[Purposes] The randomness of the distribution of cavern locations and cavern development in karst areas, coupled with the inherent spatial variability of geotechnical parameters, renders it challenging to predict the impacts of tunnel construction in such environments. [Methods] The Xinhua Batang pipe jacking project was used as a case study to test the theory of Markov Random Fields. This involved predicting the location of the caverns based on physical exploration data with small data spacing and sparse borehole data. The resulting stratigraphic model was then validated. The model was constructed on the basis of the stratigraphic type model. The geotechnical parameter random fields were then generated based on the Karhunen-Loève transform and imported into the FLAC3D numerical model using the non-invasive finite difference method. The FLAC3D numerical model is employed to examine the spatial variability of geotechnical parameters and their influence on surface deformation in karst areas. This is achieved by setting several sets of spatial variability parameters for geotechnical parameters and analysing the impact of these parameters on surface deformation using the stochastic finite analysis method. [Findings] The results indicate: 1) The distribution of cavities can be effectively predicted by using borehole and physical exploration data combined with Markov random fields; 2) The traditional numerical model analysis of homogeneous stratigraphy and the stochastic analysis considering the spatial variability of the geotechnical parameters by using the random field theory can provide confidence intervals. In comparison with the results of the conventional numerical model of homogeneous stratigraphy; 3) The spatial variability of the geotechnical parameters mainly affects the size of confidence intervals of surface deformation curves, while the longitudinal settlement profile of the surface has no effect on the shape type of the surface deformation. [Conclusions] This study demonstrates that the study of construction impacts in karst areas by the theory of random field can inform the actual construction process, which is of some significance for the construction of similar projects.