The paper presents a methodology for quantifying surface crack patterns that appear in cohesive soils under drying conditions due to environmental changes, using image analysis techniques. This has practical applications in the study of many geotechnical problems related to soil cracking such as the impact of permeability changes due to cracking in clay barriers, development of preferential flow paths for contaminant transport along cracks, decreasing bearing capacity, and others. The study of soil cracking may become even more relevant with the current climate change that may induce more frequent and severe droughts in many parts of the world, increasing the areas at risk of cracking. Qualitative and quantitative characterization of the crack patterns is needed to study the mechanical behavior of a cracking soil, how cracks generate and propagate. For this purpose a simple laboratory set-up has been developed for continuous monitoring of the processes of formation and propagation of cracks due to desiccation, and to study the final crack pattern. The paper describes a simple technique to process sequences of images obtained during the laboratory tests, and how image analysis can be used to quantify the parameters that characterize the evolving and final crack patterns.

This article presents the results of an experimental research carried to investigate the mechanics of cracking of soil layers under drying conditions. The tests were conducted under controlled laboratory conditions and in an environmental chamber with circular and rectangular specimens to investigate the effect of the boundary conditions (size, shape, and aspect ratio of the specimens and containers) on the process of initiation and propagation of cracks and on the final crack pattern at the end of desiccation. The tests in the environmental chamber were conducted with imposed temperature and relative humidity and provided new insight into the mechanics of the formation of cracks in a drying soil, and they showed that cracks can initiate either at the top, bottom, or at both surfaces of the drying specimen. The results also reveal how the crack patterns are controlled by the existing mechanical and hydraulic boundary conditions. The cracks seem to form sequentially in patterns that can be explained by three key factors: stresses higher than the tensile strength, the direction of the generated stresses, and the stress redistribution in the vicinity or inside the newly formed domain. In order to substantiate the sequential nature of the crack pattern formation, experimental evidences showing the existence of a cracking sequence during the laboratory desiccation experiments are presented and analyzed.

Abstract:Ground-based hyperspectral imaging is fairly new for studying near-vertical rock exposures where airborne or satellite-based imaging fail to provide useful information. In this study, ground-based hyperspectral image analysis was performed on a roadcut, where diagenetic tripolite facies is observed in southwestern Missouri. Laboratory-based reflectance spectroscopy and hyperspectral image analyses were also performed on collected samples. Image classification was performed using Spectral Feature Fitting (SFF) and Mixture-tuned Match Filtering (MTMF) with laboratory-and image-derived end-member spectra. SFF provided thorough yet detailed classification, whereas MTMF provided information on the relative abundances of the lithologies. Ground-based hyperspectral imaging demonstrated its potential to aid geological studies providing valuable information on mineralogical and lithological variations rapidly and with two-dimensional continuity in inaccessible rock faces of near-vertical outcrops. The results showed decreasing tripolite abundance going downward in the investigated vertical succession. Also, a leaching pattern has been observed such that persistent and continuous limestone layers become lenses and patches towards the upper portion of the outcrop. These observations show that the effect of tripolitization decreases when going deeper in the succession, suggesting that the fluid responsible for the weathering of siliceous precursors may have been flowing from top to bottom and thus have had a meteoric origin.

The inability to accurately resolve subseismic-scale structural discontinuities such as natural fractures represents a significant source of uncertainty for subsurface modeling practices. Fracture statistics collected from outcrop analogs are commonly used to fill the knowledge gap to reduce the uncertainty related to fracture-induced permeability anisotropy. The conventional methods of data collection from outcrops are tedious, time consuming, and often biased due to accessibility constraints. Recent advances in virtual outcrop-based methods in fracture characterization enhance conventional methods by streamlining data collection and analysis. However, certain limitations and challenges exist in virtually obtained fracture data sets. The ability to identify fractures that are both exposed as lineations and as planes from a digital outcrop model depends heavily upon the fidelity and resolution of its surface display of RGB color, reducing the capacity of light detection and ranging (lidar) to the resolution of the scanner-attached camera. In the present study, we adopted a hybrid approach, combining lidar-based digital outcrop models and georeferenced high-quality photomosaics, providing improved texture maps in terms of pixel density compared to maps generated from on-scanner camera images. With this approach, the effects of truncation on digital outcrop models were limited, giving the ability to detect fractures that would otherwise be aliased from on-scanner camera imagery. The fracture system developed within the exposures of the Mississippian Boone Formation, an outcrop analog for age-equivalent reservoir objectives in Mississippi Lime hydrocarbon play, was characterized using conventional and virtual outcrop-based techniques. To test the fidelity of the virtual fracture extraction approach, fracture orientation statistics generated from lidar are compared with equivalent data sets collected using traditional surveys. The results suggest that terrestrial lidar, coupled with referenced gigapixel photomosaics, provide an effective medium for fracture identification with the capacity of resolving fracture characteristics with sufficient fidelity to potentially act as conditioning data for discrete fracture network models, making it an attractive alternative tool for fracture modeling workflows.

No abstract