震后GPS坐标时序中对数弛豫时间的估计

全球定位系统（global positioning system, GPS）坐标时序去除同震形变和震前稳态速度场后,采用加权非线性最小二乘估计震后对数弛豫时间,可更准确地提取震后对数弛豫项,从而可以分析震后弛豫项对测站位移的独立物理贡献,并为震后余滑和黏滞性松弛效应等现象的分析提供参考。以日本2009—2019年GPS坐标时序为例,估计2011年Mw 9.0地震震后对数弛豫时间,发现不同站点的对数弛豫时间与其震中距关系显著,且服从高斯分布。据此,构建高斯函数加常数模型,可由震中距概略估计震后对数弛豫时间。高斯分布曲线的峰值、峰值位置、半宽度信息、最低位置分别为3.5 a、0 km、262 km、0.5 a,由此得出震后对数弛豫项影响时间大于0.5 a的站点主要集中在震中距约524 km范围内。震后弛豫效应区域分布的差异性显著,对数弛豫时间越长的区域,弛豫项水平位移表征越大,其中存在两个平均弛豫时间2.5 a的中心区域,与震后余滑的中心区域及时间相吻合。     

尼泊尔Mw 7.9级地震同震垂直位移与断层运动模型

2015年4月25日尼泊尔地区发生了Mw 7.9级地震,发震断层位于印度板块与欧亚板块碰撞边界带,此次地震是一次典型的板块逆冲型事件。利用中国境内加密的GPS同震观测资料,融合ALOS-2卫星L波段的InSAR（interferometric synthetic aperture radar）同震形变数据,基于最小二乘方法获得了此次地震的同震垂直位移场。同震垂直位移结果表明,此次地震造成尼泊尔加德满都地区抬升约0.95 m,珠穆朗玛峰地区受地震的影响有所下降,其主峰的沉降量为2~3 cm,中国境内的希夏邦马主峰沉降约为20 cm。地区利用改进的二维弹性半空间位错模型反演了发震断层运动参数,本文模型显示此次地震的断层面破裂宽度约为60 km,平均滑动量达到4 m,相当于Mw 7.89级。     

The July 11, 1995 Myanmar–China earthquake: A representative event in the bookshelf faulting system of southeastern Asia observed from JERS-1 SAR images

On July 11, 1995, an Mw 6.8 earthquake struck eastern Myanmar near the Chinese border; hereafter referred to as the 1995 Myanmar–China earthquake. Coseismic surface displacements associated with this event are identified from JERS-1 (Japanese Earth Resources Satellite-1) SAR (Synthetic Aperture Radar) images. The largest relative displacement reached 60 cm in the line-of-sight direction. We speculate that a previously unrecognized dextral strike-slip subvertical fault striking NW–SE was responsible for this event. The coseismic slip distribution on the fault planes is inverted based on the InSAR-derived deformation. The results indicate that the fault slip was confined to two lobes. The maximum slip reached approximately 2.5 m at a depth of 5 km in the northwestern part of the focal region. The inverted geodetic moment was approximately Mw = 6.69, which is consistent with seismological results. The 1995 Myanmar–China earthquake is one of the largest recorded earthquakes that has occurred around the “bookshelf faulting” system between the Sagaing fault in Myanmar and the Red River fault in southwestern China.     

InSAR数据约束下2016年和2022年青海门源地震震源参数及其滑动分布

中国青海省门源县于2016年和2022年分别发生了Mw 5.9和Mw 6.7地震,相距不足40 km。利用欧洲空间局Sentinel-1A升降轨雷达影像,采用合成孔径雷达干涉测量(interferometric synthetic aperture radar, InSAR)技术分别获取两次地震的同震地表形变场,进而利用弹性半空间的位错模型确定上述事件的震源参数,基于分布式滑动模型反演确定两次地震断层面上的滑动分布,并探讨2016年门源地震对2022年门源地震的发震影响及触发机制。结果表明,2016年门源地震为逆冲型地震,并未破裂到地表,升、降轨同震形变场沿视线向的最大形变量分别为6.7 cm和7.0 cm,断层的最大滑动量为0.53 m,主要集中在地下4~12 km区域滑动。2022年门源地震同震形变场沿NWW-SEE向破裂,降轨影像最大视线向地表形变量为78 cm,断层的最大滑动值达到3.5 m,处于地下4 km左右,断层滑动分布模型揭示此次地震为左旋走滑型地震;结合冷龙岭断裂的运动性质和几何特征,可初步判定发震断层主要为冷龙岭断裂的西段、且极有可能破裂到了其西北端西侧的托莱山断裂。静态库仑应力触发关系显示,2016年门源地震对2022年门源地震的发生有一定的促进作用。     

门源Mw5.9级地震形变InSAR观测及区域断裂带深部几何形态

利用Sentinel-1A卫星升轨、降轨合成孔径雷达影像数据,提取了2016年门源Mw5.9级地震的高精度合成孔径雷达干涉同震形变场,利用单纯形法和非负最小二乘法反演确定了地震断层几何和滑动分布,并构建了区域断裂带的深部几何形态模型。结果表明,门源Mw5.9级地震同震形变以地表抬升为主,沿升轨、降轨视线向的最大值分别为5.3 cm、7.1 cm;地震断层走向、倾角分别为133°、43°;地震滑动以逆冲为主,主要发生在地下6.14~12.28 km处,最大滑动量约0.5 m,平均滑动角为66.85°,地震矩为1.0×1018 N·m（Mw5.94）;形变观测拟合残差均方根为0.36 cm;区域断裂带的深部几何形态以花状构造为特征,整体倾向南西,门源地震发震断裂为花状构造中未出露地表的盲断层。相关成果能够为研究区域地壳运动与变形、活动断裂与地震孕育发生等提供参考。     

台湾集集大地震断层非均匀滑动分布的反演

利用台湾1999集集大地震前后GPS测量获得的地表位移数据,反演断层面上的非均匀滑动分布.反演中采用二次样条函数作为滑动分布的基函数,通过引入阿卡克贝叶斯准则,获得了反演问题的惟一解.反演得到的地表位移与实测位移的差在几个厘米,较好地解释了观测到的GPS位移.另一方面,用反演结果计算得到的地表形变场同INSAR得到的地表实测位移场较相近.     

联合地震位错模型和InSAR数据构建2017年九寨沟Mw 6.5地震同震三维形变场

利用哨兵（Sentinel）-1A卫星升、降轨影像,在地震位错模型约束下获取了2017年九寨沟Mw 6.5地震的高质量三维形变场。首先,利用合成孔径雷达干涉测量技术（interferometric synthetic aperture radar,InSAR）提取九寨沟地震升、降轨同震形变场;然后,通过“两步法”反演获取该地震发震断层的几何参数和分布式滑动模型,以此为约束,采用方差分量估计算法联合解算九寨沟地震三维形变场。结果表明,九寨沟地震同震三维形变场以水平位移为主,垂向形变较弱;南北向形变呈拉张趋势,断层上盘向南、下盘向北滑动,最大位移分别为－19.81 cm和14.38 cm;东西向形变不对称性明显,断层上盘西北部向东水平运动,最大位移为18.37 cm,下盘东南部向西运动,最大位移不足8 cm。将南北、东西向形变与6个全球导航卫星系统（global navigation satellite system,GNSS）台站观测数据进行比较,两者一致性较好且均方根误差较小,分别为1.44 cm和1.77 cm,表明联合升、降轨InSAR观测和地震位错模型约束构建同震三维形变场方法具有较高可行性,显著降低了大地测量数据不足、InSAR观测对南北向形变不敏感等问题的影响。     

汶川地震震后GNSS形变分析

采用国际通用高精度数据处理软件对1999—2015年汶川地震震后龙门山断裂两侧分布的109个GNSS连续站和区域站资料进行数据处理与分析,得到各测站点水平向坐标时间序列。考虑2013年芦山地震同震位移影响,通过时间序列分析,获取震后该区域109个测站在2010—2015年间累计震后形变场,断裂带上盘靠近震中的近场区域,水平向最大震后位移达到5~7 cm。基于震后黏弹性松弛模型,通过格网搜索,根据3种不同破裂模型分别反演龙门山断裂上盘中上地壳最佳弹性层厚度估值与下地壳/上地幔黏弹性层最佳黏滞系数,获取3种不同破裂模型在2010—2015年间震后形变观测值与模拟值分布。根据模型二反演得到的模型参数,推估汶川地震在未来几十年内对周边区域地壳形变影响,其中仅2018—2058年40年尺度累计震后位移最大能达到19 cm。     

2015年尼泊尔地震同震滑动及震后余滑的三角位错模型反演

2015年尼泊尔Mw 7.9级地震发生在印度板块向欧亚板块低角度俯冲的喜马拉雅断裂带上。对该地震的滑动模型进行精化,对于理解喜马拉雅断裂带的孕震模式具有重要意义。采用三角形位错元构建主喜马拉雅断裂“双断坡”几何模型,联合全球定位系统（Global Positioning System,GPS）和合成孔径雷达干涉测量（interferometric synthetic aperture radar,InSAR）资料反演2015年尼泊尔地震同震滑移及震后余滑。结果表明,尼泊尔地震最大同震滑移达到7.8 m,深度为15 km,位于中地壳断坡和浅层断坪的接触部位。不考虑中地壳断坡结构会使反演的最大滑移量偏低。震后余滑主要分布在同震破裂区北侧,释放的地震矩为1.02×1020 N·m,相当于一次Mw 7.3级地震,约占主震释放地震矩的12%。同震库伦应力变化和震间断层闭锁分布均表明,尼泊尔地震破裂区南部宽约60 km的区域仍具有较高的地震危险性。     

Subsidence in the Kathmandu Basin,before and after the 2015 Mw 7.8 Gorkha Earthquake,Nepal Revealed from Small Baseline Subset-DInSAR Analysis

Land subsidence in densely urbanized areas is a global problem that is primarily caused by excessive groundwater withdrawal. The Kathmandu Basin is one such area where subsidence due to groundwater depletion has been a major problem in recent years. Moreover, on 25 April 2015, this basin experienced large crustal movements caused by the Gorkha earthquake (Mw 7.8). Consequently, the effects of earthquake-induced deformation could affect the temporal and spatial nature of anthropogenic subsidence in the basin. However, this effect has not yet been fully studied. In this paper, we applied the SBAS-DInSAR technique to estimate the spatiotemporal displacement of land subsidence in the Kathmandu Basin before and after the Gorkha earthquake, using 16 ALOS-1 Phased Array L-band Synthetic Aperture Radar (PALSAR) images during the pre-seismic period and 26 Sentinel-1 A/B SAR images during the pre- and post-seismic periods. The results showed that the mean subsidence rate in the central part of the basin was about ?8.2 cm/year before the earthquake. The spatial extents of the subsiding areas were well-correlated with the spatial distributions of the compressible clay layers in the basin. We infer from time-series InSAR analysis that subsidence in the Kathmandu basin could be associated with fluvio-lacustrine (clay) deposits and local hydrogeological conditions. However, after the mainshock, the subsidence rate significantly increased to ?15 and ?12 cm/year during early post-seismic (108 days) and post-seismic (2015–2016) period, respectively. Based on a spatial analysis of the subsidence rate map, the entire basin uplifted during the co-seismic period has started to subside and become stable during the early-post-seismic period. This is because of the elastic rebound of co-seismic deformation. However, interestingly, the localized areas show increased subsidence rates during both the early-post- and post-seismic periods. Therefore, we believe that the large co-seismic deformation experienced in this basin might induce the local subsidence to increase in rate, caused by oscillations of the water table level in the clay layer.     

Very high-rate GPS for measuring dynamic seismic displacements without aliasing: performance evaluation of the variometric approach

Very high-rate global positioning system (GPS) data has the capacity to quickly resolve seismically related ground displacements, thereby providing great potential for rapidly determining the magnitude and the nature of an earthquake’s rupture process and for providing timely warnings for earthquakes and tsunamis. The GPS variometric approach can measure ground displacements with comparable precision to relative positioning and precise point positioning (PPP) within a short period of time. The variometric approach is based on single-differencing over time of carrier phase observations using only the broadcast ephemeris and a single receiver to estimate velocities, which are then integrated to derive displacements. We evaluate the performance of the variometric approach to measure displacements using 50 Hz GPS data, which were recorded during the 2013 M_W 6.6 Lushan earthquake and the 2011 M_W 9.0 Tohoku-Oki earthquake. The comparison between 50 and 1 Hz seismic displacements demonstrates that 1 Hz solutions often fail to faithfully manifest the seismic waves containing high-frequency seismic signals due to aliasing, which is common for near-field stations of a moderate-magnitude earthquake. Results indicate that 10–50 Hz sampling GPS sites deployed close to the source or the ruptured fault are needed for measuring dynamic seismic displacements of moderate-magnitude events. Comparisons with post-processed PPP results reveal that the variometric approach can determine seismic displacements with accuracies of 0.3–4.1, 0.5–2.3 and 0.8–6.8 cm in the east, north and up components, respectively. Moreover, the power spectral density analysis demonstrates that high-frequency noises of seismic displacements, derived using the variometric approach, are smaller than those of PPP-derived displacements in these three components.     

2022年泸定Mw 6.6地震InSAR同震形变与滑动分布

2022-09-05,青藏高原东缘的鲜水河断裂上发生了泸定Mw 6.6地震,该地震是鲜水河断裂上40年来发生的最大地震,研究该地震的运动学和同震破裂模式对理解青藏高原东缘构造形变机制和评估鲜水河断裂以及安宁河断裂的地震危险性具有重要意义。利用Sentinel-1和ALOS-2卫星雷达影像,采用合成孔径雷达干涉技术获取了泸定地震的同震形变场,进而基于弹性半空间的位错模型,确定了本次地震发震断层的几何参数和滑动分布。结果表明,泸定地震是一次典型的左旋走滑事件,发震断层西倾,倾角约为72°,走向沿NNW-SSE方向,约为167°;断层破裂主要集中在0~10 km深度,最大滑动发生在约5.8 km深度,约为2.23 m;同震释放的地震矩约为8.74×1018 N·m,相当于矩震级Mw 6.59。通过对震后光学影像解译,发现此次地震诱发的滑坡多集中分布在发震断层西侧,该现象与余震主要集中在断层西侧的结果相一致,可认为是地震上盘效应的体现。     

地震破裂模型约束的中国阿里地震三维形变场

利用环境卫星（Environmental Satellite,Envisat）的升、降轨数据在地震破裂模型约束下获取2007年阿里地震的高质量同震地表三维形变场。首先,对合成孔径雷达（synthetic aperture radar,SAR）影像进行差分干涉处理,得到地震造成的视线向同震地表形变场;然后,以地震破裂模型为约束条件,采用赫尔默特方差分量估计法来解算阿里地震的高质量地表三维同震形变场。结果表明,震中区域最大下沉达约4.7 cm,东西向位移较小,南北向呈挤压趋势。总体上,三维形变的特征表明阿里地震是一个以正倾滑为主兼少量右旋走滑运动的地震事件。     

2008年大柴旦M_w6.3级地震的InSAR同震形变观测及断层参数反演

2008年11月10日,青海省大柴旦地区发生了Mw6.3级地震。本文利用EnviSat卫星升降轨SAR数据和差分干涉测量技术提取了同震形变场,基于均匀位错模型反演确定了地震断层参数,然后利用格网迭代搜索法确定了较优断层倾角,同时基于非均匀位错模型获得了精细滑动分布。结果表明,地震使得上盘区域沿降轨、升轨视线向分别产生最大约8.5cm、10cm的抬升;较优断层倾角为47.9°;地震滑动未延伸至地表,主要发生在地下8.2~23.7km深度范围内,最大和平均滑动量为0.5m和0.19m,平均滑动角为104.9°。反演的地震矩为3.74×1018 N·m,矩震级为Mw6.35。     

Sentinel-1 observations of the 2016 Menyuan earthquake: A buried reverse event linked to the left-lateral Haiyuan fault

Knowledge on the interaction of active structures is essential to understand mechanics of continental deformation and estimate the earthquake potential in complex tectonic settings. Here we use Sentinel-1A radar imagery to investigate coseismic deformation associated with the 2016 Menyuan (Qinghai) earthquake, which occurred in the vicinity of the left-lateral Haiyuan fault. The ascending and descending interferograms indicate thrust-dominated slip, with the maximum line-of-sight displacements of 58 and 68 mm, respectively. The InSAR observations fit well with the uniform-slip dislocation models except for a larger slip-to-width ratio than that predicted by the empirical scaling law. We suggest that geometric complexities near the Leng Long Ling restraining bend confine rupture propagation, resulting in high slip occurred within a small area and much higher stress drop than global estimates. Although InSAR observations cannot distinguish the primary plane, we prefer the west-dipping solution considering aftershocks distribution and the general tectonic context. Both InSAR modelling and aftershock locations indicate that the rupture plane linked to the Haiyuan fault at 10 km depth, a typical seismogenic depth in Tibet. We suggest that the earthquake more likely occurred on a secondary branch at a restraining bend of the Haiyuan fault, even though we cannot completely rule out the possibility of it being on a splay of the North Qilian Shan thrusts.     

汶川地震后龙门山断层中段愈合过程的GPS时间序列反演

利用汶川地震后、芦山地震前3年时间(2010.30—2013.30)的GPS连续站时间序列结果,采用TDEFNODE负位错反演程序,对龙门山断层中南段闭锁程度和滑动亏损速率动态演化进行了反演计算与分析,并讨论了汶川地震后断层中段的愈合过程和西南段的大震背景。结果表明:①汶川地震震中周边破裂区域的闭锁程度逐渐增强,由2010年基本处于蠕滑状态,至2013年处于较强闭锁状态;闭锁范围也逐渐增大,由汶川震中西南方向逐渐趋近震中位置,表明该部分断层正在快速愈合。汶川震中北东方向的大部分破裂区域依然处于蠕滑状态,表明该部分断层还未开始愈合。②芦山地震震中附近及其西南方向断层一直处于强闭锁状态,且完全闭锁区域的滑动亏损速率逐年减小,可能表明愈合区域分担了部分巴颜喀拉块体对四川盆地的挤压作用。以上结果表明,龙门山破裂断层不同段震后愈合过程和活动特征都有显著差异;处于强闭锁状态的龙门山断层西南段,在挤压弹性应变快速积累的情况下,发生大震的背景有所增强。     

Improved detection of earthquake-induced ground motion with spatial filter: case study of the 2012 M = 7.6 Costa Rica earthquake

High-rate GPS measurements of earthquake-induced strong crustal movements reveal important information on large amplitude displacements, which cannot be obtained by other seismic monitoring equipment. However, obtaining accurate measurements of these strong movements can be challenging, because large magnitude earthquakes (M > 7) affect a wide area surrounding the epicenter. As a result, the GPS recorded movements are calculated with respect to distant sites (relative positioning), or with satellite parameters estimated from distant sites (precise point positioning). In order to improve the accuracy of the strong motion GPS measurements, we developed a new method, based on a spatial filtering technique. The method calculates the displacement of a high-rate monitoring network with respect to a moving near field site and uses a stacking technique to remove the movements of the reference site from all the time series. We applied the new method to the analysis of 5 Hz data acquired by the Nicoya Peninsula network, which recorded strong crustal movements induced by the 2012, M = 7.6 Costa Rica earthquake. The results were successfully tested with respect to 1 Hz time series calculated with a far field reference site. The spatial filtering method also removes other systematic common noise from the time series, possibly due to atmospheric delay or orbital errors and, hence, produces more accurate solutions that those based on far fields sites, or on near field site experiencing earthquake-induced action.     

利用InSAR技术观测台湾花莲地震断层滑动与运动机理分析

基于Sentinel-1卫星升降轨SAR数据,采用D-InSAR技术提取了2018年台湾花莲县Mw6.4地震的同震形变场。结果表明,2018年花莲地震造成的最大地表形变量为38.2 cm,以隆升为主,断层上下盘最大相对位移为50 cm。利用InSAR观测得到的升降轨地表形变数据,分别构建2018年与2021年台湾花莲两次地震的断层三维滑动分布模型。结果表明,2018年花莲Mw6.4地震主震断层为靠近米伦断层的西倾隐伏断层,断层最大滑动量为1.8 m,以左旋走滑为主兼具少量逆冲分量,断层破裂传播至米伦断层西侧,影响了苓顶断层和米伦断层的地震活动性。2021年花莲Mw6.0地震发生在苓顶断层北段,断层最大滑动量为0.38 m,断层滑动以左旋走滑为主,两次地震事件均具有高倾角滑动特征。综合两次地震静态库仑应力的重新分布和M-T图发现,2018年花莲地震对2021年地震起触发作用,应力沿断层从高纬度向邻近低纬度传输累积,花莲地区及近海海域短周期内地震活动性仍强烈,主要表现为小震频发、中强震孕育周期短等特点。     

基于多源SAR数据的2022年门源Ms 6.9地震同震破裂模型反演研究

2022年1月8日青海省海北州门源县发生Ms 6.9地震,震中位于青藏高原东北缘祁连-海原断裂中段,属历史地震空区,基于多源合成孔径雷达（synthetic aperture radar,SAR）遥感数据研究该地震的破裂模式对理解青藏高原东北缘构造变形机制、应变释放过程以及地震危险性评估具有重要意义。首先利用Sentinel-1数据和合成孔径雷达差分干涉测量（differential interferometry synthetic aperture radar,D-InSAR）技术获取了门源地震的同震形变场,视线（line of sight,LOS）向形变场显示此次地震造成了约20 km长的地表破裂,最大形变约0.75 m;然后基于Sentinel-2卫星数据,利用光学影像配准和相关技术获取了本次地震的东西向同震形变场,最大同震位移达2.5 m;最后基于均匀弹性半无限位错模型,以LOS向形变场为约束反演了断层的滑动分布模型。结果显示,门源地震是一次典型的左旋走滑型地震,地震破裂主要集中在0~10 km深度范围,最大滑动量3.25 m,滑动角10.44°,对应深度4.89 km;反演给出的矩震量为1.07×1019 N·m,对应矩震级Mw 6.6。结合野外考察和地质资料,初步判定发震断裂为冷龙岭断裂,并引起托莱山断裂发生同震滑动。同震库仑应力结果显示,冷龙岭断裂东段和托莱山断裂西段应力状态为加载,未来具有发生强震的风险。     

东日本Mw9.0大地震前、同震及震后地壳水平运动

利用“中国地壳运动观测网络(二期)”245个GPS连续观测站及其他站的大量观测结果,采用离震中很远的7个GPS站作为位错参考框架,得到2011年3月11日东日本Mw9.0大地震的同震水平位移和应变。此种参考框架的位移场解,符合弹性位错模型离震中很远且水平位移接近于0的要求,不仅有利于位错模型反演,也有利于合理分析此次大地震同震水平位移的影响范围。地震前后区域参考框架下远场GPS连续观测位移时间序列和同震位移弹性位错模型反演结果表明,东日本大地震同震水平位移既是地震断层破裂的结果,也是远场震前数月或数年水平位移的弹性回跳,据此可以进一步确认这些站震前水平位移前兆异常。由于这些震前异常位移中亦包含了同震位移量(方向相反),因此,弹性位错模型为定量研究大地震前远场地壳运动异常提供了一种理论模型。位错参考框架中的同震位移及其位错模型和区域参考框架的位移时间系列的综合研究有助于对此次大地震地壳运动异常的认识。这种持续时间为数月至数年的弹性形变异常,可以为大地震的中短期预测提供依据。