Atmospheric Delay To Improve The Application Of Permanent Scatterers

Synthetic Aperture Radar(SAR) is a very effective technique for measuring crustal deformation.It can monitor an area over ten thousands kilometers just with a few images,and the accuracy could achieve a level of submillimeter[Ferretti,2007].No conventional detection methods could compare with InSAR technique considering the cost.Among other techniques, persistent scatterers(PS) method is a typical means which has great applicable potential to overcome temporal and spatial decorrelation and reduce atmospheric effects.Permanent Scatterers processing over a good many of multi-temporal SAR images only takes into account pixels with high quality signal levels,over long period,in terms of amplitude and coherence values,and form a small imaging subset.Then it analyses phase change of the pixels inside the subset.At last we can get high reliable deformation values and use the results to detect minor move of the earth's surface.Such individual with stable phase is called Permanent Scatterer.PS could play the role of natural nodes of GPS surveying.In the areas where lack good correlated environment,we can also obtain reliable data of altitude and deformation. Furthermore if the actual size of the PS pixels are smaller than the resolution cell of the image, even the length of the interferential baseline beyond the so called critical length,the coherence between image pairs within the PS could be well kept.By this technique we can make use of all the SAR data within the interested area.PS technique addresses the problems of decorrelation and atmospheric delay which has been obsessed InSAR for long.One of the kernels of PS technique is how to identify PS effectually. Existing methods for for selecting PS candidates could be roughly divided into two groups,that is,Correlation Coefficient(CC) and Amplitude Dispersion Index(ADI).CC method should take account of two key points,one is to decide the size of analyzing window,and the other is to confirm identifying threshold value.The correlation coefficient is calculated by partially moving the analyzing window,whose size has direct effect on estimating CC value.Colesanti etc. showed when the threshold value taken 0.3,most of the area tagets,as waters and areas coverd with vegetation,which behaved unsteady radar scatter properties would be filtered.Theoretic study by Ferretti and others revealed that,for high signal to noise ratio(SNR),amplitude dispersion is an accurate proxy for phase standard deviation and thus the method has a high success rate at picking bright PS.However,for low SNR scatterers,the simple relationship between amplitude dispersion and phase stability breaks down and the method is no longer effective.In this paper,we introduced a new PS method developed by Stanford University.This method used spatial correlation of interferogram phase to find a network of stable pixels in all terrains, with or without buildings.Prior knowledge of temporal variations in the deformation rate is not required.We refer to these pixels as persistent scatterers(PS).A major source of error for repeat-pass InSAR is the phase delay(especially the part due to water vapour) in radio signal propagation through the atmosphere.In the year 1997,Zebker and others formulated that a 20%spatial or temporal change in relative humidity was estimated to cause an error of the order of 10-14 cm in the case of deformation estimates(Zebker el al.1997). Beijing is located to the north of North China Plain,surrounded by mountainous areas in the northwest.Such topography makes vapor condensing more serious.As a result,the atmospheric effects on detectation accuracy of InSAR in Beijing can not be ignored.GPS synchronous data can provide high accurate Precipitable Water Vapour(PWV) in troposphere.Thereby we can use GPS PWV product to reduce water vapour effects on InSAR interferograms.But this method has some limitations.It requires suffice GPS stations and the observation must be synchronous with SAR images.At present such conditions would be hard to meet.Using MODIS PWV product to correct atmospheric delay bear a critical setback:because MODIS data was taken two hours later than ASAR data,so there are time difference between them.This paper analysed the theories and method of atmospheric delay comprehensively and proposed an applicable mothod,which used MERIS PWV product to correct atmospheric delay. The Medium Resolution Imaging Spectrometer(MERIS) was launched together with the Advanced Synthetic Aperture Radar(ASAR) on the ESA ENVISAT spacecraft on 1 March 2002. We use MERIS reduced resolution near-infrared water vapour products to correct atmospheric effect and applied it to the new PS mothod.The method was tested in Beijing.