Monitoring Surface Elevation Changes of Kerman City and Prediction of High-Risk Areas Using ASAR and SENTINEL 1 Radar Images

1

One of the fundamental and increasing problems in most human societies, which occurs mostly through human activities, is the phenomenon of subsidence. The subsidence refers to the surface depression caused by various natural factors such as the dissolution, the watering of the ice, human activities such as mining, unconventional withdrawal from groundwater or oil. Growing population has led to irregular use of water in domestic, industrial and agricultural uses which, in turn, has led to undesirable quantitative and qualitative effects on water resources. In many plains in Iran, excessive exploitation of groundwater has led to a subsidence event. In recent decades, the increase in the world population, especially in urban areas as an important phenomenon, has created complexities and problems in various fields. Urban areas are particularly vulnerable due to population congestion, buildings and communication arteries. Land subsidence in urban areas is one of the most important issues in urban planning and design. Space-based geodetic techniques that can measure changes in the land-surface position have been significantly advanced over the past two decades with the development of satellite-borne differential interferometric synthetic aperture radar (InSAR) techniques. The InSAR techniques can measure sub-centimeter ground displacements at high spatial detail over regions spanning. The purpose of this research is to monitor the phenomenon of subsidence in Kerman over the course of 15 years, and identify high-risk areas across the city.
2

InSAR is a remote sensing technique that uses radar imagery to provide spatially dense measurements of surface displacements the satellite line of sight (LOS) with millimeter to centimeter accuracy. Multiple SAR images are used to generate sets of interferograms to form a time series after a joint inversion. InSAR time series analysis helps reduce the impact of several noise sources (decorrelation, orbital and DEM errors, atmospheric delays, phase unwrapping errors) on displacement rates estimates during the time period spanned by the full dataset with an accuracy for surface displacement velocity at the mm/yr scale. We used 6 ESA ASAR C-band radar images acquired by the Envisat satellite between March 2003 and August 2012 from Track 163 (along descending orbits) and we prepared 2 SENTINEL1 radar images acquired by Soyuz satellite. A small-baseline approach  was used to process interferograms and invert for average displacement rates and evolution through time with the New Small Baseline Algorithm Subset chain, as described in detail by 4 individual interferograms were generated using a modified version of the SARScape and the STRM 30-global DEM.
3

The interferometric image was provided by using the radar interference method (Fig. 4). As shown in Fig. 4, number of fringes were created. Since the used satellite (Envisat) works in the C band, and each fringe obtained is 2.2 λ which equals8.2 centimeters, the amount of displacement in the direction of the satellite's view is given by counting the number of fringes. Depending on how the color cycles are observed, the displacement rate also varies, so that if the cycle is yellow-blue-red, moving away from the radar and if the cycle is yellow-red-blue, the shift to the radar has occurred. What is observed in different periods is the increasing trend of subsidence and the decreasing trend of uplift over these years. So, as shown in Fig. 5 subsidence rate has been changed from 7.3 to 7.6 cm per year and uplift rate has been changed 6 to 6.6 cm per year in the study area, between 2004 and 2010. So, as shown in Fig. 6. Subsidence rate has been changed from 9 to 10.6 cm per year and uplift rate has been changed 5 to 4.8 cm per year in the study area, between 2010 and 2017. The spatial pattern of displacements has also changed over the course of the study, so that, the subsidence ranges have increased and that have been transferred to urban areas from around of city. In order to identify high risk areas in Kerman city, a high risk areas map was produced. Accordingly, the city of Kerman divided into four zones including very high risk, high risk, and relatively risky and relatively low risk areas. The results showed that about 700 hectares of the city's territory are in very high-risk and high-risk areas.
4

The results of the radar interferometry method show that the different areas of Kerman have had different behaviors over the past 15 years in terms of subsidence and uplift. The subsidence areas were concentrated around the city of Kerman in the period of 2004 to 2007, and the subsidence rate was relatively low. Over the years, more areas of city have been abandoned, and so far. Considering the existence of destructive evidence on the city buildings with high ranges of uplift, it can be concluded that earth's uplift can be as risky as subsidence.