Assessment, correction and impact of the dry bias in radiosonde humidity data during the MAP SOP

2006 
This study reports on the characteristics of humidity measurements in the Mesoscale Alpine Programme (MAP) radiosonde dataset. During the Special Observing Period (SOP) of MAP from 9 September 1999 to 15 November 1999, about 10 100 soundings from 51 sites were collected in the so-called ‘high-resolution data domain’. Of these, 39 used Vaisala RS80A radiosondes, which delivered 81% of all soundings. Additionally, five sites providing about 10% of all soundings used either Meteolabor SRS-400, Graw or VIZ sondes equipped with carbon hygristors for humidity measurements, six sites (8% of all soundings) used Meteolabor temperature sondes without humidity measurement, and one site used Vaisala RS90 sondes. Quality control and careful analysis of the MAP upper-air dataset revealed several characteristics including the dry bias in Vaisala RS80 radiosondes. Since about two thirds of the stations are part of the routine synoptic network and were operated during the MAP SOP without major procedural changes, this evaluation also allows us to characterize the backbone of the operational upper-air observations. The dry bias of the Vaisala RS80 radiosonde was corrected using two different algorithms. The first algorithm raised the values of relative humidity of this sonde's profiles by about 4 measurement units (% relative humidity) in the lower troposphere and by 2 in the upper levels. The results of the second algorithm highly depend on the sonde's age. The relative humidity values are raised by 10% in the lower parts of the troposphere and by about 4% in the upper levels. These results are comparable to previous studies which mainly treated soundings from research stations in tropical regions. As in these studies, the increase of humidity in the soundings results in an increase of convective available potential energy and precipitable water, as well as a decrease of convection inhibition. Copyright © Royal Meteorological Society, 2006.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    32
    References
    14
    Citations
    NaN
    KQI
    []