Patrick Matgen

A Change Detection Approach to Flood Mapping in Urban Areas Using TerraSAR-X

Very high resolution synthetic aperture radar (SAR) sensors represent an alternative to aerial photography for delineating floods in built-up environments where flood risk is highest. However, even with currently available SAR image resolutions of 3 m and higher, signal returns from man-made structures hamper the accurate mapping of flooded areas. Enhanced image processing algorithms and a better exploitation of image archives are required to facilitate the use of microwave remote-sensing data for monitoring flood dynamics in urban areas. In this paper, a hybrid methodology combining backscatter thresholding, region growing, and change detection (CD) is introduced as an approach enabling the automated, objective, and reliable flood extent extraction from very high resolution urban SAR images. The method is based on the calibration of a statistical distribution of “open water” backscatter values from images of floods. Images acquired during dry conditions enable the identification of areas that are not “visible” to the sensor (i.e., regions affected by “shadow”) and that systematically behave as specular reflectors (e.g., smooth tarmac, permanent water bodies). CD with respect to a reference image thereby reduces overdetection of inundated areas. A case study of the July 2007 Severn River flood (UK) observed by airborne photography and the very high resolution SAR sensor on board TerraSAR-X highlights advantages and limitations of the method. Even though the proposed fully automated SAR-based flood-mapping technique overcomes some limitations of previous methods, further technological and methodological improvements are necessary for SAR-based flood detection in urban areas to match the mapping capability of high-quality aerial photography.

Water Level Estimation and Reduction of Hydraulic Model Calibration Uncertainties Using Satellite SAR Images of Floods

Exploitation of river inundation satellite images, particularly for operational applications, is mostly restricted to flood extent mapping. However, there lies significant potential for improvement in a 3-D characterization of floods (i.e., flood depth maps) and an integration of the remote-sensing-derived (RSD) characteristics in hydraulic models. This paper aims at developing synthetic aperture radar (SAR) image analysis methods that go beyond flood extent mapping to assess the potential of these images in the spatiotemporal characterization of flood events. To meet this aim, two research issues were addressed. The first issue relates to water level estimation. The proposed method, which is an adaptation to SAR images of the method developed for water level estimation using flood aerial photographs, is composed of three steps: (1) extraction of flood extent limits that are relevant for water level estimation; (2) water level estimation by merging relevant limits with a Digital Elevation Model; and (3) constraining of the water level estimates using hydraulic coherence concepts. Applied to an ENVISAT image of an Alzette River flood (2003, Grand Duchy of Luxembourg), this provides plusmn54-cm average vertical uncertainty water levels that were validated using a sample of ground surveyed high water marks. The second issue aims at better constraining hydraulic models using these RSD water levels. To meet this aim, a traditional calibration using recorded hydrographs is completed via comparison between simulated and RSD water levels. This integration of the RSD characteristics proves to better constrain the model (i.e., the number of parameter sets providing acceptable results with respect to observations has been reduced). Furthermore, simulations of a flood event of a different return period (2007) using the model calibrated for the 2003 flood event shows the reliability of the latter for flood forecasting.

Inferring catchment precipitation by doing hydrology backward: A test in 24 small and mesoscale catchments in Luxembourg

The complexity of hydrological systems and the necessary simplification of models describing these systems remain major challenges in hydrological modeling. Kirchners (2009) approach of inferring rainfall and evaporation from discharge fluctuations by “doing hydrology backward” is based on the assumption that catchment behavior can be conceptualized with a single storage-discharge relationship. Here we test Kirchners approach using a densely instrumented hydrologic measurement network spanning 24 geologically diverse subbasins of the Alzette catchment in Luxembourg. We show that effective rainfall rates inferred from discharge fluctuations generally correlate well with catchment-averaged precipitation radar estimates in catchments ranging from less than 10 to more than 1000 km2 in size. The correlation between predicted and observed effective precipitation was 0.8 or better in 23 of our 24 catchments, and prediction skill did not vary systematically with catchment size or with the complexity of the underlying geology. Model performance improves systematically at higher soil moisture levels, indicating that our study catchments behave more like simple dynamical systems with unambiguous storage-discharge relationships during wet conditions. The overall mean correlation coefficient for all subbasins for the entire data set increases from 0.80 to 0.95, and the mean bias for all basins decreases from –0.61 to –0.35 mm d?1. We propose an extension of Kirchners approach that uses in situ soil moisture measurements to distinguish wet and dry catchment conditions.

Unlocking the full potential of Earth observation during the 2015 Texas flood disaster

Intense rainfall during late April and early May 2015 in Texas and Oklahoma led to widespread and sustained flooding in several river basins. Texas state agencies relevant to emergency response were activated when severe weather then ensued for 6 weeks from 8 May until 19 June following Tropical Storm Bill. An international team of scientists and flood response experts assembled and collaborated with decision-making authorities for user-driven high-resolution satellite acquisitions over the most critical areas; while experimental automated flood mapping techniques provided daily ongoing monitoring. This allowed mapping of flood inundation from an unprecedented number of spaceborne and airborne images. In fact, a total of 27,174 images have been ingested to the USGS Hazards Data Distribution System (HDDS) Explorer, except for the SAR images used. Based on the Texas flood use case, we describe the success of this effort as well as the limitations in fulfilling the needs of the decision-makers, and reflect upon these. In order to unlock the full potential for Earth observation data in flood disaster response, we suggest in a call for action(i) stronger collaboration from the onset between agencies, product developers, and decision-makers;(ii) quantification of uncertainties when combining data from different sources in order to augment information content; (iii) include a default role for the end-user in satellite acquisition planning; and(iv) proactive assimilation of methodologies and tools into the mandated agencies.

From SAR-derived flood mapping to water level data assimilation into hydraulic models

This paper describes a fully automatic processing chain that makes use of SAR images for retrieving water stage information to be assimilated into a hydraulic forecasting model. This chain is composed of three steps: flood extent delineation, water stage retrieval and data assimilation of stage information into a hydraulic model. The flood-mapping step is addressed with a fully automatic algorithm, based on image statistics and applicable to all existing SAR datasets. Uncertainty on the flood extent map is represented with an ensemble of flood extent maps, obtained following a bootstrap methodology. Water stage observations are then retrieved by intersecting the flood shoreline with the floodplain topography. The ensemble of flood extent maps allows extracting multiple water levels at any river cross section of the hydraulic model, thereby taking into account the uncertainty associated with the floodmapping step. Finally, data assimilation consists in integrating uncertain observations, i.e. SAR-derived water stages, with uncertain hydraulic model simulations. The proposed processing chain was applied to two case studies. For the test case of June 2008 on the Po River (Italy), only low resolution but freely available satellite data were used. For the January 2011 flood on the Sure River (Luxembourg), higher resolution data were used and obtained at a cost. The results show that with the assimilation of SAR-derived water stages significant improvements can be achieved in the forecasting performance of the hydraulic model.

Support for Multi-temporal and Multi-mission data processing: The ESA Research and Service Support

The ESA Research and Service Support (RSS) offers services to enable the Earth Observation (EO) data exploitation. RSS users are Principal Investigators, institutions and SMEs needing support to progress in their research and/or development activity. For these EO data users it is nowadays becoming more and more important the exploitation of data time series from historical ESA missions combined with the new Sentinel data. In order to provide support to the EO user community interested in the exploitation of long time series of data, the ESA RSS makes available ad-hoc support services that can be easily tailored on the user needs.

A new automatic synthetic aperture radar-based flood mapping application hosted on the European Space Agency's Grid Processing of Demand Fast Access to Imagery environment

This paper introduces an automatic flood mapping application that is hosted on the Grid Processing on Demand (GPOD) Fast Access to Imagery (Faire) environment of the European Space Agency. The main objective of the online application is to deliver operationally flooded areas using both recent and historical acquisitions of SAR data. Having as a short-term target the flooding-related exploitation of data generated by the upcoming ESA SENTINEL-1 SAR mission, the flood mapping application consists of two building blocks: i) a set of query tools for selecting the “crisis image” and the optimal corresponding “reference image” from the G-POD archive and ii) an algorithm for extracting flooded areas via change detection using the previously selected “crisis image” and “reference image”. Stakeholders in flood management and service providers are able to log onto the flood mapping application to get support for the retrieval, from the rolling archive, of the most appropriate reference image. Potential users will also be able to apply the implemented flood delineation algorithm. The latter combines histogram thresholding, region growing and change detection as an approach enabling the automatic, objective and reliable flood extent extraction from SAR images. Both algorithms are computationally efficient and operate with minimum data requirements. The case study of the high magnitude flooding event that occurred in July 2007 on the Severn River, UK, and that was observed with a moderateresolution SAR sensor as well as airborne photography highlights the performance of the proposed online application. The flood mapping application on G-POD can be used sporadically, i.e. whenever a major flood event occurs and there is a demand for SAR-based flood extent maps. In the long term, a potential extension of the application could consist in systematically extracting flooded areas from all SAR images acquired on a daily, weekly or monthly basis.