Ground motion detection in a salt solution mining area, an application of Multi-Temporal Satellite Interferometry

Abstract

<p>Subsurface mining is one of the human activities with the highest impact in terms of induced ground motion. The excavation of the mining layers creates a geotechnically and hydrogeologically unstable context. The generation of chimney collapses and sinkholes is the most evident surface consequence of underground mining which, in general, creates the optimal conditions for the development of subsidence bowls. Considering this, the need for ground motion monitoring tools is evident. Topographic measurements have been the obvious choice for many years. Nowadays, the flourishing of Multi-Temporal Satellite Interferometry (MTInSAR) algorithms and techniques offers a new way to measure ground motion in mining areas. MTInSAR fully covers the accuracy requirements asked by mining companies and authorities, adding new potentialities in term of area coverage and number of measurement points. The technique has some intrinsic limitations in mining areas, e.g. coherence loss, but the algorithms are being pushed to their technical limits in order to provide the best coverage and quality of measures.</p><p>This work presents a detailed scale MTInSAR approach designed to characterize ground deformation in the salt solution mining area of Saline di Volterra (Tuscany Region, central Italy). In summary, salt solution mining consists in the injection at the depth of interest of a dissolving fluid and in the extraction of the resultant saturated brine. In Saline di Volterra, this mining activity created ground motion, sinkholes and groundwater depletion. The MTInSAR processing approach used is based on the direct integration of interferograms derived from Sentinel-1 images and on the phase splitting between low and high frequency components. Phase unwrapping is separately performed for the two components that are then recombined to avoid error accumulation. Before generating the final deformation map, a classical atmospheric phase filtering is applied to remove the residual low frequency signal. The results obtained reveal the presence of several subsidence bowls, sometimes corresponding to sinkholes formed in the recent past. These moving areas register velocities up to -250 mm/yr with different spatial and temporal patterns according to the distribution and age of formation of sinkholes. This is the first time an interferometric analysis is performed here. It is hoped that such information could increase the awareness of local entities on the ground effects induced by this mining activity.</p>