Comparing Methods of Estimating Crack Volume in Shrink-Swell Soils
MetadataShow full item record
Predicting soil crack formation and closure in high shrink-swell soils is crucial for modeling water movement and solute transport. However, soil cracking, as it occurs in natural conditions, is not well understood. The objectives of this research are to 1) compare shrinking and swelling of two Vertisols with different mineralogy and 2) compare two methods for estimating soil crack volume. The first soil monitored, Ships Clay (Very-fine, mixed, thermic Chromic Hapluderts) has mixed mineralogy and is located in Snook, TX. The second soil monitored is a Burleson Clay (Fine, smectitic, thermic Udic Haplusterts), which has smectitic mineralogy and is located 1 mile southwest of the Ships site. During two drying events, vertical subsidence and cracking were measured on both soils. Vertical subsidence was measured at three locations in each soil with rods fixed at four depths. Rod movement was converted to soil crack volume by assuming equi-dimensional shrinkage. A second method for estimating crack volume used direct measurements of cracks in the field. This second method was time consuming and the accuracy is unknown. A neutron moisture meter access tube was installed to measure soil moisture at each subsidence location. A total of 20 levling and moisture measurements were completed and 10 hand measurements of cracking were made. At the completion of the study, full characterization of each measurement location was performed, including COLE measurements. These COLE measurements were used as a third method of estimating crack volume in relation to soil moisture. Total crack volume will be compared at both sites to show the differences in the drying and cracking patterns. Crack volume as a function of moisture will be used to show how soil cracking and moisture relate. The two methods of estimating crack volume will be compared to see where possible cracking was measured with subsidence but not seen visually. The results showed the leveling-predicted crack volume was ten times that of the hand measured crack volume but followed the same temporal trend. The leveling-predicted crack volume is the most accurate of the three methods. COLE was also shown to be a good predictor a soils shrink-swell potential.
height change measurments
Coefficient of Linear Extensibility
leveling-predicted crack volume
Rivera, Leonardo D. (2011). Comparing Methods of Estimating Crack Volume in Shrink-Swell Soils. Available electronically from