Extracting Urban Morphology for Atmospheric Modeling From Multispectral and Sar Satellite Imagery

Abstract

Commission I, WG I/8KEY WORDS: Urban Morphology, Land Cover Classification, Digital Surface Model, Sentinel-2, TanDEM-X, Satellite Remote Sensing ABSTRACT:This paper presents an approach designed to derive an urban morphology map from satellite data while aiming to minimize the cost of data and user interference. The approach will help to provide updates to the current morphological databases around the world. The proposed urban morphology maps consist of two layers: 1) Digital Elevation Model (DEM) and 2) land cover map. Sentinel-2 data was used to create a land cover map, which was realized through image classification using optical range indices calculated from image data. For the purpose of atmospheric modeling, the most important classes are water and vegetation areas. The rest of the area includes bare soil and built-up areas among others, and they were merged into one class in the end. The classification result was validated with ground truth data collected both from field measurements and aerial imagery. The overall classification accuracy for the three classes is 91%. TanDEM-X data was processed into two DEMs with different grid sizes using interferometric SAR processing. The resulting DEM has a RMSE of 3.2 meters compared to a high resolution DEM, which was estimated through 20 control points in flat areas.Comparing the derived DEM with the ground truth DEM from airborne LIDAR data, it can be seen that the street canyons, that are of high importance for urban atmospheric modeling are not detectable in the TanDEM-X DEM. However, the derived DEM is suitable for a class of urban atmospheric models. Based on the numerical modeling needs for regional atmospheric pollutant dispersion studies, the generated files enable the extraction of relevant parametrizations, such as Urban Canopy Parameters (UCP).

A Cloud-Based Mapping Approach Using Deep Learning and Very-High Spatial Resolution Earth Observation Data to Facilitate the SDG 11.7.1 Indicator Computation

“…Experiment 2 for obtaining UOS, proved that in addition to VHR multispectral data, SAR data, even in a relatively coarse spatial resolution (i.e., Sentinel-1 at 10 m), through the vertical information offered in its polarized wave bands, is crucial in mapping open areas in the spectrally complex urban space. Although SAR data were used in studies focused on urban morphology [62,63], its use in urban open area extraction has not been previously tested. Notwithstanding the fact that higher accuracy metrics are observed in other studies, it is important to note that these results mostly refer to urban green spaces as opposed to UOS which were studied herein.…”