Most plains in Iran are subject to land subsidence due to over-exploitation of groundwater mainly for agricultural purposes. Synthetic Aperture Radar (SAR) interferometry has shown its ability to provide precise measurements of the ground surface displacement at high spatial and temporal resolution. In SAR interferometry, the processed interferograms are combined together via interferogram stacking or time series analysis. Stacking is a temporal averaging of the interferograms which results in mean displacement velocity. However, time series analysis of a significant number of interferograms enables us to study the short-term as well as ling-term behavior of the subsidence. In this research, three different case studies were accomplished for subsidence monitoring. The subsidence in the Varamin plain was studied using 13 ENVISAR ASAR images spanning between 2003/08/03 and 2005/11/20. The maximum subsidence rate extracted from Small Baseline Subset (SBAS) time series was estimated as 0.4 m/year. The second case study was to monitor the subsidence in the Neyshabour plain. In this area, the interferogram stacking using 9 ENVISAT ASAR images spanning between 2004/01/10 and 2005/06/18 was applied. The maximum subsidence rate was estimated as 0.16 m/year. Groundwater level measurements made at piezometric wells were applied to compare to the interferometry results. The piezometric wells mostly show the increase in water level depth caused by over-exploitation of groundwater. The groundwater information jointly with stratigraphic profiles highly correlate with subsidence in the area. In the last case study, the Persistent Scatterer Interferometry (PSI) which is a proper method of time series analysis in areas with high decorrelation effects, was used in the Shahriar plain. A hybrid method of conventional and PSI was proposed in order to address the problem of monitoring the high-rate deformation. There are 22 ENVISAT ASAR images available in the study area spanning between 2003 and 2008. The maximum subsidence rate was estimated as 0.25 m/year. The time series analysis results were then compared to the groundwater level information at piezometric wells. Due to the low correlation between water level decline and subsidence rate at some piezometric wells, it can be concluded that other geology and hydrogeological factors play important role in controlling the subsidence occurrence. To show this, two data mining methods including Multi-Layer Perceptron (MLP) neural network as well as Support Vector Regression (SVR) were applied to model the subsidence in Shahriar plain using 6 different geology and hydrogeology factors as input and the subsidence rate extracted from interferometry as output of the model. These models can be further applied to estimate the subsidence rate in pixels in which the interferometry technique cannot measure the deformation due to some reasons including insufficient correlation.