Meteorologische Zeitschrift, Vol. XX, No. X, 1-3 (April 20XX) c  by Gebr ¨ uder Borntraeger 20XX Article Intercomparison of integrated water vapour measurements LORENZ MARTIN 1,∗ ,CHRISTIAN M ¨ ATZLER 1 ,TIM J. HEWISON 2 ,DOMINIQUE RUFFIEUX 3 1 Institute of Applied Physics, University of Bern, Switzerland 2 Met Office, University of Reading, UK 3 MeteoSwiss, Payerne, Switzerland (Manuscript received June XX, XXXX; in revised form October XX, XXXX; accepted November XX, XXXX) Abstract Measurements of tropospheric integrated water vapour (IWV) made with two microwave radiometers (AS- MUWARA, TP/WVP-3000), GPS, and radiosondes (SRS 400) during the Temperature, hUmidity, and Cloud (TUC) profiling campaign under mid-latitude conditions in Payerne, Switzerland, in winter 2003/2004 are compared. All methods provide robust IWV retrievals in clear sky and cloudy situations. The mean differ- ence between radiometric and radiosonde IWV is less than 0.15 kgm −2 being not significant with respect to the standard deviation and to the theoretical accuracy. The GPS IWV measurements have a persistent sig- nificant dry bias of approx. 0.5 kgm −2 with respect to radiometers and radiosondes. The different temporal and spatial resolutions of the instruments were found to have a strong influence on the standard deviation. A characteristic diurnal cycle of the GPS and radiometric IWV was observed. Zusammenfassung Der troposph¨ arische integrierte Wasserdampf (IWV) wurde mit zwei Mikrowellenradiometern (ASMU- WARA, TP/WVP-3000), GPS und Radiosonden (SRS 400) w¨ ahrend der Temperature, hUmidity, and Cloud (TUC) profiling Kampagne in mittlerer geographischer Breite in Payerne, Schweiz, im Winter 2003/2004 gemessen und verglichen. Alle Methoden haben sich sowohl bei klarem Himmel wie auch bei Bew¨ olkung bew¨ ahrt. Die mittlere Differenz zwischen radiometrischem und Radiosonden-IWV ist weniger als 0.15 kgm −2 . Dies ist nicht signifikant mit Bezug auf die Standardabweichung und auf die theoretische Genauigkeit. Der mit GPS bestimmte IWV hat einen best¨ andigen systematischen Fehler von ca. −0.5 kgm −2 mit Bezug auf die Radiometer und auf die Radiosonden. Es wurde festgestellt, dass die unterschiedliche zeitliche und ¨ ortliche Aufl¨ osung der Instrumente einen starken Einfluss auf die Standardabweichung hat. Ein charakteristischer Tagesgang des radiometrischen und des GPS-IWV wurde beobachtet. 1 Introduction The concentration of water vapour in the troposphere is typically approx. 1%, but its impact on physical pro- cesses in the troposphere is outstanding. Water vapour is the most important greenhouse gas, and the change between its gaseous, liquid and solid state is an effec- tive means of energy storage and transport in local and global weather. Thus it is crucial in meteorology to know its amount and its distribution precisely. This is one of the reasons why the COST 720 Tem- perature, hUmidity, and Cloud (TUC) profiling cam- paign (RUFFIEUX et al. (2006)) was held at the aero- logical station of MeteoSwiss in Payerne, Switzerland, in winter 2003/2004. The station is located close to the Lake of Neuchˆ atel at 490 m above sea level where weather situations with fog and temperature inversions occur frequently during winter. A database of meteo- rological measurements made with in-situ and various remote sensing methods was produced to asses their po- ∗ Corresponding author: Lorenz Martin, Institute of Applied Physics, University of Bern, Sidlerstrasse 5, CH-3012 Bern, Switzer- land, e-mail: lorenz.martin@mw.iap.unibe.ch tential for tropospheric monitoring under these condi- tions. Whereas CIMINI et al. (2006) investigated the retrieval of humidity and temperature profiles, this pa- per focuses on integrated water vapour (IWV). Three methods were used for the retrieval of IWV in the TUC campaign: radiosondes, the global positioning system (GPS), and two different types of microwave radiome- ters. Since the radiometers also measure the integrated liquid water (ILW), these data are included in the inter- comparison. Similar studies were made by EMARDSON et al. (1998) and by KOPKEN (2001) within the Baltic Sea Ex- periment (BALTEX) and by REVERCOMB et al. (2003) within the Atmospheric Radiation Measurement (ARM) programme. However, this study is the first to in- volve IWV measurements made with the SRS 400 radiosonde (RICHNER (1999)) and with the All-Sky MUlti WAvelength RAdiometer (ASMUWARA, MAR- TIN et al. (2006a)). Furthermore, the all-sky scanning capabilities of ASMUWARA are used to asses the influ- ence of inhomogeneities in the water vapour field on the IWV measurement. Fig. 1 shows an overview of the IWV dataset gained DOI: 10.1127/0941-2948/20XX/00XX-XXXX 0941-2948/XX/00XX-XXXX $ X.XX c  Gebr¨ uder Borntraeger, Berlin, Stuttgart 20XX