Radar-Cover

Background

ESA is currently developing the Sentinel-1 satellite. The planned constellation of two C-band radar satellites will provide close to daily (depends on geographical latitude) acquisition of C-Band SAR data in the Interferometric Wide Swath Mode with 250 km swath width and 5x20 m spatial resolution. SENTINEL-1 is designed to meet the requirements of the European Union (EU) / European Space Agency (ESA) Global Monitoring for Environment and Security (GMES) programme for which three priorities, so-called fast-track services, have been identified. One of them is called 'land monitoring services'.

The availability of reliable land-cover information is crucial for a wide range of applications including monitoring of land-use change and land degradation as well as administrative matters. Currently, SAR data are not the primary data source for the classification of basic land-cover classes (forest, agriculture, settlement, etc.) if multi-channel optical imagery with high spatial resolution is available. The independence of cloud cover, however, makes SAR data an attractive data source in regions of frequent cloud cover, where consistent high-resolution multi-temporal optical remote sensing datasets can hardly be acquired.

A significant improvement of land-cover classification performance can be expected from SENTINEL-1, which will provide close to daily coverage and acquisitions in two polarizations. RADAR-Cover aims at the simulation of SENTINEL-1 Level-2 land-cover products. By means of multi-temporal datasets of ERS-1/2 and ENVISAT C-Band SAR data for two test areas located in North Rhine-Westphalia and Thuringia, Germany, the performance of the SENTINEL-1 mission will be simulated. The dual-polarization capabilities of SENTINEL-1 will be simulated using ENVISAT ASAR Alternating Polarization data and the frequent revisit will be simulated based on ERS-1 ICE-Phase data from 1994, when ERS-1 acquired every three days over the same area. The full processing chain, comprising preprocessing of the SAR data in different formats, speckle filtering and the classification algorithms, will be provided to ESA in form of a prototype processing software.