Collaborative Research: Elements: Monitoring Earth Surface Deformation with the Next Generation of InSAR Satellites: GMTSAR

了解地震、火山和水文变化的过程及其相关危害是固体地球研究界和美国地质调查局的首要任务，因为如果没有做好充分准备，可能会造成社会混乱、经济后果，甚至可能造成人员伤亡。这不仅需要在社会相关时间范围内对此类过程及其危害进行长期估计，还需要评估人类活动对地球表面和内部的影响。这些估计和评估取决于准确测量地球表面随时间变化和变形的能力。例如，了解圣安德烈亚斯断层系统地震矩累积的速度非常重要，因为这将告诉我们预计下一次破坏性地震将在何时何地发生。这要求我们能够以0.5毫米/年的精度测量跨越数百公里的变形，分辨率优于10公里。干涉合成孔径雷达 (InSAR) 是完成这项关键任务的最佳技术，因为目前支持该技术的遥感卫星观测具有广泛的覆盖范围、低成本、不受天气影响且定期进行。然而，即将到来的新 InSAR 任务提出了新的挑战：如何有效处理急剧增加的数据量（NISAR 任务每天约 80 TB）。为了应对这一挑战，免费提供的 InSAR 处理软件 GMTSAR 正在开发强大而高效的方法，以充分利用卫星生成的数据进行科学研究和社会应用。该项目的主要创新点是启用云计算功能，转移到新一代编程语言，并不断吸引更多用户使用该软件构建自己的数据处理策略。开发人员将确保全球用户获得获得最先进处理技术所需的支持，并将继续改进文档、示例数据集和教程，以加强空间大地测量学领域的教育基础。干涉合成孔径雷达 (InSAR) 是一种强大的技术，用于测量地球表面的小位移（1-10 厘米），包括由构造载荷、地震、火山、山体滑坡、冰川、地面流体注入/抽取和地下核试验引起的位移。在过去的十年中，开发了一种免费的开源软件来利用这些有价值的数据集，称为 GMTSAR。在过去的调查中，该软件能够利用欧洲航天局运营的 Sentinel-1 卫星每年约 1200 TB 的免费数据，并作为强大的研究工具提供给用户群。 NASA 和 ISRO 即将执行的 NISAR 任务将大幅增加可用的 SAR 数据量，每年超过 30,000 TBytes。虽然这对 InSAR 科学来说是一个福音，但它提出了两个主要的处理障碍：如何利用这些不断增加的大型数据集实现最大生产力，同时仍然保持测量的准确性，以及如何最好地促进广大用户访问这一数据宝库。该项目通过 (1) 使 GMTSAR 能够在云计算环境中快速处理非常大的数据集，以及 (2) 通过与 Python 集成并简化处理模块以简化用户交互，进一步扩大这些 InSAR 数据在研究和学生社区中的使用。利用增强的 GMTSAR 软件促进大型 InSAR 数据集的处理将使固体地球和冰冻圈科学家能够利用大量 InSAR 数据集来推进他们的跨学科研究，包括全球火山观测、地震灾害估计等。应变率测绘、监测城市基础设施、跟踪冰盖运动以及检测海岸沉降。 InSAR 科学的全球影响力将通过专家和学生都可以使用的简化处理模块来推进。此外，我们建议对特定模块进行的改进，包括开发与全球导航卫星系统 (GNSS) 数据的常规集成以及将右视 Sentinel-1 和左视 NISAR 卫星的视距 (LOS) InSAR 测量相结合，将提高测量的准确性，从而实现完整的 3D 矢量位移时间序列分析。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Understanding the processes of earthquakes, volcanoes and hydrological changes, and their associated hazards is a top priority of the solid earth research community and USGS, due to the potential for societal disruption, financial consequences, and possible loss of life if not prepared for adequately. This requires not only long-term estimates of such processes and their hazards within a socially relevant timeframe, but also an evaluation on the impact of human activities over the Earth’s surface and interior. These estimates and evaluations hinge on the capability of accurately measuring how the Earth’s surface changes and deforms over time. For example, it is important to know how fast the seismic moment is accumulating over the San Andreas fault system, as that will tell us where and when will we be expecting the next destructive earthquake. This requires us to be able to measure the deformation that spans hundreds of kilometers at an accuracy of 0.5 mm/yr with resolution better than 10 km. Interferometric Synthetic Aperture Radar (InSAR) is the best technique for this crucial task, as the current remote sensing satellite observations that inform this technique come with broad-scale coverage, at low-cost, regardless of weather and on a regular basis. However, the upcoming new InSAR missions are raising a new challenge: how to efficiently handle drastically increasing amounts of data (~80 TB per day for the NISAR mission). To answer this challenge, the freely-available InSAR processing software GMTSAR is developing robust and efficient approaches to take full advantage of the satellite-generated data for both scientific research and societal applications. The main innovations of this project are to enable the cloud computing capabilities, transfer to newer generation of programing language, and keep engaging more users to build their own data processing strategies using this software. The developers will ensure that users from across the globe have the support they need for access to state-of-the-art processing techniques, and will continue improving the documentation, example datasets and tutorials to strengthen the foundation for education in the field of space geodesy. Interferometric Synthetic Aperture Radar (InSAR) is a powerful technique for measuring small displacements (1-10 cm) of the surface of the earth including those caused by tectonic loading, earthquakes, volcanoes, landslides, glaciers, ground fluid injection/withdrawal and underground nuclear tests. Over the past decade, a freely available, open-source software has been developed to harness these valuable datasets, which is called GMTSAR. During past investigations, this software has been equipped with the power to capitalize on the freely available ~1200 TB per year of data from Sentinel-1 satellite operated by the European Space Agency, and was provided as a robust research tool to the user base. The upcoming NISAR mission operated by NASA and ISRO, will dramatically increase the amount of available SAR data to over 30,000 TBytes per year. While this is a boon for InSAR science, it presents two main processing hurdles: how can one achieve maximum productivity with these increasing large datasets, while still preserving the accuracy of the measurements and how can one best facilitate broad user access to this trove of data. This project addresses these challenges by (1) enabling GMTSAR to permit rapid processing of very large data sets in a cloud computing environment, and (2) further expanding the usage of these InSAR data in both research and student communities by integrating with Python and streamlining processing modules to simplify user interactions.Facilitating the processing of large InSAR datasets with enhanced GMTSAR software will allow solid earth and cryosphere scientists to utilize the massive InSAR data sets to advance their interdisciplinary investigations, including global observations of volcanoes, estimates of seismic hazard through strain-rate mapping, monitoring urban infrastructure, tracking ice sheet movements, and detecting coastal subsidence. The global reach of InSAR science will be advanced by streamlined processing modules that are accessible to both specialists and students alike. In addition, the improvements we propose to make to specific modules, including the development of routine integration with Global Navigation Satellite System (GNSS) data and the combination of line-of-sight (LOS) InSAR measurements from both right-looking Sentinel-1 and left-looking NISAR satellites, will improve the accuracy of measurements to enable full 3D vector displacement time series analyses.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.