Atmospheric VLBI: A method to validate long time series of water vapour content G. Elgered, R. Haas, T. Nilsson Department of Radio and Space Science, Chalmers University of Technology, Onsala Space Observatory, SE - 439 92 Onsala, Sweden Abstract. We assess the possibility to vali- date time series of the atmospheric integrated water vapour (IWV) from GPS observations us- ing geodetic VLBI at the Onsala Space Obser- vatory in Sweden. An overall motivation is to determine the relation—and its uncertainties— between trends in the IWV with trends in the ground temperature. We find that the frequency of VLBI experiments is too low in order to vali- date estimated linear trends using data acquired over a ten year period. On the other hand, the VLBI method provides an accuracy of the same order, compared to using GPS or nearby radiosonde launches for validation of the IWV on an absolute scale. Further assessments using other, as well as larger, data sets are called for. Keywords. VLBI, atmospheric water vapour, global wetting, global warming 1 Introduction Water vapour is an important atmospheric gas in climate models. A specific question at issue is the relation between changes in the tempera- ture and possible corresponding changes in the integrated water vapour (IWV). Following the Clausius-Clapeyron relation, assuming conserva- tion of relative humidity, we obtain a relation for the IWV changes due to a change in the tempera- ture of approximately 6 [%/K] (Trenberth et al., 2003). Analyses and comparisons of trends in the temperature and the IWV using the ERA40 model do not give an accurate assessment of this relation. This is argued to be due to artifacts in the global observing system of water vapour (Bengtsson et al., 2004). Accurate observations, and especially with a high long term stability, of the IWV as well as of the temperature are there- fore important. The possibility to infer the IWV from space geodetic observations has been demonstrated, with increasing quality, during the last twenty years. First by using geodetic VLBI data (e.g. Herring et al., 1990; Heinkelmann et al., 2007) and thereafter also based on GPS (e.g. Tralli & Lichten, 1990; Elgered et al., 2005). Ground- based GPS receiver stations provide a relatively high spatial resolution and results in close to real time, at least compared to VLBI. Therefore, the IWV estimates from GPS networks are nowa- days used, and their impact assessed, in assimi- lation into numerical models for weather predic- tion. The absolute accuracy, and especially the long term stability of the GPS results is of funda- mental importance for applications in climate re- search. In order to assess the stability of the GPS results an independent, preferably a more stable and accurate, method is called for. We here in- vestigate to what extent geodetic VLBI is an ap- propriate method for this task. For this purpose we use geodetic VLBI and GPS data from the Onsala Space Observatory and radiosonde data from the G¨ oteborg-Landvetter Airport. In Section 2 we present an overview of the re- sults recently obtained in terms of linear trends in the IWV using data acquired with the Swedish and Finnish ground-based GPS networks from 1996 to 2006. Section 3 describes the analy- sis of the geodetic VLBI experiments at Onsala. In Section 4 we compare the results of the esti- mated IWV using VLBI, GPS, and radiosondes. Section 5 contains the conclusions. 2 GPS estimates of IWV 2.1 Summary of estimated trends Linear trends in the IWV were recently esti- mated using ten years of GPS data from the Proceedings of the 19th European VLBI for Geodesy and Astrometry Working Meeting, 24-25 March 2009, Bordeaux