Abstract: The dispersivity results of certain soil samples are often inconsistent when tested using the five common methods of pinhole, double hydrometer, crumb, pore water soluble cation, and exchangeable sodium ion; a variety of other comprehensive methods for discriminating soil dispersivity have been developed in previous studies. This paper presents an analysis of the advantages and disadvantages of these five methods based on 45 sets of soil samples collected from 8 water conservancy projects, focusing on comparison of the qualitative discrimination results of the original authors with our comprehensive quantitative results obtained using weight analysis and a rating system. The results show that the weight analysis and the rating system are reliable and the rating system leads to comparatively low safety. We also verify the accuracy of an empirical formula. This formula features simple parameters and clear physical meanings and its results have a tendency toward higher reliability and security. Therefore, we suggest that it could be adopted for preliminary discrimination and then more accurate dispersive soil characteristics are examined through weight analysis or a rating system.

Key words: dispersive soil, test method, comprehensive discrimination, matching degree, safety