A new IT tool to identify and classify non-perennial rivers

Quantifying the intensity of flow intermittency and the spatial and temporal extension of dry riverbeds remains one of the most important gap in non-perennial rivers (NPRs) conservation and management. In this context, satellite images and remote sensing data can be used to identify and classify flow intermittency, detecting flow occurrence along river reaches. Satellite data have a short revisit time (about one week), and the recent availability of free of charge, high spatial resolution data (e.g., ESA Sentinel-2 mission) has already opened up the possibility of innovative applications for NPRs. Based on the results of the RIVERTEMP Erasmus+ project, we present a new web platform that analyses Sentinel-2 multispectral images. The false color composition of the bands SWIR, NIR and RED is automatically generated by the platform. False color images can be used to clearly distinguish water presence and identify the hydrological conditions of river reaches, that can be: “flowing” (F), “ponding” (P) or “dry” (D) conditions. The web tool generates time series of hydrological conditions, allowing the automatic classification of the NPRs based on the frequency and duration of F, P and D classes. A user manual and training materials for university students are also freely available on the website of the project.