Multi-frequency SAR data help improving the monitoring of intertidal flats on the German North Sea coast Martin Gade a, * , Sabrina Melchionna a , Kerstin Stelzer b , Jörn Kohlus c a Universität Hamburg, Institut für Meereskunde, Bundesstr. 53, 20146 Hamburg, Germany b Brockmann Consult, Max-Planck-Str. 2, 21502 Geesthacht, Germany c LKN, Nationalparkverwaltung, Schlossgarten 1, 25832 Tönning, Germany article info Article history: Received 31 May 2013 Accepted 21 January 2014 Available online 31 January 2014 Keywords: satellite remote sensing radar imagery tidal flats sediment distribution oyster beds Integral Equation Model abstract We demonstrate that Synthetic Apertur Radar (SAR) data have great potential to improve an existing monitoring system based on optical data for intertidal flats and to complement the classification of sediments, macrophytes, and mussels in the German Wadden Sea. Multi-satellite SAR data acquired at different radar bands (L, C, and X band, from ALOS PALSAR, from ERS SAR, Radarsat-2 and ENVISAT ASAR, and from TerraSAR-X, respectively) were used to investigate whether they can be jointly used for crude sediment classification on dry-fallen intertidal flats and for detecting benthic fauna such as blue mussel or oyster beds. In this respect, we show that both multi-satellite and multi-temporal analyses provide valuable input for the routine monitoring of exposed intertidal flats on the German North Sea coast, the latter already improving the identification of the spatial extent of mussel (oyster) beds. In addition, we demonstrate that high-resolution SAR is capable of detecting residuals of historical land use in areas that were lost to the sea during major storm surges in the 14th and 17th centuries. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction Worldwide, the largest intertidal flats can be found on the German, Danish and Dutch North Sea coast (CWSS, 2008) and on the western coast of South Korea (Kellermann and Koh, 1999), in a distance of up to 10 Km offshore. Those areas fall dry once during each tidal cycle and consist of fine sediments such as (fine) sand and mud, and they are only partly vegetated. Not only because they are impacted by the stress of the global sea level rise and the ex- pected increasing frequency of storm events, a frequent surveil- lance is of great importance, though this is a difficult task by boat, foot, or land vehicles. This is when remote sensing techniques come into play. Optical sensors are already being used for sediment and habitat classification on intertidal flats (Dennert-Möller, 1982; Kohlus, 1998; Stelzer et al., 2007), and promising results have been ach- ieved through the classification of different sediment types, vege- tation, and mussel beds (Brockmann and Stelzer, 2008). Existing classification systems for the different surface types of intertidal flats are usually based upon optical remote sensing data, since the hyperspectral data allow for a classification of surface types (Brockmann and Stelzer, 2008). However, because of the strong dependence on daylight and cloud conditions, and because of the short time window around low tide (approx. 3 h), useful optical data acquired at low tide, daytime, and sunny weather conditions from the German North Sea coast are rare. A classification system based on spaceborne remote sensing data would therefore strongly benefit from the additional utilization of synthetic aperture radar (SAR) data. Gade et al. (2008) suggested using multi-frequency SAR data for a sediment classifi- cation on exposed intertidal flats. They demonstrated that pairs of simultaneously acquired L-, C- and X-band SAR images from the Spaceborne Shuttle Imaging C/X-Band SAR (SIR-C/X-SAR) cam- paigns in 1994 can be used for a crude sediment classification based on the inversion of the Integral Equation Model, IEM (Fung et al., 1992; Fung and Chen, 2004). However, whereas SIR-C/X-SAR was providing multi-frequency SAR imagery acquired simultaneously, current spaceborne SAR sensors operate at single frequencies, and as a consequence, SAR data from different satellites have to be used for multi-frequency SAR classification purposes. Because they are usually acquired with a considerable time lag in between, a pro- found knowledge of the radar backscatter properties of the sedi- ment types, and their dependence on weather conditions, tidal cycle, and imaging geometry is needed, which can only be gained from a joint analysis of multi-satellite SAR data and optical remote sensing data, together with a-priori knowledge gained during in- situ campaigns. The sub-project 4 of the German national joint * Corresponding author. E-mail address: martin.gade@uni-hamburg.de (M. Gade). Contents lists available at ScienceDirect Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss 0272-7714/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ecss.2014.01.007 Estuarine, Coastal and Shelf Science 140 (2014) 32e42