GNSSseg, a Statistical Method for the Segmentation of Daily GNSS IWV Time Series

Abstract

Homogenization is an important and crucial step to improve the usage of observational data for climate analysis. This work is motivated by the analysis of long series of GNSS Integrated Water Vapour (IWV) data, which have not yet been used in this context. This paper proposes a novel segmentation method called segfunc that integrates a periodic bias and a heterogeneous, monthly varying, variance. The method consists in estimating first the variance using a robust estimator and then estimating the segmentation and periodic bias iteratively. This strategy allows for the use of the dynamic programming algorithm, which is the most efficient exact algorithm to estimate the change point positions. The performance of the method is assessed through numerical simulation experiments. It is implemented in the R package GNSSseg, which is available on the CRAN. This paper presents the application of the method to a real data set from a global network of 120 GNSS stations. A hit rate of 32% is achieved with respect to available metadata. The final segmentation is made in a semi-automatic way, where the change points detected by three different penalty criteria are manually selected. In this case, the hit rate reaches 60% with respect to the metadata.

A statistical method for the attribution of change‐points in segmented Integrated Water Vapor difference time series

“…Nearby stations were searched in the NGL data set, with a distance limit of 200 km in horizontal and 500 m in vertical. Data from the ERA5 reanalysis are extracted at the location of each GNSS station and the difference series, G-E and G'-E', are formed and segmented using the GNSSseg package (Quarello et al, 2022). The segmentation results are post-processed to remove clusters of change-points which occur occasionally in regions where the GNSS data and reanalysis data have a significant representativeness difference (Bock & Parracho, 2019).…”

“…The GNSS minus reanalysis difference series show usually strong heteroscedasticity and periodic (seasonal) biases, along with weak autocorrelation (Quarello et al, 2022). In the following, a series is modelled using the following regression model:…”

“…The segmentation results are post-processed to remove clusters of change-points which occur occasionally in regions where the GNSS data and reanalysis data have a significant representativeness difference (Bock & Parracho, 2019). Change-points within clusters are either completely removed, or only one is kept (see an example in Figure 3 of [Quarello et al, 2022]).…”

“…We use integrated water vapour (IWV) derived from Global Navigation Satellite System (GNSS) measurements (Bock, 2019) and from the fifth ECMWF reanalysis (ERA5; Hersbach et al, 2020). Because the global GNSS data set is relatively sparse, the reanalysis is used as a reference to form the target minus reference difference series Nguyen et al, 2021;Ning et al, 2016;Quarello et al, 2022;Van Malderen et al, 2020). The GNSS minus reanalysis data are segmented with the 'GNSSseg' method developed by (Quarello, 2020) which is based on a penalized likelihood approach.…”