Statistical analysis of thaw index for thaw weakening design purposes

In frost action research and design most focus has been put on frost heave. There are several models available to predict frost heave and the concept of statistical re-occurrence times of frost index is commonly used in design. Less effort has been put on thaw weakening modeling and design even though thaw weakening causes a lot of damage on infrastructure such as road and railways. In e.g. Sweden the Transport Administration collects temperature data during the freezing season for frost heave design purposes on project levels but thawing effects are conventionalized by reference tables on a regional level.The aim of this study is to define and quantify the thawing event in such way that it could be applied on a project level in a similar way as temperature measurements are used for frost design. In this study thaw index has been analysed statistically based on air temperature data from 50 stations, collected from year 1951 and onwards by the Swedish Meteorological and Hydrological Institute (SMHI), in Sweden. The temperature thaw seasons have been correlated to the Swedish Transport Administrations frost depth measurement stations.A reference table has been developed based on reoccurrence times of thawing seasons based on the magnitude of thawing index. Based on the statistical distribution of thawing indexes Sweden was divided into three regions in this study. It seems, by statistical means, to be fully possible to develop reference values for thawing periods where thawing index is the working unit. In order to implement the results in e.g. pavement design thawing temperature measurements on a local or semi-regional level are required.Godkänd; 2016; 20160530 (tomede