Determining a representative elementary area for soil desiccation cracking

Laboratory tests involving soil desiccation cracking are subjected to geometrical boundary constraints that are not always present in field conditions. To better understand the effects of sample geometry on desiccation cracks, several researchers have used controlled climatic conditions coupled with image analysis to accurately quantify the crack characteristics of soil samples subjected to laboratory studies. However, to date, no known studies in the literature present a simple method for determining the appropriate sample size for laboratory desiccation cracking testing that ensures the effects of boundary conditions are minimized. The primary objective of this research is to address this gap by proposing an experimental approach to determine a representative elementary area (REA) for soil desiccation cracking studies in the laboratory. To achieve this, we conducted four series of tests using an environmental chamber. One preliminary test series was carried out to identify the optimal sample preparation and REA testing techniques. Subsequently, three additional series of REA tests were conducted with sample thicknesses of 2.5, 5.0, and 7.5 mm to explore the impact of sample size on the REA. The REA for each sample thickness was determined by examining the surface area at which the surface crack ratio (RSC) for the sample was unaffected by boundary constraints. The results indicate that as sample thickness increases, REA increases. Further research is needed to determine how the REA is affected by sample shape and soil type.