Study on the Soil Seismic Liquefaction Probability Distribution Model for Environmental Factors

Soil liquefaction caused by earthquakes poses a serious threat to the safety of engineering structures. Assessing liquefaction likelihood is the primary task in liquefaction research. The establishment of a liquefaction assessment method is related to the groundwater table, buried depth, and stress conditions. However, most existing methods regard influencing factors as fixed values, ignoring uncertainties caused by measurement errors and other factors. Probability distributions for various soil environmental factors were constructed based on field liquefaction investigation cases, and the optimal distributions were determined. Two methods for calculating liquefaction probability were proposed based on the distributions, and their effectiveness was validated through an actual case. The results reveal that the groundwater table, buried depth, total stress, and effective stress follow a lognormal distribution, a chi-square distribution, a Gaussian distribution, and a gamma distribution, respectively. High groundwater tables, shallow burial depth, low total stress, and low effective stress are more likely to induce liquefaction. The two methods proposed in this paper for calculating liquefaction probability based on influencing factor distributions were verified to be effective, with results comparable to existing probability models through real-world case validation. The methods proposed in this paper not only determine whether soil will liquefy but also calculate liquefaction probability. Additionally, the impact of regional differences in actual samples can be considered by incorporating a normal distribution error term. The results of this study could provide guidance for soil seismic liquefaction probability assessment.