Using in situ soil measurements for analysis of a polarimetric SAR-based classification of levee slump slides in the Lower Mississippi River

There is a critical need to improve levee monitoring techniques to enable the implementation of a cost-effective multi-scale monitoring system. In this study, polarimetric synthetic aperture radar (PolSAR) imagery was used as input in an automated classification system for characterizing areas in the levee having anomalies, with an emphasis on slump slides. In addition to the remotely sensed data, a set of in-situ soil data was collected to provide detailed soil properties over the study area. In-situ soil data were collected via a series of sensors and soil corings to obtain information regarding soil texture, moisture content, hydraulic conductivity and penetration resistance. While the automated classification system correctly identified the location of active slides, there were some points, referred to as “false positive points”, which were classified as anomalous but no failure was found at those locations. The goal of this study was to compare in-situ soil properties of the false positive points with those of the true positive points (i.e., the slide area) and true negatives (i.e., the healthy areas) by two methods, one pixel-based and one object-based, to determine how similar they are to either the healthy or slide areas. The results from the object-based method showed a strong relationship between the false positive points and the slide area in the top 30–45 cm of soil. The observed relationship and the later discovery of cracks in one of the dense false positive zones indicate the importance of this study.