Collaborative Research: Detection and mechanics of sinkhole activities in central Florida

佛罗里达中部的天坑活动是一种主要的自然灾害。在过去几年中，有两个事件特别引起媒体的注意。2013年3月，一个天坑在塞夫纳的一所房子下面倒塌，从卧室里“吞下”了一个人。2013年8月，一个天坑坍塌摧毁了迪士尼世界附近的一个度假区。然而，这些戏剧性的事件只是极端的例子，从每年约2亿美元的财产损失索赔中脱颖而出，这些索赔在人为抽取地下水时激增。探测初期的天坑活动是一项具有挑战性的任务，因为大多数活动发生在地下。然而，在某些情况下，天坑活动也会引起地面沉降，这可以通过天基干涉合成孔径雷达（干涉合成孔径雷达）观测来探测。该项目是由NSF资助的概念验证项目的后续项目，在该项目中，研究人员证明了干涉合成孔径雷达观测可以检测到佛罗里达中部地表变化的局部模式。然而，使用更先进的干涉合成孔径雷达时间序列技术检测缓慢沉降需要比在初始项目的有限持续时间内获得的更长的采集跨度。在这一更新项目中，我们建议继续利用国际合成孔径雷达探测佛罗里达中部的天坑活动，将观测期再延长两年。较长的时间序列将使我们能够减少测量噪声的影响，并检测由天坑活动引起的局部缓慢沉降。拟议的项目还包括地球物理调查，以绘制天坑结构图，并在选定的实地进行土壤取样。此外，机械建模工作将检查是否，在现场，InSAR探测到的沉降是兼容的粘性土壤，现场数据的测量特性，以及天坑沉降的简单模型。这种建模方法可用于评估所观察到的沉降是否代表可能的崩溃前兆。

Sinkhole activity in central Florida is a major natural hazard. Over the past several years, two events in particular attracted media attention. In March 2013, a sinkhole collapsed beneath a house in Seffner and "swallowed" a person from his bedroom. In August 2013 a sinkhole collapse destroyed a resort complex near Disney World. These dramatic events, however, are just extreme examples that stand out from a steady stream of about $200 million per year in property damage claims, which spike at times of anthropogenic groundwater withdrawals. Detecting incipient sinkhole activity is a challenging task, because most of the activity occurs in the subsurface. However, in some cases sinkhole activities also induce surface subsidence, which can be detected by space-based Interferometric Synthetic Aperture Radar (InSAR) observations. This project is a follow up of a proof-of-concept project funded by NSF, in which the researchers demonstrated that InSAR observations can detect localized patterns of surface changes in central Florida. However, the detection of slow subsidence using more advanced InSAR time series techniques requires longer span of acquisitions than obtained during the limited duration of the initial project. In this renewal project, we propose to continue the InSAR-based detection of sinkhole activity in central Florida by extending the observation period for two additional years. The longer time series will allow us to reduce the effect of measurement noise and detect localized slow subsidence induced by sinkhole activity. The proposed project also includes geophysical surveys to map sinkhole structure and soil sampling at selected field sites. In addition, a mechanical modeling effort will examine whether, at the field sites, InSAR-detected subsidence is compatible with measured properties of cohesive soils, field data, and simple models for sinkhole subsidence. This modeling method could be used to assess whether observed subsidence represents a likely precursor to collapse.