Houston GNSS Network for Subsidence and Faulting Monitoring: Data Analysis Methods and Products

Abstract

Harris-Galveston Subsidence District (HGSD), in collaboration with several other agencies, has been operating a dense Global Navigation Satellite System (GNSS) network for subsidence and faulting monitoring within the Greater Houston region since the early 1990s. The GNSS network is designated HoustonNet, comprising approximately 250 permanent GNSS stations as of 2021. This paper documents the methods used to produce position time series, transform coordinates from the global to regional reference frames, identify outliers and steps, analyze seasonal movements, and estimate site velocities and uncertainties. The GNSS positioning methods presented in this paper achieve 2-4-mm RMS accuracy for daily positions in the north-south and east-west directions and 5-8-mm accuracy in the vertical direction within the Greater Houston region. Five-year or longer continuous observations are able to achieve submillimeter-per-year uncertainties (95% confidence interval) for both horizontal and vertical site velocities. Two decades of GNSS observations indicate that Katy in Fort Bend County, Jersey Village in northwestern Harris County, and The Woodlands in southern Montgomery County have been the areas most affected by subsidence (1-2 cm=year) since the 2000s; the overall subsidence rate and the size of subsiding area (>5 mm=year) have been decreasing as a result of the groundwater regulations enforced by HGSD and other local agencies. HoustonNet data and products are released to the public through HGSD. The primary products are the daily East-North-Up (ENU) position time series and site velocities with respect to the International GNSS Service (IGS) Reference Frame 2014 (IGS14), the stable Gulf of Mexico Reference Frame (GOM20), and the stable Houston Reference Frame (Houston20). The ENU position time series with respect to Houston20 are recommended for delineating subsidence and faulting within the Greater Houston region. The ENU time series with respect to GOM20 are recommended for studying subsidence and faulting within the Gulf coastal plain and sea-level changes along the Gulf Coast. The entire HoustonNet data set is reprocessed every a few years with updated positioning software, IGS and regional reference frames, and data analysis tools. We recommend that users use the most recent release of HoustonNet data products and avoid mixing old and new positions.

A methodology for long-term offshore structural health monitoring using stand-alone GNSS: case study in the Gulf of Mexico

“…Bertiger et al 47 reported that the GipsyX-PPP daily resolutions achieve approximately 2-3 mm global RMS accuracy in the horizontal direction and 6-7 mm in the vertical direction. According to the previous study in the GOM region, 15,50 the accuracy of the PPP solutions in the GOM coastal region is slightly worse than the global average, mostly because of the humid climates. The RMS accuracy for the daily positions at DEV1 is 4 mm in the NS direction, 3 mm in the EW direction, and 7 mm in the vertical direction (Figure 4).…”

“…The detailed setup for the PPP processing is the same as the setup for the HoustonNet GNSS data routine processing. 50…”

“…The detailed method is addressed in a recent publication. 50 About 50 daily positions among 2948 days (1.5%) are removed as outliers. Further analysis indicates that the majority of outliers occurred in those days that only had a few hours of observations (e.g., less than 4 h).…”

New Preconsolidation Heads Following the Long‐Term Hydraulic‐Head Decline and Recovery in Houston, Texas

“…Only the datasets from two extensometers (Pasadena, Texas City) among these 13 extensometers show a clear trend of land rebound. There are over one hundred GPS stations in the HGSD Regulatory Areas 1 and 2 (Wang et al 2022); only about a dozen GPS recorded steady land rebound (1 to 2 mm/year), as illustrated in Figure 6. The locations of these GPS stations are marked in Figure 1.…”

“…There is a pair of shallow and deep extensometer boreholes at each of the Clear Lake and Baytown sites. The contour lines indicate the average subsidence rates derived from GPS measurements (2010‐2020) (Wang et al 2022). The dots represent GPS stations that recorded a steady land rebound.…”