Abstract: [Objective] The widely distributed discontinuous loess stratum in the Guan zhong Plain area of China necessitates, due to its unique stratigraphic structure, the indoor calculated value of the self-weight wetting amount in evaluating the wetting property of the foundation has a large difference with the measured value in the field. [Methods] Therefore, taking the loess stratum on the north bank of Wei he River as the research object, indoor wetting test and on-site large-scale test pit immersion test were carried out to compare the influencing factors of the difference between on-site and indoor wetting amount. [Results] The following results were obtained from the study: (1) The ratio of self-weight wet depressions between the field test and the indoor test at the site was less than 0.1, and the reasons for this discrepancy include the discontinuity and inhomogeneity of loess layers, the sampling disturbance factor of the indoor test, and the difference in immersion conditions of the field test. (2) The layer bow effect formed by the discontinuity of loess is the main reason for the difference between the indoor test and the field test. It weakened part of the upward transmitted deformation, hindered the downward transmitted gravity stress, and caused the discontinuity of the percolation process. (3) Combining the results of the stratification calculations of the four test sites, it can be seen that most of the self-gravitational wetting in the field test occurred in the Q₃ soil layer. For the Q₂ loess, there is a large difference between the field measured values and the indoor test values. Therefore, the Q₂ loess layer is considered to have little or no wetting. [Conclusion] The shape of the saturated zone range obtained by numerical simulation is basically consistent with the field test results, and the numerical simulation method is more advantageous when observing the experimental results. When calculating the amount of wet subsidence by self-weight, the calculation method of stratification according to the age of strata can be adopted. In Q₃ stratum, the method of correction coefficient in the specification is selected, and the Q₂ loess stratum is determined by on-site test pit immersion test. [Significance] The research methodology of this paper can provide theoretical guidance for future engineering construction in the Guan zhong Plain.