Some Aspects of the Thermal Energy Exchange on the South Polar Snow Field and Arctic Ice Pack'

MAY 1961 MONTHLY WEATHER REVIEW 173

SOME ASPECTS OF THE THERMAL ENERGY EXCHANGE ON THE SOUTH POLAR SNOW FIELD AND ARCTIC ICE PACK'

KIRBY J. HANSON Po!ar Meteorology Research Project, US. Weather Bureau, Washington, D.C. [Manuscript received September 13, 1960; revised February 21, 1961 ]

ABSTRACT Solar and terrestrial radiation measurements that were obtained at Amundsen-Scot>t (South Pole) Station and on Ice lsland (Bravo) T-3 are presented for representative summer and winter months. Of the South Polar net radiation loss during April 1958, approximately20 percent of the energy came from the snow and80 percent from the air. The actual atmospheric cooling rate during that period was only about lj6 of the suggested radiative cooling rate. The annual net radiation at various places in Antarctica is presented. During 1058, the South Polar atmosphere transmitted about 73 percent of the annual extraterrestrial radiation, while at T-3 the Arctic atmosphere transmitted about 56 percent. The albedo of melting sea ice is discussed. Measurements on T-3 during July 1958 indicate that the net radiation is positive on both clear and overcast days but greatest on overcast days. Refreezing of the surface with clear skies, as observed by Untersteiner and Badglep, is discussed.

1. INTRODUCTION Ocean.ThisIsland wasabout by 5 11 miles in size and The elliptical orbit of the earthbrings it about 3 million about 52 meters thick (Crary et al. [2]) in 1953 when it miles farther from the sun at aphelion than at perihelion; drifted near 88' N., looo W. In the years that followed, consequently,during midsummer, about 7 percent less this it drifted southward and in July 1958 was located solar radiation impinges on the top of the Arctic atrnos- 79.5' N., 118' W. phere than on the Antarct'ic atmosphere during a corn- Solar radiationmeasurements at bot,h stations were parableperiod. This difference is enhancedas solar obtainedwith Eppley pyranometers. The data are energy penetrates into both polar atmospheres. Absorp- corrected for the temperature response of the instrument tion, scattering, nnd reflection of the solar rays gives each (MacDonald [Ill) and are presented in the International polarregion its own particularradiation environment. Pyrheliometric Scale of 1956. At both stations, Beckman There are also other notable digerences between the heat andWhitley (Gier andDunkle type) radiometers were budgets of these two areas; forexample, the conduction of used tomeasure the combinedsolar andterrestrial heatthrough the ice is distinctlydifferent. Annually, radiation streams. heat from the Arctic Ocean is conducted upward through 2. WINTER MONTH AT THE SOUTH POLE the thin ice pack to the relatively cold surface where the temperatureaverages about "20' C. In contrast,the With the exception of a few weeks of twilight, sunset flux of heat through the ice layers of central Antarctica at the time of the March equinox marks the beginning of is quite small. Because of the heat budget differences, the 6 months of continuousdarkness atthe South Pole. annual temperature near the North Pole is about 30' C. During the first of the dark months, April, the tempera- warmer than that at the South Pole. It is the purpose of ture a few metersabove the snowaveraged "58' C. this paper to discuss some aspects of the thermal energy (1957-59)"the same as the average temperature during budgets of these regions. the entire dark period. The datawhich are presented were obtained during the In April 1958 the long-wave radiation from the snow International Geophysical Year and later yearsat Amund- surface averaged 229 ly. day" (table l),while the atmos- sen-Scott Station, located within a mile of the geographic pheric (back) radiation returned 76 percent of this energy South Pole, and at Ice Island T-3, drifting in the Arctic (175 ly. day") tothe surface. The net radiation averaged "54 ly. day". 1 Paper presented at the International Antarctic Symposium at Buenos Aires, No- vember 1959. 2 Terrestrial radiation, with mostof the energy between wavelengthsof 3 to 30 microns.

Unauthenticated | Downloaded 09/26/21 02:19 AM UTC 174 REVIEWWEATHERMONTHLY MAY 1961

TABLE1.-Thermal energy exchange at the snow surface, Amundsen- Scott (South Pole) Station, Antarctica AT (DEG c) TEMPERATURE (DEG C) -6 -2 +2 +6 I I -65 -60 -55 -50 -45 -40 January 1 1958 0

Incident solar radiation (ly. day-I)* ...... 770 0 266 Reflected solar radiation (ly. day-1) ...... 674 0 218 2 Albedo (percent) ...... nn ...... 82 Snow surface radiation (ly. day")...... 440 229 301 Atmospheric (back) radiation (ly. day-1)"...... 309 175 218 Return (percent) ...... 70 76 72 ...... -4 Net radiation (ly. day-1) -35 "54 -35 07 Thermal energy fromsnow (ly. day-1) ...... "". 11 ...... Thermal energy frommr (ly. day-])...... "" 43 ...... -65 *One Langley (ly.) equals one cal. cm.3 E 38 The total thermal energy in the top 12 meters of snow at the beginning and end of April 1958 wascalculated, 10 using the equation 12 Q=S" cpTdz FIGURE1.-Snow temperature profiles of April 1 and 30, 1958, and 0 monthlytemperature-change profile, Amundsen-Scott(South Pole) Station, Antarctica. where c is the specific heat in cal. gm" deg." (List [9]), p is thedensity in gm. ~m.-~(Giovinetto [4]), and T is thetemperature (fig. 1) in OK. The calculations indicate thatduring the month heat wasconducted to thesurface atthe rate of 11 ly. day". This suggest,s averaged 770 ly. day". Measurementsindicate that, of that, of the net radiation loss during April, about 20 per- this amount', 88 percent (674 ly. day") was reflected from cent of the energycame from the snow and 80 percent the snow, while the remainder (96 ly. day") was absorbed. fromthe air. During clear, coldperiods atthe South, During the same period, the snow surface radiated440 ly. Pole the snow and air supply about equal amount's of en- day", while the atmospheric (back) radiation returned 70 ergy to make up the surface radiation loss (Hanson [SI). percent (809 ly. day-l) of that energy. The net radiation Liljequist [8] found that with clearskies at hlaudheirn, averaged "35 ly. day". along the coast of Antarct'ica, roughly 40 percent of the required energy comes from the snow and 60 percent from 4. ANNUALENERGY EXCHANGE, ANTARCTICA the air. Some idea of the heat budget of the atmosphere during During the 6 months of sunlight at the South Pole, the this period can be obtained from the airborne radiation solar radiation which wasincident on the snowtotaled measurementswhich were takenwith Suorniairborne 9.71 x lo* ly. The snow reflected 7.96X lo4 ly., indicat- radiometers (Suomi et al. [14]). Data from the clear-sky ing an average albedo of 82 percent for the sunlit period. flight on April 27, 1959 (fig. 2) indicat'e a radiative loss of Unlike solar radiation, the emission of long-wave radia- 240 ly. day-l at 50 mb. Assuming this loss is representa- t,ion by the snow is continuous throughout the year. Dur- tive of April 1958, and adding the small amount of heat ing 1958, the snow surfaceradiation averaged301 ly. day", which was conducted from the snow (11 ly. day"), the of which about 72 percent (218 ly. day") was returned by net cooling rate from the surfaceto 50 mb. becomes atmospheric(back) radiation. This percentage is rela- 1.49' C. day". This is about 6 times greater than t'heob- tively unchanged from summer to winter even though t,he served cooling rate (0.26'C. day"). Presumably, subsi- sky is much clearer duringwinter (fig. 3). Withother dence and advection provide the necessary energy t80ac- things being equal,clear skies would certainly tend tolower count for the discrepancy. thispercentage. Apparently, a compensatingfactor is that t,hesurface temperature inversion is moreintense 3. SUMMER MONTHAT THE SOUTH POLE during winter; this would allowa greater return of bhe Eventhough the South Polar plateau receives more surface radiation. solar radiation at midsummerthan any other area on The net mdiation at theSouth Pole (2800 m.) averaged earth, the temperature of the snow surface remains well about "35 ly. day" during 1958. Liljequist [7] found an below freezing. InJanuary, the warmest month of the annual loss of 25 ly. day" at Maudheim, and Loewe [9] summer, thetemperature averages near -27O C., and found a loss of 20 ly. day-1 at Port Martin. Both stations rarely exceeds about -17' C. are located along the coast of Ant,arctica. Rusin [13] has During January 1958, with continuous sunlight at the reported an annual net radiation of -6 to "8 ly. day" at South Pole, the incomingsolar (sun andsky) radiation Mirny, anot,her coastal station, and -19 to -22 ly. day"

Unauthenticated | Downloaded 09/26/21 02:19 AM UTC &1AY 1961 MONTHLY WEATHER REVIEW 175

LANGLEY/DAY TEMPERATURE ("C)

1: 1: JAN FE0 MAR APR MAY JUN JUL AUG SEP OCT NOV DEC FIGERE3.-Average cloud coverage during the period January1957 through L)ecemhc:r 1960 at Amundsen-Scott(South Pole) Station, Antarctica.

FIGURE2.--Net radiation profile, temperature profile, and temperature-change profile computedfrom radiometersonde (Suomi et al. [14]) ascent at Amundsen-Scott(South Pole) Station, Antarctica, OB00 GMT April 27, 1059. The €act that a smaller percentage of the extraterrestrial radiation reaches the surface in the Arct'ic may be due to a number of factors: possibly there are thicker cloud systems in the Arct,ic, t'hc surface albedo is less in the Arctic, at Pionerskaya (2700 m.) in the interior of eastern Antr anti theoptical thickness of the Arct,ic atmosphere is arctica. There is 1it't)ledoubt that, as a whole, the surface greater because of the comparatively lower surface eleva- of Antarctica has a negative net radiation, although cer- tion of theArctic. Precisely how thesevariables affect tain snow-free areas on the continent and portions of the the amount of solarenergy, incident on the Arctic and Palmer Peninsula are probably posit'ive. Considering the Antarctic, is difficult to determine withouta better knowl- magnit'ude of the individual incoming and outgoing radia- edge of the surfacealbedo and thickness of the cloud tion streams, the slight variation of net radiation, as ob- systems (Fritz [3]). served in Antarctica, seems rather remarkable.

5. ANNUALENERGY EXCHANGE, ARCTIC OCEAN 6. SUMMER MONTHSIN THE ARCTIC BASIN The snow that covers much of the Arctic sea-ice in Although the e1lipt)ical orbit of the earth causes com- early spring is gradually melted during May and June. parat,ively less midsummerextraterrestrial radiation in As a result, pools of melt water form on the ice floes in the Arctic, it also provides 9 additional days of sunlight late June and remain during July and sometimes August. at the North Pole each year compared to the South Pole. These pools aid in lowering the surface albedo during this The net result is that during thecourse of their respective midsummer ablation period. Sverdrup [15] has indicated sunlit' periods, t.qual amounts of solar energy impinge on that thealbedo of melting Arctic sea-ice is between 60 and the top of the North and South Polar atmospheres. This 65percent. Recent Soviet investigations (Briazgin [l]) was pointed out by Milankovitch[12], who indicat'etl that have shown a similar albedo value, 60 percent. The fact at both geographic poles the annual extraterrestrial radia- that pools of melt water on the floes lower the albedo is tion is of equal intensity, and, assuming a solar constant indicat'ed by the results of an aerial albedo survey over of 2.00 ly. min.-l, totals 13.33X lo4 ly. during the year. the Arct'ic Ocean. This survey indicated that the albedo This equalityis not maintained as the solar rays penetrateof melting ice witha rnaxirnurn amount of puddling is the polaratmospheres, however. The previouslymen- about 46 percent (Hanson [SI). t'ioned measurements atthe South Pole indicatet,hat, Measurements on T-3 indicate that during July 1958 annually, about73 percent of the extraterrestrial radiation theincident solar radiation averaged 524 ly. day". was incident on the snow surface. At Ice Island T-3, on Assuming an albedo of 60 percent,the ice would have the other hand, only about 56 percent of the extraterres- gained 210 ly. day". Measurenlents also indic,ate that, trialradiation wasincident at the surface during 1958. with an average 6.7-tenthsdoud coverage, the atmospheric - long-wave radiationreturned about 88 percent (578 ly. 8 Located at 79.5" N., the extraterrestrial radiation atT-3 totaled 13.81XlO 4 ly. during 1958. day") of the 653 ly. day" which were emitted as long-

Unauthenticated | Downloaded 09/26/21 02:19 AM UTC 176 MONTHLY WEATHER REVIEW MAY 1961 waveradiat,ion from the surface. This gives an average explanation would require some idea of the rates of evap- net radiation of $135 ly. day” during July. orationand turbulent heat exchangein theboundary Currently, one of the most important questions in polar layer. heat-budgetinvestigations is: How is the “surplus” There is also an additional point of interest. In view thermal energy, available from radiative exchange, used of the observed refreezing of the surface with clear skies, in warming or melting the ice, evaporation, or warming it is rather surprising that the net radiation should be the lower atmosphere? During the previously mentioned positive. An examination of thedata revealed thatan period, for example, the net radiation could have melted albedo of 72 percent or higher would be required in order 58 cm. of surface ice, assuming a density of 0.9 gm. crn.r3 tohave a negative net radiation with clear skies. This and a latent heat of fusion of 80 cal.gm.” The actual seems excessive. It seemsreasonable that iffreezing of ablation was probably somewhat less, however, as a small the surface occurs with a positive radiation balance, either amount of energy is lost in evaporation (Untersteiner and oue of twothings is happening.Either the total heat Badgley [16]), and possibly, as Fritz [3] and Yakovlev 1171 budget is negative because of losses by evaporation and have suggested, some energymay be lost to the atmosphercturbulence aud therefore refreezing of the surface occurs; byturbulence. Because ablation, evaporation, and or,the total heat budget is positive andthe surplus” 1L temperature profile measurements are not available, t’he energy is used in mrltingjust below the surface while heatand water budget during this irnport’ant sulnmer freezing cont,irlues onthe surface. The latter may be ablation period cannot be determined precisely. possible as t,he ice is partially transparent t’o solar radia- An interestingobservation regarding the heat budget tion and opaque t’o t’he longer wavelengths of terrestrial was made by Unterst,einer and Badgley [161 on Floating origin. In order t’o present a realistic model of the proc- Ice Station “A” : esses involved, additional field studies are desirable. Duringthe summer, melting occurred mostly when there was overcastand strong atmospheric radiation. Radiosonde observa- ACKNOWLEDGMENTS tionsreveal frequent inversions, with comparatively high cloud temperatures.Temporary breaks in the overcast were frequently Tlle writergratefully acknowledges the many helpful accompaniedby freezing at the surface even thoughdirect solar cornment>s from Drs. H. Wexler and S. Fritz. This work radiation was relatively larger in such periods. was made possible by support from the National Scie,nce In order to investigate, therole of radiat’ive heatin nlelting, Foundation of t’he Kational Academy of Sciences. we have examined t’he radiation data for clear and overcast conditions at T-3 during July 1958. From measure- REFERENCES ments of the incoming solar and atmospheric (back) radi- 1. X. N .Briazgin, “K Voprosu ob Al’bedo Poverkhnosti dreifulll- ation, assuming an albedo of 60 percent which is probably shchikh l’dov,” [On the Questionof the Albedo on the Surface representative of the melting pack ice, it was found (table of Drifting Ice], ProblemyArktiki, i Antarktiki KO.1, 1959, 2) that on 6 clear days the net radiation averaged +7G pp. 33-39, ly. day“. On 9 overcastdays it averaged $147 ly. 2. A. P. Crary, J. L.Kulp, and E. W. Marshall,“Evidences of Climatic Change from Ice Island Studies,” Science, vol. 122, day”--.an increase by a factor of about 2 providing the 1955, pp. 1171-117:3. albedo is unchanged. Theradiation data, as presented 3. S. Fritz, “Solar Radiation Measurements in the Arctic Ocean,” here, lend some support in explaining these observations pp. 159-166 of Proceedings,Polar Atmosphere Symposium, made by Untersteiner and Badgley [iG]. However, a full Oslo, July 1956, Part I, Meteorology Section, Pergamon Press, h’ew York, 1958. 4. M. Giovinetto,Unpublished document, 1958. (Presents measurements of snowdensity from the surface to -12 meters for the South Pole.) TABLE2.-Thermal energy exchange at the surface of Arctic sea-ice 5. K. J. IIanson, “The Albedo of Sea Ice and Ice Islands in the duringclear, overcast, and average sky conditions as determaned Arctic Ocean Basin,” 1960. from measurements of ancident solar and atmospheric (back) radia- tionat T-3, assumingan albedo of 60 percentand ice surface 6. K. J. Ilanson,“Radiation Measurement on the Antarctic temperature of 0’ C. Snowfield, APreliminary Report,” Journal of Geophysical Research, vol. 65, No. 3, Mar. 1960, pp. 935-946. July 1958 7. G. 11. Liljequist, “Energy Exchange of an Antarctic Snowfield, I Long-Wave Radiation and Radiation Balance,” Norwegian- Clear sky Overcast Average British-SwedishAntarctic Expedition, 1949-59, Scientijic condition sky condi- sky condi- <“nth’ tion, 10- tion, 6.7- Itesults, vol. 11, Part IB, Oslo, 1956, pp. 113-183, (p. 174). &y cover tenths tenths (on 6 days) sky cover sky cover 8. G. I€. Liljequist, “Energy Exchange of an Antarctic Snowfield, (on 9 days) (31 days) WindStructure in the Low Layer,” Norwegian-British- SwedishAntarctic Expedition, 1949-52, Scientijic Results,

Incident solar radiation (ly. day-l)..- ~ ...... 643 401 524 Reflected solar radiation (assuming an albedo of vol. 11, Part lC, Oslo, 1957, pp. 187-233, (p. 224). 60 percent) (ly. day-1) - ...... ~..-~ ~ ~-...... 386 241 314 9. R. J. List (ed.), SmithsonianMeteorological Tables, 6th revised Ice surface radiation (ly. day-1) ._...... ~~~ ...... 653 653 653 Atmospheric (back) radiation (ly. day-1) _...... 472 640 578 cdition,Smithsonian Institution, Washington. D.C. 1951, Return (percent) ___...._...... _...~~~~.~...... 72 98 88 (see p. 343). Net radiation(ly. day-1)”...... ~~_._~...... -.. +76 +147 +135 10. F. Loewe, “Etudesde Glaciologie enTerre Adi.lie, 1951-52,”

Unauthenticated | Downloaded 09/26/21 02:19 AM UTC MAY 1961 MONTHLY WEATHER REVIEW 177

Expdditions Polaires Francaises, KO. 9, Actualitds Scientifiques Quarterly Journal of the Royal Meteorological Society, vol. 84’ et Zndustrielles, No. 1247, Hermannet Cie., Paris, 1956, No. 360, Apr. 1958, pp. 134-141. 159 pp. 15. H. U. Sverdrup,“Meteorology, Part I, Discussion,” The 11. T. H. MacDonald,Unpublished document, 1959. (Presents Norwegian North PolarExpedition with the “Maud”, 1918- theresults of temperatureresponse tests on the Eppley 1925,Scientific Results, vol. 11, Geofysisk Institutt, Bergen, pyranometers which were used at United States stations in 1933, pp. 1-328. theArctic and Antarct,ic during the International Geo- 16. X. Untersteinerand F. I. Badgley,“Preliminary Results of physical Year.) Thermal Budget Studies on Arctic Pack Ice During Summer 12. M. Milankovitch, “Mathernatische Klimalehre,” Handbuchder and Autumn,” pp. 85-95 of Proceedings, Arctic Sea Ice Con- Klimatologie, W. Koppenund R. Geiger, Band I, Teil A, ference,Easton, Maryland, Feb. 24-28, 1958, NationalRe- Berlin, 1930, (pp. A30-A31 andtable 3). searchCouncil, Washington, D.C., Publication No. 598, 13. K. P. Rusin, “Radiatsionnyl Balans Snezhnoi Poverkhnosti v Dec. 1958. Antarktide,”[Radiation Balance of the Snow Cover in the 17. G. N. Yakovlev, “Solar Radiation as the Chief Component of Antarctic], Informatsionnyi BzZIleten’ SovetskoiAntarkti- the IIeat Balance of the Arctic Ice,” pp. 181-189 of Proceed- cheskoiExpeditsii, No. 2, Lcningrad, 1958, pp. 25-30. ings, Arctic Sea Ice Conference, Easton, Maryland, Feb. 24-28, 14. V. E. Suorni, D. 0. Staley, and P M. Kuhn, “A Direct Measnre- 1%8, National Research Council, Washington, D.C. Publica- ment of Infra-RedRadiation Uivergerlce to 160 mb.,” tion No. 598, Dec. 1958.

Unauthenticated | Downloaded 09/26/21 02:19 AM UTC