Site scale slow moving landslides: characterization by monitoring and modelling
Resumen Abstract Índice Conclusiones
Bru Cruz, Guadalupe
MEJOR TESIS EN GEOFÍSICA
2019-A
Descargar PDF 
Los deslizamientos de ladera son un peligro natural a nivel global. Pueden ocurrir bajo multitud de condiciones y son capaces de provocar una gran cantidad de pérdidas de vidas humanas e importantes daños económicos. La definición general de deslizamiento incluye una amplia variedad de fenómenos con características y comportamientos mecánicos propios. Entre esta diversidad de tipologías, los deslizamientos lentos de ladera se caracterizan por alcanzar velocidades de hasta varios centímetros/año y porque la rotura se desarrolla a lo largo de planos de cizalla. Pueden llegar a producir un gran impacto en zonas urbanas e infraestructura y ser precursores de movimientos más rápidos y catastróficos.
La heterogeneidad y complejidad de los fenómenos de inestabilidad de laderas exigen una nomenclatura y sistema de clasificación universales que puedan ser utilizados por toda la comunidad científica y técnica. La identificación preliminar del tipo de movimiento y principales características, tanto espaciales como temporales, es fundamental para poder diseñar sistemas de monitorización óptimos y para realizar la hipótesis del sistema a modelar.
Monitorizar los desplazamientos de un deslizamiento sirve para estudiar y controlar su actividad. En la actualidad es posible medir desplazamientos de la superficie terrestre mediante técnicas de teledetección, entre las cuales destaca la interferometría diferencial avanzada radar de satélite (A-DInSAR), que utiliza imágenes radar adquiridas de forma periódica bajo cualquier condición climática y permite monitorizar deformaciones complejas con precisión sub-centimétrica. En este trabajo se examina la integración de esta técnica en los sistemas de monitorización geotécnica de laderas.
Los modelos físicos se emplean para elaborar análisis retrospectivos (back analysis) del movimiento de deslizamientos y predicciones cuantitativas. Este trabajo explora los beneficios de realizar análisis en 3 dimensiones, aprovechando la versatilidad de los programas de elementos finitos y la información obtenida de los sistemas de monitoreo para calibrar y validar los modelos. Se presenta una metodología avanzada que reproduce la cinemática de deslizamientos lentos de ladera activos en 3D.
En conjunto, esta tesis doctoral presenta una guía práctica integral para estudiar deslizamientos lentos de ladera a escala local, basada en la clasificación detallada del fenómeno, monitorización mediante técnicas A-DInSAR y modelos numéricos en 3D.Site scale slow moving landslides: characterization by monitoring and modelling
Guadalupe Bru Cruz
Landslides are a worldwide widespread natural hazard that occur under a wide range of environments and conditions, and that can result in enormous fatalities and huge economic losses. Landslide generic definition includes a large variety of phenomena, with their own peculiar features and mechanical behavior. Among the broad diversity of landslide a types, slow moving slides are characterized by motion rates reaching several centimeters per year and occurring dominantly on surfaces of rupture. These slides are capable of producing a dramatic impact on properties and infrastructures in urbanized areas, and may be precursors for faster and catastrophic mass movements.
The heterogeneity and complexity of unstable slope phenomena brings the need for a universal terminology and a system of classification to be used by the landslide community. Preliminary identification of the type of movement and main features, both spatially and temporally, serves to design the most optimal monitoring system and to provide inputs for hypothesizing the problem that wants to be modelled.
Monitoring displacements of unstable slopes is fundamental for both studying and controlling their activity. Nowadays measurements of superficial displacements can be achieved with remote sensing techniques, such as Advanced Differential SAR Interferometry (A-DInSAR), which uses satellite radar images acquired regularly under any meteorological conditions. These data allow to monitor complex deformation with sub-centimetric precision. Its integration to the operational geotechnical monitoring of landslides is examined in this work.
Physically based models are used to perform quantitative landslide motion back analysis and predictions. This work explores the benefits of carrying out 3D analysis taking advantage of the versatility of the finite element method and the monitoring data to calibrate and validate it. This PhD presents an advanced 3D numerical methodology to reproduce the kinematics of slow active landslides.
Overall, this work provides an integral practice for site-scale slow moving landslide analysis, based on accurate phenomena classification, A-DInSAR surface monitoring and 3D numerical modelling.
Site scale slow moving landslides: characterization by monitoring and modelling
Guadalupe Bru Cruz
This PhD Thesis has been committed to the study of site scale slow moving landslides affecting urban areas or infrastructures. An integral practice for analysis, based on accurate landslide classification, A-DInSAR surface monitoring and 3D deterministic numerical modelling, has been presented. The formulated objectives have been addressed by the analysis of three real landslide case studies which have been published in international journals included in the quartile Q1 of the Journal Citations Reports (JCR) Science Edition.
The general conclusions raised from the publications of each case study are valid for similar slow moving landslides local analysis.
Regarding the first objective, the importance of preliminary landslide classification has been remarked as the first step to approach landslide studies. The case study described in Chapter 2, occurring in Arcos de la Frontera town, presents two different geotechnical processes (a slow earth slide and the compaction of an artificial earth fill) that have been detected with A-DInSAR monitoring. The remote sensing surface motion data was correctly interpreted only after the two different mechanisms were identified. Additionally, it was demonstrated that even a short set of SAR images are capable of detecting surface displacement (i.e. active areas) if a multi-master method is used.
Related to the second objective, it has been demonstrated that A-DInSAR techniques can be applied in operational geotechnical monitoring of site scale slow moving landslides. A practical guideline has been presented in Chapter 3 that includes an appropriate dataset selection taking into account the scenario characteristics, a high resolution visibility analysis using Sentinel-2 optical data and considerations when comparing advanced differential SAR interferometry (A-DInSAR) results with other monitoring techniques.
Finally, as regard to the third objective, 3D landslide kinematics modelling has been successfully achieved taking advantage of the versatility of the finite elements method (FEM) programs. In Chapter 4 it has been demonstrated that short to medium term predictions of the evolution of real site scale slow earth slides in 3D are achievable, taking advantage of the high spatial density of surface displacement data from advanced remote sensing techniques that can be used for calibration and validation. It has been proven that, for the purposes of reproducing slow velocity landslides in a more realistic way, 3D analyses are preferable to 2D ones, mainly for two reasons: it catches the 3D features that all landslides exhibit, differential movements and prevents from testing different profiles to find the most critical one. The 3D model has been adjusted to determine material parameters uncertainties by a stability back analysis and by reproducing the time dependent past behavior, using a coupled formulation to model solid skeleton and pore fluids interaction (taking into account daily rainfall intensity). It has been confirmed that 3D models are complex and demanding, yet at the same time they offer precise quantitative results, which are very valuable for engineering solutions calculations and hazard assessment.
Site scale slow moving landslides: characterization by monitoring and modelling
Guadalupe Bru Cruz