Coastal environments and, in particular, ground motions in coastal areas are in practice often poorly known, and in many cases, only little information is available about the relevant patterns and time evolution. For this reason, it is strategic the continuous monitor of coastal delta regions by the use of advanced Earth Observation (EO) systems that are capable to detect and monitor the evolution of surface deformation phenomena, recovering their spatial extent on the ground and following their temporal variability. This is beneficial for the subsequent interpretation of natural/anthropogenic processes causing surface motion.

The activities performed within this present Dragon IV project have mostly been focused on the analysis of modification processes that characterize two important Delta river areas in China: the Yangtze and the Pearl River Deltas. Principally, we focused on Shanghai area but also a few experiments have been carried out on PRD. Both delta regions are significantly affected by sea-level rise and natural/anthropogenic deformation phenomena, making it clear the need of extended analyses for a better understanding of the mechanisms responsible for the observed surface modifications, and for the planning of actions devoted to risk prevention for populations living in coastal areas.

More specifically, the main aim is to retrieve long-term displacement time-series from EO data, specifically satellite Synthetic Aperture Radar (SAR), of the investigated areas through advanced differential interferometric synthetic aperture (DInSAR) techniques [1]-[2].

To evaluate the combined risk of sea level rise, storm surges, and ground subsidence, the availability of high-resolution digital elevation models (DEM) of monitored coastal areas has also resulted mandatory. Added-value EO data-products, such as the updated DEMs of coastal areas subject to sea-level rise, the time-series of terrain displacement, mean displacement velocity maps, and time-series of SAR backscattering maps, have been obtained by exploiting archives of SAR data at different spatial resolution spanning more than 10 years, from 2007 to 2018.

A few experiments have been conducted. In particular, the combined use of ENVISAT, Cosmo-SkyMed, and Sentinel-1 data have permitted to recover long-term displacement time-series of the ocean-reclaimed lands. New combination methods for the retrieval of the components of surface deformations from InSAR-driven LOS-projected measurements have been applied, and the most relevant results have been published on peer-reviewed journals [3]-[5].

Further investigations are currently being in progress to assess the risk of flooding of the coastal region of Shanghai, by benefiting from the retrieved InSAR deformation maps and a digital elevation model (DEM) of the area. The latter has been generated by using 2012 TanDEM-X bistatic SAR data [6]. The results of all these investigations will also be presented in separate communications at the mid-term D4 meeting.

We would like to remark that these studies are the result of a strict cooperation between the European and Chinese research institutions involved in the project.

Finally, some results evidencing the current ground deformation of the Pearl River Delta (PRD) region, obtained using Sentinel-1 acquisitions acquired over the last two years will be presented. In particular, we have selected this test-site area as a laboratory to evaluate the performance of a new multi-grid phase unwrapping approach. We moved from the observation that new-generation satellite data are characterized by larger spatial coverage and/or improved spatial resolutions, thus leading to augmented computational problems. In particular, the number of observation points in each SAR scene tends to considerably increase, thus posing new challenges. To overcome this problem, some multi-grid phase unwrapping methods, based on partitioning a scene in several overlapped, multi-resolution grids of pixels, and on their proper recombination [7]-[8], can be profitably adopted. In our experiments we focused on the PRD region and we provided InSAR-based analyses over multi-grids of pixels characterized by different spatial pixel spacing (i.e., from 500m x 500 m to 25m x 25m). Further developments consist in adaptively identifying the correct (most adequate) spatial spacing grid in each area, separately, depending on the observed spatial rate of deformation, so as to use finer grids in areas with significantly large rates of deformation and worse pixel spacing where deformation has a low spatial rate. Noteworthy, the efficient use of multiple grids of resolution can permit both to unwrap/process large interferograms (even on a continental basis) and, then, progressively “zoom in” given regions in conformity with the Nyquist sampling condition. Very preliminary results will be presented at the D4-meeting. A hybrid multi-scale experiment has also been performed on the Shanghai coastal area.

[1] Berardino, P., G. Fornaro, R. Lanari, E. Sansosti (2002), A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms, IEEE Trans. Geosci. Remote Sens., 40(11), 2375-2383.

[2] A. Ferretti, C. Prati, and F. Rocca (2001), Permanent scatterers in SAR interferometry, IEEE Trans. Geosci. Remote Sens., 39(1), 8-20.

[3] Zhao Q., Pepe A., Gao W., Lu Z., Bonano M., He M.L., Wang J., Tang X. (2015) A DInSAR investigation of the ground settlement time evolution of ocean-reclaimed lands in Shanghai, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8, 1763-1781.

[4] A. Pepe, M. Bonano, Q. Zhao, T. Yang, H. Wang, “The Use of C-/X-Band Time-Gapped SAR Data and Geotechnical Models for the Study of Shanghai’s Ocean-Reclaimed Lands through the SBAS-DInSAR Technique, “ Remote Sensing 2016, 8, 911; doi:10.3390/rs8110911.

[5] Lei Yu, Tianliang Yang, Qing Zhao, Min Liu and Antonio Pepe, “The 2015-2016 Ground Displacements of the Shanghai Coastal Area Inferred from a Combined COSMO-SkyMed/Sentinel-1 DInSAR Analysis,” Remote Sens. 2017, 9, 1194.

[6] Krieger G., Moreira A., Fiedler H., Hajnsek I., Werner M., Younis M., Zink M. (2007) TanDEM-X: A satellite formation for high resolution SAR interferometry. IEEE Tans. Geosci. Remote Sens., 45, 3317-3340.

[7] M. D. Pritt, “Multigrid phase unwrapping for interferometric SAR,” in IGARSS 95, Florence, Italy

[8] Antonio Pepe, L. D. Euillades, M. Manunta, R. Lanari: "New Advances of the Extended Minimum Cost Flow Phase Unwrapping Algorithm for SBAS-DInSAR Analysis at Full Spatial Resolution," IEEE Transaction on Geoscience and Remote Sensing, vol. 49, n° 10, October 2011, pp. 4062-4079.

Global mean sea-levels have risen during the 20^th century, and they will accelerate rising by up to ~60 cm by 2100 (Nicholls and Cazenave, 2010). However, the projections remain uncertain in estimating the rate of increase in melting of glaciers, Greenland and Antarctic ice sheets. The fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC) suggests the higher values for sea level rise base on newer ice-sheet observations (IPCC, 2014). This will assuredly increasingly submerge risk in the low-lying areas of coastal zones throughout this century. Assessing and mapping the coastal inundation risk under sea level rise has been conducted in Charlestown, RI, USA (Grilli et al., 2017), Italian coastal plains (Antonioli et al., 2017), coastal zones of Poland (Paprotny et al., 2017), the Southeast Queensland (Mills et al., 2016), the Venice city of Italy (Sperotto et al., 2015), and Shanghai (Wang et al., 2012).

Moreover, non-climate-related anthropogenic processes, such as ground subsidence due to groundwater extraction, ground settlements due to large scale land-reclamation, and fast and non-linear subsidence phenomena of artificial sea wall, will exacerbate the risk to coastal zones and megacities and amplify local vulnerability. Making the situation worse is the combination of sea-level rise resulting from climate change, local sinking of land resulting from anthropogenic and natural hazards. Previous study has already stressed the significance of relative sea level rise in increasing coastal flood frequency (Karegar et al., 2017; Little et al., 2015; Cayan et al., 2008; Carminati et al., 2002; Shi et al., 2000).

The coastal vulnerability of mega-city, Shanghai, which is located at the Yangtze River Delta, is currently being amplified by the compounding effects of the time-dependent ground subsidence and the accelerated rate of sea level rise (Yin et al., 2013). The provided examples of delta regions affected by the combination of sea-level rise and significant modifications over time make clear the need of extended analyses for the understanding of the mechanismsat the base of the surface modifications of coastal areas, estimating of future regional relative sea level change, and evaluating the potential submerged land area. The main goals of this study are to provide a full characterization of the scene modifications over time and causes of the coastal region environments, to provide estimates of future regional sea level change, and to project coastal submerged area.

In this study, the use of well-established remote sensing technologies, based on the joint exploitation of multi-spectral information gathered at different spectral wavelengths, the advanced Interferometric Synthetic Aperture Radar (InSAR) techniques, and the hydrodynamic model-FloodMap projections will be employed for these purposes. The results obtained in this study represents an asset for the planning of present and future scientific activities devoted to the monitoring of such fragile environments. These analyses are essential to assess the factors that will continue to amplify the vulnerability of the low-elevation coastal zones.

In order to evaluate the combined risk of sea level rise and ground subsidence, the availability of high-resolution digital elevation models (DEM) of monitored coastal areas is generated with InSAR. The time-series of terrain deformation and mean deformation velocity maps, will be obtained by exploiting archives of Synthetic Aperture Radar (SAR) data with different levels of spatial resolution spanning a long time interval of about 10 years since the beginning of 2007 to 2017. SAR data will be collected by the former (i.e., the ESA ENVISAT-ASAR) and the new generation of radar sensors (the Cosmo-SkyMed constellations and Sentinel-1A). Large scale DInSAR analysis over wide areas will be performed by exploiting the Small BAseline Subset (SBAS) algorithm. The InSAR derived DEM with different spatial resolution and a 2D hydrodynamic model-FloodMap will be employed to investigate the evolving flood risk in the eastern coastal area of Shanghai and to derive coastal submerged area.

As global warming problem is becoming serious in recent decades, the global sea level is rising continuously. This will cause damages to the coastal deltas with the characteristics of low-lying land, dense population, and developed economy. Continuously reclamation costal intertidal and wetland areas are making Shanghai, the mega city of Yangtze River Delta, more vulnerable to sea level rise. Previous studies have shown that there is severe land subsidence in the reclamation area in the east of Shanghai [1-3]. Land subsidence greatly exacerbates the risk of sea level rise. How to obtain land subsidence data efficiently is crucial.

DInSAR technology can quickly obtain a large range of land subsidence information, and the accuracy can reach the millimeter level. Today, land subsidence monitoring using DInSAR technology has been widely used in coastal cities such as Shanghai, Guangzhou and Hong Kong [1-5]. However, limited by the satellite launch time and life cycle, it is difficult to obtain a long time series of land subsidence. Antonio et al. [2,3]. used the subsidence model derived from laboratory centrifuge tests and the Singular Value Decomposition (SVD) to joint the land subsidence time series of the three satellites of ENVISAT/ASAR, COSMO-SkyMed, and Sentinel-1A. In this way, a subsidence time series of up to ten years is obtained.

Based on this, we have found that the deviation of land subsidence time series obtained by using different joint strategies (Using a different order to ioint three satellite platform subsidence time series) at some high coherence points is larger. In this paper, By exploiting a set of 35 SAR images acquired by the ENVISAT/ASAR from February 2007 to May 2010 , a set of 61 SAR images acquires by the COSMO-SkyMed (CSK) sensors from December 2013 to March 2016, and a set of 33 SAR images acquires by the Sentinel-1A (S1A) sensors from December 2013 to March 2016, coherent point targets identified by using the Small Baseline Subset (SBAS) algorithm. Then, the subsidence time series of high coherence points was obtained.

We use the algorithm proposed in [1,2] to joint the subsidence time series of the three satellite platforms. We adopt different joint strategies: Strategy 1 is to first combine the subsidence time series of CSK and S1A, and then combine the CSK_S1A subsidence time series with the ENV subsidence time series; Strategy 2 is to first combine the subsidence time series of ENV and CSK, and then combine the ENV_CSK subsidence time series with the S1A subsidence time series. We set a threshold for the Euclidean distance of the subsidence time series obtained by the two joint strategies, and call the high-coherence point with a Euclidean distance greater than the threshold as "bad pixel" and the high-coherence point with a Euclidean distance less than the threshold as " Good pixel". Then, through the consistency matching algorithm of the un-joint subsidence time series of three satellite platforms between the bad point and the good point, the joint subsidence time series of bad pixels is corrected.

Meanwhile we use the monthly mean tide level series from Lvsi Station (1959 ~ 2011), and combine the Ensemble Empirical Mode Decomposition(EEMD), Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Back Propagation (BP) Neural Network to propose two improved EEMD-GA-BP and EEMD-PSO-BP method for regional sea level change prediction. The use of GA and PSO optimize BP Neural Network can improve the accuracy, and PSO is superior to GA. Multi-platform long-term land subsidence time series and precise sea level predict time series provides a realistic meaning for the impact of relative sea level change on the coastal areas.

[1]Zhao Q, Pepe A, Gao W, et al. A DInSAR Investigation of the Ground Settlement Time Evolution of Ocean-Reclaimed Lands in Shanghai[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2015, 8(4):1763-1781.

[2]Pepe A, Bonano M, Zhao Q, et al. The use of C-/X-band time-gapped SAR data and geotechnical models for the study of Shanghai's ocean-reclaimed lands through the SBAS-DInSAR technique[J]. Remote Sensing, 2016, 8(11):911.

[3]Yu L, Yang T, Zhao Q, et al. The 2015–2016 Ground Displacements of the Shanghai Coastal Area Inferred from a Combined COSMO-SkyMed/Sentinel-1 DInSAR Analysis[J]. Remote Sensing, 2017, 9(12):1194.

[4]Zhao Q, Lin H, Jiang L, et al. A Study of Ground Deformation in the Guangzhou Urban Area with Persistent Scatterer Interferometry[J]. Sensors, 2009, 9(1):503-18.

[5]Zhao Q, Lin H, Gao W, et al. InSAR detection of residual settlement of an ocean reclamation engineering project: a case study of Hong Kong International Airport[J]. Journal of Oceanography, 2011, 67(4):415-426.

Continuing investigations of tidal variability at multiple tide gauges in Hong Kong and the South China Sea (SCS) have identified correlations with the potential to amplify extreme water levels and nuisance flooding at certain locations. Observed changes are hypothesized to be due to mechanisms active on multiple spatial scales. The regional behaviour of the SCS may have changed tidal evolution via MSL rise, upper-ocean warming (and hence, stratification), or modulations in the baroclinic conversion at the entrance of the SCS (the Luzon Strait). The baroclinic tidal signal can be enhanced at the northern shelf of the SCS and can generate multiple PSI-type interactions that yield amplifications in minor tides such as M₃ that can be observed in Hong Kong. Additionally, the enclosed regions of Hong Kong have undergone massive land reclamation projects that may have changed the resonant and/or frictional response of the harbors to the regional dynamics. Previous works reported on the tidal anomaly correlations (TACs) to detrended MSL fluctuations, shown to be most important in harbour regions such as Victoria Harbor in Hong Kong. In this work, we highlight the intertidal correlations of diurnal (D₁) tides to semidiurnal (D₂) tides, which are positively reinforced through the northern SCS, and the correlations of overtide (OT) fluctuations to D₁ and D₂, shown to be negatively reinforced (i.e., anti-correlated) across the same region. The consideration of all water level variabilities may help explain the large TACs previously reported and may have serious implications for future water levels in Hong Kong.

We present an updated method for semi-empirical determination of significant wave height (H_s) estimation applied to Sentinel-1 SAR data. The method is based on existing semi-empirical algorithms, intending to identify a narrow-band swell-wave spectrum on the open-ocean waters and focused on linear imaging mechanism of tilt modulation on radar-looking directional surface wave. Utilizing the radar backscatter cross-section, we develop and evaluate a linear method to estimate H_s in various environment conditions without any prior knowledge from external input variables. Our method is divided into two components; first, the estimation of dominant wavelength using a 2-Dimensional Fast Fourier Transform (2D-FFT) and dimensionless coefficient, second, an estimation of ocean surface roughness and slope variation. The routines are aided by: adaptive filtering of the radar cross-section using median filters and Gaussian filters in different domains, linear fitting, and a detailed dependency analysis based on wave type and varying wind speed. Standard meteorological buoy data from the National Buoy Data Center (NDBC) is used for validation of the estimated Hs and input for the dependency analysis. These stations are selected to test differing water depths, wave properties, and wind conditions. This study employs Level-1 GRD Sentinel-1A and 1B SAR images from 2016 to 2017 covering locations of in-situ NDBC stations located near Hawaii, used for validation. Results show that the method performs well in estimating H_s under low to moderate wind forcing conditions (4 – 10 ms^-1) for any wave type in open-water areas. Lower performances are found under very low and strong wind conditions, and in wind-wave dominant environments.

This study focuses on the C-band radar backscatter features of the shallow water topography of Subei Bank in the Southern Yellow Sea using 25 ENVISAT (Environmental Satellite) ASAR (advanced synthetic aperture radar) and ERS-2 (European Remote-Sensing Satellite-2) SAR images acquired between 2006 and 2010. Under different sea states, SAR imagery shows different bathymetric features: the wide bright patterns with an average width of 6 km are shown under low to moderate wind speeds and correspond to sea surface imprints of tidal channels formed by two adjacent sand ridges, while the sand ridges appear as narrower (only 1 km wide), fingerlike, quasi-linear features on SAR imagery in high winds. Two possible SAR imaging mechanisms of coastal bathymetry are proposed in the case where the flow is parallel to the major axes of tidal channels or sand ridges. Two vortexes will converge at the central line of the tidal channel in the upper layer and form a convergent zone over the sea surface when the surface Ekman current is opposite to the mean tidal flow, therefore the tidal channels are shown as wide and bright stripes on SAR imagery. For the SAR imaging of sand ridges, all the SAR images were acquired at low tidal levels. In this case, the ocean surface waves are possibly broken up under strong winds when propagating from deep water to the shallower water, which leads to an increase of surface roughness over the sand ridges.

To satisfy the growing land demand for industrial and urban development, man-made lands, reclaimed from the sea, are used to build airports, harbors, and industrial areas. However, in such reclaimed areas, foundation settlements caused by unconsolidated soils are of public concern, and may induce severe damage to buildings and infrastructures. In such a context, Differential Synthetic Aperture Radar (SAR) Interferometry (DInSAR) technique [1] is able to retrieve ground displacements, with centimeter to millimeter accuracy, by exploiting the phase difference between two SAR images acquired over the investigated area at different times and from different orbital positions.

Advanced DInSAR approaches, such as the Persistent Scatterer Interferometry (PSI) [2] and the Small BAseline Subset (SBAS) technique [3], nowadays represent effective tools for remotely detecting, mapping and monitoring surface deformation phenomena, thanks to their capability to produce spatially dense velocity maps as well as long-term displacement time-series corresponding to coherent targets location.

This study is focused on the retrieval of deformation signals over the ocean-reclaimed lands of Shanghai, China, and it is mostly devoted to the development of an ad-hoc procedure based on the combination of multiple-scale of resolution information. Over the last recent years, several investigations [4]-[5] have been carried out to study the deformation of the coastal area of Shanghai. In particular, the time evolution of ground deformation occurring over the coastal zone was derived from 2007 to 2017 [5] by jointly analyzing sequences of X-band (COSMO-SkyMed) and C-band (Sentinel-1A and ENVISAT/ASAR) SAR images. To achieve this task, a novel approach to link the time-gapped COSMO-SkyMed and ENVISAT/ASAR data was applied and an SVD-based combination approach to link time-overlapped COSMO-SkyMed and Sentinel-1A SAR data was developed. More precisely, the temporal evolution of the sensor-line-of-sight terrain deformation occurring over the coastal area of Shanghai was retrieved by independently processing the available sets of SAR images. This was done by analyzing sequences of multilooked differential SAR interferograms generated from the stacks of SAR images collected from 2007 to 2010 by the ENVISAT sensor, from 2013 to 2016 by the CSK sensors’ constellation, and from 2015 to 2016 by the Copernicus S1A radar instrument. The well-established Small BAseline Subset (SBAS) differential interferometry technique [3] was used for retrieving the LOS deformation time-series for each SAR sensor. Subsequently, for each coherent pixel found both in the ENVISAT and CSK datasets, the time-gapped ENVISAT and CSK LOS deformation time series were preliminarily converted in vertical (subsidence) deformation and, then, linked by using a time-dependent geotechnical centrifuge model.

Starting from December 2017 a new set of CSK data is available relevant to the same track used in the previous investigations; however (due to the lack of a regular planning of SAR acquisitions over the investigated area) there is a big lapse of more than one year between the old (from February 2014 to March 2016) and the new CSK SAR dataset. This leads us to the impractability of obtaining long-term SBAS deformation time-series by exclusively using CSK data. This drawback, at the same time, lead us the possibility to conduct an experiment based on the application of a hybrid multi-scale SBAS strategy. The idea is to focus on a group of highly coherent point-wise targets that preserve their coherence, also after one year, and to generate the CSK 2014-2018 surface displacement time-series related to those pixels. The applied advanced approach, originally presented in [6], for the identification of the highly-coherent point-wise scatterers and for the subsequent generation of the displacement time-series will be adopted. Subsequently, the achieved time-series will be compared with those obtained by linking the two independent (time-gapped) sequences of CSK data (similarly to what done in [5]) by the model adopted in [4]-[5] through the solution of a non-linear optimization problem based on the use of the Levenberg-Marquadt technique. The goal of this present investigation is to prove that, at least in correspondence to the highly coherent targets on the ground, the expected deformations behavior dictated by the used model is in general agreement with the achieved results. This finding is interesting to check the validity of the model used in our previous investigations [4]-[5]. The preliminarily results will be presented and discussed at the next Dragon-IV mid-term meeting.

[1] R. Bürgmann, P. A. Rosen, and E. J. Fielding, “Synthetic aperture radar interferometry to measure Earth’s surface topography and its deformation,” Annu. Rev. Earth Planet. Sci., vol. 28, no. 1, pp. 169–209, May 2000.

[2] A. Ferretti, C. Prati, and F. Rocca, “Permanent scatterers in SAR interferometry,” IEEE Trans. Geosci. Remote Sens., vol. 39, no. 1, pp. 8–20, Jan. 2001.

[3] P. Berardino, G. Fornaro, R. Lanari, and E. Sansosti, “A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms,” IEEE Trans. Geosci. Remote Sens., vol. 40, no. 11, pp. 2375–2383, Nov. 2002

[4] Q. Zhao, A. Pepe, W. Gao, Z. Lu, M. Bonano, M. He, X. Tang, “A DInSAR Investigation of the Ground Settlement Time Evolution of Ocean-Reclaimed Lands in Shanghai,” IEEE Selected Topics in Applied Earth Observations and Remote Sensing, vol. 8, no. 4, pp. 1763-1781, April 2015.

[5] Y. Lei, Y. Tianliang, Qing Zhao, Min Liu and Antonio Pepe, “The 2015-2016 Ground Displacements of the Shanghai Coastal Area Inferred from a Combined COSMO-SkyMed/Sentinel-1 DInSAR Analysis,” Remote Sens. 2017, 9, 1194.

[6] Antonio Pepe, L. D. Euillades, M. Manunta, R. Lanari: "New Advances of the Extended Minimum Cost Flow Phase Unwrapping Algorithm for SBAS-DInSAR Analysis at Full Spatial Resolution," IEEE Transaction on Geoscience and Remote Sensing, vol. 49, n° 10, October 2011, pp. 4062-4079.

Abstract

Due to large-scale infrastructure construction and land reclamation, the problem of land subsidence in Shanghai is becoming more and more serious, which will have a major impact on urban public safety. Shanghai has established leveling networks and GPS networks to detect land subsidence, but due to cost constraints, the resolution is relatively insufficient (Amighpey et al. 2016). In recent years, InSAR technology has been widely used to monitor urban land subsidence due to its low cost and high precision (Sansosti et al. 2010; Hooper et al. 2012). Sentinel-1A data were available in the single-look-complex (SLC) format and acquired through the interferometric wide swath (IW) mode by employing the terrain observation by progressive scans (TOPS) acquisition mode, which provides large swath widths of 250 km at ground resolutions of 5 m x 20 m. In order to study the distribution and spatial pattern of land subsidence in Shanghai, a set of 33 SAR images acquired by the Sentinel-1A from July 2015 to August 2017 (ascending passes, VV polarization, with a side-looking angle of about 39˚and a satellite heading angle of about 348˚) were exploited to get coherent point targets as long as land subsidence velocity maps and time series which were identified by using the Small Baseline Subset (SBAS) algorithm (Berardino et al. 2002; Lanari et al. 2007). SBAS is based on the use of multiple-master multilook interferograms generated after a proper selection of Small Baseline (SB) SAR data pairs. LOS displacement time-series are computed by solving a least-squares (LS) minimization problem, based on the application of the singular value decomposition (SVD) method, to the sequence of unwrapped multilook interferograms. In this paper Sentinel-1A data were processed by the SBAS toolbox implemented within the commercial ENVI’s SARScape modules from EXELIS VIS Information Solutions with the coherence threshold of 0.35 and the maximum temporal baseline set to 180 days.

Urban land subsidence is mainly caused by infrastructure construction and groundwater extraction (Galloway et al. 2011; Wang et al. 2017). We use Landsat optical satellite data to analyze the changing trend of coastline in Shanghai since the 1990s (Landsat5、7、8, resolutions of 30m x 30m, every five years collected in winter). It is found that the coastline is expanding dramatically and most of the coastal areas with obvious settlement are the regions which has been reclaimed in the recent decade. In the central city, subsidence areas are densely located along subway lines. Since Shanghai began to reduce the exploitation of groundwater from the 1970s, the current settlement in Shanghai is mainly due to the unstable geological structure of the reclamation area and the construction of a large number of above ground and underground infrastructure in the city. By focusing on the land subsidence trend in high-rise buildings in urban areas and coastal areas, we found that the settlement in the coastal area is still significant, and the high-rise buildings in the area along the Huangpu River also have a subsidence of up to 2 cm/y.

Reference

Amighpey, M., & Arabi, S. (2016). Studying land subsidence in Yazd province, Iran, by integration of InSAR and levelling measurements. Remote Sensing Applications: Society and Environment, 4, 1-8. doi: 10.1016/j.rsase.2016.04.001

Sansosti, E., Casu, F., Manzo, M. & R, L., 2010. Space-borne radar interferometry techniques for the generation of deformation time series: an advanced tool for earth’s surface displacement analysis, Geophys. Res.Lett., 37, L20305, doi:10.1029/2010GL044379.

Hooper, A., Bekaert, D., Spaans, K. & Arikan, M., 2012. Recent advances in SAR interferometry time series analysis for measuring crustal deformation, Tectonophysics, 514-517, 1–13.

Berardino, P.; Fornaro, G.; Lanari, R.; Sansosti, E. A New Algorithm for Surface Deformation Monitoring Based on Small Baseline Differential SAR Interferograms. IEEE Trans. Geosci. Remote Sens. 2002, 40, 2375–2383.

Lanari, R.; Casu, F.; Manzo, M.; Zeni, G.; Berardino, P.; Manunta, M.; Pepe, A. An overview of the small baseline subset algorithm: A DInSAR technique for surface deformation analysis. Pure Appl. Geophys. 2007,164, 637–661.

Galloway, D. L., & Burbey, T. J. (2011). Review: Regional land subsidence accompanying groundwater extraction. Hydrogeology Journal, 19(8), 1459-1486. doi: 10.1007/ s10040-011-0775-5

Wang, H., Feng, G., Xu, B., Yu, Y., Li, Z., Du, Y., & Zhu, J. (2017). Deriving Spatio-Temporal Development of Ground Subsidence Due to Subway Construction and Operation in Delta Regions with PS-InSAR Data: A Case Study in Guangzhou, China. Remote Sensing, 9(10). doi: 10.3390/rs9101004

In this study, the possible submergence area of the Yangtze River Delta (YRD) under the background of sea level rise is investigated combining both satellite data and numerical models. The sea level rises (SLRs) in the East China sea in the middle and end of 21^st century are first predicted based on the statistical analysis of historical satellite altimeter data. The mean SLR values in 2050 and 2100 are 20 cm and 35cm, respectively. Then a regional tidal wave model of the East China sea is constructed using the Finite-Volume, primitive equation Community Ocean Model (FVCOM), and a new storm surge inundation model of the YRD is developed (here we take typhoons Fung-wong and Wipha as examples) to analyze the possible submergence area. The results show that if there is no coastal protection, the maximum possible inundation caused by SLR through tidal wave propagation in 2100 is 2.5×10³ km², 87.2% larger than that at the current sea level, and the maximum submergence area during the two storm surges is 8.3×10²~2.7×10³ km². Considering a 4 m-high breakwater along the coastlines, there is no submergence in the above cases under the SLR in 2100, while the inundation is about 15.4 km² during typhoon Wipha when the breakwater is 2 meter high. The submergence mainly occurs in Jiangsu Province, especially Yan Cheng and Lian Yungang cities. It is suggested that the height of the breakwater should be not less than 2 m considering the impact of sea level rise and storm surges.