A Scalable Earth Observation Service to Map Land Cover in Geomorphological Complex Areas beyond the Dynamic World: An Application in Aosta Valley (NW Italy)

Abstract

Earth Observation services guarantee continuous land cover mapping and are becoming of great interest worldwide. The Google Earth Engine Dynamic World represents a planetary example. This work aims to develop a land cover mapping service in geomorphological complex areas in the Aosta Valley in NW Italy, according to the newest European EAGLE legend starting in the year 2020. Sentinel-2 data were processed in the Google Earth Engine, particularly the summer yearly median composite for each band and their standard deviation with multispectral indexes, which were used to perform a k-nearest neighbor classification. To better map some classes, a minimum distance classification involving NDVI and NDRE yearly filtered and regularized stacks were computed to map the agronomical classes. Furthermore, SAR Sentinel-1 SLC data were processed in the SNAP to map urban and water surfaces to improve optical classification. Additionally, deep learning and GIS updated datasets involving urban components were adopted beginning with an aerial orthophoto. GNSS ground truth data were used to define the training and the validation sets. In order to test the effectiveness of the implemented service and its methodology, the overall accuracy was compared to other approaches. A mixed hierarchical approach represented the best solution to effectively map geomorphological complex areas to overcome the remote sensing limitations. In conclusion, this service may help in the implementation of European and local policies concerning land cover surveys both at high spatial and temporal resolutions, empowering the technological transfer in alpine realities.

Earth Observation Data and Geospatial Deep Learning AI to Assign Contributions to European Municipalities Sen4MUN: An Empirical Application in Aosta Valley (NW Italy)

“…The reference year 2020 was considered. Land cover (LC) complies with EAGLE guidelines and is realized according to the following methods [30] and it is based on Sentinels missions (S1-S2) [31][32][33][34] adopting Aosta Valley land cover [25,30]. It is worth to note that, to assist the semantic and technological foundation of a European harmonized information management capability for land monitoring, the EAGLE Group has been working on a solution and proof of concept since 2008.…”

Snow Metrics as Proxy to Assess Sarcoptic Mange in Wild Boar: Preliminary Results in Aosta Valley (Italy)

“…Geospatial analyzes were carried out by adopting Earth Observation data from the USGS NASA's mission Landsat 8 OLI. In particular, the input files were Landsat-8 level-2A products downloaded from the CNES Theia Land Data Centre (https://theia.cnes.fr/ atdistrib/rocket/#/home, last accessed on 30 January 2023) and the Digital Surface Model (DSM) retrieved from the Aosta Valley SCT Geoportal (https://geoportale.regione.vda.it/, last accessed on 30 January 2023) resampled at 30 m (staring from a native resolution product of 2 m) [44][45][46][47]. Landsat 8 data ranging from the hunting seasons 2013-2014 and 2014-2015 respectively, were temporally aggregated and stacked in order to compute Snow Metrics (here in after called SM) thanks to the Orfeo Toolbox (OTB) remote module developed by CESBIO and CNES called Let-It-Snow (LIS).…”

“…It is worth noting that, FSC were computed per each hunting season stack of Theia product Landsat 8 OLI Level-2A. To do this in Let-It-Snow algorithm the following optional parameters were included: the tree cover density (TCD) retrieved from Land Copernicus platform (https://land.copernicus.eu/, last accessed on 30 January 2023), the water mask retrieved from Aosta Valley Land cover [47], and the relief shadow mask obtained from the Aosta Valley DSM. The output generated was two FSC stacks (Level-2B snow product).…”

Evaluation of Multiple Classifier Systems for Mapping Different Hierarchical Levels of Forest Ecosystems in the Mediterranean Region Using Sentinel-2, Sentinel-1, and ICESat-2 Data

“…This is a non-parametric classification technique where a class is determined based on the most frequent class among its 'k' nearest neighbors, with 'k' being a positive, usually small, integer value. If k equals 1, the pixel or object is classified according to the class of its nearest neighbor [50]. In the current study, KNN was implemented using the Mahalanobis distance, and k was set to 1 following the work of Qian et al [51] and Thanh Noi and Kappas [52].…”