Abstract
We developed operators to assimilate Global Navigation Satellite System (GNSS) Zenith Total Delays (ZTDs) and horizontal delay gradients into a numerical weather model. In this study we experiment with refractivity fields derived from the Global Forecast System (GFS) available with a horizontal resolution of 0.5°. We begin our investigations with simulated observations. In essence, we extract the tropospheric parameters from the GFS analysis, add noise to mimic observation errors and assimilate the simulated observations into the GFS 24h forecast valid at the same time. We consider three scenarios: (1) the assimilation of ZTDs (2) the assimilation of horizontal delay gradients and (3) the assimilation of both ZTDs and horizontal delay gradients. The impact is measured by utilizing the refractivity fields. We find that the assimilation of the horizontal delay gradients in addition to the ZTDs improves the refractivity field around 800 hPa. When we consider a single station there is a clear improvement when horizontal delay gradients are assimilated in addition to the ZTDs because the horizontal delay gradients contain information that is not contained in the ZTDs. On the other hand, when we consider a dense station network there is not a significant improvement when horizontal delay gradients are assimilated in addition to the ZTDs because the horizontal delay gradients do not contain information that is not already contained in the ZTDs. Finally, we replace simulated by real observations, that is, tropospheric parameters from a Precise Point Positioning solution provided with the G-Nut/Tefnut software, in order to show that the GFS 24h forecast is indeed improved when GNSS horizontal delay gradients are assimilated in addition to GNSS ZTDs; for the considered station (Potsdam, Germany) and period (June and July, 2017) we find an improvement in the retrieved refractivity of up to 4%.
Highlights
Radio signals which are transmitted by the Global Navigation Satellite System (GNSS)constellation and received by a ground-based station allow the estimation of the Zenith Total Delay (ZTD) [1] and the horizontal delay gradient [2]
24h forecast is improved when GNSS horizontal delay gradients are assimilated in addition to GNSS ZTDs; for the considered station (Potsdam, Germany) and period (June and July, 2017) we find an improvement in the retrieved refractivity of up to 4%
In order to make sure that what we model is what we measure, we show in the second sub-section how ZTDs and horizontal delay gradients are estimated with the Precise Point Positioning (PPP) method [13]
Summary
Radio signals which are transmitted by the Global Navigation Satellite System (GNSS). Douša et al presented in Reference [10] a comparison study on ZTDs and horizontal delay gradients for hundreds of stations located in central Europe. For several stations they find clear artificial signals in horizontal delay gradients which are caused by problems to receive low-elevation observations They show that the standard deviations between GNSS and NWM ZTDs and horizontal delay gradients have a clear seasonal dependency, that is, the standard deviation is larger in summer than in winter. This can be explained by the fact that NWMs have difficulties to predict the high water vapour variability in summer It appears that the standard deviation for the ZTD does not change over the years whereas the standard deviation for the horizontal delay gradient decreased over the years.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
