Implemented, based on practical engineering projects, is a statistical analysis of strata soil characteristics and geological columns sampled from the main embankment of Yangtse Rive in southern Jingzhou zone of China. Furthermore, it was verified under Kolomogorov-Smirnov theory that this assumption, namely, the statistical hypothesis of Gaussian probability distribution for random permeability tensors of soils to be submitted into the stochastic seepage field simulation would be accepted by the statistical testing. By the help of numerical analysis for stochastic seepage fields simulations using three dimensional finite element method as well as the rationality of simulated stochastic results, investigated were various statistical characteristics and random natures in random seepage fields of the main embankment of Yangtse Rive in southern Jingzhou zone of China. Meanwhile, the developed model was also examined through statistical testing. Especially, as a stochastic boundary condition, the random variation of upstream and downstream water level was taken into account of three-dimensional finite element model in practical analysis for the random seepage field simulations, by which, it has been realized deeply to describe disturbed effects of the boundary randomness due to variability of downstream and upstream water level over variabilities of simulated results presented by vector series of the random seepage field. Furthermore，the function of some safeguards to set up against the seepage problem，such as sheet-pile breakwater and catch-water，as the turbulence of complex boundary condition together with the effects of variability of soil permeability has been analyzed for description of influence of them on the hydraulic potential distributed in the random seepage field and the random evolution of stochastic seepage field under these comprehensive aspects has also been compared with the result obtained by determinate analysis for the corresponding steady seepage field problem. Philosophy deduced in this paper presents the necessity，feasibility and practicability of employing stochastic studies in seepage field problems. Meanwhile, sensitivities of random seepage field hydraulic gradient and downstream exit height were studied upon the variation of permeability and boundary conditions of the main embankment of Yangtse Rive in southern Jingzhou zone of China, by which, broadened is the research domain on seepage field evolution under uncertain numerical coverage.