Code, Canada, 1970. Chart 7 sho shows ws th e gen general eral distribution of sno snow w loads loads on th e ground, that is is,, th e load due to snow which will be exceeded on the average once in 30 years, plus the load due to the maximum oneday rainfall in the late winter or early spring. Values of the snow loads on the ground were read from fro m the t he larg large-s e-scal cale e original of Chart and are listed in th e Table. The snow snow loads are tabulated in whole pounds per square foo t but are not reliable t o this accuracy. Charts on such a small scale as those in this Supplement cannot show local differences in the th e weather elements, eve even n where these are known to exist. All the t he weather obse.wationsuse used d in preparing Chart were, of necessity, taken at inhabited locations, and hence th e charts apply only t o permanently popula populated ted areas. This is particularly significant iin n mountainou moun tainouss areas where the lines on the chart apply only only t o the t he populated valley valleyss and not t o the mountain mountain slopes, slopes, where, in some cases, much greater snow depths are known to accumulate and must be taken into account in the design design of roofs. roofs.

WIN

EFFECTS (CHART 10)

All structures should be built to withstand the pressures and suctions caused by the strongest gust of wind that is likely to blow at the site in many years. For many buildings this is th e only wind effect that needs t o be considered, considered, bu t tall o r slender structures should should also also be designed to limit their vibrations to acceptable levels. Wind induced vibrations may require several minutes to build up to their maximum amplitude and hence wind speeds averaged over severall minutes o severa orr longer should be used for design. The hourly average wind speed is th the e value value available in Canada. The provision provision of veloc velocity ity pressu pressures res for bot h avera average ge wind speeds and gust gust speeds for estimating pressures, pressures, suctions and vibrations involves th e following steps:

1. The annual maximu maximum m hourly wind speeds were analysed analysed t o obtain the hourly wind wind

speedss tha speed thatt will have one chance in 10, 30 and 10 100 0 of being exceeded in any on e year.

Copyright NRC-CNRC  

2.

averag rage e air density was assumed in order t o compute comp ute the n ave

velocity pressures pressures for th e

hourly wind speeds. 3.

value of two was was assumed for the th e pressures pressur es for th e gust speeds.

gu gust st effect factor

t o com put e the

velocity velocity

The actual wind pressure pressure on a structure increas increases es with height and varies varies with the shape of the structure. T he factors needed t o allow for these effects effects are tabulated in Section 4.1 4.1 of the National Building Building Code of Canada 1 970 and in Supplement No. No. 4. The Th e other thr ee steps are discussed discuss ed in more detail de tail in t he following following paragraphs. paragraphs. Until recently the only wind speed record kept at a large number of wind-measuring stations in Canada was the number of miles of wind that pass an anemometer head in each hour, or the th e hourly average wind speed. Many Many stations stati ons are now recording only spot s pot readings of the t he wind speed each hour and these may have to be used for design at some future time. For the present, however, the older hourly mileages are the best data on which to base a statistical analysi anal ysis. s. The annual maximum hourly mile mileage agess for over over 10 100 0 stations for periods from 1 0 t o 22 years were analysed usin using g Gumbel's Gumbel's extreme extre me value value method meth od t o estimat es timate e th e hourly mileage mileagess that would have have one chance chance.. in 10 ,3 0 and 100 of being being exceeded in any one year. The 1 in 30 hourly mileages were used to prepare Chart 10. Values of the 1 in 30 hourly mileages mileages for the additiona addit ionall 50 500 0 locations in the Table could have been b een estimated estimate d from the th e large-s large-scal cale e original of Chart 10 10.. However, However, to ensure ensu re consist-

ency with w ith the wind gust g ust speeds published in earlier earlier editions of this supplem sup plement ent it was was necessary necessary to compute hourly mile mileage agess from the th e published gusts gusts usi using ng the equation:

where V is the hourly mileage and G is th e gust speed speed in miles per hour. hour . This equation equa tion was based on a comparison of over 15 00 hourly mileag mileages es of 30 miles miles or over (as recorded by cup cu p anemometers) with the t he corresponding maximum gust speeds speeds (as recorded by Dines pressure tube tub e anemometers). Values of the hourly mile Values mileag ages es for annual annual probabilities probabilities of 1 /1 0 and 1/1 00 were were readily readily computed for the 100 stations in the original Gumbel analysis. For the other 500 locations it was necessar necessary y to t o estimate th e value of the th e parameter l a which is a measure of the dispersion of the individual individual annual maximum hourly mile mileage ages. s. To do th is t he 10 100 0 known values values were plotted on a large large scale scale map from which estimates estimates were made made for the th e other locations. locations. Knowing the 1 in 30 hourly mile mileag ages es and the values values of l / a , th e 1 in i O and 1 and 100 values could be computed. Pressures, suctions and vibrations caused by the wind depend not only on the speed of the wind but also also on th e air density density and hence on t he air temperature and an d atmospheric atmospheric pressure. The pressure, in turn tu rn , depends on o n elevation above sea sea level level and varies with changes in the weather systems. systems. If V is the th e design wind speed in miles per hour, hou r, then the t he velocity pressure, P, in pounds per square foo t is given given by the equation:

where C depends on air temperature and atmospheric pressure as explained in detail by Boyd (14). The Th e value 0.0027 is within 1 0 per per cent of the monthly mont hly average value of C for most of Canada Cana da in the windy part of the year. This value value (0.0027) has has been been used to compute co mpute all th e velocity pressures corresponding t o tthe he hourly ho urly mileag mileages es with annual an nual probabilities of of being exceeded of 1 lo , 1/30 and and 11100. The pressures are shown in the Table in columns headed only by the numerical values of these probabilities. In t he 19 1970 70 edition of t he National N ational Build Building ing Code the desig design n gust pressures for structural struc tural elements are twice the th e hourly mileage pressures. pressures. Becau Because se wind speeds are ar e squared t o get pressures, pressures, the above statement is equivalent to saying that the gust factor is the square root of two. he table below shows that the 1970 requirements increase the wind loads by less than 8 per cent ov over er those of 1965 computed from gusts give given n by the eq uati on:

Copyright NRC-CNRC  

For buildings over 4 0 feet high the gust velocity pressur pressures es and suctions must be increased according to a table in Section 4 1 of the National Building Code 1 9 7 0 which is based on the assumption that the gust speed increases in proportion to the one-tenth power of the height. The average average wind wind speeds used in com puting the vibrations vibrations of a building building are more depe nde nt on the roughness roughness of the underlying underlying surfac surface. e. me thod of estimati estimating ng their their dependen ce on roughness and height is given given in Sup plem ent No. 4 The calculations for building vibrations in Supplement No. 4 hav havee been drawn up for wind speeds measured in feet per second. The table below may be used for converting the wind pressure pressuress in the m ain Table to wind speeds in in feet per second. It is based o n the equation:

P vsf

Copyright NRC-CNRC  

PERMAFROST (CHART 11 The li lines nes on C hart 11 ind indicate icate th e approximate southern limit of permafrost and the boundary between th e discont discontinuous inuous and continuous permafros permafrostt zone zoness in Cana Canada. da. The distri distribubution of permafrost varies from continuous in the north to discontinuous in the south. In the continuous zone permafrost occurs everywhere under the ground surface and is generally hundreds of feet thick. Southw ard, the c ~ n ti n u u u s one gi give vess way gradua graduall llyy to the discont discontinuous inuous zone where permafrost exist existss in combination with some areas of unfrozen material material.. Th e dis dis-continuous zone is zne of broad transition between continuous permafrost and ground having no permafrost permafrost.. In th is zo ne, permafrost may vary fr om a widespread distribution with isolated isolated patches of unfrozen ground to predominantly thawed material containing islands of ground that remain frozen. In the southern area of this discontinuous zone, permafrost occurs as scattered patches and is only a few fee t thick. It is emphas emphasize izedd that th e lin lines es on this m ap must be cons conside idered red as th e approximate localocation of broa d transition b ands m any m il iles es wid wide. e. Permafrost also exis exists ts a t high high altitudes in th e

mountains of West Western ern Canada a gr great eat dist distance ance sou th of th e sou thern limit shown on th e map. Information Informati on on the occurrence and dist distribut ribution ion of permafrost in Cana Canada da has been compiled by th e Divi Divisio sionn of Build Building ing R esearch, National Research C ouncil (1 5 , 16). SEISMIC ZONES (CHART 12 Th e param eter used as th e basi basiss for establ establishi ishing ng th e seismic seismic zones iiss A100 defined as the ground acceleration acceleration with an annual probabilit probabilityy of bei being ng equalle equalledd or exce exceeded eded of 1 in 10 0 (17). This map is based on the statistical computer analysis of past earthquakes throughout the coun try for this ce ntury . I t is cor robo rated by t he results fro m a lar larger ger b u t lles esss reliable reliable seis seismic mic sample dating back to 1638 (18). The map reflects the opinion of experts in the fields of seismology, geology, and engineering from industry, government and universities comprising mem the Canadian Na tional Com mitte eittee on Earth quak e Engineer Engineering ingngand va various rious relevant relevant commbers'of ittees rresponsi esponsible ble t o the A ssoci ssociate ate Comm on the N ational Bui Buildi lding Code. The zones and their res respect pective ive R-fa R-facto ctors rs are shown in th e table o n Chart 12. The zone boundaries boundari es iinn terms of A100 are sshown hown in Tabl Tablee 2 of t he Com mentary on Loads Due t o Earthquakes (17). In the Arctic Region and other parts of the Northwest Territories, there are insufficient data for a statistical study. The zone boundaries have been drawn by the Seismologists of the Department of Energy, Mines and Resources from their knowledge of earthquake activity in these areas. REFERENCES (1)

Miness an d Technical Surveys, Geographical Branch, ATLAS O F CAN AD A-, Dept. of Mine

Ottawa 1957. (2) THO MA S, Climatol Climatological ogical Atlas of Canada . Na tional Research Cou ncil, Division Division of Building Research, and Dept. of Transport, Meteorological Branch. Ottawa $953. NRC No. 3151. Also in in National Building Cod e of Canada 1 95 3, Part 2: Cl Climate imate , National Research Council, Asso Assoc. c. Com mitte e on th e National Building Code. Ottawa 1953 . NRC No. 3 188. HOURLY DATA SUMMARIES-. ept. of Transport, Meteorological Branch, various (3) dates from from Ma Mayy 1967 to December 1 968. C.C. Perc Percent entage age Frequency of Dry- an d Wet-b Wet-bulb ulb Tem peratures from Ju ne (4) B OUGHNE R , C.C. to Septem ber at Selecte Selectedd Canadian Cit Citie ies. s. Dept. of Transpo rt, Canadian Meteorologi Meteorological cal Memoirs, Memoir s, No. 5 , To ron to, 1960. Design ign Co ndition s for American Society of Heati Heating, ng, (5) CRO W, L.W. Study of Weather Des Refrigeratin Refriger atingg and Ai rco ndi tion ing Engineers, Inc. Inc. Research Proj Project ect N No. o. 23, Janu ary 31 1963. (6) %OM, H C S The Ratio nal R elationship betw een Heating Degre Degree-Day e-Dayss and Temp era7)

tu ture re . Monthly Weather Revi Review. ew. Vol. 82, No. 1 , p. 1 4 , Jan. 1954. THOMAS, M nd D.W. B O Y D Standard Period Heating Degree-Day Normal Normalss . Dept. of Tran sport , Meteor Meteorologi ological cal Branch, CIR-2849, CLI-16, Dec. 19 56.

Copyright NRC-CNRC  

  Heating Heating Deg Degre reee-Da Day y Norm Normals als below 6 5 ' ~ Base Based d on the Period 1931-1960. 1931-1960. Dept. Dept. of Transport, Transp ort, Meteorolog Meteorological ical Branch, Climatic Data Sheets S heets No. 5-64, October 30, 1964. BRUCE, JQ Rainfall Intensity Inten sity Duration Frequency Maps Maps for Canada Canada . Dept. of Transport, Transp ort, Meteorological Meteorological Branch, CIR-3243, TEC-308, Aug. Aug. 1959. Maximum Precipitation Reported on any One Observation Day 1931 1958 . Dept. of Transport, Transp ort, Meteorologi Meteorological cal Branch, Climatic Climatic Data Sheets No. No. 9-59, Oct. 1959. ~ R S H F I E L D , M. M . nd W T WILSON. Generalizin Generalizing g of Rainfall Intensity Frequency Data Data . Internationa Intern ationall Association Association of Scientific Hydrology, General Assembly, Assembly, Toronto, Toron to, 1957, Vol. Vol. 1, p. p. 499-506. 499-506.

Temperature and Precipitation Normals for Canadian Weather Stations Based on the Period 1921-1950 . Dept. of Transport, Trans port, Meteorological Branch, CIR-3208, CLI-19, CLI-19, June 1959. BOYD, BOY D, D.W. Maximu Maximum m Snow Depths and Snow Loads on Roof Roofss in Canada Canada . ProceedProcee dings, 29th Annual Meeting, Western Snow Conference, Spokane, Wash., April 1961. BOYD, BOY D, D.W. Variations Variat ions in Air Air Density over Canada Canada . National Natio nal Research Rese arch Council, Counci l, Divisi Div ision on of of Building Research, Technical Note No. 486, 486 , June 1967. Permafrost Perma frost Map of Canada Canada (a joint production produ ction of the Geological Survey of Canada Canada and DBRINRC). DBRIN RC). August 1967 196 7 NRC NRC 9769. BROWN, R J E Permaf Permafrost rost Map Map of of Canada Canada . Reprin Rep rintt from fro m Canadian Canadi an Geographical Geogra phical Journal, February 1968, pp. 56-63 NRC NRC 10326. R H FER FERAHI AHIAN, AN, Commentary Commentary on Loads Loads due t o Earthquakes Earthquakes , Supplement No. No. 4 t o the National Building Building Code 1970. W.G. MILNE nd k G . DAVENP D AVENP ORT, Distri Distribut bution ion of Earthquake Risk in Can Canad adaa , Bulletin of Seismologica Seismologicall Society of America, Vol. Vol. 59, No. No. 2, pp. 729-754, April 1969, also Fourth Four th Worl World d Conference on o n Earthqua Ea rthquake ke Engineering, Santiago, Chile, January, 1969.

Copyright NRC-CNRC  

CHARTS

to 2

Copyright NRC-CNRC  

 

Copyright NRC-CNRC

Copyright NRC-CNRC  

Copyright NRC-CNRC  

Copyright NRC-CNRC  

Copyright NRC-CNRC  

Copyright NRC-CNRC  

Copyright NRC-CNRC  

Copyright NRC-CNRC