Search

Land Use/Land Cover (LULC) maps can be effectively produced by cost-effective and frequent satellite observations. Powerful cloud computing platforms are emerging as a growing trend in the high utilization of freely accessible remotely sensed data for LULC mapping over large-scale regions using big geodata. This study proposes a workflow to generate a 10 m LULC map of Europe with nine classes, ELULC-10, using European Sentinel-1/-2 and Landsat-8 images, as well as the LUCAS reference samples. More than 200 K and 300 K of in situ surveys and images, respectively, were employed as inputs in the Google Earth Engine (GEE) cloud computing platform to perform classification by an object-based segmentation algorithm and an Artificial Neural Network (ANN). A novel ANN-based data preparation was also presented to remove noisy reference samples from the LUCAS dataset. Additionally, the map was improved using several rule-based post-processing steps. The overall accuracy and kappa coefficient of 2021 ELULC-10 were 95.38% and 0.94, respectively. A detailed report of the classification accuracies was also provided, demonstrating an accurate classification of different classes, such as Woodland and Cropland. Furthermore, rule-based post processing improved LULC class identifications when compared with current studies. The workflow could also supply seasonal, yearly, and change maps considering the proposed integration of complex machine learning algorithms and large satellite and survey data. [ABSTRACT FROM AUTHOR]

This study analyzes inter- and intra-zonal seasonal variability of surface urban heat islands (SUHIs) within the methodological framework of local climate zones (LCZs) in three central European cities (Prague, Brno and Novi Sad). These cities differ in urban area and structure as well as in topography and hinterland land-cover features. LCZs were delineated on the basis of a GIS-based classification method. Land surface temperature (LST) was derived from LANDSAT-8 scenes in the period 2013–2018. The first step was to detect seasonal SUHI intensity differences for built LCZ types and LST for land-cover types of LCZ. The results revealed the highest differences in summer and spring, and lowest in winter. The highest SUHI intensity values occur in densely built-up and industrial zones, and the lowest in sparsely-built city outskirts. The coolest LCZs based on LST were dense trees and water areas in spring and summer. The second step aimed to analyze the effects of vegetation on SUHI formation. Hence, 11 land cover subclasses (from dense trees to bush/scrub) were defined in order to research intra-zonal seasonal LST variability. The height and density of vegetation have substantial effects on intra-zonal variability of LST in land-cover types of LCZ, whereas differences between forest subclasses were relatively low. Finally, the character of the vegetation had a substantial influence on intra-zonal LCZ variability of LST and SUHI formation. Further research in this field could contribute to better understanding of micro- and mezzo-climate-scale patterns, as well as better adaptation to climate change in urban areas. Image 1 • Statistically significant seasonal SUHI variability between LCZs. • Highest SUHI differences between built and land cover classes occurred in summer. • Regional climate and city surrounding affect SUHI during winter, spring and autumn. • Statistically significant SUHI variability within LCZ land cover classes (A−C). • Density and tree/scrub pattern, together with land cover, affect seasonal SUHI. [ABSTRACT FROM AUTHOR]