Assessing Spatial Variability of Piezocone Penetration Resistance of Layered Soft Clays Using Geostatistics

Haifeng Zou,Guojun Cai,Songyu Liu,Tejo Vikash Bheemasetti,Anand J. Puppala
DOI: https://doi.org/10.1061/9780784480717.029
2017-01-01
Abstract:Geostatistics is one of the most effective methods to assess the spatial variability and provide the spatial distribution of a soil property over a local region. In this research, geostatistics is utilized to evaluate the spatial distribution of the cone tip resistance (q(t)) of two soft clays at a highway site in Northern Jiangsu province, China. The impact of separating different soil units on the geostatistics predictions is investigated. To achieve this objective, linear trends were removed from the measurements using ordinary least square regression to procedure stationary random fields. Semivariogram models are then constructed to describe the spatial variability of qt residuals about the corresponding linear trends for the different soft clays. The ordinary Kriging is thereafter used to provide the best linear unbiased estimates of residuals at the unsampled locations and simultaneously to quantify the uncertainties associated with predictions. Finally, actual measurements are compared with the predictions to validate the geostatistics model. It is shown that if the different soil units are investigated separately in the geostatistical analysis, the accuracy of predictions at the unsampled locations can be improved.
What problem does this paper attempt to address?