Recently, the research team led by Prof. Chaosheng Tang from the School of Earth Sciences and Engineering at Nanjing University has developed an innovative stochastic crack generation model for generating and predicting soil desiccation crack networks. By establishing a five-step simulation mechanism—comprising "crack initiation point selection, random path propagation, dynamic intersection judgment, secondary crack derivation, and crack network optimization"—the researchers achieved dynamic simulation of the full soil desiccation cracking lifecycle. A major innovation lies in its ability to predict future crack development and final network morphology based on partially formed cracks. This advancement offers strong theoretical and technical support for analyzing the engineering behavior of desiccated soils and mitigating drought-induced geohazards. The relevant research results were published in the journal Engineering Geology under the title "Generating and predicting soil desiccation cracking patterns utilizing a stochastic model based on geometric parameters."
Source: https://doi.org/10.1016/j.enggeo.2025.108122
Figure 1. Flowchart of the soil desiccation crack simulation program (DCSP)
Figure 2. Prediction of soil desiccation cracking network morphology based on DCSP
