Mapping Damage-Affected Areas after Natural Hazard Events Using Sentinel-1 Coherence Time Series

Abstract

The emergence of the Sentinel-1A and 1B satellites now offers freely available and widely accessible Synthetic Aperture Radar (SAR) data. Near-global coverage and rapid repeat time (6-12 days) gives Sentinel-1 data the potential to be widely used for monitoring the Earth's surface. Subtle land-cover and land surface changes can affect the phase and amplitude of the C-band SAR signal, and thus the coherence between two images collected before and after such changes. Analysis of SAR coherence therefore serves as a rapidly deployable and powerful tool to track both seasonal changes and rapid surface disturbances following natural disasters. An advantage of using Sentinel-1 C-band radar data is the ability to easily construct time series of coherence for a region of interest at low cost. In this paper, we propose a new method for Potentially Affected Area (PAA) detection following a natural hazard event. Based on the coherence time series, the proposed method (1) determines the natural variability of coherence within each pixel in the region of interest, accounting for factors such as seasonality and the inherent noise of variable surfaces; and (2) compares pixel-by-pixel syn-event coherence to temporal coherence distributions to determine where statistically significant coherence loss has occurred. The user can determine to what degree the syn-event coherence value (e.g., 1st, 5th percentile of pre-event distribution) constitutes a PAA, and integrate pertinent regional data, such as population density, to rank and prioritise PAAs. We apply the method to two case studies, Sarpol-e, Iran following the 2017 Iran-Iraq earthquake, and a landslide-prone region of NW Argentina, to demonstrate how rapid identification and interpretation of potentially affected areas can be performed shortly following a natural hazard event.

Applications of SAR Interferometric Coherence Time Series: Spatiotemporal Dynamics of Geomorphic Transitions in the South‐Central Andes

“…Though a limited number of independent soil moisture estimates exist (e.g., SMAP and CCI), they have proven unreliable in ground tests and for the spatial resolution required for this study (e.g., Dorigo et al, ; Moran et al, ; Ray et al, ; Scott et al, ). We estimate relative soil moisture changes by calculating amplitude decrease between SAR images (Olen & Bookhagen, ). Because the presence of soil moisture, and particularly standing water, will result in a decrease in received radar amplitude, a decrease in amplitude ( δAmp ) between adjacent images may indicate an increase in soil moisture for a given date n: …”

“…Coherence has been widely used in the natural disaster community as a proxy for post-event damage following a catastrophic event, for example, an earthquake or a landslide (e.g., Matsuoka & Yamazaki, 2004; e.g., Olen & Bookhagen, 2018;Yun et al, 2015). Coherence images selected from specific dates before and after a known event have been used to detect and map the spatial extents of displacement or damage.…”

DuPLO: A DUal view Point deep Learning architecture for time series classificatiOn

“…As state of the art standard machine learning tool, we selected a Random Forest (RF ) classifier [ As concerns DuP LO, we perform a preprocessing phase in order to associate each pixel to its surrounding area (i.e., to force the learning process to take into account the spatial context). We consider patches with a spatial extent equals to 5 × 5, where each patch represents the spatial context of the pixel in position (2,2). This means that for each timestamp we have a cube of information of size (5 × 5 × 5), since 5 is the number of raw bands and indices involved in the analysis.…”

“…Our hypothesis is that, since CNNs and RNNs capture different aspects of the data, a combination of both models would produce a more diverse and complete representation of the information for the underlying land cover classification task. Experiments carried out on two study sites characterized by different land cover characteristics (i.e., the Gard site in France and the Reunion Island in the Indian Ocean), demonstrate the significance of our proposal.Classification, Sentinel-2 toring and land management planning [1,2,3,4,5,6,7,8,9]. In the context of Land Use/Land Cover (LULC) classification, exploiting SITS can be fruitful to discriminate among classes that exhibit different temporal behaviors [10], i.e., with the respect to the results that can be obtained using a single image.…”

“…Classification, Sentinel-2 toring and land management planning [1,2,3,4,5,6,7,8,9]. In the context of Land Use/Land Cover (LULC) classification, exploiting SITS can be fruitful to discriminate among classes that exhibit different temporal behaviors [10], i.e., with the respect to the results that can be obtained using a single image.…”