Over fifty small sinkholes (~1 meter in depth and width) were found in conjunction with structural damage to homes in an area south of Cleveland, TX. The local geology lacks carbonate and evaporite deposits associated with normal sinkhole development through dissolution. The morphology and distribution of sinkholes, and the geologic setting of the site are consistent with piping erosion. However, the site lacked the significant hydraulic gradient or exit points for sediment associated with traditional piping erosion. In areas of sinkholes, geophysical measurements of apparent electrical conductivity delineated anomalously high conductivity levels that are interpreted as a brine release from a nearby oil-field waste injection well. The contaminated areas have sodium adsorption ratios (SAR) as high as 19, compared to background levels of 3. Sodium has been shown to cause dispersion of soil colloids, allowing for sediment transport at very low velocities. Thus, subsurface erosion of dispersed sediment could be possible without significant hydraulic gradients. This hypothesis is backed by the observation of the depletion of colloidal particles within the E-horizon of sinkholes. However, there is a lack of precedence of waste brines initiating colloid dispersion. Also, sodium dispersion is not thought to be an important process in piping erosion in humid settings such as this one. Therefore, laboratory experiments on samples from the site area, designed to simulate field conditions, were conducted to measure dispersion verses pH, SAR and electrical conductivity (EC). Analysis of the experimental data with neural networks showed that an increase in SAR did increase dispersion. A dispersion prediction map, constructed with the trained neural network and calibrated geophysical data, showed correlation between sinkhole locations and increased predicted dispersion. This research indicates that a contaminant high in sodium content has caused colloidal dispersion, which may have allowed nontraditional subsurface erosion to occur in an area lacking a significant hydraulic gradient.

Over fifty small sinkholes (~1 meter in depth and width) were found in conjunction with structural damage to homes in an area south of Cleveland, TX. The local geology lacks carbonate and evaporite deposits associated with normal sinkhole development through dissolution. The morphology and distribution of sinkholes, and the geologic setting of the site are consistent with piping erosion. However, the site lacked the significant hydraulic gradient or exit points for sediment associated with traditional piping erosion. In areas of sinkholes, geophysical measurements of apparent electrical conductivity delineated anomalously high conductivity levels that are interpreted as a brine release from a nearby oil-field waste injection well. The contaminated areas have sodium adsorption ratios (SAR) as high as 19, compared to background levels of 3. Sodium has been shown to cause dispersion of soil colloids, allowing for sediment transport at very low velocities. Thus, subsurface erosion of dispersed sediment could be possible without significant hydraulic gradients. This hypothesis is backed by the observation of the depletion of colloidal particles within the E-horizon of sinkholes. However, there is a lack of precedence of waste brines initiating colloid dispersion. Also, sodium dispersion is not thought to be an important process in piping erosion in humid settings such as this one. Therefore, laboratory experiments on samples from the site area, designed to simulate field conditions, were conducted to measure dispersion verses pH, SAR and electrical conductivity (EC). Analysis of the experimental data with neural networks showed that an increase in SAR did increase dispersion. A dispersion prediction map, constructed with the trained neural network and calibrated geophysical data, showed correlation between sinkhole locations and increased predicted dispersion. This research indicates that a contaminant high in sodium content has caused colloidal dispersion, which may have allowed nontraditional subsurface erosion to occur in an area lacking a significant hydraulic gradient.

Le travail présenté dans ce mémoire de thèse porte sur la modélisation discrète de l’évolution régressive du front d’un conduit d’érosion qui peut se produire dans les barrages ou digues en remblai. Des outils numériques ont été développés en se basant sur le couplage entre la méthode des éléments discrets (DEM) et la méthode de Boltzmann sur réseau (LBM) pour la description, respectivement, des phases solide et fluide. L’implémentation de la méthode DEM suit une approche standard de type dynamique moléculaire (DM) et les interactions intergranulaires sont modélisées par des contacts unilatéraux visco-élastiques frottants ou bilatéraux (ponts solides) viscoélastiques, afin de permettre la modélisation d’un sol légèrement cohésif. La méthode LBM est implémentée ici avec des temps de relaxation multiples (MRT) et une condition de rebondissement interpolée pour les frontières solides en mouvement, afin d’améliorer la stabilité numérique des calculs. Le schéma du couplage entre les deux méthodes, ainsi que les critères pour le choix des paramètres numériques des deux méthodes. Pour étudier le phénomène visé, un échantillon représentatif de sol granulaire situé au front d’un conduit d’érosion est d’abord assemblé par une procédure de préparation “à sec”, puis testé dans des conditions saturées sous un chargement hydraulique monotone croissant. L’érosion régressive se produit par amas de grains au niveau du front d’érosion ayant subi au préalable une dégradation due à de nombreuses ruptures de ponts solides en traction. L’autre phénomène important observé est la présence d’arcs associés a` des chaines de forces en compression qui parviennent à maintenir parfois totalement, parfois partiellement l’intégrité du matériau non érodé. / The work reported in this thesis consists in a discrete modelling of the backward front propagation of an erosion pipe, as can take place in embankment dams or dikes. Some numerical tools have been developed to this end, based on the coupling between the Discrete Element Method (DEM) and the Lattice Boltzmann Method (LBM) for the representation of the solid and uid phases, respectively. The implementation of DEM follows a standard molecular dynamics approach and the interaction among grains are regulated by unilteral frictional visco-elastic and breakable visco-elastic bonds, in order to take into account a slightly cohesive soil behaviour. The LBM was implemented according to the Multiple Relaxation Time (MRT) scheme along with an interpolated non-slip conditions for moving boundaries, in order to improve the numerical stability of the calculations. The coupling scheme is described along with the criteria for the numerical parameters of the two methods. A representative specimen of a granular soil located at the front of an erosion pipe is first assembled by a \dry" preparation precedure and then tested under fully-saturated conditions and increasing hydraulic load over time. Backward erosion is takes place in the form of clusters of grain being eroded at the erosion front after a degradation of the material due to the breakage of tensile bonds. The other interesting feature that was observed is the creation of arches of compressive force chains. These arches enabled the specimen to maintain a stable or metastable configuration under the increasing hydraulic load.