3D surface velocities and strain rate fields for the central-eastern Altyn Tagh fault (NW Tibet)

This dataset includes surface velocities (at 1 km resolution) and strain rate fields for a total area of ~ 600,000 km2 (~ 1,300 km × 450 km, Longitude 84°E-100°E, Latitude 36°N-42°N) around the central and eastern segment of the Altyn Tagh fault using Sentinel-1 InSAR rate maps. This dataset can be used to study interseismic strain accumulation and its termination on the central-eastern Altyn Tagh fault and surface deformation of the northwestern Tibet.

We used the LiCSAR system (Lazeckỳ et al., 2020) to derive interferogram networks. We processed 11 LiCSAR frames on 7 ascending tracks and 9 LiCSAR frames on 6 descending tracks. For each frame, we used ~170 acquisition epochs between October 2014 and July 2022. To reduce the impact of phase biases and nontectonic seasonal signals, we combine both short temporal (< 4 months) and 1-year to 7-year long summer-to-summer baseline interferograms in the network, which generates an average of nearly 2000 interferograms in each LiCSAR frame.

We carried out time-series analysis using LiCSBAS (Morishita et al., 2020), during which the tropospheric phases for each epoch were removed based on the Generic Atmospheric Correction Online Service (GACOS) (Yu et al., 2018). As the remaining interferograms may still have unwrapping errors, we chose to reduce the impact of such errors by nullifying (removing the values) all the displacements of interferograms associated with an unclosed loop in the time series for each pixel. After obtaining displacement time series of each pixel in a frame, the average linear velocities were calculated based on the standard approach in LiCSBAS. Final line-of-sight (LOS) velocity uncertainties are calculated from the standard deviation (STD) of 100 velocities based on resampled datasets of displacement time series using the bootstrap method. We masked relatively unreliable pixels using several noise evaluation indices such as the average coherence and number of network gaps for each pixel (Morishita et al., 2020).

We derived reference frame adjustment parameters for each frame using VELMAP (Wang & Wright, 2012; Wang et al., 2019) to tie all the LOS velocity maps to the same Eurasian reference system as the GNSS data. We decomposed the final LOS velocity field accounting for the local radar incidence and azimuth angle. We inverted for the east and vertical velocity , as well as their associated uncertainties, for each pixel based on a method (Watson et al., 2022) in which the solution of north velocities from the VELMAP inversion was projected in local LOS direction and subtracted from the original LOS velocities first.

We calculated our horizontal strain rates follow the method from Ou et al. (2022), in which the horizontal strain rates are calculated based on the median filtered east velocities at the InSAR resolution and interpolated north velocities from GNSS (the north velocities from VELMAP in this study).

本数据集基于Sentinel-1合成孔径雷达干涉测量（Sentinel-1 InSAR）速率图，涵盖了阿尔金断裂带中东段周边约60万平方千米（约1300千米×450千米，经度84°E~100°E，纬度36°N~42°N）区域的地表流速（分辨率为1千米）与应变率场。该数据集可用于研究阿尔金断裂带中东段的震间应变积累及其终止过程，以及西藏西北部的地表形变。

本研究采用LiCSAR系统（Lazeckỳ等，2020）构建干涉图网络。共处理了7条升轨上的11景LiCSAR影像帧，以及6条降轨上的9景LiCSAR影像帧。每景影像帧使用2014年10月至2022年7月间约170个采集时段的数据。为削弱相位偏差与非构造季节性信号的影响，本研究在网络中同时结合了短时间基线（<4个月）与1~7年长时间夏到夏基线的干涉图，最终每景LiCSAR影像帧平均生成近2000幅干涉图。

本研究使用LiCSBAS工具（Morishita等，2020）开展时间序列分析，在此过程中基于通用大气校正在线服务（Generic Atmospheric Correction Online Service, GACOS）（Yu等，2018）去除每个采集时段的对流层相位延迟。由于剩余干涉图仍可能存在解缠误差，我们通过将每个像素时间序列中未闭合环路对应的干涉图位移全部置零（移除其数值），以削弱此类误差的影响。获取单景影像帧中每个像素的位移时间序列后，基于LiCSBAS的标准方法计算平均线性流速。最终的视线向（line-of-sight, LOS）流速不确定性，通过自举法（bootstrap method）对位移时间序列进行重采样得到100组流速，再基于其标准差（standard deviation, STD）计算得出。本研究使用平均相干性、每个像素的网络间隙数量等多项噪声评估指标，对相对不可靠的像素进行掩膜处理（Morishita等，2020）。

本研究使用VELMAP工具（Wang & Wright, 2012; Wang等，2019）为每景影像帧推导参考框架校正参数，以将所有视线向流速图统一至与全球导航卫星系统（GNSS）数据一致的欧亚参考框架。结合雷达局部入射角与方位角，对最终视线向流速场进行分解。基于Watson等（2022）提出的方法，对每个像素的东向与垂向流速及其相关不确定性进行反演：首先将VELMAP反演得到的北向流速投影至局部视线向，再从原始视线向流速中减去该投影值。

本研究参照Ou等（2022）的方法计算水平应变率场：基于InSAR分辨率下经过中值滤波的东向流速，以及从GNSS数据插值得到的北向流速（本研究中采用VELMAP反演得到的北向流速）进行计算。