applied sciences

Article Subsidence Evolution of the Peninsula, , Based on InSAR Observation from 1992 to 2010

Yanan Du 1,2, Guangcai Feng 1,2,*, Xing Peng 1,2 and Zhiwei Li 1,2

1 School of Geosciences and Info-Physics, Central South University, 410083, China; [email protected] (Y.D.); [email protected] (X.P.); [email protected] (Z.L.) 2 Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Central South University), Ministry of Education, Changsha 410012, China * Correspondence: [email protected]; Tel.: +86-182-7486-7449

Academic Editor: Saro Lee Received: 2 March 2017; Accepted: 26 April 2017; Published: 29 April 2017

Abstract: Over the past two decades, the has suffered from many geological hazards and great property losses caused by land subsidence. However, the absence of a deformation map of the whole peninsula has impeded the government in making the necessary decisions concerning hazard prevention and mitigation. This study aims to provide the evolution of land deformation (subsidence and uplift) in the whole peninsula from 1992 to 2010. A modified stacking procedure is proposed to map the surface deformation with JERS, ENVISAT, and ALOS1 images. The map shows that the land subsidence mainly occurs along the coastline with a maximum velocity of 32 mm/year and in a wide range of inland arable lands with a velocity between 10 and 19 mm/year. Our study suggests that there is a direct correlation between the subsidence and the surface geology. Besides, the observed subsidence in urban areas, caused by groundwater overexploitation for domestic and industrial use, is moving from urban areas to suburban areas. In nonurban areas, groundwater extraction for aquaculture and arable land irrigation are the main reason for land subsidence, which accelerates saltwater intrusion and coastline erosion if regular surface deformation measurements and appropriate management measures are not taken.

Keywords: The Leizhou Peninsula; land subsidence; stacking Interferometric SAR; groundwater pumping

1. Introduction The Leizhou Peninsula (LZP), one of the three largest peninsulas in China, is located in the southwest of province, China, covering an area of 8500 km2 [1] (see Figure1). It plays an important role in the economic development of Guangdong province with its considerable agriculture contributions. However, over the past two decades the fast growing economy has accelerated groundwater over-pumping, causing large scale land subsidence and other geological hazards. Until 2001, the subsidence area in city reached 220 km2, with the maximum cumulative subsidence up to 17.9 cm [2]. Subsidence induced geological hazards have resulted in substantial property damage every year. For example, in 2006 only, the loss in Zhanjiang city was greater than 90 million RMB [3]. However, large-scale land subsidence measurement is still scarce in the LZP apart from a few leveling measurements carried out in the urban area of Zhanjiang city in 1984, 1989, 1999, and 2001, whose results indicated that the surface subsidence increased by >10 mm/year [2,4]. A surface deformation map presenting the temporal-spatial pattern of land subsidence of the whole peninsula is urgently needed to help the government make the right decisions in geological disaster prevention and mitigation, as well as to understand the mechanism of land subsidence caused by over-pumping groundwater.

Appl. Sci. 2017, 7, 466; doi:10.3390/app7050466 www.mdpi.com/journal/applsci Appl. Sci. 2017, 7, 466 2 of 18 decisionsAppl. Sci. 2017 in, 7geological, 466 disaster prevention and mitigation, as well as to understand the mechanism2 of 18 of land subsidence caused by over‐pumping groundwater.

Figure 1. BackgroundBackground setting setting (left) (left) and and Synthetic Synthetic Aperture Aperture Radar Radar (SAR) (SAR) coverages (right) (right) of the Leizhou Peninsula. Peninsula. The The yellow, yellow, red, red, and and black black rectangles rectangles represent represent the the coverage coverage of of JERS, JERS, ENVISAT, ENVISAT, and ALOS1 respectively.

Interferometric Synthetic Aperture Radar (InSAR), a satellite based technique, offers high Interferometric Synthetic Aperture Radar (InSAR), a satellite based technique, offers high precision precision (cm even to mm), high spatial resolution, and wide coverage measurements of surface (cm even to mm), high spatial resolution, and wide coverage measurements of surface deformation, deformation, reducing the use of in‐situ traditional techniques [5,6]. Furthermore, with the reducing the use of in-situ traditional techniques [5,6]. Furthermore, with the development of the SAR development of the SAR satellite missions, more and more newly acquired TerraSAR, satellite missions, more and more newly acquired TerraSAR, COSMO-SkyMed, Sentinel-1, ALOS-2 and COSMO‐SkyMed, Sentinel‐1, ALOS‐2 and archived SAR data (ERS1/2, JERS‐1, ENVISAT, ALOS1) archived SAR data (ERS1/2, JERS-1, ENVISAT, ALOS1) can be used to analyze the spatial-temporal can be used to analyze the spatial‐temporal characteristics of large area surface deformation with the characteristics of large area surface deformation with the temporal InSAR technique [7]. Therefore, temporal InSAR technique [7]. Therefore, the InSAR technique can serve as a good option in the InSAR technique can serve as a good option in mapping the surface deformation of a large-scale area. mapping the surface deformation of a large‐scale area. Actually, it has been widely used and its Actually, it has been widely used and its superiorities shown in detecting time-series surface superiorities shown in detecting time‐series surface deformation over many peninsulas and delta deformation over many peninsulas and delta areas, such as the Mexico City [8], [9], areas, such as the Mexico City [8], Shanghai [9], and Indonesia [10]. Traditional temporal InSAR and Indonesia [10]. Traditional temporal InSAR methods, such as Persistent Scatterer Interferometry methods, such as Persistent Scatterer Interferometry (PS‐InSAR), IPTA (Interferometric Point Target (PS-InSAR), IPTA (Interferometric Point Target Analysis), and Small Baseline Subset (SBAS-InSAR) Analysis), and Small Baseline Subset (SBAS‐InSAR) have been applied in detecting surface have been applied in detecting surface deformation, but they need abundant SAR images in data deformation, but they need abundant SAR images in data processing [11–13]. As the datasets are processing [11–13]. As the datasets are very limited in our study area, we used a modified stacking very limited in our study area, we used a modified stacking method to measure the surface method to measure the surface deformation evolution over the whole LZP with three kinds of SAR deformation evolution over the whole LZP with three kinds of SAR images (JERS, ENVISAT and images (JERS, ENVISAT and ALOS1) acquired from 1992 to 2010. Based on those results, the reasons ALOS1) acquired from 1992 to 2010. Based on those results, the reasons and mechanisms of land and mechanisms of land subsidence can be explained and are investigated in this study. subsidence can be explained and are investigated in this study. Here, we map the surface deformation of three different sensor platforms and illustrate Here, we map the surface deformation of three different sensor platforms and illustrate the the spatial-temporal evolution patterns in urban and agriculture areas. Based on the surface spatial‐temporal evolution patterns in urban and agriculture areas. Based on the surface hydrogeological and land use, we analyze the reason for land subsidence. The consequences and hydrogeological and land use, we analyze the reason for land subsidence. The consequences and implications of our land subsidence measurement are discussed in the last section. implications of our land subsidence measurement are discussed in the last section. 2. Hydrogeological Setting 2. Hydrogeological Setting The study area is the most southerly tip of China, with to the south, separating the peninsulaThe study from area is the Island.most southerly The elevation tip of of China, this area with ranges Qiongzhou from a fewStrait meters to the along south, the separatingcoast to 275 the m peninsula in the south from of thisHainan peninsula, Island. seeThe Figure elevation1. The of this Suixi area Fault ranges borders from the a areafew meters in the alongnorth, the see coast Figure to2 275a. Terranes m in the ofsouth Cambrian of this peninsula, age and Cretaceous see Figure age 1. The constitute Suixi Fault the basementborders the rocks area inof thisthe north, peninsula. see Figure Tertiary 2a. semi-consolidated Terranes of Cambrian sandy age clay and and Cretaceous Quaternary age unconsolidated constitute the sediments, basement rockssuch as of alluvial this peninsula. and lacustrine, Tertiary overlie semi the‐consolidated basement rocks. sandy The clay unconsolidated and Quaternary sediments unconsolidated include (1) Appl. Sci. 20172017,, 77,, 466 466 3 of 18 sediments, such as alluvial and lacustrine, overlie the basement rocks. The unconsolidated sedimentsZhanjiang Groupinclude of (1) Lower Zhanjiang Pleistocene Group age of Lower with an Pleistocene upper layer age of claywith and an upper a lower layer layer of of clay scattered and a lowerlenses layer of clay of andscattered coarse lenses sand of with clay gravel and coarse (Qlz), sand (2) Beihaiwith gravel group (Qlz), of middle (2) Pleistocene group of age middle with Pleistocenea surface layer age ofwith clayey a surface sand andlayer a of lower clayey layer sand of sandand a with lower gravel layer (Q2b), of sand (3) with Recent gravel sediments (Q2b), (3) of RecentQuaternary sediments age with of siltyQuaternary sand and age clayey with coarse silty sandsand (Q4), and (4)clayey basalt coarse of Himalayan sand (Q4), Period (4) basalt occurs of in HimalayanQuaternary Period sediments occurs (β) within Quaternary lots of dead sediments craters [14 (β],) aswith shown lots inof Figuredead 2cratersa. In order [14], toas obtain shown the in Figurerelationship 2a. In between order landto obtain subsidence the relationship and unconsolidated between sediments, land subsidence the soft soiland thickness unconsolidated derived sediments,from 22 boreholes the soft withsoil thickness an ordinary derived Kriging from interpolation 22 boreholes were with generated an ordinary (see Kriging Figure 2interpolationb). The soft weresoil thickness generated for (see each Figure borehole 2b). was The calculated soft soil thickness as the total for thickness each borehole of silt, was sand calculated and gray-clay as the which total thicknessoverlaid the of uppersilt, sand layer and of gray gravels‐clay above which the overlaid basement the [ 15upper,16]. layer of gravels above the basement [15,16].

Figure 2. Geology and land use maps of the Leizhou Peninsula (LZP) (modified (modified from [3,15,16]). [3,15,16]). ( a) is the geology map with four kinds of compressible deposits: Beihai group of middle Pleistocene age (Q2b); basalt of Himalayan Period ( β)) with lots lots of dead craters represented by black dots; Zhanjiang group ofof Lower Lower Pleistocene Pleistocene age age (Qlz); (Qlz); recent recent sediments sediments of Quaternary of Quaternary age (Q4); age the (Q4); black the dots black represent dots representthe dead craters the dead in LZP; craters (b) isin the LZP; distribution (b) is the of distribution soft soil thickness of soft derived soil thickness from the interpolationderived from of the 22 interpolationboreholes represented of 22 boreholes by black represented dots [16]; ( cby) is black the land dots use [16]; map (c) derivedis the land from use Zhanjiang map derived Municipal from ZhanjiangBureau of LandMunicipal and Resources Bureau of (ZMBLR) Land and [3 ],Resources the inset is(ZMBLR) an enlarged [3], the view inset of Zhanjiang is an enlarged city. view of Zhanjiang city. The aquifer system in this peninsula can be grouped into three subsystems, shallow aquifer, The aquifer system in this peninsula can be grouped into three subsystems, shallow aquifer, middle aquifer, and deep aquifer [4,14]. The shallow aquifer is an unconfined aquifer and mainly middle aquifer, and deep aquifer [4,14]. The shallow aquifer is an unconfined aquifer and mainly located at Q2b, Qlz, and Q4. Its maximum depth is 30 m. The middle aquifer occurs approximately located at Q2b, Qlz, and Q4. Its maximum depth is 30 m. The middle aquifer occurs approximately from 30 to 200 m. The depth of the deep aquifer ranges from 200 to 500 m. The latter two kinds are from 30 to 200 m. The depth of the deep aquifer ranges from 200 to 500 m. The latter two kinds are confined aquifers, extending laterally to the coast. The recharge of the unconfined aquifer is mainly confined aquifers, extending laterally to the coast. The recharge of the unconfined aquifer is mainly derived from precipitation with an annual rainfall of 1500 mm [4,14], and infiltration of canal and derived from precipitation with an annual rainfall of 1500 mm [4,14], and infiltration of canal and reservoir. Besides the leakage recharge of confined aquifer, the deep aquifer can also be recharged reservoir. Besides the leakage recharge of confined aquifer, the deep aquifer can also be recharged directly from unconfined aquifer [14]. This is because the wide range distribution of β and dead craters directly from unconfined aquifer [14]. This is because the wide range distribution of β and dead (3694 km2), as shown in Figure2a, has a high water content up to a depth of 310 m, deep enough to craters (3694 km2), as shown in Figure 2a, has a high water content up to a depth of 310 m, deep connect to the deep aquifer [14]. The discharge of the aquifer system is of two types, one is subterranean enough to connect to the deep aquifer [14]. The discharge of the aquifer system is of two types, one discharge into the sea and the other is artificial pumping. Groundwater is extracted from aquifers is subterranean discharge into the sea and the other is artificial pumping. Groundwater is extracted through production wells for water supply, such as agricultural irrigation, together with industrial from aquifers through production wells for water supply, such as agricultural irrigation, together and residential water [2,4]. In addition, we also collected the land use map, see Figure2c, to better with industrial and residential water [2,4]. In addition, we also collected the land use map, see interpret the relationship between land subsidence and land use in the following discussion [3]. Figure 2c, to better interpret the relationship between land subsidence and land use in the following discussion [3].

3. Data Coverage and Method

3.1. Data Coverage and InSAR Processing Appl. Sci. 2017, 7, 466 4 of 18

3. Data Coverage and Method

3.1. Data Coverage and InSAR Processing Here we use the SAR data from the L-band JERS, ALOS, and the C-band ENVISAT satellites to map the line-of-sight (LOS) surface deformation. There are three kinds of SAR images, including 42 ALOS1 images acquired from two parallel ascending-orbit tracks from December 2006 to July 2010, 22 JERS images acquired from October1992 to September 1998, and 23 ENVISAT images acquired from October 2003 to May 2007 (See Figure1 and Table S1–S3). Those InSAR datasets provide deformation of a 200 × 100 km area covering the whole peninsula (see Figure1). Detailed SAR data information is shown in Table S1–S3. These SAR data are processed by the conventional two-pass differential interferometry methods of GAMMA software package [17]. We applied a multilook operation of 6 × 16 pixels for the ALOS1, JERS data and 4 × 20 for the ENVISAT data in the range and azimuth directions to reduce the phase noise as well as calculation burdens before the phase unwrapping. Using the 90-m resolution SRTM v4.2 digital elevation model, we removed the phase relating to topography. The refined ESA VOR orbits were used to remove the phase ramps for the ENVISAT acquisitions. However, there was no precise orbit for the JERS or ALOS1 data. So we adopted a refined method based on fringe frequency to correct the phase ramp in differential interferograms [18]. Before phase unwrapping, a modified Goldstein filter was used to reduce noise phases [19] and a manual mask applied to eliminate its impact on phase unwrapping from some low coherence areas such as isolated islands and sea areas. In some studies, ionospheric disturbance error of SAR interferogram also needed to be considered [20,21], but we did not find any SAR images contaminated by ionospheric disturbances in this study. Finally, we used minimum cost flow algorithm to unwrap pixels whose coherence value was over a threshold (0.7). Combination of ascending and descending modes can be used to separate vertical and horizontal ground motions. However, here only the ascending ALOS1 or the descending JERS and ENVISAT acquisitions can be used in every time span. Thus vertical and horizontal components of the deformation are impossible to be retrieved independently [22]. Our results show that most ground displacement rates are <30 mm/year and the vertical subsidence is the main deformation component [23]. Thus, we assumed that the horizontal ground displacement is negligible and projected the results in LOS direction to vertical direction by dividing by the cosine of the incidence angles of each platform.

3.2. Modified Stacking Method The stacking technique can mitigate atmospheric delays and measure uniform surface deformation by averaging data with multiple interferograms [24]. However, there are very limited images for every frame (≤11 images) in this study (see Table S1–S3). What is worse, lots of SAR images have significant atmospheric delays due to strong evaporation in the LZP, especially in summer. For example, four of the 10 observations have serious atmospheric effects for track466/frame400 of ALOS1 acquisitions (see Figure S1 and Table S1). In order to make full use of all available images in the stacking process, we firstly obtained unwrapped differential interferogram pairs frame by frame. Secondly, we manually identified the images with serious atmospheric delays and then used pair wise logic instead of the traditional stacking method which combine pairs in a chronological order. It means every identified image should be treated as master and slave in turn for the final differential pairs used, see Figure S2. This process can not only reduce the atmospheric effect but also strengthen the effective value decided by mean coherence in the combined interferogram. Figure3 shows the differential interferogram of ALOS1 before and after using this strategy and the coherence histograms of Figure3c,d are also calculated in Figure3e. We can find that the mean coherence of stacking in Figure3c is obviously better than that of Figure2d obtained by direct interferometry (see Figure3e). The atmosphere effects of image 20090209 represented by black dashed lines in Figure3a,b are also mitigated in Figure3c by the combination of Figure3a,b. Moreover, the cumulative perpendicular baseline of each frame should Appl. Sci. 2017, 7, 466 5 of 18 be close to zero to reduce topography error in the final stacking results. Finally, we calculated the linearAppl. Sci. displacement 2017, 7, 466 rate with the ratio between the cumulative unwrapped phase and cumulative5 time. of 18 It should be noted that the method above is different from the one proposed by Biggs [25], in which everyevery SAR SAR acquisition acquisition can can be be usedused only only twicetwice throughthrough the the chainchain stacking.stacking. WeWe used used SAR SAR images images with with atmosphereatmosphere contaminated contaminated twice twice only only if if the the frequency frequency of of master master and and slave slave were were the the same. same. In In addition, addition, toto analyze analyze the the development development of subsidenceof subsidence in urban in urban areas moreareas scientifically, more scientifically, we also used we aalso traditional used a SBAS-InSARtraditional SBAS to calculate‐InSAR to the calculate deformation the deformation time-series intime the‐series capital in city the ofcapital LZP, city Zhanjiang of LZP, city Zhanjiang [12]. city [12].

FigureFigure 3. 3.Atmospheric Atmospheric delay delay and and differential differential interferograms interferograms before andbefore after and pair after wise logicpair calculation.wise logic (calculation.a) interferogram (a) interferogram between image between 20090209 image and 20070204;20090209 (andb) interferogram 20070204; (b between) interferogram image 20100212 between andimage 20090209; 20100212 (c) and stacking 20090209; interferogram (c): stacking after interferogram summing Figure after3a,b, summing aiming to Figure reduce 3a the and atmosphere Figure 3b, effectsaiming of to image reduce 20090209; the atmosphere (d) interferogram effects betweenof image image 20090209; 20100212 (d) andinterferogram 2070204 after between using directimage interferometry;20100212 and 2070204 (e) the coherenceafter using histogram direct interferometry; of Figure3c,d. (e) Blackthe coherence dashed lines histogram in Figure of Figure3a,b show 3c–d. areasBlack with dashed serious lines atmospheric in Figure 3a–b delays show and areas red with rectangles serious in atmospheric Figure3c,d represent delays and zones red selectedrectangles for in coherenceFigure 3c–d comparison. represent zones Red andselected blue for solid coherence lines in comparison. Figure3e represent Red and the blue coherence solid lines histogram in Figure of 3e Figurerepresent3c,d. the (Date coherence format: histogram yyyymmdd). of Figure 3c,d. (Date format: yyyymmdd).

3.3.3.3. UncertaintyUncertainty EstimationEstimation The cumulative phase∑K ∅is the sum of phases, including surface deformation∑K ∅,, The cumulative phase ∑i=1 ∅i is the sum of phases, including surface deformation ∑i=1 ∅de f ,i, topography errors∑K ∅ , atmospheric delays∑K ∅ , satellite orbit errors ∑K ∅ , and topography errors ∑= ∅,topo,i, atmospheric delays ∑= ∅,atm,i, satellite orbit errors ∑ = ∅,orb,i, and i 1 i 1 i 1 thermal noise∑K ∅, in the stacking process: thermal noise ∑i=1 ∅noise,i in the stacking process:

∑ ∅ K∑ ∅ K ∑ ∅ K ∑ ∅ K ∑ ∅ K∑ ∅ , K ∑i=1∅i =,∑i=1 ∅de f ,i,+ ∑i=1 ∅topo,,i + ∑i=1 ∅atm,,i + ∑i=1∅orb,,i + ∑i=1 ∅noise,i, ⋅∆ (1) ≅ ⋅=∼ −∑4π V · K t −∑4π·∆z K∑B ∅ + K ∑ ∅ + K∑ ∅ + , K 1) λ ∑i=1 i λRsinθ ∑i=1⊥i,∑i=1 ∅atm,,i ∑i=1∅orb,,i ∑i=1 ∅noise,i, where is the unwrapped phase, K is the number of differential interferograms, t and B⊥ are the where∅ ∅i is the unwrapped phase, K is the number of differential interferograms,i andi are timethe time span span and and perpendicular perpendicular baseline baseline of one of interferogram,one interferogram,λ is the is wavelength the wavelength of SAR of SAR images, images,R is the is distance the distance between between satellite satellite and groundand ground and ∆andz is ∆ the is residue the residue DEM DEM error. error. In order In order to obtain to obtain the precisethe precise surface surface deformation, deformation, the potential the potential spurious spurious phase, phase, i.e., orbital i.e., orbital errors, errors, topographic topographic errors, errors, and atmosphericand atmospheric delays, delays, should should be taken be intotaken account into account and minimized. and minimized. The orbital The error orbital can beerror corrected can be bycorrected a 2-D linear by a or 2‐ quadraticD linear or model quadratic in unwrapped model in interferograms unwrapped interferograms [15] The atmospheric [15] The delays atmospheric relating todelays topography relating canto topography be neglected can in be this neglected study duein this to study the flat due terrain, to the andflat terrain, the variance and the of variance neutral atmosphericof neutral atmospheric delay has been delay mitigated has been by mitigated our modified by our stacking modified method. stacking The method. topographic The topographic error also canerror be also neglected can be here neglected because here the because cumulative the cumulative perpendicular perpendicular baseline is closebaseline to 0 is m close [26], to which 0 m [26], can causewhich an can error cause of 0.6an mm,error 1.5of 0.6 mm, mm, and 1.5 3.5 mm, mm and with 3.5 a 5mm m DEM with errora 5 m for DEM JERS, error ENVISAT for JERS, and ENVISAT ALOS1 and ALOS1 respectively. We used several reference points to calibrate the results of three different platforms, assuming these points are stable with respect to the main movement areas. So we obtained the mean velocity () after considering the source errors listed above. Appl. Sci. 2017, 7, 466 6 of 18 respectively. We used several reference points to calibrate the results of three different platforms, assuming these points are stable with respect to the main movement areas. So we obtained the mean Appl.velocity Sci. 2017 (V), 7 after, 466 considering the source errors listed above. 6 of 18

λ K K V= − ∑ ∅ ∅/ ∑i/ , ti,(2) (2) 4π ∑i=1 ∑i=1 WeWe calculatedcalculated the the variances variances of of residual residual error error in eachin each selected selected interferogram interferogram and theseand these results results (up to (up99, 183to 99, and 183 140 and mm 1402 for mm² ENVISAT, for ENVISAT, JERS and JERS ALOS1) and ALOS1) indicated indicated that the atmosphericthat the atmospheric delay is sensitivedelay is 2 −dmn/α sensitivein this study in this (see study Figure (see S3). Figure So we S3). adopted So we a adopted 1-D exponential a 1‐D exponential covariance covariance function Cov functionmn = σ e / 2 to estimate to the estimate uncertainty the uncertainty of mean velocity of mean [27 velocity], where [27],σ is where the variance, is the Covvariance,mn is the covariance, is the covariance,dmn is the distance is between the distance pixels, between and α is pixels, a variable and window is a variable size of window each acquisition size of each named acquisition e-folding namedwavelength. e‐folding Figure wavelength.4 shows the Figure 1-D 4 covariance shows the function 1‐D covariance of deformation function velocityof deformation in three velocity different in threesensors different and their sensors variance and for their JERS, variance ENVISAT, for andJERS, ALOS1 ENVISAT, are 3.0, and 5.1, ALOS1 and 4.1 are mm 3.0,2/year 5.1,2 withand the4.1 mm²/year²e-folding wavelengths with the e‐folding as 2.0, wavelengths 2.5, and 8.8 km as respectively.2.0, 2.5, and 8.8 Furthermore, km respectively. we also Furthermore, utilized the residualwe also utilizederror between the residual the estimated error between velocity the and estimated the single velocity interferograms and the single to estimate interferograms velocity uncertaintyto estimate velocityand found uncertainty that both methodsand found gave that very both similar methods results. gave So very the similar uncertainty results. we estimatedSo the uncertainty in this study we estimatedis considered in this to be study reliable. is considered to be reliable.

FigureFigure 4. Exponential covariance covariance function function of of velocity velocity of of each each interferogram interferogram before before (black) (black) and and after after (red)(red) atmosphereatmosphere phase phase screen screen (APS) (APS) correction correction with with our our modified modified stacking stacking method. method. (a–c) show(a–c) resultsshow resultsof the frame of the covering frame covering Zhanjiang Zhanjiang city with city JERS, with ENVISAT, JERS, ENVISAT, and ALOS1, and respectively. ALOS1, respectively. Every black Every curve blackrepresents curve the represents statistical the result statistical of an interferogramresult of an interferogram in the stacking in the for stacking each platform. for each platform.

4.4. Results: Results: Spatial—Temporal Spatial—Temporal Evolution of Subsidence

4.1.4.1. Rates Rates of of Land Land Subsidence Subsidence TheThe averageaverage vertical vertical velocity velocity map map of theof LZPthe LZP was generatedwas generated by the by processes the processes described described in Section in3 Sectionwith three 3 with kinds three of SAR kinds images of SAR acquired images fromacquired 1992 from to 2010. 1992 They to 2010. are They shown are in shown Figure in5 withFigure the 5 withorder the of theirorder acquisition of their acquisition times. Figure times.5a showsFigure the5a velocityshows the map velocity of JERS. map The of main JERS. subsidence The main subsidenceareas distribute areas along distribute the coastline along the (the coastline black dashed (the black elongated dashed lines). elongated The peaklines). land The subsidencepeak land subsidenceoccurred in occurred the southeast in the of southeast Zhanjiang of withZhanjiang a maximum with a maximum velocity 26 velocity mm/year. 26 mm/year. In urban areasIn urban the areasland subsidencethe land subsidence occurred mainlyoccurred in Zhanjiangmainly in city,Zhanjiang with a city, velocity with of a 6–15 velocity mm/year. of 6–15 Figure mm/year.5b is Figurethe velocity 5b is map the velocity of ENVISAT. map of Compared ENVISAT. with Compared the JERS with data, the the JERS ENVISAT data, the data ENVISAT has a much data largerhas a muchcoverage larger but withcoverage comparatively but with poorercomparatively coherence poorer due to coherence the serious due temporal to the decorrelation serious temporal of the decorrelationC-band satellite. of the There C‐band are three satellite. obvious There subsidence are three areas obvious shown subsidence in Figure areas5b, such shown as the in ZhanjiangFigure 5b, suchcity representedas the Zhanjiang by red city dashed represented rectangle, by and red two dashed insets rectangle, named ‘Nansan and two Island’ insets andnamed ‘ZhaoZiLin’. ‘Nansan Island’The maximum and ‘ZhaoZiLin’. subsidence The rate maximum (19 mm/year) subsidence appears rate in the (19 coastal mm/year) areas appears of Nansan in the Island. coastal In additionareas of Nansanto the land Island. subsidence, In addition there to are the also land 1 to 2subsidence, mm/year upliftsthere are in the also southwest 1 to 2 mm/year of the LZP. uplifts For the in land the southwest of the LZP. For the land subsidence from 2006 to 2010, we collected the ALOS1 images to map. As shown in Figure 5c, the main subsidence zones also distribute along the coastline (about 19 mm/year) defined by a black dashed line. Another feature of the subsidence is its discrete distribution in agricultural areas, see Figure 2c, such as farmlands in Leizhou and Suixi. Land uplift was observed in the southwest of the peninsula ranging from 1–3 mm/year, in accordance with results of ENVISAT data. Appl. Sci. 2017, 7, 466 7 of 18 subsidence from 2006 to 2010, we collected the ALOS1 images to map. As shown in Figure5c, the main subsidence zones also distribute along the coastline (about 19 mm/year) defined by a black dashed line. Another feature of the subsidence is its discrete distribution in agricultural areas, see Figure2c, such as farmlands in Leizhou and Suixi. Land uplift was observed in the southwest of the peninsula ranging from 1–3 mm/year, in accordance with results of ENVISAT data. Appl. Sci. 2017, 7, 466 7 of 18

Figure 5. Average subsidence velocity of the LZP from 1992 to 2010. (a) Mean velocity map of JERS from October 1992 to September 1998; (b) mean velocity map of ENVISATFigure from 5. OctoberAverage 2003 to subsidence May 2007; (c): mean velocity velocity of map the of ALOS1 LZP from December 1992 to 2006 2010. to July (2010.a) Mean Positive velocityvelocities (yellow) map represent of JERS uplift, from while the negative values (purple) indicate subsidence. Brown squares are representative of city’s locations. The red dashed rectangle is the representative subsidence area coveringOctober Zhanjiang 1992 city to shown September in Figure 7. 1998;Rectangle (b A,) meanB, C, D, and velocity E in Figure map 5c are of five ENVISAT representative from areas in October Figure 10. 2003The locations to May are abbreviated 2007; as follows:(c) mean the Leizhou velocity Bay (L.Z.Bay), map ofthe ALOS1Zhanjiang Bay from (Z.J.Bay). December 2006 to July 2010. Positive velocities (yellow) represent uplift, while the negative values (purple) indicate subsidence. Brown squares are representative of city’s locations. The red dashed rectangle is the representative subsidence area covering Zhanjiang city shown in Figure 7. Rectangle A, B, C, D, and E in Figure5c are five representative areas in Figure 10. The locations are abbreviated as follows: the Leizhou Bay (L.Z.Bay), the Zhanjiang Bay (Z.J.Bay).

Based on the results in Figure5, we can conclude that the major area of the LZP was stable during the period from 1992 to 2010, in spite of a small scale land subsidence in the urban and agricultural areas along the coastline. The subsidence in agricultural areas was moderate with a rate of ~19 mm/year. Furthermore, the subsidence area was spreading instead of being still during the image coverage time. For example, the subsidence center in the overlapped coverage area of JERS and ALOS1 was moving from the northeast of Xuwen to the junction areas of Leizhou and Xuwen. Also, the subsidence rate and range in urban areas were changing. This is further discussed in Section 4.2. The subsidence in coastal areas, increased both in rate and range, especially the area marked with black dashed lines (from 8–24 to 12–32 mm/year) in Figure4a,c. This subsidence area is general consistent with that of the compressible deposit (Q4) distribution in Figure2a. In order to quantitatively analyze the relationship between the depth of compressible sediments and land subsidence [28], we generated the distribution of recent soft soil thickness of the Holocene age with 22 boreholes by the method of [29], see Figure2b. The statistical results of mean subsidence rate and pixel numbers of ALOS1 are shown in Figure6 due to a more complete coverage compared with the other two satellites. We can see that the majority of the area is underlain by soft-soil thicknesses of 15–20 m, and the mean vertical velocity is growing with the increase of thickness, which agrees with the results of [30], i.e., maximum average deformation velocities are found on the thickest soft soils (>35 m), conversely, the mean velocity values calculated for the thinnest soft soils are satisfactorily within the precision range. The relationship between these two factors can help to interpret the mechanism of land subsidence. Appl. Sci. 2017, 7, 466 8 of 18

Based on the results in Figure 5, we can conclude that the major area of the LZP was stable during the period from 1992 to 2010, in spite of a small scale land subsidence in the urban and agricultural areas along the coastline. The subsidence in agricultural areas was moderate with a rate of ~19 mm/year. Furthermore, the subsidence area was spreading instead of being still during the image coverage time. For example, the subsidence center in the overlapped coverage area of JERS and ALOS1 was moving from the northeast of Xuwen to the junction areas of Leizhou and Xuwen. Also, the subsidence rate and range in urban areas were changing. This is further discussed in Section 4.2. The subsidence in coastal areas, increased both in rate and range, especially the area marked with black dashed lines (from 8–24 to 12–32 mm/year) in Figure 4a,c. This subsidence area is general consistent with that of the compressible deposit (Q4) distribution in Figure 2a. In order to quantitatively analyze the relationship between the depth of compressible sediments and land subsidence [28], we generated the distribution of recent soft soil thickness of the Holocene age with 22 boreholes by the method of [29], see Figure 2b. The statistical results of mean subsidence rate and pixel numbers of ALOS1 are shown in Figure 6 due to a more complete coverage compared with the other two satellites. We can see that the majority of the area is underlain by soft‐soil thicknesses of 15–20 m, and the mean vertical velocity is growing with the increase of thickness, which agrees with the results of [30], i.e., maximum average deformation velocities are found on the thickest soft soils (>35 m), conversely, the mean velocity values calculated for the thinnest soft soils are satisfactorily withinAppl. Sci. the2017 ,precision7, 466 range. The relationship between these two factors can help to interpret8 ofthe 18 mechanism of land subsidence.

Figure 6.6. RelationshipRelationship between between soft soft soil soil thickness thickness and meanand mean vertical vertical velocity. velocity. The white The bars white represent bars representthe number the of number pixels andof pixels the red and diamonds the red diamonds represent represent the calculated the calculated mean velocity mean velocity for each for soft each soil softthickness soil thickness interval. interval.

4.2. Land Land Subsidence in the Urban Areas Zhanjiang is is the largest city, city, the political and economic center of of the the LZP, LZP, with a population of 7.45 millionmillion until until 2009. 2009. It has It ahas long a historylong history of land subsidence,of land subsidence, but only a fewbut levelingonly a measurementsfew leveling measurementsin the LZP from in 1984 the to LZP 2001, from which 1984 can to help 2001, validate which the can InSAR help measurement. validate the Hence,InSAR wemeasurement. selected the Hence,area (latitude: we selected 21.105 the◦–21.330 area ◦(latitude:, longitude: 21.105°–21.330°, 110.281◦–110.435 longitude:◦) with historic 110.281°–110.435°) leveling measurements with historic as levelingthe experiment measurements area. The as surfacethe experiment deformation area. ofThe the surface chosen deformation area is shown of the in Figure chosen7a–c. area We is shown chose innine Figure representative 7a–c. We subsidencechose nine bowlsrepresentative with a segmentation subsidence bowls based with on a a threshold segmentation of mean based velocity on a threshold(0.9 mm/year), of mean marked velocity by red (0.9 dashed mm/year), lines inmarked Figure 7by, to red demonstrate dashed lines the in spatial-temporal Figure 7, to demonstrate features of theland spatial subsidence.‐temporal The features land use of of land this city subsidence. is showed The by land the inset use inof Figurethis city2c, is where showed we canby the see thatinset six in Figurebowls (bowl2c, where 1, 2, we 5, 6, can 7, and see 8)that are six in bowls the urban (bowl areas, 1, 2, and 5, 6, two 7, and bowls 8) are (bowl in the 3 and urban 4) are areas, in the and nearby two bowlsagricultural (bowl area 3 and of 4) Xiashan are in andthe nearby Mazhang. agricultural Bowl 9 appeared area of Xiashan in the suburb and Mazhang. area. The Bowl subsidence 9 appeared areas inof the these suburb nine bowls area. The are calculatedsubsidence and areas shown of these in Table nine1 bowls. are calculated and shown in Table 1.

Table 1. Area of subsidence bowls in Figure7.

B1 (U) B2 (U) B3 (A) B4 (A) B5 (U) B6 (U) B7 (U) B8 (U) B9 (S) Data Areas: km2 (Bowl: B, Urban area: U, Agricultural area: A, Suburban area; S) JERS 3.3 1.9 1.8 1.6 2.3 9.7 1.4 2.1 0.9 ENVISAT - - - - - 10.7 2.8 10.8 4.3 ALOS1 3.0 0.8 4.8 6.1 2.0 6.1 2.7 6.1 12.2 Appl. Sci. 2017, 7, 466 9 of 18

Appl. Sci. 2017, 7, 466 9 of 18

Figure 7.FigureThe spatial-temporal7. The spatial‐temporal evolution evolution of land of subsidence land subsidence in Zhanjiang in Zhanjiang city. (city.a) Average (a) Average subsidence rate of JERSsubsidence from rate Octorber of JERS 1992 from to Octorber September 1992 1998;to September (b) average 1998; subsidence(b) average subsidence rate of ENVISAT rate of from ENVISAT from October 2003 to May 2007; (c) average subsidence rate of ALOS1 from December October 2003 to May 2007; (c) average subsidence rate of ALOS1 from December 2006 to July 2010. 2006 to July 2010. Zhanjiang city consists of four districts, namely Chikan, Xiashan, Potou, and Zhanjiang city consists of four districts, namely Chikan, Xiashan, Potou, and Mazhang. Black dashed Mazhang. Black dashed lines indicate the district border. Areas marked with red dashed lines are lines indicateland subsidence the district bowls; border. Black Areas circles marked are the leveling with red stations dashed and lines two areblack land solid subsidence lines are selected bowls; Black circles areprofiles the leveling of the subsidence stations bowl and twoby aa’ black and bb’. solid lines are selected profiles of the subsidence bowl by aa’ and bb’. Table 1. Area of subsidence bowls in Figure 7.

In additionData to theB1 (U) average B2 (U) subsidence B3 (A) rate,B4 (A) we usedB5 (U) SBAS-InSARB6 (U) B7 to(U) generate B8 (U) theB9 deformation (S) Areas: km² (Bowl: B, Urban area: U, Agricultural area: A, Suburban area; S) time-series of these bowls. Four bowls with large subsidence rates, i.e., bowls 1, 4, 8, and 9, were JERS 3.3 1.9 1.8 1.6 2.3 9.7 1.4 2.1 0.9 selected toENVISAT analyze ‐ ‐ ‐ ‐ ‐ the development of subsidence in the city, see10.7 Figure 82.8. Pixels 10.8 located at4.3 or near the central ofALOS1 the bowl observed3.0 by0.8 each satellite4.8 were6.1 selected2.0 and6.1 plotted2.7 in Figure6.1 7. In12.2 urban areas, such as bowl 1 and bowl 8, the deformation time-series approximates to linear, as observed in other In addition to the average subsidence rate, we used SBAS‐InSAR to generate the deformation areas [10]. The continual groundwater extraction will change the stress between confined aquifers time‐series of these bowls. Four bowls with large subsidence rates, i.e., bowls 1, 4, 8, and 9, were and leadselected to the to compaction analyze the development of the compressible of subsidence sediments, in the city, resulting see Figure finally 8. Pixels in located the land at or subsidence. near However,the the central magnitudes of the bowl of subsidenceobserved by each rate insatellite these were two selected bowls are and different. plotted in BowlFigure 1 7. experienced In urban two major changes,areas, such the as subsidence bowl 1 and bowl rate first8, the slowed deformation down time from‐series 13 (Figureapproximates8e) to to 4 linear, mm/year as observed (Figure in8 i) and then increasedother areas from [10]. 4 (FigureThe continual8e) to 29groundwater mm/year extraction (Figure8 willm), change whereas the bowl stress 8 between increased confined during the periodsaquifers (Figure 8andg,k,o). lead Unliketo the compaction the results of of the bowls compressible 1 and 8, sediments, the subsidence resulting of finally bowl 9in is the complicated. land subsidence. However, the magnitudes of subsidence rate in these two bowls are different. Bowl 1 In the first stage (Figure8h), the deformation time-series trend is nonlinear, which is mainly caused experienced two major changes, the subsidence rate first slowed down from 13 (Figure 8e) to 4 by the seasonalmm/year groundwater(Figure 8i) and then pumping increased for from irrigation 4 (Figure or 8e) aquaculture to 29 mm/year along (Figure the 8m), coastline. whereas For bowl example, the uplift8 increased trends occur during at the summer periods (May(Figure 1995, 8g,k,o). July Unlike 1995, the and results August of bowls 1998) 1 and while 8, the the subsidence fall trends of occur at springbowl (February 9 is complicated. 1997 and In March the first 1998). stage (Figure The fall 8h), trends the deformation occurring time at non-irrigation‐series trend is nonlinear, periods in this case maywhich be caused is mainly by caused the delay by the from seasonal aquifer groundwater compaction pumping to deformation. for irrigation or During aquaculture the second along stage the coastline. For example, the uplift trends occur at summer (May 1995, July 1995, and August 1998) (Figure8l,p), the deformation time-series is close to linear. The main reason for this phenomenon is while the fall trends occur at spring (February 1997 and March 1998). The fall trends occurring at the exploitationnon‐irrigation of groundwater periods in this from case the may deep be aquifercaused ofby thethe newdelay waterworks from aquifer named compaction Lindong to from 1999 [31],deformation. aiming to During mitigate the the second over-pumping stage (Figure 8l,p), groundwater the deformation in urban time‐ areas.series is More close detailsto linear. about The bowl 9 are depictedmain reason by the for profile this phenomenon bb' in Figure is the9b exploitation showing the of continuedgroundwater development from the deep of aquifer subsidence of the rates. Similarly,new bowl waterworks 4 in the named agricultural Lindong from area 1999 experienced [31], aiming a to nonlinear mitigate the deformation. over‐pumping Moregroundwater importantly, the periodic deformation time-series of Figure8f,j,n are observed, indicating the seasonal groundwater pumping for irrigation. Appl. Sci. 2017, 7, 466 10 of 18

in urban areas. More details about bowl 9 are depicted by the profile bbʹ in Figure 9b showing the continued development of subsidence rates. Similarly, bowl 4 in the agricultural area experienced a Appl.nonlinear Sci. 2017 deformation., 7, 466 More importantly, the periodic deformation time‐series of Figure 8f,j,n10 of are 18 observed, indicating the seasonal groundwater pumping for irrigation.

FigureFigure 8.8. Deformation time time-series‐series of of central central points points in in four four selected selected bowls, bowls, i.e., i.e., bowl bowl 1, 4, 1, 8, 4, and 8, and 9, for 9, foreach each platform. platform. The The first first row row (a– (ad–)d represents) represents the the location location of of central central points points with with optical backgrounds derivedderived fromfrom GoogleEarth.GoogleEarth. The The circles, diamonds, and and triangles represent the location of JERS, ENVISAT,ENVISAT, andand ALOS1ALOS1 respectively.respectively. BlackBlack wordswords aroundaround thethe opticaloptical imagesimages areare thethe adjacentadjacent landland classes.classes. The The red red star star in in (d ()d is) isa waterworks a waterworks named named Lindong Lindong started started in 1999; in 1999; The Thesecond second to fourth to fourth rows rowsrepresent represent the deformation the deformation time‐series time-series of four of points four points for JERS, for ENVISAT, JERS, ENVISAT, and ALOS1. and ALOS1. The red The arrows red arrowsrepresent represent the movement the movement direction direction of land of landsubsidence. subsidence. The dashed The dashed lines lines in (e in–p (e) –arep) arethe the results results of oflinear linear fitting fitting of the of the deformation deformation time time-series.‐series. The word The word ‘V=’ in ‘V=’ (e–p) in is (e the–p) mean is the velocity mean velocity in the vertical in the verticaldirection. direction.

WeWe alsoalso calculatedcalculated thethe spatial-temporalspatial‐temporal evolution of land subsidence area showed in Table 11.. FiveFive bowlsbowls (Bowl(Bowl 1,1, 2,2, 3,3, 4,4, andand 5)5) ofof ENVISATENVISAT datadata areare excludedexcluded duedue toto theirtheir unclear boundaries. TheThe subsidencesubsidence areaarea ofof bowlsbowls decreaseddecreased withwith timetime inin thethe urbanurban area,area, forfor instanceinstance itit decreaseddecreased fromfrom 3.33.3 toto 3.03.0 kmkm²2 inin bowlbowl 1,1, fromfrom 1.91.9 toto 0.80.8 kmkm²2 inin bowlbowl 2,2, fromfrom 10.810.8 toto 6.16.1 kmkm²2 inin thethe newnew bowlbowl mergedmerged fromfrom bowlbowl 66 andand bowlbowl 8.8. Conversely, inin suburbsuburb andand agriculturalagricultural areas,areas, thethe subsidencesubsidence areaarea increasedincreased continuously:continuously: from 0.9 toto 4.34.3 andand thenthen toto 12.212.2 kmkm²2 inin bowlbowl 9,9, fromfrom 1.81.8 toto 4.84.8 kmkm²2 inin bowlbowl 33 andand fromfrom 1.61.6 toto 6.16.1 kmkm²2 inin bowlbowl 44 (see(see TableTable1 1).). In In order order to to analyze analyze the the spatial spatial evolution evolution of of bowl bowl 6 6 and and bowl bowl 8, 8, anotheranother profile profile aa’ aa’ is is plotted plotted in in Figure Figure9. 9. We We can can see see that that the the two two separated separated bowls bowls (red (red line line in Figure in Figure9a) merged9a) merged into into one one bowl bowl (green (green line line in in Figure Figure9a), 9a), and and then then became became two two connected connected bowls bowls (blue (blue line line inin FigureFigure9 a)9a) from from a a to to a’, a’, which which can can also also be be visually visually observed observed in in Figure Figure7a–c. 7a–c. In In addition addition to to the the subsidencesubsidence area,area, thethe centralcentral bowlbowl selectedselected forfor bowlsbowls 1,1, 4,4, 8,8, andand 99 aforementionedaforementioned isis regardedregarded asas thethe movementmovement indicationindication inin thisthis study.study. TheThe resultsresults areare shownshown inin FigureFigure8 8a–da–d with with red red arrows, arrows, we we can can see see thatthat thethe surfacesurface subsidencesubsidence extendsextends fromfrom urbanurban areas areas to to suburban suburban and and agricultural agricultural areas. areas. Appl. Sci. 2017, 7, 466 11 of 18 Appl. Sci. 2017, 7, 466 11 of 18

Figure 9. Two bowl subsidence profiles (aa’ and bb’) in Figure 7. (a) aa’ profile going through Figure 9. Two bowl subsidence profiles (aa’ and bb’) in Figure7.( a) aa’ profile going through subsidence subsidence bowl 6 and bowl 8 in the urban area. Two separated bowls, i.e., Bowl 6 and Bowl 8, bowl 6 and bowl 8 in the urban area. Two separated bowls, i.e., Bowl 6 and Bowl 8, represented by red represented by red merged into one bowl represented by green, and then became two connected mergedbowls into represented one bowl by represented blue; (b) bb’ byprofile green, going and through then becamesubsidence two bowl connected 9 in the suburban bowls represented area, the by blue;mean (b) bb’ velocity profile grows going with through increase subsidence of time. The bowl red, 9green in the and suburban blue solid area, lines the represent mean velocitythe mean grows withvelocity increase of of JERS, time. ENVISAT The red, and green ALOS1 and datasets. blue solid lines represent the mean velocity of JERS, ENVISAT and ALOS1 datasets. It is known that subsidence in Zhanjiang city is the result of over‐extraction of groundwater for domestic and industrial use [4,32]. Groundwater exploitation from middle and deep aquifers started It is known that subsidence in Zhanjiang city is the result of over-extraction of groundwater for in the1980s and accelerated in the 1990s with the development of the economy and urbanization. domesticHowever, and as industrial our InSAR use measurements [4,32]. Groundwater suggest from exploitation 1992 to 2010 from the subsidence middle and rate deep decreased aquifers after started in the1980sthe government and accelerated prohibited inpumping the 1990s groundwater with the in development the urban areas of [32] the (see economy Figure 7). and Meanwhile, urbanization. However,the subsidence as our InSAR rate started measurements increasing suggestin the suburban from 1992 areas to 2010as the the groundwater subsidence was rate extracted decreased to after the governmentsupply the growing prohibited population pumping and groundwater expanding industries in the urban in the areasurban [ 32areas] (see accompanied Figure7). Meanwhile,by the the subsidencemoving of the rate waterworks. started increasing Besides, inwe the take suburban three small areas cities as of the the groundwater LZP into measurement, was extracted Leizhou to supply the growingcity, Xuwen population County, andand expandingSuixi County. industries Unlike Zhanjiang in the urban city, areas the subsidence accompanied in these by the cities moving is of the waterworks.comparatively Besides, small with we the take average three smallsubsidence cities rate of the 0.6, LZP0.7, and into 1.5 measurement, mm/year, respectively. Leizhou city, , and Suixi County. Unlike Zhanjiang city, the subsidence in these cities is comparatively small 4.3. Land Subsidence in Agriculture Areas with the average subsidence rate 0.6, 0.7, and 1.5 mm/year, respectively. Besides urban areas, there is also some subsidence in agricultural areas caused by natural and 4.3. Landanthropogenic Subsidence processes. in Agriculture Here Areas we only analyze the result of ALOS1 data because of the decorrelation and uncompleted coverage of ENVISAT and JERS data in those areas. From Figure 5c, weBesides find that: urban (1) there areas, is therea wide is range also of some discrete subsidence subsidence in in agricultural the arable land areas with caused the rate by of natural 10–19 and anthropogenicmm/year, such processes. as the junction Here we area only between analyze Xuwen the and result Leizhou, of ALOS1 the southwest data because of Leizhou; of the (2) decorrelation rapid andsubsidence uncompleted occurred coverage along of the ENVISAT coastline andas shown JERS by data the in black those dashed areas. elongated From Figure circles.5c, The we peak find that: (1) theremaximum is a wide subsidence range in of this discrete area is subsidence up to 32 mm/year. in the arableIn order land to get with more the details, rate of we 10–19 chose mm/year, five suchrepresentative as the junction area between denoted Xuwenby red rectangles and Leizhou, (A–E) the in southwestFigure 5c to of analyze Leizhou; the (2) characteristics rapid subsidence occurredand causes along of the subsidence coastline (see as Figure shown 10). by the black dashed elongated circles. The peak maximum subsidence in this area is up to 32 mm/year. In order to get more details, we chose five representative regions denoted by red rectangles (A–E) in Figure5c to analyze the characteristics and causes of subsidence (see Figure 10). Appl. Sci. 2017, 7, 466 12 of 18

Appl. Sci. 2017, 7, 466 12 of 18

Figure 10. Average deformation rate of the five representative coastal areas for ALOS1 dataset from Figure 10. Average deformation rate of the five representative coastal areas for ALOS1 dataset from December 2006 to July 2010. (a)–(e) are enlarged views of the red rectangles in Figure 4c. (a) is an area Decembercovered 2006 with to large July‐scale 2010. farmlands, (a–e) the are maximum enlarged subsidence views of rate the is red 18 mm/year; rectangles (b) in is located Figure at4c. the ( a) is an areasouth covered of LZP withwith a large-scale maximum subsidence farmlands, rate the of maximum 13 mm/year subsidence and a uplift raterate of is 3 18 mm/year; mm/year; (c–e) ( b) is locatedare atareas the with south large of‐scale LZP aquaculture, with a maximum the maximum subsidence subsidence rate of rates 13 are mm/year 23, 20, and and 23 a mm/year, uplift rate of 3 mm/year;respectively. (c–e) are areas with large-scale aquaculture, the maximum subsidence rates are 23, 20, and 23 mm/year, respectively. Subsiding areas shown in Figure 10a are farmlands characterized with highly compressible deposits, consisted of clay, silt, and silty sand (see Figure 2a), which is vulnerable to land subsidence Subsiding areas shown in Figure 10a are farmlands characterized with highly compressible due to sediment loading [8,33,34]. Shallow aquifers are the main supplier of these agricultural areas, deposits,which consisted can be recharged of clay, silt, seasonally and silty from sand surface (see Figurerunoff,2 a),such which as reservoirs is vulnerable and canals. to land However, subsidence due toduring sediment the observation loading [8 ,period,33,34]. the Shallow LZP experienced aquifers are lots the of main droughts, supplier resulting of these in the agricultural continuous areas, whichdeclining can be recharged of the shallow seasonally aquifers’ from level surface [35]. Moreover, runoff, suchunguided as reservoirs drilling, such and as canals. mixed However, exploitation during the observationof water from period, shallow the‐ and LZP deep experienced‐aquifers [35], lots and of droughts, illegal drilling resulting both accelerated in the continuous the shrinking declining of of the shallowgroundwater aquifers’ level level and [ 35caused]. Moreover, a wide range unguided of land drilling, subsidence. such The as mixed bowl exploitationin Figure 10a of (marked water from shallow- and deep-aquifers [35], and illegal drilling both accelerated the shrinking of groundwater level and caused a wide range of land subsidence. The bowl in Figure 10a (marked with a red solid line) has an average velocity of 13 mm/year over an area of 2.2 km2. Coincidently, a collapse happened Appl. Sci. 2017,, 7,, 466 13 of 18 with a red solid line) has an average velocity of 13 mm/year over an area of 2.2 km². Coincidently, a collapsein this area happened in March in this 2008 area [35 ].in FigureMarch 112008a,b,d [35]. show Figure the 11a,b,d differential show subsidence the differential of a subsidence building and of a pumpingbuilding and facility a pumping resulting facility from over-exploitationresulting from over of‐ groundwaterexploitation of in groundwater this area. A in wide this range area. ofA widecrops range were soakingof crops inwere the soaking flood because in the flood of the because rapid subsidenceof the rapid in subsidence the farmland in the (see farmland Figure 11(seec). FigureIn addition, 11c). In the addition, bowl 4 in the Figure bowl7 4c ofin ALOS1Figure 7c is of also ALOS1 located is also in the located agricultural in the agricultural area, see the area, optical see theimage optical of Figure image8b, of whose Figure deformation 8b, whose deformation time-series (Figuretime‐series8n) trend (Figure presents 8n) trend an obvious presents seasonality. an obvious Theseasonality. seasonal The deformation seasonal deformation is mainly caused is mainly by irrigation caused by for irrigation agricultural for use.agricultural use.

Figure 11. Field photos of non‐uniform subsidence of a building (a) and a pump facility (b), cracking Figure 11. Field photos of non-uniform subsidence of a building (a) and a pump facility (b), cracking and settlement of a channel (d) caused by over‐exploiting groundwater in Fucheng town. (c) Flooded and settlement of a channel (d) caused by over-exploiting groundwater in Fucheng town. (c) Flooded crops caused by rapid subsidence. crops caused by rapid subsidence. Some uplift areas on the southwest coast of the LZP are shown in Figure 10b. These uplifts may Somebe caused uplift by areas the onintermittent the southwest tectonic coast movement. of the LZP The are shownuplift rate in Figure is about 10b. 3 These mm/year, uplifts bigger may thanbe caused the results by the (~2 intermittent mm/year) tectonicobtained movement. from the tide The station uplift [23]. rate isThis about uncertain 3 mm/year, level is bigger consistent than withthe results the uncertainty (~2 mm/year) of ALOS1 obtained measurement from the tide in stationthis study [23 ].(~2.0 This mm/year). uncertain levelFigure is 10c–e consistent represent with landthe uncertainty subsidence of areas, ALOS1 such measurement as tidal flats in this and study marshes, (~2.0 mm/year).distributed Figurealong 10thec–e coastline represent of land the peninsulasubsidence which areas, are such covered as tidal with flats compressible and marshes, deposits distributed of the along Quaternary the coastline age of(see the Figure peninsula 2a). whichThese areas are covered are dominated with compressible by aquaculture deposits facilities, of the such Quaternary as shrimp age ponds (see Figure and sunfish2a). These ponds areas and are a fewdominated farmlands, by aquaculture greatly dependent facilities, on such fresh as water, shrimp see ponds the insets and sunfish of Figure ponds 5c. The and coast a few zone farmlands, in the LZPgreatly is the dependent most important on fresh aquaculture water, see thebase, insets providing of Figure aquatic5c. The products coast to zone the in Pearl the LZPRiver is Delta the marketsmost important and many aquaculture countries base,in America providing and aquaticEurope products[1]. The great to the exploitation of Delta groundwater markets andhas ledmany to countriesthe groundwater in America level and drop below [ 1the]. Thesea level, great which exploitation in turn of has groundwater caused shoreline has led erosion to the andgroundwater saltwater level intrusion drop [1,32,36]. below the sea level, which in turn has caused shoreline erosion and saltwater intrusion [1,32,36]. 4.4. Validations 4.4. Validations The leveling was conducted only in Zhanjiang city from 1989 to 2001 [2]. We collected five The leveling was conducted only in Zhanjiang city from 1989 to 2001 [2]. We collected five stations’ stations’ leveling data with the uncertainty of 1 mm/year (JC18, JC72, JC53, JC26, and JC55) from leveling data with the uncertainty of 1 mm/year (JC18, JC72, JC53, JC26, and JC55) from 1989 to 1999 in 1989 to 1999 in Figure 7. We compare the results of leveling and InSAR (JERS) data in the overlap time span. Because of their different resolution, we calculated the rate of leveling points by Appl. Sci. 2017, 7, 466 14 of 18

Figure7. We compare the results of leveling and InSAR (JERS) data in the overlap time span. Because of their different resolution, we calculated the rate of leveling points by estimating the mean value of a small square around them. The difference between leveling and InSAR observation displayed in Table2 are 2, 1.8, 2.4, 1.5, and 1.9 mm/year respectively, which is also consistent with our uncertainty estimated. In order to assess the relative precision among overlapped InSAR data, we selected points in the overlapped area at adjacent paths uniformly and calculated the root mean square (RMS) value of these misfits. The relative precision of JERS was not estimated due to the deficiency of the adjacent path. The relative precision of the ALOS1 and ENVISAT are 0.8 mm/year and 0.9 mm/year, which again confirmed the accuracy of our modified stacking method.

Table 2. Comparison between Interferometric Synthetic Aperture Radar (InSAR) and leveling results.

Comparison JC18 JC72 JC53 JC26 JC55 (mm/year) Leveling1989~1999 −5 −3.2 −3.6 −2.5 −3.1 InSAR_JERS1992~1998 −7 −5 −6 −4 −5 Difference 2 1.8 2.4 1.5 1.9

5. Discussion Before this study, land subsidence measurements were almost absent in the LZP. With three kinds of SAR images acquired from 1992 to 2010, we achieved the first deformation map of the whole peninsula with a modified stacking method. The InSAR measurement indicates that the land subsidence is mainly distributed along the coastline and the arable lands with the maximum velocity as 32 and 19 mm/year, respectively. The land subsidence evolution in Zhanjiang city moved from urban areas to suburban areas with an increasing trend along the coastal areas. We can say with proper evidence that the land subsidence in the LZP will continue if the groundwater extraction is not regulated. Similarly, land subsidence caused by over-extraction of groundwater occurs in many delta areas in China, such as the Yellow River Delta and the Yangtze River Delta [37,38]. Thus regular land subsidence monitoring using the InSAR technique is very necessary to regulate groundwater extraction and land subsidence. Land subsidence is a common problem in most peninsulas and delta areas over the world [39]. Similar to the LZP, over-exploitation of groundwater is also the main reason for land subsidence in the Chalco basin located southeast of Mexico City and Jakarta in Indonesia (see Table3). The obvious difference is the speed of groundwater extraction and the water level declining in the urban areas, although the use of groundwater in these cities is similar. The extraction speed and drawdown velocity of groundwater in the three cities are 7.8 m3/s, 1.5 m/year, 1.5 m3/s, 0.1–1.9 m/year and 6.7 m3/s, 0.1–0.8 m/year, respectively [4,8,10,40–42]. So we can see that the groundwater level in LZP is comparatively steady although the groundwater extraction speed is high. Interestingly, the maximum velocity of Mexico City and Jakarta exceeds that of the LZP by an order of magnitude (see Table3). Based on these differences, we can conclude that the following two factors will help understand the mechanism of surface deformation: (1) Geology and aquifers. The LZP has a wide range (~3694 km2) of a mixture of confined basalt and volcanic, which has stronger water storage capacity. The strong water storage capacity can promote leakage between aquifers. More importantly, lots of dead craters located at these areas with local depth up to 310 m, can directly reach to the deep aquifer and help leakage between the shallow and deep aquifers directly as described in Section2. Furthermore, the LZP has much thicker aquifer (500 m) than the other two places. (2) Compressible sediments determine the land subsidence velocity in agricultural areas. The subsidence of agricultural area in the LZP (<25 mm/year) is much smaller than that of the other two places (>100 mm/year), although the former has a larger agriculture area (see Table3). In Mexico City and Jakarta, the agricultural lands are mainly located above the highly compressible sediments, which leads to the compaction of aquifers and finally causes land subsidence. In the LZP, the compressible sediments are mainly distributed along the coastline Appl. Sci. 2017, 7, 466 15 of 18 with large thickness, whereas in agricultural lands, especially in Xuwen, the thickness is less than 10 m, see Figure2b. Furthermore, wide range arable lands overlie the thick basalt and volcanic whose compressibility is low. Because of this, the LZP has luckily enjoyed comparatively slow subsidence.

Table 3. Comparison of subsidence parameters in Mexico City [8,40,42], Jakarta of Indonesia [10,41], and the Leizhou Peninsula (LZP) [4].

Urban areas Mexico City (Chalco) Jakarta of Indonesia LZP Aquifers Lacustrine aquitard Upper aquifer (<40 m) Shallow aquifer (<30 m) Granular aquifer Middle aquifer (40 < d < 140 m) Middle aquifer (30 < d < 200 m) Volcanic aquitard Lower aquifer (140 < d < 250 m) Deep aquifer (200 < d < 500 m) (0 < d < 300m) Sediments Lacustrine deposits and Quaternary sediments overlying Unconsolidated alluvial and Alluvial deposits of Tertiary sediments lacustrine overlaying the Quaternary basement rocks of Neogene and Quaternary age. Ava. extraction speed 1990s: 7.8 m3/s 1995: 1.5 m3/s Until 2001: 6.7 m3/s Max. velocity <40 cm/year <22 cm/year <3.2 cm/year Water level decline 1.5 m/year 0.1–1.9 m/year 0.1–0.8 m/year speed Correlation with Thickness of deposits Land use Thickness of soft soil/Land use surface geology Agricultural areas Mexico City(Chalco) Jakarta of Indonesia LZP Max. velocity <18.4 cm/year 15.1 cm/year <2.5 cm/year Surface geology Compressible deposits Surficial deposits Volcanic basalt Correlation with Type of compressible Correlate with land use Thickness of soft soil geology deposits

Whether the location of maximum subsidence corresponds to the location of lowest groundwater level is an interesting question and it can help us understand the mechanism of land subsidence caused by groundwater pumping. The rapid declining level of the middle and deep aquifers can result in land subsidence due to the damage to the balance of the hydraulic structure [33,34]. In Zhanjiang city, the water level of middle-deep and deep aquifers started declining in the late 1980s and resulted in two obvious subsidence zones centered on [2,32] until 2001. The lowest point of groundwater of the first subsidence zone is generally consistent with the center of subsidence bowl 5 and 6 in Figure7a. For the second subsidence zone, the cone of bowl 1 in Figure7 is located within the place whose drawdown depth is less than 10 m during the observation time span [4], but not in the center. So we can conclude that the location of maximum land subsidence is often, though not always, associated with maximum decline of the groundwater level, whereas sometimes it may be located away from the center of the groundwater depression. The rapid land subsidence along the coastline reveals the potential disaster. The coastline of the LZP is long enough (1180 km) to support complex ecosystems, including mangroves, coral reefs, bedrocks, and fluvial deltas [1]. However, the rapid land subsidence caused by aquaculture facilities near the coastline not only leads to intrusion of the saltwater and coastline erosion but also damages the ecosystems [35,36]. Fortunately, owing to the slow tectonic uplift of the peninsula, until now only a small part of the coastline has undergone intrusion [23,32]. However, the land subsidence may spread to the rest of the coastal areas in future if not addressed. Furthermore the sea level of the Sea rises 2.6 mm/year [1] and it may aggravate the pollution of shallow-water near the coast. Taking no action, the land subsidence (32 mm/year) of coastline will reach 64 cm and the sea level will rise 5.2 cm in 20 years, which will lead to an area disappearance of 55,000 m2 for every 100 km coastline, and will devastate the ecosystem along the coastline as well as the aquaculture economy. Therefore, grasping the spatial patterns, e.g., the deformation time-series in a wide coverage, and progress of these hazards caused by natural and anthropogenic process, is particularly important to Appl. Sci. 2017, 7, 466 16 of 18 help the government plan aquaculture scientifically and mitigate hazards caused by groundwater over-exploitation. This is particularly important in China as it has a great national territory area, varied relief, complex climate, and rapid development of urbanization which will cause many kinds of hazards.

6. Conclusions We obtained the average subsidence rate map of the LZP from 1992 to 2010 with a modified stacking technology. A relationship between land subsidence and thickness of soft soil of the Holocene age was found. The main cause of land subsidence in this peninsula is the overexploitation of groundwater for different usages, domestic and industrial use in urban areas and agricultural use in arable and coastal areas. Land subsidence with a velocity up to 32 mm/year in Zhanjiang city is propagating from urban areas to suburban areas which is coincident with the variation of groundwater level. The subsidence velocity of a wide range of inland arable lands is mainly between 10 and 19 mm/year. In addition, continuous increasing land subsidence is observed in coastal areas between the Leizhou Bay and the Zhanjiang Bay with an average velocity over 19 mm/year, caused by rapid exploitation of groundwater for aquaculture. If not addressed, the subsidence will spread and evolve into more serious disasters such as saltwater intrusion. Our work can help guide the rational exploitation of groundwater, and urban planning as well as offer technical suggestions for government decision making.

Supplementary Materials: The following are available online at www.mdpi.com/2076-3417/7/5/466/s1, Figure S1: The atmospheric phase of ALOS1, track466 and frame400 in SAR coordination. The coverage of these images is represented by a black rectangle on the right-hand side in Figure1. The wrapped phase of the differential interferogram whose master and slave images are the same SAR image with serious atmospheric delay. For example, 20070905, 20080723, 20090910 and 2010313 for (a–d) respectively. The red dashed lines in (a–d) represent the obvious atmospheric delay of each SAR image, Figure S2: Perpendicular baselines and temporal intervals of the selected InSAR image pairs from three different sensors, JERS (a), ENVISAT (b) and ALOS-1 (c). The black dots and diamond represent SAR image contaminated with serious atmospheric delay used as a master and a slave, respectively, Figure S3: 1-D exponential covariance function of residual errors between the estimated displacement phase, calculated by the multiplication between velocity and time, and the single differential interferometric phase for each platform. The primary component in residual errors is caused by atmosphere, which can be estimated with 1-D exponential covariance function. a–c Show statistical results of the frame covered Zhanjiang city with JERS, ENVISAT and ALOS1, respectively. Every black curve in (a–c) represents a statistical result of residual error for a selected pair, Table S1: The parameters of the ascending ALOS-1 images selected, Table S2: The parameters of the descending JERS images selected, Table S3: The parameters of the descending ENVISAT images selected. Acknowledgments: The ENVISAT data were obtained through the ESA project (No. C1P.14838 and C1P.16734). The ALOS data and JERS data were provided by the Japan Aerospace Exploration Agency (JAXA) through projects P1229002 and P1390002. This work was supported by the National Natural Science Foundation of China [41574005], [41474007] and Shenghua Yuying fund of Central South University and The National Research Foundation for the Doctoral Program of Higher Education of China under Grant No. 20130162110015. Author Contributions: Yanan Du and Guangcai Feng performed the experiments and produced the results. Yanan Du drafted the manuscript. Xing Peng and Zhiwei Li contributed to the discussion of the results. All authors conceived the study and reviewed and approved the manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Li, Z.Q. Vulnerability of Ecological Environment in Leizhou Peninsula Coastal Zone and Counterneasures. In Proceedings of the 2012 2nd International Conference on IEEE Remote Sensing, Environment and Transportation Engineering (RSETE), , China, 1–3 June 2012. 2. Li, G.Q. Analysis of regional subsidence of Zhanjiang city. West-China Explor. Eng. 2009, 12, 108–110. (In Chinese) 3. Zhanjiang Municipal Bureau of Land and Resources (ZMBLR). Available online: http://www.zjlr.gov.cn/ newDetail-2000-20111204.htm (accessed on 17 January 2013). Appl. Sci. 2017, 7, 466 17 of 18

4. Zhou, X.; Yan, X.; Li, J.; Yao, J.M.; Dai, W.Y. Evolution of the groundwater environment under a long-term exploitation in the coastal area near Zhanjiang, China. Environ. Geol. 2007, 51, 847–856. 5. Zhang, L.; Lu, Z.; Ding, X.L.; Jung, H.S.; Feng, G.C.; Lee, C.W. Mapping ground surface deformation using temporarily coherent point SAR interferometry: Application to Los Angeles Basin. Remote. Sens. Environ. 2012, 117, 429–439. [CrossRef] 6. Yang, Z.; Li, Z.; Zhu, J.; Yi, H.; Hu, J.; Feng, G. Deriving dynamic subsidence of coal mining areas using InSAR and logistic model. Remote Sens. 2017, 9, 125. [CrossRef] 7. Du, Y.N.; Zhang, L.; Feng, G.C.; Lu, Z.; Sun, Q. On the Accuracy of Topographic Residuals Retrieved by MTInSAR. IEEE Trans. Geosci. Remote Sens. 2017, 55, 1053–1065. [CrossRef] 8. Chaussard, E.; Wdowins, S.; Cabral-Cano, E.; Amelung, F. Land subsidence in central Mexico detected by ALOS InSAR time-series. Remote Sens. Environ. 2014, 140, 94–106. [CrossRef] 9. Chen, J.; , J.C.; Zhang, L.N.; Zou, G.X.; Liu, G.X.; Zhang, R.; Yu, B. Deformation Trend Extraction Based on Multi-Temporal InSAR in Shanghai. Remote Sens. 2013, 5, 1774–1786. [CrossRef] 10. Chaussard, E.; Amelung, F.; Abidin, H.; Hong, S.H. Sinking cities in Indonesia: ALOS PALSAR detects rapid subsidence due to groundwater and gas extraction. Remote Sens. Environ. 2013, 128, 150–161. [CrossRef] 11. Ferretti, A.; Prati, C.; Rocca, F. Permanent scatterers in SAR interferometry. IEEE Trans. Geosci. Remote Sens. 2001, 39, 8–20. [CrossRef] 12. Berardino, P.; Fornaro, G.; Lanari, R.; Sansoto, E. A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms. IEEE Trans. Geosci. Remote Sens. 2002, 40, 2375–2383. [CrossRef] 13. Xu, B.; Feng, G.C.; Li, Z.W.; Wang, Q.J.; Wang, C.C.; Xie, R.A. Coastal Subsidence Monitoring Associated with Land Reclamation Using the Point Target Based SBAS-InSAR Method: A Case Study of , China. Remote Sens. 2016, 8, 652. [CrossRef] 14. Wen, H.H. Study on Circulation Pattern and Numerical Modeling of Groundwater Flow in Leizhou Peninsula. Ph.D. Thesis, China University of Geosciences, , China, 2013. 15. Zhou, X.; Chen, M.; Liang, C. Optimal schemes of groundwater exploitation for prevention of seawater intrusion in the Leizhou Peninsula in southern China. Environ. Geol. 2003, 43, 978–985. 16. Shen, J.H.; Wang, R.; Zhu, C.Q. Research on spatial distribution law of gray clays of Zhanjiang Formation. Rock Soil Mech. 2013, 34, 331–339. (In Chinese) 17. Wegmuller, U.; Werner, C.L. Gamma SAR processor and interferometry software. In Proceedings of the 3rd ERS Symposium Europe Space Agency Special, Florence, Italy, 14–21 March 1997; ESA SP-414: Auckland, New Zealand; pp. 1686–1692. 18. Ding, C.; Feng, G.C.; Li, Z.W.; Shan, X.J.; Du, Y.N.; Wang, H.Q. Spatio-Temporal Error Sources Analysis and Accuracy Improvement in Landsat 8 Image Ground Displacement Measurements. Remote Sens. 2016, 8, 937. [CrossRef] 19. Du, Y.N.; Xu, Q.; Zhang, L.; Feng, G.C.; Li, Z.W.; Chen, R.F.; Lin, C.W. Recent Landslide Movement in Tsaoling, Taiwan Tracked by TerraSAR-X/TanDEM-X DEM Time Series. Remote Sens. 2017, 9, 353. [CrossRef] 20. Feng, G.C.; Li, Z.W.; Xu, B.; Shan, X.J.; Zhang, L.; Zhu, J.J. Coseismic Deformation of the 2015 Mw 6.4 Pishan, China, Earthquake, Estimated from Sentinel1 and ALOS2 Data. Seismol. Res. Lett. 2016, 87, 800–806. [CrossRef] 21. Feng, G.C.; Li, Z.W.; Shan, X.J. Geodetic Model of the April 25, 2015 Mw 7.8 Gorkha Nepal Earthquake and Mw 7.3 Aftershock Estimated from InSAR and GPS Data. Geophys. J. Int. 2015, 203, 896–900. [CrossRef] 22. Wright, T.J.; Parsons, B.E.; Lu, Z. Toward mapping surface deformation in three dimensions using InSAR. Geophys. Res. Lett. 2004, 31, L07607. [CrossRef] 23. Xu, Q.H. A ground crack long itudinally penetrating the holocene sand dyke in Gangzi, Leizhou Peninsula is discovered. Quaty. Sci. 2008, 28, 712–720. (In Chinese) 24. Strozzi, T.; Wegmuller, U.; Werner, C.; Wiesmann, A. Measurement of slow uniform surface displacement with mm/year accuracy. In Proceedings of the IEEE 2000 International Geoscience and Remote Sensing Symposium, Honolulu, HI, USA, 24–28 July 2000; pp. 2239–2241. 25. Biggs, J.; Wright, T.; Lu, Z.; Parsons, B. Multi-interferogram method for measuring interseismic deformation: Denali Fault, Alaska. Geophys. J. Int. 2007, 170, 1165–1179. [CrossRef] 26. Du, Y.N.; Feng, G.C.; Li, Z.W.; Zhu, J.J.; Peng, X. Generation of high precision DEM from TerraSAR-X/TanDEM-X. Chin. J. Geophys. 2015, 58, 1634–1644. Appl. Sci. 2017, 7, 466 18 of 18

27. Parsons, B.; Wright, T.; Rowe, P.; Andrews, J.; Jackson, J. The 1994 Sefidabeh (eastern Iran) earthquake revisited: New evidence from satellite radar interferometry and carbonatedating about the growth of an active fold above a blind thrust fault. Geophys. J. Int. 2006, 164, 202–217. [CrossRef] 28. Cigna, F.; Osmano˘glu,B.; Cabral-Cano, E.; Dixon, T.H.; Ávila-Olivera, J.A.; Garduño-Monroy, V.H.; Wdowinski, S. Monitoring land subsidence and its induced geological hazard with Synthetic Aperture Radar Interferometry: A case study in Morelia, Mexico. Remote Sens. Environ. 2012, 117, 146–161. [CrossRef] 29. Bonì, R.; Herrera, G.; Meisina, C.; Notti, D.; Béjar-Pizarro, M.; Zucca, F.; Fernández-Merodo, J.A. Twenty-year advanced DInSAR analysis of severe land subsidence: The Alto Guadalentín Basin (Spain) case study. Eng. Geol. 2015, 198, 40–52. [CrossRef] 30. Herrera, G.; Fernández, J.A.; Tomás, R.; Cooksley, G.; Mulas, J. Advanced interpretation of subsidence in Murcia (SE Spain) using A-DInSAR data-modelling and validation. Nat. Hazard. Earth Sys. 2009, 9, 647. [CrossRef] 31. Yu, J.; Ding, L.G.; Li, W.H.; Lu, W.F.; Cai, L.X. Characteristics of the evolution of groundwater flow field in Zhanjiang. Hydrogeol. Eng. Geol. 2006, 33, 95–98. (In Chinese) 32. Luo, Q.; Su, Z.H. The characteristic analysis and the prevention and control methods from the sea water invasion in Naozhou island. J. Geol. Hazards Environ. Preserv. 2007, 18, 28–32. (In Chinese) 33. Terzaghi, K. Erdbaumechanik auf Bodenphysikalischer Grundlage; Franz Deuticke: Wien, Austria, 1925; p. 399. 34. Hoffmann, J.; Howard, A.Z. Seasonal subsidence and rebound in Las Vegas Valley, Nevada, observed by synthetic aperture radar interferometry. Water Resour. Res. 2001, 37, 1551–1566. [CrossRef] 35. Li, W.H. Causes and prevention of ground subsidence in Shaoshan village, Leizhou city. West-China Explor. Eng. 2011, 5, 121–124. (In Chinese) 36. Zhou, P.P.; Li, G.M.; Li, M.; Dong, Y.H. Numerical modeling of tidal effects on groundwater in the coastal aquifer of Donghai Island. New Front. Eng. Geol. Environ. Springer Geol. 2013, 9, 131–134. 37. Higgins, S.; Overeem, I.; Tanaka, A.; Syvitski, J.P.M. Land subsidence at aquaculture facilities in the Yellow River delta, China. Geophys. Res. Lett. 2013, 40, 3898–3902. [CrossRef] 38. Shi, X.Q.; Xue, Y.Q.; Wu, J.C.; Ye, S.J.; Zhang, Y.; Wei, Z.X.; Yu, J. Characterization of regional land subsidence in , China: The example of Su-Xi-Chang area and the city of Shanghai. Hydrogeol. J. 2008, 16, 593–607. [CrossRef] 39. Syvitski James, P.M. Sinking deltas due to human activities. Nat. Geosci. 2009, 2, 681–686. [CrossRef] 40. Carrera-Hernandez, J.J.; Gaskin, S.J. The basin of Mexico aquifer system: Regional groundwater level dynamics and database development. Hydrogol. J. 2007, 15, 1577–1590. [CrossRef] 41. Abidin, H.Z.; Andreas, H.; Gumilar, I.; Fukuda, Y.; Pohan, Y.E.; Deguchi, T. Land subsidence of Jakarta (Indonesia) and its relation with urban development. Nat. Hazards 2011, 59, 1753–1771. [CrossRef] 42. Ortega-Guerrero, A.; Rudolph, D.L.; Cherry, J.A. Analysis of long-term land subsidence near Mexico City: Field investigations and predictive modeling. Water Resour. Res. 2010, 35, 3327–3341. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).