SCIENT. PAP. NAMIB DESERT RES. STN, NO. 38
THE CLIMATE OF GOBABEB
by B. R. SCHCnZE Weather Bureau, Pretoria
(With 4 figures and 8 tables)
INTRODUCTION AIR TEMPERATURE
The Namib Desert Research Station in South As measured in a Stevenson screen, the air West Africa is situated at Gobabeb on the northern temperature has varied from an absolute maximum bank of the dry Kuiseb River some 60 miles of 42.3"C to an absoIute minimum of 2.I0C, giving southeast of Walvis Bay and about 35 miles from an absoIute range of 40.2"C. Mean daily maxima the Atlantic coast. The geographical coordinates are: are about 32 "C in the summer months December to Latitude 23"34'S; Longitude 15"03'E ; Height above April and about 27°C in mid-winter. Mean daily mean sea level 407 m (1335 ft.). minima vary from about 15°C in summer to 10°C )
1 in winter, whilst the average daily aperiodic range
1 The instrumental equipment at the climatological
0 (mean &. - mean min.) remains fairly constant
2 station makes provision for the measurement of throughout the year and is of the order of 16 to d atmospheric pressure, surface air temperature, soil e 18.5"C. mean and t Details on extreme temperatures a temperature, hurnjdity of the air, wind direction and as well as of other elements are contained in Table 1. d
( force, precipitation, evaporation from a class A
r The frequency of days with temperatures exceed- e evaporation pan, and sunshine duration. Except for h ing or falling under certain threshold values is s the latter WOparameters and for soil temperature, i l climatologically of special interest. Referring again b autographic instruments are also provided. Observa- u tions are made daily at three terminaI hours, viz. to Table 1, we notice that, in March, hot days with a P maximum temperature of at: least 35°C (95°F) oc- e 0800, 1400 and 2000 South African Standard Time h
t cur on the average ten times and that such high
which is 2 hours ahead of Greenwich Mean Time. y temperatures are less frequent in other months and, b At present five years' complete records are available
d namely from October 1962 to September 1967 and as yet, have not been measured in June or July. e t During the five years of records daiIy minima have
n thee form the basis of the tables and discussion a not fallen below freezing point, but minima below r presented in the following pages. g
5°C have been measured mainly during the months e of c The aim this paper is to present in short the June to September; these occur on the average on n main clirnatological features obtaining at the e three days in July and are less frequent in the other c i l Research Station without, however, embarking on an months mentioned (see Table 1)."Tropical nights"
r analysis of all the interesting details that could be e d extracted from the present records. A more detailed n u
analysis is no doubt desirable but had best be post- y (9 W~eIljngton,J. H.: Southem Africa: A Geographfral a poned untiI at least 10 years' continuous records are l; 1955. w available. Study, Cambridge e t I=) Logan, R. F.: The Central Narnib Desert, S.W. Afrira. a Further most interesting reading on the climate, U.S. Not. Acad. 8cf.-Nnt. Res. Coun., Pzlb7. 758; G Washington 1960. t weather and other features of the Namib region e (l), South Africa (Union of) : South African Air Force n may be found in works by WeIIington 'toffan (2) i Met. Service of the Royal Navy. "Weatha on the b and in "Weather on the Coasts of Southern a Coasts of Sozttlrent Africa". 2: 1-16, plate 1: Cape
S (a). Africa" Town 1M4. y b d e c u d o r p e R 6 DR. FITZSPMONS COM3TEMORATIVF: VOLUME
with a minimum temperature exceeding 20°C are to 4.7". Some idea of the daily heating of the soil likely in a11 months except dune, July and August is provided by observing at the three terminal hours and are most frequent during March, April and 8 am.,2 pm. and 8 pm. It will be seen that near May when roughly three to five such nights are the surface the highest temperatures occur at 9 p.m. indicated. Cool days during which the maximum and one would expect the maximum soil temperature temperature does not rise above 17.5"C (63.5"F) to occur normally somewhere between 2 and 8 pm., are rather infrequent but do occasionaIly occur probably between 4 and 5 p.m., and the minimum from June to September; the coldest day yet before 8 a.m. The highest soil temperature at 10 cm recorded (viz. in August) showed a maximum depth, measured on a single day, was 42.Z°C in temperature of 145°C (58°F). In the hot season, January and the lowest 10.4"C in August. November to March, maximum temperatures on the cooIest days are not likely to fall below 22 "C (72°F). HUMIDITY OF THE AIR The mean diurnal variation of temperature for every month and for the year as gained from On the whole the air is very dry. Mean monthly thermograph traces appears in Table 2, the last relative humidity varies from 607; in February to column of which shows the mean monthly tempera- 3GC; in May as shown in the last column of Table 4. tures from which the annual variation becomes It is only in the early morning hours of the summer apparent. This table shawx that March is the warm- months that the mean humidiry rises above 70 or est month with a mean temperature of 24.Z°C and 80:;. During the afternoon in summer it drops to July the coldest with 17.7'C. The annual range is about 30% and in winter to about 20';. Extremely therefore small (6.5"C) as is the case in most IOW values below 5% have been observed on coastal climates. On the other hand the diurnal wave individual days whilst during foggy weather 100:; of temperature shows a large amplitude comparable is reached. with continental conditions in middle latitudes. This The diurnal variation has a large amplitude in summer and a much smaller amplitude in winter. This is mainly due to the fact (as will be shown later) that in summer northwesterly winds transport moisture landwards at night but during the day the intense heating of the atmosphere causes the relative )
1 humidity to drop considerably; in winter on the 1
0 other hand, the wind is mainly from the southeast to 2 northeast, transporting initially dry air, so that, d e t even at night the relative humidity barely exceeds a
d 50r;. These facts are well illustrated in Figure 2, (
r which shows the diurnal variation for the moistest e and h driest months. s i 1 : Mean diurnal variation of Temperature l Figurc b r March and Julyr. u P e h 1 shows t is illustrated in Figure which the diurnal
y variation for the warmest and coldest months. b Especially in winter the temperature rises extremely d e t rapidly from 8.00 a.m. to 11.00 am. and is due to n
a the intense heating of the desert in the absence of r g cloud during the day as well as intense terrestrial
e seen mean c radiation at night. It will he that the n is mere than e diurnal range twice the mean annual c i
l range. r e d SOIL TEMPERATURE n u
y Average soiI temperatures at various depths below Figure 2: Mean diurnal variation of Relative Humidit). a (February and May). w the surface and at three terminal hours are cont.ained e t in TabIe 3. From about 30 m depth all thm a
G terminal hours show (within about 1'C) the same DURATION OF SUNSHINE t
e temperatures manth for month, which indicates n i that the diurnal wave dies out just beyond 30 cm. A summary of the sunshine duration at Gobabeb, b as a The annual variation however, is still apparent at gained from a Campbell-Stokes sunshine recorder, S 120 cm depth, though its amplitude has decreasd appears in Table 5. Firstly it will be seen that in y b d e c u d o r p e R ST-3ERT RES. STY, YO. 'W ..
px+ery month the average duration per day is rht. rnonth ~vithlcasi winrl; Table G also shows that ~~ractically10 or I1 hours. Sinc~days arc long^^. in April has Ihe ~reatrslfrequency of calms. summer the percentage of the a~tronnmicallg T~Pdiurnal crarlaljol1 of wind dirtclion as nbtaincd duration is less in summer than in wintpr: by calculating thr he~~r'lywind vcctor ( for Jannary ~huswe find that, in mid-summer, the averagp and July) shows some inter~stingf~atur~s. T'g1 U~PS tll~l-alionis SO(& of the possible, whilst in urinlcr the 3 and 4 indicate the results graphicaIIy. In January percentage is well ahove 90";. This indicates that thc dircclion shuws little change thsoi~ghout the Furnmer, as will be ~stablishedlater, is the more day: Thp diwctinn (or air displacement is from the cloudy or foggy season and that, in winler, full north up to ahout 1000h, th~nstarts backing io sunshine is ihe usual condition. This is shown quitc west-south-tv~st tiII 1900 (7 p.rn.1 and evcntuiilly clearly in the frequency of days falling within vpcrs hack tn north; th~velocity is failJy low al certain categories of sunshine fluration. It is nnlth- night hr11 shows a vrry ~lrollo~lnccdmaximum nr 1vorih)- thal days with no sunshinc (overcast days) and days with less than 30'; of the llnssibl~sun- shine occur very rarely. By far I hp greater ntlrnb~r nf days enjoy more than 50'; of th~pnssihlc sun- shine and in winter practically all days (more than ?5 per month) receive very nearly I~Pfull quota of the possihl~sunshine durar ion. The diurnal variation shows quite definitely that the mornings are cloudier (havin~less sunshi~~e) than the aft~nloonsas is also evident from the cloud statistics in Tahlc 1. An rxamination of thr hourly values [Table 5) shows that, from about 10 am. more lhan nine tenths of the pnss~blehourb sunshinc occurs in pmcticalIy all months. which indical~sthat most fog and cloutl cIpars at ahnul that time. ) 1 1
0 The main chmacteristics crt Lhe wind r~qim~at 2 Gohabeb appear in Tahlc 6. We nnticc thc irnporkant d 1600 !o 1700h. duc 10 thr syvonji so~ithwes!rrlysra- e t fact that rast~rlyto so~~theastrrlywinds are at l>rc~*zcwhich ~icnctrates itlland at least as far as a maximum frcq~lency in winter anrl have inean d (:nl~abcb a1 an altiturl~of ,100 inelrrs; how far in- ( vc1o:itics up tn 21 km pet hour; on thr? 91 her hand, r lat~rland to what altitude the sea-bree~eppnetrates e winds from the NW srctor show their highrst fre- h is unfortunately not known on account of insuffi- s i quency in sumrncr hut have cnnsirlcrahly Inwrr- l cipnt data. b ~rlocitics. u In July, on thc othrr hand. the direction is south- P nl 100h and e easterly night up till about the11 starts h t backing round I he clock. hecomiltg snuthc~~stcrJ'
y again at rnidniphl. Tn July, hnu-cva., The vrlncily b maximum d rraches a primary at 1100h, rhen drups e t tn almost calm and attains a sccnndary ~naxirnum n
a at 1900h, thr latter bring again rFu~to thc spa-blrlrze r g which, hart-ever. is much ICSF pronounced in winter. e
c Tablr 7, which gives !fir variation of speed only. n
e also qlinws most nf thew charact~ristics. c i l r e I-. I .I, I I I I r l d n Figur~3: JTtaarl hnurl? ind I rcrnr.; - January. u
y Although five ycars is ton short a period on a The diurnal variation of wind speed is niven in which an average prccipitatinn can bc hrlsctl, w e csprcially iotn d~serlsl a1 ion, t hr annuat figurr t TabIc 7. Thc main fcatt~rcsarc thar, in summer, thc a a maximum sped is reached in thp late afternoon at shnwn in Table 1, viz. 24 mm, is in rcasonablc G
t ahout l8OOh and th~mlnimum at about 060011. agrwmcnt with those lor Swak~pmuncland IValvis e
n whilst in the winter months May. Jun~and July. Ray. Precipitation dors occur the ypar rounrl, but i b the maximum is reachpd at nr just h~forenoon srrrns to bc mainly conc~ntratetlaround rhc sum- a
S and the minimum at midni~ht.April seems to be mer months .Tan~iary to March. This is in agrec- y b d e c u d o r p e R 8 DR. FITZSLMONS COMMEMORATIVE VOLUME
ment with the wind regime and the humidity data. A display of lightning is quite often observed in The frequency of days with precipitation (Table 1) the evenings against the great western escarpment also shows that precipitation is most likely in sum- in a northeasterly or easterly direction from Goba- mer, though, taking all the evidence into account, beb. Cumulo-nimbus clouds apparently discharge there seems to be some slight indication of a double their moisture against the escarpment but usually maximum in the rainfall regime, namely in March produce only insignificant precipitation at Gobabeb. and September. Lightning displays have been observed from about October to May; they are mcst common in February and March but rarely seen in ApriI, May or June. EVAPORATION
The avaiIable records indicate that the average HAIL annual evaporation from a free water surface in a class A evaporation pan is about 138 inches (3505 During the five years of observations, hail has mm) per year. This is about on a par with the occurred only once, and surprisingly enough. at 4.00 evaporation at Windhoek. Also, the annual march a.m. in September 1965, when hailstones of :A*'' of average evaporation shown in Table 8 agrees diameter fell and the total precipitation, measured fairly well with conditions at Windhoek. at 8 am., was 11.7 mm. It would appear that the individual monthly amounts can vary quite considerably, due probably to the prevalence or otherwise of easterly berg SNOW winds. Thus, for instance, two different Juiys During the period of observations at Gobabeb showed evaporation totals of 11.1 and 7.1 inches. snow has not been observed, though snow is known to have fallen in that area, albeit rarely. FOG AND OTHER PHENOMENA Fog, due both to advection from the cold ocean DUST AND SANDSTORMS off the coast and to terrestrial radiation, occurs As shown in Table 1, sand or duststonns, probably almost exclusively at night and dears at
) the most unpleasant weather phenomena at Goba-
1 approximately 10 a.m. in the morning. It is mainly
1 beb, occur practically in all months of the year, but
0 associated with light to moderate northerly to ncrth-
2 are most prevaIent in winter from May to about westerly winds. Table 1 shows the frequency of fog d October. They occur on an average of 15 days per e at 8 a.m.; however, it sometimes clears before t year and are associated with easterly berg winds as a 0800h and very exceptionally occurs in the afternoon d well as with the strong southwesterly sea-breeze (
or evening, so that the number of clays with fog r Csee Figues 3 and 4). e will be somewhat in excess of the frequency at. h s i 8 am. It will be seen from Table 1that fog is main- l b ly a summer phenomenon, having the greatest fre- u
P quency in September and October, whilst May Is the e month with least fog. From the average monthly cloud coverage shown h t in Table 1, it is quite evident that there is a marked y b
annual as well as diurnal variation in cloud amount.
d THUNDER
e Summer is the main cloudy season whilst in winter t n find skies are usually clear or only slightly cloudy. a It is most surprising to a relatively high r g frequency of days with thunder so near to the west Summer mornings are very often overcast with a
e shows thunder c coast. Table 1 that is heard on an layer of stratus or stratocumulus, which clears away n average of 23 days per year, occurring mainIy in forenoon, e in the course of the leaving the afternoon c i January to March. Associated precipitation. however, early l and evening only sIightly cloudy, very often
r is very slight and has not, as yet, exceeded 16.5 mm with remnants of altocumulus and cirrus, the latter e d on a single day. very likely the remains of cumulo-nimbus. n u y a w e t a G t e n i b a S y b d e c u d o r p e R SCIENT. PAP. NAMIB DESERT RES. STN, NO. 38
TABLE 1: Mmn monthly ternpemtmw ('C), dond (oktas) and precipitation (mm) data.
C .-E No. of days No. of days with max. temp. with min. temp.
Sc 0 '$E ~1~~0~~~~~1559 12200 2 l&O 050 31.6 14.9 23.2 16.7 22.2 40.4 10.4 21.6 24.1 4.4 21.2 0 0 0 0 2,O F 31.6 4 23.2 16.8 22.1 39.7 9.4 21.5 23.6 4.4 19.4 0 0 0 0 0.6 M 33.5 16.0 24.8 17.5 24.2 42.3 9.9 25.4 25.9 10.2 25.8 0 0 0 0 5.0 A 31.5 14.4 23.0 17.1 22.2 39.4 5.1 19.5 24.3 53 20.8 0 0 0 0 4.4 M 3Q.7 14.5 22.6 16.2 21.8 37.0 5.2 18.6 21.4 7.6 20.4 0 0 0 0 2.4 J 26.4 11.2 19,l 15.7 18.1 33.0 2.2 14.8 20.0 0 9.4 0.6 0 0 1.8 0 J 26.9 10.1 18.5 16.8 17.7 34.8 2.1 17.2 19.1 0 10.0 0.4 0 0 3.0 0 A 27.6 10.4 19.0 17.2 17.9 37.4 3.0 14.5 18.8 1.8 12.4 0.6 0 0 2.2 O S 29.1 10.6 19.9 18.5 18.9 40.3 3.6 16.9 23.0 5.6 12.6 0.2 0 0 1.0 0.4 0 29.5 11.0 20.2 18.5 19.3 40.6 5.4 18.0 12.5 2.8 12.6 0 0 0 0 0.6 N 30.5 1.7 21.1 15.8 20.3 39.6 23.9 22.2 2.2 16.8 0 0 0 0.2 0.4 D 31.2 13.2 22.2 18.0 21.3 41.9 24.2 19.8 4.2 18.6 0 0000
Year130.1 12.7 21.4117.4 20,5142.31 2.1(14.5125.9 49.01200.0 1.8 8.2 15.8 0 I 0 ) 1 1 0
2 0
.Md d $ precipitation:- EdZd e 0 t W.; .E d S 0
a 6- g L 'C 2 " d z 8 ( 2 4 cl :g 2 0.1 12 1.0 10.0 25.0 iUsL$2 oc** Y r 5
P 0.2 0 0 3.5 2.4 M 5.4 16.5 1.8 1.4 4.0 3.6 0.0 2.2 e h A 15 5.0 1.2 0.2 0.0 0 1.4 0 2.4 0.7 2.3 1.3 1.1 t
y M 2.5 5.5 1.0 0.6 0.0 0 0.6 0 0.8 1.5 1.6 1.2 1.0 b
d J 0.7 2.4 0.6 0.2 0.0 0 0.2 0 1.4 2.0 1.5 0.8 0.9 e t 0.0 0 0.2 0 2.0 1.0 1.4 0.7 0.6 n J 0.5 1.9 0.6 0.2 a r A 2.6 12.0 0.6 0.2 0.2 0 0.4 0 3.4 1.5 2.5 0.9 1.0 g
e S 2.7 11.7 1.0 0.4 0.2 0 1.2 0.2 5.8 2.0 3.5 1.2 1.2 c n 0.8 0 0 1.7 1.6 e 0 3.6 0.4 0.2 0.0 2.2 5.4 1.6 3.9 c i l N 1-0 2.0 0.4 0.4 0.0 0 1.0 0 4.4 1.6 4.1 1.7 1.6 r e D 1.2 3.3 0.6 0.2 0.0 0 2.4 0 3.8 0.7 4.6 1.3 1.4 d n u Year 123.8 1 16.5 11.8 5.0 0.6 1 0 1 23.0 1 0.2 40.2 1 14.6 3.3 1.4 1.4 y a
w 'I Fog frequency at 0800 h. only. e t a G t e n i b a S y b d e c u d o r p e R DR. FITZSLWONS COMMEMORATIVE VOLUME
TABLE 2 : Mean monthly dIurnaE variation of temperature ( "C).
I I-Iours, S.A. standard time Month 11 161 B 1101214 20 1 22 1 24 1 Mean
J 17.3 15.9 15.4 16.2 20.6 25.7 29.8 F 17.2 16.0 15.3 15.9 20.3 26.0 30.0 M 19.6 18.0 16.9 17.0 22.9 28.8 32.4 A 17.5 16.3 15.3 15.2 21.2 27.2 30.4 M 16.8 15.9 15.4 15.3 21.3 27.1 29.7 J 13.6 12.8 12.4 12.2 17.0 23.0 25.8 J 12.8 11.8 11.3 11.0 15.9 22.6 25.9 A 13.1 12.0 11.2 11.0 16.6 22.7 26.0 S 13.6 12.3 11.6 11.8 17.5 23.7 27.5 0 13.9 12.6 11.5 12.7 18.3 24-1 21.8 N 14.8 13.4 12.4 13.4 19.2 25.4 28.9 D 15.7 14.3 13.6 14.9 20.4 26.1 29.6 Year 115.5 14.3 13.5 / 13.9 / 19.3 / 25.2 128.3 )
1 TABLE 1 3: Mean monthly soil temperatures ("C)at various depths (cm). 0 2 d e 0800 t I at S.A.S.T. I at 1400 S.A.S,T. 1 at 2000 S.A.S.T. a d ( r e h s i l b u P e h t y b d e t n a r g e c n e c i l r e Year 22.6 ( 26.7 27.0 27.4 d 1 1 1 n u Range 1 9.0 1 10.4 1 9.9 8.0 y 1 a w e t a G t e n i b a S y b d e c u d o r p e R SCIENT. PAP. S,\SITR DESERT RES. STY. NO. 38
TABLE 4 : Mean monthly diurnal variation of relative ht~midity(Yr).
Hour, S.A. staildard time Month Mean
Year
TABLE 5 : A.c.em~emnnthl s~~nshineduration (hours) and frequency of days fafling ~~itlli~~certain mt~gories of dncntion. ) 1
1 I-lour ending t 1oc:ll npllnrent t inw) I KO.of days with:- 0 2 d e
t ?
a -
d 6 7 8 9 lU 11 12 13 14 15 l6 17 1s ( r e h s J ,137 .3& .59 .S5 .92 .94 .97 .99 3'1 .g7 .05 .g5 .86 .33,10.T SO 0.0 0.0 1.0 22.3 7.7 i l b F .06 .a6 .62 .SO .91 .94 .94 .94 .37 .98 .D7 .93 .SLI -20 10.6 S3 0.0 0.4 1.3 14.3 12.0 u
P - ,613 10.5 0.0 0.3 11.4 19.0 M .,58 .83 .91 .93 .g5 .98 .96 .95 .93 .90 .86 - 86 0.4 e h A - -38 -78 .M .90 .D5 -97 .9s1 .W .95 .g6 .95 .54 - 10.2 39 0.3 0.3 0.3 7.0 22.0 t
y M - .24 .g1 .!l4 .g? .+l8 1.00 1.00 .FIS .W .PS .g4 .36 - 10.3 04 0.0 0.0 0.3 4.0 26,7 b
d J - .IT1.% .S& .94 .9G -98 9 .g9 .99 .F19 .97 .23 - 9.9 93 0.3 0.0 0.0 4.0 25.7 e t - .g9 1.00 1.00 - 10.1 93 0.0 0.0 0.0 3.7 27.3 n J .IS .m .93 .95 .g6 1.00 1.00 .nn .zs a r A - .23 .72 -82 .E! .g7 1.00 1.06 1.00 1.N .99 .96 .37 - 9.9 88 0.0 I 0.0 0.0 11.7 19.3 g
e S .33 .66 .76 .E8 .94 '95 .97 .97 .X .g7 .g3 .!is - 9.9 P3 0.3 0.0 2.0 11.0 16.7 c l- n -84 .OX 10.6 84 0.0 0.3 1.0 14.0 15.7 e 0 .Q4 .56 .72 .83 .90 .93 .96 .g7 .g5 .g4 .95 .9.5 c i l .90 .91 .95 ,971 .97 .96 .95 -95 .94 .34 11.3 86 0.0 0.0 1.4 12.3 16.3 N .l5 .S7 .13 .W r e D .15 -49 -67 .X8 .96 .g8 .g? .97 -98 .98 .g7 .g6 .Cl5 .45 11.4 84 0.0 0.0 0.7 18.3 12.0 d n u Ye0 3 7 861 .g2 -86 97 .g8 98 -08 .OS 95 / 62 .28 IIIIIl10.5 87 1.3 l 1.0 5.3 134.0 1 220.4 y a w e t a G t e n i b a S y b d e c u d o r p e R DR. FITZSLMONS COMMEMORATIVE VOLUME
TAB= 6: MonthIg direction frequency (%) and mean velocity (V, in km/hr) for each of eight directions.
N NE E SE S SW W NW Calm %v(%V C Mth. l~lYIFY%V %V %V %V l- J 31 14 2 10 18 1 7 2 9 15 17 14 14 24 13 10 F 24 12 2 8 18 1 9 4 12 16 16 16 13 22 12 14 M 18 11 3 10 2 17 5 10 8 13 14 17 13 13 13 12 24 S 11 3 7 14 10 9 20 12 15 15 10 12 8 11 29 A 13 M 7 13 T 12 12 17 21 10 10 12 10 15 6 11 5 11 16 J 6 14: S L9 14 21 25 12 11 , 13 10 13 5 10 6 12 15 J 7 14 6 21 13 20 29 If 9 11 10 13 4 10 5 11 17 A 11 12 4 14 7 17 19 11. 10 13 14 14 8 10 7 11. 20 S 14 11 3 9 4 16 9 13 10 15 18 16 10 11 11 11 21 0 13 11 2 10 3 17 3 9 13 14 22 17 12 12 12 11 20 N 19 12 2 5' 1 14 2 12 7 13 20 18 14 13 16 12 19 D 26 13 2 7 1 16 1 7 3 12 15 17 16 13 22 13 14 Year I 15 12 1 4 12 1 5 15 1 11 1 10 1 8 / 12 1 15 16 11 l2 1 13 12 18
TABLE 7:Mean monthly diurnal variation of wind speed in km/hr.
) Hours, S.A. standard time 1
1 Month 0 2 4 6 8 10 12 14 16 18 20 22 24 1Mean 2 1 1 1 1 1 1 1 1 1 1 1 d e 11.9 10.4 12.3 t J 10.0 9.1 8.6 8.6 9.5 12.1 15.2 19.3 18.6 14.6 a 7.1 5.9 6.4 8.2 11.3 14.5 18.0 18.1 14.6 10.5 9.3 11.0 d F 8.5 ( M 6-0 5.7 4.7 6.9 6.2 11.2 12.9 17.1 18.8 14.7 4.4 6.7 10.0 r e A 4.3 4.4 5.0 5.1 7.2 9.0 11.7 14.4 11.4 12.7 1.5 5.1 8.6 h s i M 6.9 9.5 10.1 11.0 13.1 13.8 12.4 12.5 13.6 10.2 7.2 6.1 10.5 l b J 10.0 10.5 11.6 11.9 16.6 17.0 14.2 13.3 12.7 10.5 9.4 9.0 12.2 u
P 8.2 9.4 11.4 12.2 13.4 16.0 14.7 12.2 13.1 10.1 8.4 6.9 11.3
J e 6.7 7.5 8.1 8.1 9.9 12.6 12.2 13.5 15.1 11.7 8.5 6.0 10.0
h A t S 6.3 6.0 6.7 7.2 8.2 10.7 13.3 17.1 17.1 12.4 9.3 7.3 10.1 y b 8.8 12.2 15.0 19.3 14.0 10.5 7.9 10.6 0 6.7 4-7 4.4 5.3 18.8 d 9.2 12.7 16.2 19.8 15.1 10.3 7.6 11.1 e N 6.2 5.3 4 6.3 20.4 t 18.5 15.1 10.9 11.6 n D 6.3 6.9 5.9 7.5 9.213.5 26.1 19.0 8.8 a r g
e Year 7.3 7.2 7.3 8.0 10.0 12.7 14-0 16.3 16.9 13.0 9.5 7.6 50.8 c n e c i l TABLE & : Avemge Evapomtion in mm. r e d n u AI M J JI AI S[ 01 N D Year y m a Windhoek w e t Gobabeb a G t e n i b a S y b d e c u d o r p e R