Analysing Time Series of Land Subsidence in the West of Tehran Province (Shahriar Plain) and its Relation to Groundwater Discharge by InSAR Technique

Land subsidence, as one of the natural and geological hazards, can be caused by human activities such as long-term discharge of groundwater and traditional irrigation farming. In the last two decades, climate change and successive droughts, unsustainable management of water resources and overexploitation of groundwater as well as population growth have caused land subsidence in Shahriar plain in the west of Tehran province. The present study aims to monitor and estimate land subsidence using Interferometric Synthetic Aperture Radar (InSAR) approach. Moreover, it tries to investigate the relationship between groundwater level and subsidence rate. Therefore, the time series of the ENVISAT-ASAR satellite images from 2003 to 2010 were used. The Small BAseline Subset (SBAS) algorithm was applied to analyze the time series of land surface displacement and to generate the mean displacement velocity map. The findings from time series analysis of InSAR data indicate a continuous subsidence occurrence in the agricultural lands of the region. The mean velocity of deformation along the satellite line of sight (LOS) in the time period of study, shows the displacement at an average rate of -10 cm / year and a maximum rate of -27 cm / year in the Shahriar plain in the west of Tehran province. Over this time period, groundwater level decreases about 0.5 to 1.5 m in the aquifer storage at 12 observational wells located in the study area. The overall correlation between changes in groundwater level and subsidence rate was estimated circa 89.45 percent, which indicates a strong relationship between subsidence and groundwater exploitation in the region.      
Extended Abstract
1-Introduction
Land subsidence, as a natural phenomenon, is defined as the gradual subsidence or abrupt sinking of the ground surface due to the subsurface material’s compression. One of the common causes of the land subsidence formation is the overexploitation of underground aquifers. The occurrence of land subsidence due to groundwater extraction from aquifers has been studied in several researches and documented in various regions of the world. This phenomenon is known as a global problem and leads to many environmental consequences such as damaging human structures like buildings, streets, bridges and power lines, creating holes on the earth surface, intensifying floods and flooding and reduction of aquifer capacity for water storage and ultimately it poses social and economic risks for human societies. In regions with bounded groundwater aquifers, groundwater discharge causes the reduction of pore pressure and subsequently sedimentary layers are compacted and condensed. This process leads to the downward movement of the ground surface and so-called land subsidence. Surface deformation is often measured using Interferometric Synthetic Aperture Radar (InSAR) technique. Generally, InSAR technique measures the phase difference of radar waves caused by the deformation created on the earth surface in time interval between two satellite pass.
2-Materials and Methods
In this study, radar interferometry techniques were used to monitor, estimate and analyze time series of land displacement in the west of Tehran province. Based on the historical archive and free accessibility, ENVISAT images were identified as the best dataset to estimate the land subsidence. Therefore, the level 2 time series data product (IMS) of ENVISAT-ASAR sensor related to frame number of 2889 from descending track of 149 were acquired. The time series selection of SAR data was determined according to the availability of well depth data. In the first step, the radar interferometry method was applied by analysing all possible differential interferograms with respect to temporal and spatial baselines to detect deformation signals (in particular, land subsidence). In the next step, by selecting a set of optimal generated interferograms regarding spatial and temporal baseline and using the SBAS algorithm, the land surface displacement time series were evaluated and the mean displacement velocity map for the region was produced.
3-Results and Discussion
The results of using InSAR approach indicate some subsidence event in the eastern part of Shahriar and northwest and west of Eslamshahr with mean velocity of -10 cm per year. Also, the estimated maximum subsidence rate in this region is -27 cm per year. The results of time series analysis, using the SBAS algorithm, showed that the subsidence signals before 2005 occurred at a faster rate compared to the 2005 to 2010. Groundwater level data of study area for the period of 2003 to 2010 generally represents drop in groundwater levels. Due to the slope and elevation (topographic status), the wells located in the northern part of the study area are much deeper than those in the southern part. Besides, the trend of water level decline in this area has not always been descending and sometimes the water level has increased limitedly at certain times. The results indicate that the water level decline has been more severe in the north and west than in the south and east part of region. The results of correlation analysis between changes in groundwater level and land subsidence indicate a high correlation between land subsidence signals and groundwater extraction with an average estimate of 89.45%. The lowest correlation was observed in well No. 2 with a correlation of 52.11%. The highest correlation of 99.96% was observed in well No. 7. The poor correlation between water depth in observational wells and land surface subsidence signals is related to the geological properties of the area and the type of soil which ultimately causes a time delay between groundwater exploitation and subsidence signals. The results also indicate that an average subsidence of 5 to 12 cm have been occurred per 1 meter drop in groundwater level.
The results of previous studies related to InSAR processing in Shahriar plain were examined to confirm the accuracy of the obtained results. The results of the time series analysis of land displacement signals to extract the spatial-temporal pattern of the region's subsidence, despite minor differences, are very similar to the results of other studies in the same region. Minor discrepancies in the estimated subsidence rate (sometimes up to 3 cm per year) can be attributed to the type of applied algorithm and radar images, as well as the considered time period.
4-Conclusion
Having recognized land subsidence phenomenon in the current study, time series images were estimated using InSAR technique in the west of Tehran province. Then, the relationship between subsidence rate and groundwater level changes in the area was investigated. The high correlation indicated that the main cause of the subsidence in this area was the overexploitation of groundwater. Also, the conformity of the spatial patterns of subsidence signals to agricultural lands in the study area indicates the relationship between subsidence event and land use. In this study, for the first time with a deeper look, the relationship between land subsidence and water underground level changes in observational wells was evaluated separately. Due to the observation of nonlinear patterns in relation to the correlation analysis between subsidence and groundwater levels in some areas of the study area, it is suggested that the geological structures of each well be separately examined and compared with the amount of subsidence in future research.