Multi-Temporal Image Analysis for Fluvial Morphological Characterization with Application to Albanian Rivers

• Published in: ISPRS international journal of geo-information Vol. 7; no. 8; p. 314
• Main Authors: Spada, Daniele, Molinari, Paolo, Bertoldi, Walter, Vitti, Alfonso, Zolezzi, Guido
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01-08-2018
• Subjects: Albanian rivers; Anthropogenic factors; Data; Data processing; Digital Elevation Models; Digital imaging; Ecosystems; Evolution; Fluvial morphology; geomatics; Gravel; Hydroelectric power; Image acquisition; Image analysis; Image processing; Imagery; Information processing; Landsat; Landsat satellites; Missions; Monitoring systems; Morphology; Remote sensing; River ecology; river morphological trajectories; river morphology; River networks; Rivers; Satellite imagery; Satellites; Sediment transport; Sentinel-2; Spaceborne remote sensing; Studies; Temporal resolution; Vegetation; Italy; Albania
• ISSN: 2220-9964; 2220-9964
• DOI: 10.3390/ijgi7080314

A procedure for the characterization of the temporal evolution of river morphology is presented. Wet and active river channels are obtained from the processing of imagery datasets. Information about channel widths and active channel surface subdivision in water, vegetation and gravel coverage classes are evaluated along with channel centerline lengths and sinuosity indices. The analysis is carried out on a series of optical remotely-sensed imagery acquired by different satellite missions during the time period between 1968 and 2017. Data from the CORONA, LANDSAT and Sentinel-2 missions were considered. Besides satellite imagery, a digital elevation model and aerial ortho-photos were also used. The procedure was applied to three, highly dynamic, Albanian rivers: Shkumbin, Seman and Vjosë, showing a high potential for application in contexts with limitations in ground data availability. The results of the procedure were assessed against reference data produced by means of expert interpretation of a reference set of river reaches. The results differ from reference values by just a few percentage points (<6%). The time evolution of hydromorphological parameters is well characterized, and the results support the design of future studies aimed at the understanding of the relations between climatic and anthropogenic controls and the response of river morphological trajectories. Moreover, the high spatial and temporal resolution of the Sentinel-2 mission motivates the development of an automatic monitoring system based on a rolling application of the defined procedure.