Search

The damage and the eventual breach of river embankments may be due to internal erosion and instability processes in the waterside and landside slopes. Beside the progressive degradation of soil properties, different types of macro-pores inside the levee body can influence its short- and long-term performance. Among macro-pores, burrows are a widespread form of biologic erosion of earthen structures potentially producing damage or even their collapse. In fact, animal burrows are erosion tunnels, which can lead to piping phenomena. Thus, the emergent risk connected to burrowing animals in earthen levees has to be tackled, given also the environmental changes driven by human activity and climate. Remote sensing and geophysical surveys can complement data from in situ investigation campaigns in the definition of the real-embankment model, as well as in the imaging of local defects which may influence its local and/or global stability. Several techniques were integrated in the study area of the Panaro River, where an active animal burrow was detected in spring 2015. Two campaigns were carried out: Survey-1, in June 2015, just after the relocation of the animals, and Survey-2 in December 2015, after the filling of the burrow with a cement-bentonite slurry. Here, we highlight the peculiarity of each method and the choice of an integrated multi-technique approach. The results allowed the known burrow, as well as two other tunnels, to be imaged in 3D, providing specific guidelines for the best integrated strategy to detect and characterize these macro-pores in a fluvial levee. The proposed approach can advance our knowledge of embankments in space and time, so that effective remedial actions in flood risk and wildlife management can be identified.

This paper deals with the development of a new Maximum Likelihood (ML) estimator for semi-variogram parameters in ordinary Kriging, based upon the assumption of a multi-normal distribution of the Kriging cross-validation errors. The paper discusses the difference between the proposed ML formulation and previously developed algorithms, showing its advantages, also in view of an approximate analysis of the uncertainty that the semi-variogram parameter estimates may induced on the Kriging estimates.