River morphology monitoring using multitemporal sar data: Preliminary results

Abstract

In this paper, we test the capability of satellite synthetic aperture radar (SAR) images to enhance the monitoring of river geomorphological processes. The proposed approach exploits the recently introduced Level-1a products. These products are bi-temporal RGB composites in which the association color-object, being physical-based, is stable whatever the scene is considered. This favors the detection of temporary rivers’ characteristics for classification purposes in a change-detection environment. The case study was implemented on the Orco river (northwest Italy), where a set of 39 COSMO-SkyMed SAR stripmap images acquired from October 2008 to November 2014 was used to monitor channel planform changes. This preliminary study is devoted to assess the suitability of Level-1a images for geomorphologist, with particular reference to the detection of phenomena of interest in river monitoring. This is prior for semi-automatic or automatic classification activities.