A GIS‐based multicriteria decision analysis to reduce riparian vegetation hydrogeological risk and to quantify harvested biomass (Giant reed) for energetic retrieval. (November 2022)

• Record Type: Journal Article
• Authors: Sciuto, L.; Licciardello, F.; Barbera, A.C.; Cirelli, G.
• Is Part Of: Ecological indicators. Volume 144(2023)
• Publication Date: 2022-11
• Subjects: Arundo donax L. -- Riparian vegetation management -- Hydraulic risk -- Invasive species -- Remote sensing -- Biochemical methane potential; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2022.109548
• Languages: English
• ISSNs: 1470-160X

Abstract

Highlights: The uncontrolled riparian vegetation development causes flooding hydraulic risk. Remote sensing and GIS allow the monitoring of riparian ecosystem dynamics. Vegetation Red Edge composition is the most accurate for identifying Giant reed. Giant reed is a real opportunity to generate bioenergy. Abstract: The vegetation development in riverbeds creates obstructions to the regular water flow with the flooding hydraulic risk increase. Giant reed (GR) ( Arundo donax L.) is one of the most successful invasive species of riparian ecosystems in Mediterranean semi-arid climate conditions, with significant public administrations costs for its removal and disposal. At the meantime, due to its high biomass yield and adaption capacity to several conditions, GR is a very promising no-food crop to produce biogas by the anaerobic digestion (AD). The research activity provides for the involvement of territories belonging to the inner areas of Sicily, especially Calatino one, in which the SIMBIOSI Consortium operates an AD plant fed with agricultural by-products from other close agro-industrial companies. The aim of the study was to map and to quantify the actual spatial distribution of GR in watercourses embankments through Remote Sensing (RS) techniques applied in a Geographic Information System (GIS) environment. In this regard, a method based on the automatic supervised classification, was applied on three different combinations of spectral bands (True Color Image – TCI; Near-Infrared, Green and Blue – NGB; Vegetation Red Edge – VRE) of Sentinel-2 satellite images related to the summer season (11th August 2019), with the aim of identifying the most suitable classification to map the GR. The results showed that the VRE composition is the most accurate combination of spectral bands for the identification of GR, followed by the NGB image. The worst performance was obtained by using the TCI combination. A further elaboration was carried out combining the three classified images, in order to obtain a more accurate localization and quantification of GR. The final thematic map allowed to correctly classify GR for the 46 % of the cases, with an overall accuracy of 85.02 % and a high Kappa Coefficient of Agreement value equal to 0.81. Finally, the surface covered by GR in the study area (computed in the GIS environment) was about 2 km 2 and the estimated GR biomass, available for the biomethane production, obtainable from watercourses embankments maintenance interventions would amount to about 11, 780 tons year −1 . The study could contribute to the development of a watercourses maintenance plan with the aim to reduce the risk of streams flooding in valley areas and at intersections with infrastructure works. Furthermore, the proposed methodology can be used by stakeholders in marginal areas, for the watercourses management, offsetting its costs through the energy use of the collected biomass.