Global IWV trends and variability in atmospheric reanalyses and GPS observations

Abstract

Abstract. Water vapour plays a key role in the climate system. However, its short residence time in the atmosphere and its high variability in space and time make it challenging when it comes to study trends and variability. There are several sources of water vapour data. In this work we use Integrated Water Vapour (IWV) estimated from GPS observations and atmospheric 10 reanalyses. Monthly and seasonal means, interannual variability, and linear trends are analysed and compared for the period between 1995 and 2010. A general good agreement is found but this study highlights issues in both GPS and reanalysis data sets. In GPS, gaps and inhomogeneities in the time series are evidenced, which affect mainly variability and trend estimation.In ERA-Interim, too strong trends in certain regions (e.g. drying over northern Africa and Australia, and moistening over northern South America) were found. Representativeness differences in coastal areas and regions of complex topography 15 (mountain ranges, islands) are also evidenced as limitations to the intercomparison of the point observations and reanalysis data. A general good agreement is found for the means and variabilities, with the exception of a few stations where representativeness issues are suspected. Monthly IWV trends are also found to be in good sign agreement, with the exception of a handful of stations where, in addition to representativeness errors, there might be inhomogeneities in the GPS time series.Seasonal trends are found to be different and more intense than monthly trends, which emphasizes the influence of atmospheric 20 circulation on IWV trends. In order to assess strong trends over regions lacking GPS stations, a second reanalysis, MERRA-2, is introduced. The period of analysis is extended to 1980-2016 (the longest period the reanalyses have in common) and differences with the shorter period are found. This exemplifies how much IWV trends are dependent on the time period at study and must be interpreted carefully. Temperature trends are also computed for both reanalyses. The Clausius-Clapeyron scaling ratio is found to not be a good humidity proxy at seasonal and regional scales. Regions over northern Africa and 25 Australia, where ERA-Interim and MERRA-2 disagree, are investigated further. Dynamics at these regions is assessed by analyzing the wind fields at 925 hPa and is shown to be tightly linked with the trends and variability in IWV.

Impact of humidity biases on light precipitation occurrence: observations versus simulations

“…The data cover the period from January 1995 to May 2011 and include more than 400 stations globally. For the present study, the delay data were screened and converted into 6-hourly IWV estimates as described in Parracho et al (2018). The dataset was restricted to the period from Januray 1995 to December 2008 including 95 GPS stations over Europe as shown in Fig.…”

“…In this study, we make use of the same Global Positioning Sysem (GPS) IWV dataset as used by Parracho et al, (2018) which provides IWV measurements over a hundred of European sites. The GPS technique accurately measures IWV in all weather conditions including rainy situations, which is an important aspect for our study (e.g Wang et al 2007).…”

A breakpoint detection in the mean model with heterogeneous variance on fixed time intervals

“…By looking to the estimated means (Figure 8 the detection of outliers and seems to capture a periodic signal. This latter point can be due to the fact that a periodic tendency remains despite the correction by the ERAI model [31]. Table 1.…”

“…Context and data description. The GNSS-derived IWV series are used to study and verify climate model predictions of atmospheric water vapour trends and variability connected to climate change (global warming) [31]. The biases induced by the abrupt changes are small and thus difficult to distinguish from the natural climate variation of the measured IWV signal [15].…”

“…In this application we consider the two stations SYOG (Syowa, Antarctica) and ONSA (Onsala, Sweden) contributing to the International GNSS Service (IGS) network of continuously operating reference stations (www.igs.org). The IWV data retrieved from these GPS measurements are described in [31]. In the present work, the IWV data series are used with daily time sampling.…”