Statistical Evaluation of Ground Subsidence Data

Authors

  • Marija Kačavenda Faculty of Architecture, Civil Engineering and Geodesy, University of Banja Luka Author

DOI:

https://doi.org/10.63356/stes.ing.2026.001

Keywords:

statistical analysis, non-parametric tests, monitoring, ground subsidence, geotechnical engineering

Abstract

Introduction: Ground subsidence represents a complex geotechnical process that may significantly compromise the stability of structures and infrastructure.
The Aim: This study presents a comprehensive statistical analysis of subsidence measurements collected at 89 geodetic control points over the period from 2013 to 2018, with the purpose of spatial analysis of subsidence and identification of critical zones.
Material and Methods: Descriptive analysis showed that mean subsidence ranged from –0.2 mm (2014) to –18.1 mm (2018), while the maximum recorded subsidence reached –297 mm. Critical points with cumulative subsidence exceeding 100 mm were identified, most notably GT048 (–297 mm; –59.4 mm/year), PT015 (–257 mm; –51.4 mm/year), and GT007 (–242 mm; –48.4 mm/year). Correlation analysis revealed a strong positive relationship between the early and middle intervals (dH1–dH3: ρ = 0.652; p < 0.001), and a very strong relationship between the middle and final intervals (dH3–dH5: ρ = 0.860; p < 0.001). Friedman's test confirmed statistically significant differences between years (χ² = 16.011; p = 0.001), and the post-hoc Wilcoxon test identified three phases: the phase of highest subsidence (2014), a stable phase (2015–2016), and a phase of reduced intensity (2018).
Conclusion: The results indicate the need for additional geotechnical measures and continuous monitoring, particularly at critical points GT048, PT015 and GT007, where further investigations and interventions are recommended.  (–242 mm; –48.4 mm/year). Correlation analysis revealed a strong positive relationship between the early and middle intervals (dH1–dH3: ρ = 0.652; p < 0.001), and a very strong relationship between the middle and final intervals (dH3–dH5: ρ = 0.860; p < 0.001). Friedman's test confirmed statistically significant differences between years (χ² = 16.011; p = 0.001), and the post-hoc Wilcoxon test identified three phases: the phase of highest subsidence (2014), a stable phase (2015–2016), and a phase of reduced intensity (2018).  
Conclusion: The results indicate the need for additional geotechnical measures and continuous monitoring, particularly at critical points GT048, PT015 and GT007, where further investigations and interventions are recommended.

Downloads

Published

2026-11-18

Issue

Section

Articles