Detection Of Antarctic Atmospheric Structure And Data Validation

Data used in this study include Radiosonde data from Kunlun station（Dome A），radiationdata from Panda-1station and conventional atmosphere observation data from four automaticweather stations along the traverse route fom zhongshan station to Dome A,East Antarctic.Vertical structure of the troposphere over Kunlun station,radiation balance characteristics ofPanda-1station, applicability of different reanalysis data in east Antarctic are analyzed.Theresults showed that:1. The analyses of vertical structure troposphere on Kunlun station shows that: averagevertical temperature lapse rate of the Antarctic plateau middle troposphere is5.2℃/km,which is significantly lower than the global average middle tropospheric lapse rate (6.5℃/km), The average height and temperatures of Lapse Rate Tropopause are4.6km and-51.3℃,respectively. The vapor exists mainly in the lower troposphere below2km. Multilayerinversion structure exists in the planetary boundary layer, strong inversion layer mainly existsbelow500m. The mean height of planetary boundary is890m, and it shows an obviousdiurnal variation: low in the morning and high at noon. This study can provide ground truthvalue for validation and calibration of satellite data in inland Antarctic, and also lay animportant foundation for the verification of the results get from the atmospheric generalcirculation model.2.The analyses of radiation balance in Panda-1station shows that: The albedo (α) isdepends mainly on solar zenith angle，annual mean α is0.8, the daily mean α get maximumvalue in the late and early summer. The situation of Instantaneous downward shortwaveradiation higher than top of atmosphere radiation which is ture in four seasons，the frequencyis highest in spring and lowest in winter. The mean diurnal variations of the downwardshortwave radiation and the upward shortwave radiation have a single pattern peak curve.Except summer, downward long wave radiation almost has no diurnal variation. Upwardlongwave radiation appears single kurtosis and single value in each season except summer. Forthe whole year, net radiation are basically negative, only in December and January is smallpositive value, Annual mean net radiation is-8.7W.m-2. Land compared to the atmosphere iscold source, in addition to atmospheric radiation and solar radiation. Land also get heat by the way of sensible and latent heat from the atmosphere. Panda-1has similar radiation balancecharacteristics with other Antarctic inland plateau sites.3. The studies of applicability of reanalysis radiation data in panda-1station showed that:ERA interim radiation data in the Panda-1station were significantly better than the other threekinds of reanalysis data, which is mainly attributed to the application of four-dimensionalvariational (4D-VAR) data assimilation system, a new cloud prediction equation, improvedParameterization scheme and assimilated more satellite and radar data. For the downwardshortwave radiation, maximum bias (18.7W.m-2) found between NCEP-1and measured, thepossible cause is the model overestimate atmospheric transparency and underestimate thecloud coverage. The bias of net shortwave radiation is mainly caused by the bias of albedo.NCEP-1and JCDAS underestimated the surface albedo in Panda-1area, which caused surfaceground absorbed more downward shortwave radiation, finally overestimated net shortwaveradiation. Four reanalysis data all overestimated the downward longwave radiation in differentdegree, and the biasin winter is larger than in summer, the largest bias found in NCEP-1andNCEP-2(-62.6W.m-2and-37.3W.m-2respectively). All four kinds of reanalysis data can’twell reflect the annual variation of net radiation, in general, smaller bias in summer, larger biasin winter. Though the reanalysis dataset exist obvious flaws and shortcomings， in vastAntarctica plateau where observation stations are scarce, the measured data can not satisfy theresearch need, reanalysis data still an effective tool in studying Antarctic climate.4. The studies of applicability of reanalysis Data on east Antarctic plateau Shows that:(1)All four reanalysis data can reflect the annual variation of temperature on observation site. Indifferent sites and different seasons, different reanalysis temperature data show differentdeviation. Compared with the observational temperature，usually the deviation between ERAinterim and measured temperature is minimum. From the coastal to inland areas of theAntarctic ice sheet,assimilaltion data reduce gradually, so absolute deviation betweenreanalysis temperature and the measured temperature increased, RMSE increased,thevaricance is reduced,and the applicability is reduced gradually.(2) All four reanalysis data canreflect the annual variation of pressure on observation site. Terrain is the main reason lead topressure difference between different model. In coastal aeras,the topography is complex,biggerdifference between model and actual height lead to bigger pressure deviation, in inlandareas,the terrain is flat,smaller difference between model and actual height lead to smallerpressure deviation. Owe to the application of the four-dimensional variational assimilationscheme and more observational data, the applicability ERA interim in the East Antarcticplateau is significantly better than other kinds of reanalysis data.(3) In inland areas, all fourreanalysis data overestimated local humidity on different degrees.the applicability of ERAinterim is obviously better than other reanalysis data. Obvious seasonal difference existed, the devation is small in winter and big in summer in general.