Erosion and progradation in the Atrato River delta: A spatiotemporal analysis with Google Earth Engine

Abstract

The Atrato River Delta in Northwestern Colombia has experienced notable geomorphological changes in its shoreline in recent years. We analyze these changes, associated with erosion and progradation, using Landsat imagery and Google Earth Engine (GEE) algorithms to automatically identify the changes in an annual basis over 33 years (1986–2019). We compare the results with manual delineations on the same imagery using ArcGIS, obtaining similar outcomes, although GEE is much more efficient in processing large amounts of imagery compared with handmade procedures. We identify with good accuracy trends in erosion and progradation areas along the mouths and sides of the delta. Our algorithm performs well at delineating the shorelines, although special care must be taken to clean the images from clouds and shadows that may alter the definition of the shoreline. Results show that the Atrato delta has lost around 10 km2 due to erosion and has gained around 18 km2 in progradation during the period of assessment. Overall, progradation is the dominant process at the delta’s mouths, while erosion is dominant only in areas far from the mouths, which agrees with a river-dominated environment of high sediment loads and is coherent with other studies made in the region. The algorithm in GEE is a versatile tool, appropriate to assess short and long-term changes of coastal areas that do not count with land-based information.

Keywords

• coastal zones
• computer application
• remote sensing
• algorithms
• geodynamics