Flooding Water Depth Estimation With High-Resolution SAR

Abstract

Abstract-The retrieval of flooding levels with high-resolution (HR) synthetic aperture radar (SAR) images is presented in this paper. A new framework is proposed. It is based on the inversion of theoretical scattering models initially developed for nonflooded urban areas and here adapted to the flooding case. Starting from the theory, two possible retrieval approaches have been developed and are the main topic of this paper: two possible retrieval approaches have been developed and are the main topic of this paper: the local Single Image Objects Aware (SIObA) and the global Two Image Area Aware (TIArA). These two approaches are conceived to be applicable under different working conditions and consequently holding different properties and reliability. For each of them, a different algorithm is derived and tested, and the retrieval results are validated on a meaningful data set of HR TerraSAR-X images relevant to the Gloucestershire (U.K.) flooding that occurred in year 2007.

Semi-automated estimation of the local flood depth on SAR images

“…In the flowchart, both the extraction of the building's double-backscatter contribution and the estimation of the building's height rely on the urban backscattering model given in the previous section. Moreover, the same values for the roughness and dielectric properties of the ground and the building's wall materials measured in situ in [6] were used here, thanks to the fact that the flood level estimation was performed on the same dataset.…”

“…where the two variables relevant to this study are: The formula in (1) was inverted in [9] to estimate the height of a building from the intensity of its double-bounce contribution on the SAR image, by assuming that the dielectric and roughness properties of the scene materials were given or had been measured a priori. The same rationale was applied in [6] to evaluate, this time, the flood depth on SAR images, by considering the soil flooded and changing its physical and electrical properties in (2) accordingly. The strength of this method lies in the fact that only a single SAR image is needed for the estimation of the building's height, without the necessity for any additional ancillary data.…”

“…Ultimately, the water level was calculated, straightforwardly, as the absolute difference between the building heights in normal and flooded conditions [6].…”

“…In this paper, the principal objective is to improve the work done in [6] by semi-automating the retrieval of the double-bounce contribution of a given building from the SAR image, before the subsequent estimation of the water level in its vicinity. The need for automation is motivated by the fact that automated inundation detection methods outperform the cumbersome and subjective manual flood mapping performed by human experts, especially in the aspect of the mapping speed [7], which is crucial to allow the civil protection authorities to react promptly with the adequate relief efforts.…”

Water Level Measurements from Drones: A Pilot Case Study at a Dam Site

“…Water level measurements can then be used to improve hydraulic modeling of rivers to support flood risk mitigation plans. Only few studies have aimed at estimating water depth from flood data acquired during the flood event itself (e.g., [28,29]), while flood-level is often retrieved using remote-sensing data in the aftermath of the event for post-event flood simulations (e.g., [30,31]). …”