RT journal article
T1 Geospatial Clustering of GNSS Stations Using Unsupervised Learning: A Statistical Framework to Enhance Deformation Analysis for Environmental Risk Management
A1 Álvarez-Ruiz, Daniel
A1 Sánchez Alzola, Alberto
A1 Pastor Fernández, Andrés
A2 Ingeniería Mecánica y Diseño Industrial
A2 Estadística e Investigación Operativa
K1 machine learning
K1 GNSS
K1 time series analysis
K1 harmonic models
AB The global expansion of continuous GNSS networks has generated large-scale spatiotemporaldatasets whose analysis requires robust mathematical and statistical tools. This studyintroduces a geospatial, multivariate statistical framework for classifying 21,548 GNSSstations from the University of Nevada repository. The methodology integrates harmonicregression, stochastic noise modeling, quality assessment, and slope estimation into aunified feature space suitable for high-dimensional analysis. Using unsupervised learningclustering computed with our custom-developed code, based entirely on free andopen-source software, we identify homogeneous station groups that reflect dominant signalproperties—periodicity, noise structure, data quality, and long-term velocity—togetherwith their spatial context. The resulting clusters exhibit strong mathematical coherenceand reveal continental-scale patterns driven by seasonal forcing, tectonic regime, climaticvariability, and monument stability. By grouping stations with similar statistical behavior,the proposed framework improves reference-site selection, enhances deformation-field interpretation,and supports the detection of anomalous or hazard-related behavior. Overall,this approach provides a scalable, data-driven mathematical tool for analyzing complexspatiotemporal signals and contributes to more reliable deformation modeling and environmentalrisk assessment.
PB MDPI, Basel
SN 2227-7390
YR 2026
FD 2026-03-03
LK http://hdl.handle.net/10498/39182
UL http://hdl.handle.net/10498/39182
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 05-nov-2026