Search

The present study aims to discriminate four semi-arid grassland habitats in a Mediterranean Natura 2000 site, Southern Italy, involving 6210/E1.263, 62A0/E1.55, 6220/E1.434 and X/E1.61-E1.C2-E1.C4 (according to Annex I of the European Habitat Directive/EUropean Nature Information System (EUNIS) taxonomies). For this purpose, an intra-annual time-series of 30 Sentinel-2 images, embedding phenology information, were investigated for 2018. The methodology adopted was based on a two-stage workflow employing a Support Vector Machine classifier. In the first stage only four Sentinel-2 multi-season images were analyzed, to provide an updated land cover map from where the grassland layer was extracted. The layer obtained was then used for masking the input features to the second stage. The latter stage discriminated the four grassland habitats by analyzing several input features configurations. These included multiple spectral indices selected from the time-series and the Digital Terrain Model. The results obtained from the different input configurations selected were compared to evaluate if the phenology information from time-series could improve grassland habitats discrimination. The highest F1 values (95.25% and 80.27%) were achieved for 6210/E1.263 and 6220/E1.434, respectively, whereas the results remained stable (97,33%) for 62A0/E1.55 and quite low (75,97%) for X/E1.61-E1.C2-E1.C4. However, since for all the four habitats analyzed no single configuration resulted effective, a Majority Vote algorithm was applied to achieve a reduction in classification uncertainty. [ABSTRACT FROM AUTHOR]

This study focused on implementation of the Sustainable Development Goal (SDG) 11 indicators, at local scale, useful in monitoring urban social resilience. For this purpose, the study focused on updating the distribution map of the migrant population regularly residing in Bari and a neighboring town in Southern Italy. The area is exposed to increasing migration fluxes. The method implemented was based on the integration of Sentinel-2 imagery and updated census information dated 1 January 2019. The study explored a vector-based variant of the dasymetric mapping approach previously used by the Joint Research Center (JRC) within the Data for Integration initiative (D4I). The dasymetric variant implemented can disaggregate data from census areas into a uniform spatial grid by preserving the information complexity of each output grid cell and ensure lower computational costs. The spatial distribution map of regular migrant population obtained, along with other updated ancillary data, were used to quantify, at local level, SDG 11 indicators. In particular, the map of regular migrant population living in inadequate housing (SDG 11.1.1) and the ratio of land consumption rate to regular migrant population growth rate (SDG 11.3.1) were implemented as specific categories of SDG 11 in 2018. At the local level, the regular migrant population density map and the SDG 11 indicator values were provided for each 100 × 100 m cell of an output grid. Obtained for 2018, the spatial distribution map revealed in Bari a high increase of regular migrant population in the same two zones of the city already evidenced in 2011. These zones are located in central parts of the city characterized by urban decay and abandoned buildings. In all remaining city zones, only a slight generalized increase was evidenced. Thus, these findings stress the need for adequate policies to reduce the ongoing process of residential urban segregation. The total of disaggregated values of migrant population evidenced an increase of 44.5% in regular migrant population. The indicators obtained could support urban planners and decision makers not only in the increasing migration pressure management, but also in the local level monitoring of Agenda 2030 progress related to SDG 11. [ABSTRACT FROM AUTHOR]

The present study focuses on the use of Convolutional Neural Networks (CNN or ConvNet) to classify a multi-seasonal dataset of Sentinel-2 images to discriminate four grassland habitats in the "Murgia Alta" protected site. To this end, we compared two approaches differing only by the first layer machinery, which, in one case, is instantiated as a fully-connected layer and, in the other case, results in a ConvNet equipped with kernels covering the whole input (wide-kernel ConvNet). A patchwise approach, tessellating training reference data in square patches, was adopted. Besides assessing the effectiveness of ConvNets with patched multispectral data, we analyzed how the information needed for classification spreads to patterns over convex sets of pixels. Our results show that: (a) with an F1-score of around 97% (5 × 5 patch size), ConvNets provides an excellent tool for patch-based pattern recognition with multispectral input data without requiring special feature extraction; (b) the information spreads over the limit of a single pixel: the performance of the network increases until 5 × 5 patch sizes are used and then ConvNet performance starts decreasing. [ABSTRACT FROM AUTHOR]

Snow cover plays an important role in biotic and abiotic environmental processes, as well as human activities, on both regional and global scales. Due to the difficulty of in situ data collection in vast and inaccessible areas, the use of optical satellite imagery represents a useful support for snow cover mapping. At present, several operational snow cover algorithms and products are available. Even though most of them offer an up-to-daily time scale, they do not provide sufficient spatial resolution for studies requiring high spatial detail. By contrast, the Let-It-Snow (LIS) algorithm can produce high-resolution snow cover maps, based on the use of both the normalized-difference snow index (NDSI) and a digital elevation model. The latter is introduced to define a threshold value on the altitude, below which the presence of snow is excluded. In this study, we revised the LIS algorithm by introducing a new parameter, based on a threshold in the shortwave infrared (SWIR) band, and by modifying the overall algorithm workflow, such that the cloud mask selection can be used as an input. The revised algorithm has been applied to a case study in Gran Paradiso National Park. Unlike previous studies, we also compared the performance of both the original and the modified algorithms in the presence of cloud cover, in order to evaluate their effectiveness in discriminating between snow and clouds. Ground data collected by meteorological stations equipped with both snow gauges and solarimeters were used for validation purposes. The changes introduced in the revised algorithm can improve upon the overall classification accuracy obtained by the original LIS algorithm (i.e., up to 89.17 from 80.88%). The producer's and user's accuracy values obtained by the modified algorithm (89.12 and 95.03%, respectively) were larger than those obtained by the original algorithm (76.68 and 93.67%, respectively), thus providing a more accurate snow cover map. [ABSTRACT FROM AUTHOR]