EGU26-15806, updated on 14 Mar 2026
https://doi.org/10.5194/egusphere-egu26-15806
EGU General Assembly 2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Wednesday, 06 May, 08:30–10:15 (CEST), Display time Wednesday, 06 May, 08:30–12:30
 
Hall X3, X3.61
Expansive-Clay Deformation Monitoring and Rainfall-Lag Analysis in Houston Using Multi-Source Data Constraints and PGAU-Net
Zhuotan Shang1, Jinzhao Si2, and Zhong Lu1
Zhuotan Shang et al.
  • 1China University of Mining and Technology, School of Environment and Spatial Informatics, China (sztan@cumt.edu.cn)
  • 2School of Geological Engineering and Geomatics, Chang’an University(sijinzhao_chd@163.com)

Expansive clays can undergo pronounced seasonal oscillations and long-term trend deformation driven by rainfall infiltration and soil-moisture fluctuations, posing persistent potential threats to urban infrastructure. Taking Houston, a representative region with widespread expansive-clay deposits, as a case study, this paper proposes an expansive-clay hazard monitoring and interpretation framework that integrates heterogeneous-data-based atmospheric correction and signal-separation techniques to address tropospheric-delay contamination and complex deformation-signal mixing. First, we develop a Point–Grid Attention U-Net (PGAU-Net) that fuses high-temporal-resolution GNSS-ZTD observations with the spatially continuous ERA5 background field to reconstruct a high-accuracy tropospheric delay correction field, significantly suppressing atmospheric phase noise in interferometric synthetic aperture radar (InSAR) time-series analysis. Using this correction, we retrieve a five-year (2018–2023) surface deformation time series for the Houston area. The results show that expansive-clay deformation exhibits a pronounced periodic component together with a linear subsidence trend. We further apply wavelet analysis to decompose the deformation into periodic and trend components. The periodic oscillations agree well with the rainfall time series, while the overall deformation indicates an evident subsidence trend, with an average annual deformation rate of approximately −14 mm/yr. Moreover, we investigate the coupling between periodic parameters of expansive-clay deformation and rainfall cycles, estimating a deformation lag relative to rainfall of about Δt = 23 days, and discuss its implications for soil-moisture diffusion and interlayer seepage processes. Finally, we cross-validate the InSAR-derived deformation using GNSS deformation time series at different burial depths, thereby revealing differences between shallow and deep soil layers in periodic response amplitude and phase lag. Overall, the proposed framework can stably extract the periodic–linear deformation characteristics of Houston expansive clays while effectively mitigating atmospheric errors, providing a verifiable technical pathway for long-term monitoring and mechanistic analysis of urban expansive-clay hazards.

How to cite: Shang, Z., Si, J., and Lu, Z.: Expansive-Clay Deformation Monitoring and Rainfall-Lag Analysis in Houston Using Multi-Source Data Constraints and PGAU-Net, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15806, https://doi.org/10.5194/egusphere-egu26-15806, 2026.