A turbulence-based stochastic model for GNSS tropospheric delays from boundary-layer Large Eddy Simulations
Year: 2026
Type: preprint
Abstract: Accurate stochastic modelling of tropospheric delays is essential for highprecision Global Navigation Satellite System (GNSS) positioning. Empirical elevation-dependent weighting schemes ignore the temporal correlation structure induced by atmospheric turbulence. We derive analytical relationships between Zenith Wet Delay (ZWD) spectral parameters and atmospheric boundary layer (ABL) forcing from 63 idealised Large Eddy Simulations (LES) of a convective ABL, varying geostrophic wind, surface sensible heat flux and boundary-layer height. Matérn parameters, the cutoff frequency λ and the process standard deviation σ, are extracted from synthetic ZWD and slant wet delay (SWD) time series via debiased Whittle pseudolikelihood estimation. A variance-based sensitivity analysis shows that λ depends almost exclusively on wind speed, consistent with Taylor’s frozen-turbulence hypothesis, whereas σ is jointly forced by mechanical shear and buoyant production. Symbolic regression yields a linear wind dependence of λ and a dual additive dependence of σ on wind speed and heat flux. SWDs follow the same wind scaling for λ and an inverse-sine elevation law for σ. These expressions allow a variance inflation factor (VIF) to be computed from routine meteorological inputs (wind speed and surface heat flux from reanalysis or local instruments), giving GNSS least-squares adjustment a siteand time-dependent variance scheme in place of static elevation-based weighting. Confrontation with three years of GNSS ZWD at Payerne (Switzerland) yields agreement to 5% on the median λ and a median σ within a factor 1.4 of the
Institution: Leibniz University Hannover
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Keywords: GNSS tropospheric delay, zenith wet delay, atmospheric boundary layer, atmospheric turbulence, stochastic modeling, Matérn covariance, cutoff frequency, variance inflation factor, convective boundary layer, geostrophic wind, sensible heat flux, frozen turbulence hypothesis, least squares adjustment, Payerne
Open Access status: green
OpenAlex ID: W7226008905
