Water vapour trends in Switzerland from radiometry, FTIR and GNSS ground stations

Water vapour in the atmosphere is not only a strong greenhouse gas, but also affects many atmospheric processes such as the formation of clouds and precipitation. With increasing temperature, Integrated Water Vapour (IWV) is expected to increase. Analysing how atmospheric water vapour changes in time is therefore important to monitor ongoing climate change. To determine whether IWV increases in Switzerland as expected, we asses IWV trends from a tropospheric water radiometer (TROWARA) in Bern, from a Fourier Transform Infrared (FTIR) spectrometer at Jungfraujoch and from the Swiss network of ground-based Global Navigation Satellite System (GNSS) stations. In addition, trends are assessed from reanalyses data, using the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5) and the Modern-Era Retrospecitve Analysis for Research and Applications (MERRA-2). Ground-based GNSS data are well suited for IWV trends due to their high temporal resolution and the spatially dense networks. However, they are highly sensitive to instrumental changes and care has to be taken when determining GNSS based trends. We therefore use a straightforward trend method to account for jumps in the GNSS data when instrumental changes were performed. Our data show mostly positive IWV trends between 2 and 5 % per decade in Switzerland. GNSS trends are significant for some stations and have the tendency to be larger at higher altitudes. We found that IWV scales to temperature as expected in Bern, except in winter. However, the correlation between IWV and temperature based on reanalyses data is not always clear. Besides our positive IWV trends, we found a good agreement of radiometer, GNSS and reanalyses data in Bern, and a clear-sky bias of the FTIR compared to GNSS data at Jungfraujoch. Our results are generally consistent with the positive water vapour feedback in a warming climate. We show that ground-based GNSS networks provide a valuable source for regional climate monitoring with high spatial and temporal resolution, but homogeneously reprocessed data and advanced trend techniques are needed to account for data jumps