Author(s): Michael Nones, Massimo Guerrero

Linked Author(s): Michael Nones, Massimo Guerrero

Keywords: Google Earth Engine; NDVI; Po River; River dynamics; Vegetation growth;

Abstract: The cloud-based platform Google Earth Engine was used to evaluate the changes of the Normalized Difference Vegetation Index (NDVI) on a lowland reach of the Po River, Italy. Using the 30m resolution Landsat8 images, the study covered the April-October growing season in the 2013-2017 period and considered three different zones: a curve, a central bar and an island. Observing the variations of the NDVI across the areas it is possible to retrieve qualitative information about the vegetation encroachment and the correlated changes of the local river morphology. In the case of the curve, the inner bank was subjected to deposition and the vegetation improved this trend, stabilizing the bar and moving the current towards the outer bank, more stable. As for the central bar, the plants’ encroachment contributed to increasing the dimension of the bar, providing additional room for future plants and changing the bar morphology towards a stable island. In the third case, the island was already formed, and the NDVI remained almost constant during the observed period, indicating that plants have already stably colonized this area. The NDVI was correlated to the water discharges measured upstream of the study sites, pointing out a slight dependence of the riparian vegetation growth from the water flow and on the seasons. However, such a correlation is not yet well explained because of the short temporal horizon here evaluated.
This paper represents a first attempt to use the free service Google Earth Engine to characterize the medium-term behaviour of riparian vegetation establishment on different fluvial landscapes, improving our knowledge about the reciprocal feedback between hydrology, fluvial morphology and plant dynamics at the reach scale.

DOI: https://doi.org/10.3850/38WC092019-1581

Year: 2019