o1,O•bjpa; .—--

OU(;..•:..- -.

•--L

., -.v--- - -• *AMERY ICE 5HE1F

ot. t '.iJ'd* .-

b .

• - - GILLOCKMAWSON ISLAND

-:ESCARPMENT

ot

PRYDZ BAYW

\ j Figure 5. N0AA-6 AVHRR near-00 IR visible image (channel

4P 1 2) of the Amery Ice Shelf.

Pa kIi ce I

MAE KE.N7E BAY

The authors wish to acknowledge Charles Liddick of theNational Earth Satellite Service Office of Operations Sup-port Branch for calculating the satellite reception circlesfrom McMurdo; William Kosco of the U.S. GeologicalSurvey for his support of the project; and Olivia Smith forher preparation of the manuscript.

This research was supported by National Science Foun-dation grants, DPP 79-06412 and DPP 77-27010.

References

Hussey, W. J . 1979. The TIROS-N/NOAA operational satellite system.Washington, D.C.: U.S. Department of Commerce/Noi,National Environmental Satellite Service.

Schwalb, A. 1979. The TIROS-N/NOAA A-G satellite series (NoAATechnical Memorandum, NESS 95). Washington, D.C.: U.S. De-partment of Commerce/No.

Measurements of the katabaticwind in Antarctica

Giw WENDLER

Geophysical InstituteUniversity of Alaska

Fairbanks, Alaska 99702

A. Pocci

Laboratoire de GlaciologieUniversite de Grenoble

Grenoble, France

During the austral summer 1979-80, as a part of theU.S.-French joint experiment, three automatic weather sta-tions (Aws) built by Stanford University were installed in

Antarctica to investigate the katabatic wind in the Frenchsector. One was placed on Dome Charlie (3,215 meters), theother two at D17 (438 meters) and D10 (267 meters) nearDumont d'Urville Station (66°4('S 140°01'E). Measure-ments from these stations, together with those obtained bythe French scientists, are increasing our data base rapidly,giving hope for a better understanding of this windphenomenon. This joint French-U.S. experiment will estab-lish additional stations between Dome Charlie and D17, aswell as some stations around Dumont d'Urville, during thecoming season.

Stations at D10 and D17 were installed in January 1980and at Dome Charlie in February; March 1980 was the firstmonth during which we obtained a full set of data from allthree stations. The wind velocities for the three stations aregiven in figure 1. D10 and D17 have very similar windconditions, a result to be expected. The winds are steadilydownslope, changing their SSE direction very little. Windsare strong, up to 128 kilometers per hour, and are expected

1980 REVIEW 193

AWS8901March 1980AWS8900March 1980AWS8904March 1980

80 mph

80 mph

40 mph

40 mph7 40 mph \I ( \

)I,1

7

.:.

\\

Figure 1. Wind velocities for the three automatic weather stations (Aws) during March 1980. A dot indicates a wind vectorcoming from the center of the graph. Note the steady, strong downslope winds for D10 (Aws 8901) and D17 (Aws 8900); at DomeCharlie (Aws 8904) the winds are lighter (scale Is different) and more variable.

E

to increase during winter. In April 1980 winds up to 140kilometers per hour were measured at 017, and an absolutemaximum in wind speed at Dumont d'Urville was recordedat 314 kilometers per hour in midwinter (June)—a speedthat will challenge the design of any station. Dome Charlieexperiences much lighter winds and the direction is morevariable. This is understandable as there is no gravity windand the wind regime is more like that of the free atmos-phere. Temperatures drop to very low values at this high-altitude station. March had a minimum of — 66.1°C, with amean of — 51.8°C. In April the mean temperature droppedto — 60.3°C, with a minimum of — 73.4°C. An absoluteminimum for Antarctica might be measured in the nextfew months, assuming, of course, that our station staysoperational at these low temperatures.

The diurnal variation in the wind speed at D10 duringMarch is given in figure 2. 017 showed a very similarcourse. The maximum is observed during the night, and theminimum at noon. Note that local noon is around 3 a.m.Greenwich mean time. Furthermore, data were used onlywhen Dome Charlie had wind speeds below 8 kilometersper hour, that is, in the absence of strong cyclonic activities.These results are expected for a gravity wind and werepreviously reported from French observations. In figure 3the wind speed at D10 is plotted against the temperaturedifference between D10 and Dome C, the assumption beingthat the greater the temperature difference, the stronger thegravity wind. Some indication of this effect can be seen, atleast for stronger wind speeds. it is also interesting to note

the very large temperature differences which, most of thetime, are far above the adiabatic rate. Since the surfacewinds are always offshore, this can be explained by rela-tively warm maritime air, which is moved at a somewhatelevated level inland and is mixed into the surface layer,hence warming the coastal stations. This large temperaturedifference is, of course, the driving force of a gravity wind.

This work is being supported by National Science Foun-dation grant DPP 77-26379. Allen Peterson's group of Stan-ford University designed the stations. They were installedby Robert Flint (010 and 017) and Glenn Rosenberger andCal Teague (Dome Charlie). Programming was carried outby K. Chen (Stanford University) and Coert Olmsted (Uni-versity of Alaska). Our thanks to all these people.

40

Eli

20

E

30

In

GMT

Figure 2. Diurnal variation of wind speed for D10 duringMarch 1980. Note that the wind speed has Its minimum atlocal noon, around 3 a.m. Greenwich mean time, at a timewhen the temperature Is at the maximum.

N

194 ANTARcTIC JOURNAL

ILM

LLJ

(0

DELTA TEMP VS. WIND SPEEDD 10 MARCH

ui..woi,j.ujvjz W.jjj4j. wia4W. Lo MW.(jBbEJ. 0 070. 00 QI.QU290.0WINO SPEED C1'1P1)

Figure 3. Wind speed at 010 plotted against the temperature difference of D10 and Dome Charlie for 1980. Note that there issome indication that stronger winds are connected with large temperature difference..

1980 REVIEW 195