آشکارسازی فرونشست زمین جهت پایداری دشت هریس با استفاده از تصاویر راداری و روش‌های sbas و psi

از جمله مخاطرات محیطی بسیار مهم که در منطقه دشت هریس باید موردتوجه قرار بگیرد فرونشست است. پدیده فرونشست در دهه‎های اخیر مشکلات زیادی را برای زمین‎های کشاورزی، مناطق مسکونی، جاده‎ها و کانال‎های آبرسانی در برخی مناطق کشور به وجود آورده است. به‌منظور تحلیل سری زمانی جابه‌جایی سطح زمین، الگوریتم خط مبنای کوتاه موسوم به sbas به کار گرفته شد. تحلیل سری زمانی فرونشست، با استفاده از 6 تداخل نگاشت (اینترفروگرام) محاسبه شده از 21 تصویر راداری ماهواره sentinel1 در بازه زمانی 4 سال (2016-2019) انجام شد. یک عامل نرم‎کنندگی بهینه که نوفه ناشی از اتمسفر، خطای بازیابی فاز و اثرات مداری را کاهش می‎دهد و یا به‌عبارت‌دیگر با استفاده از مدل اتمسفری تاثیرات این عامل کاهش یافت تا سیگنال‎های غیرخطی به‌صورت بهینه حفظ شوند. نتایج تحلیل سری زمانی نشان داد که منطقه به طور پیوسته در حال فرونشست است. همچنین از روش psi نیز برای برآورد فرونشست و مقایسه آن با روش sbas استفاده گردید که نتایج نشان‌دهنده برتری روش sbas به روش psi در این تحقیق بود. در این تحقیق با استفاده از تصاویر راداری سنتینل 1 و همچنین تصاویر سنتینل 2 اقدام به برآورد فرونشست زمین در دشت هریس شده است. کاربری‎های موجود در دشت را از تصاویر سنتنیل 2 استخراج شده و با نتایج حاصل از تصاویر راداری انطباق داده شده تا ارتباط فرونشست زمین با هریک از کاربری‎های موجود در دشت هریس بررسی شود. نتایج حاصل از اجرای دو روش برآورد فرونشست زمین میزان جابه‌جایی سطح زمین را (0 تا 15) سانتیمتر در سال نشان داد. نتایج بررسی عمق چاه‎های پیزومتری نیز ارتباط قوی بین فرونشست و عمق چاه‎های آب منطقه را نشان می‎دهد.

revealing land subsidence in heris plain using radar images and sbas, psi method

landslides are one of the most important environmental hazards to be considered in the harris plain area. the phenomenon of subsidence in recent decades has created many problems for agricultural lands, residential areas, roads and water supply canals in some parts of the country. in order to analyze the time series of surface displacement, a short baseline algorithm called sbas was used. subsidence time series analysis was performed using 6 interferograms calculated from 21 radar images of sentinel1 satellite over a period of 4 years (20162019). it is an optimal attenuation factor that reduces atmospheric noise, phase recovery error, and orbital effects; in other words, it reduced atmospheric effects using the atmospheric model to optimally maintain nonlinear signals. the results of the time series analysis showed that the region is continuously subsiding. psi method was also applied to estimate subsidence and compare it with sbas method, which showed the superiority of sbas method with psi method in this study. in this research, using land radar images of sentinel 1 and also images of sentinel 2, the land subsidence in harris plain has been estimated. the land uses in the plain were extracted from sentinel 2 images and adapted to the results of radar images to investigate the relationship between land subsidence and each of the land uses in the harris plain. the results of two methods of estimating land subsidence showed the amount of land displacement (0 to 15) cm per year. the results of depth analysis of piezometric wells also show a strong relationship between subsidence and the depth of water wells in the region. extended abstract 1introductionland subsidence is a global problem and a morphological phenomenon. this phenomenon is affected by human activities and natural factors. geological and groundwater factors are also very effective in creating subsidence phenomenon due to improper groundwater abstraction. in recent decades, various methods for producing landslide maps have been proposed by researchers, including psi and sbas. the psi method has a variety of techniques for processing radar images, the two most important of which are psinsar and squeesar, which are used to analyze ground deformation. in the sbas time series method, only pairs of images are used in which vertical component of the baseline is less than the critical value of the baseline.  their timeline is at the same time minimal. due to the fact that the radar interference method is one of the most powerful tools for monitoring the subsidence phenomenon, this method is able to determine the changes in the earth’s surface in that time period by comparing the phases of two radar images taken from one area but at two different times. the first step in subsidence monitoring is to measure the amount of displacement caused by it at ground level. the main issue in this paper is to measure the amount of subsidenceinduced displacement in the harris plain area. the main purpose of this study is to use radar interferometry method to determine the amount of subsidence of harris plain in a period of 4 years. in this research, it has been tried with new methods to estimate the amount of subsidence, its amount in heris plain and compared with geological and land use factors as an example of environmental and human factors.2materials and methodsthe basis of measuring ground surface changes is the use of duplicate radar images. an image captured from one area at a given time is combined with an image captured at the same time by the same radar sensor. this radar satellite, which uses synthetic aperture radar sensors, is capable of imaging in any weather conditions. in the present study, a permanent dispersion method was used to estimate land subsidence. coregistration is one of the most basic steps in image interference processing at this stage; two or more images should be used. the sbas method is known as the time series analysis method of radar images to estimate land subsidence. in this regard, a network is created from interferometers, which uses the least squares method to estimate the amount of pixel displacement. the permanent scattering method (psi) is the degree of correlation of radar signals dependent on the distribution of scattered structures within a pixel. the purpose of the permanent scatter method is to identify pixels in the image, hereinafter referred to as psis that remain consistent in the time it takes to obtain the data used. the coherence of these pixels is suitable even for interference graphs with a base length greater than the critical value.3 results and discussionevaluation and estimation of subsidence obtained from sentinel1 images and psi method show that land subsidence has increased during the study period 20162019, so that the central part of the plain had the highest rate of subsidence and the eastern part had the lowest rate of subsidence. then the average subsidence rate in this time was calculated. the maximum amount of subsidence in this period is about 16 cm per year by psi method and about 10 cm by sbas method, which is related to rangelands, and the lowest subsidence is related to residential areas. in addition, increasing the amount of groundwater extraction causes subsidence in the region, so the relationship between the depth of wells in the region and the amount of subsidence in that area can be a good indicator to assess the accuracy of operations. regression correlation analysis between these two factors showed a positive correlation of 87% between them; the implication is that any place that has a lot of subsidence in that place has also deeper piezometric wells. this finding confirms the hypothesis of this research that there is a direct and strong relationship between subsidence and groundwater abstraction. in order to study the subsidence of the study area, sbas and psi methods were used and 21 sentinel images were applied in the period (2016 to 2019) and the depth of level wells was shown, which shows the study area despite seasonal fluctuations. it has a downward trend, which is presented in millimeters. finally, the time series obtained from sbas and psi methods were compared with each other. it was found that sbas method has more ability to show subsidence in the region due to less standard deviation. the results of the time series showed that most of the subsidence is related to pastures due to excessive use of groundwater. after pastures, agriculture is the second priority and residential areas have less subsidence than other areas. the reason can be attributed to the feeding of the harris plain aquifer during this period due to rainfall.4 conclusionin this study, the capability of sbas and psi interferometry time series analysis method in determining the rate and pattern of subsidence affected in harris plain has been illustrated. harris plain has subsided due to overextraction of groundwater. to calculate the subsidence time series using sbas method, 21 sentinel 1 radar images were used in the time interval (20162019) on which sbas time series and psi were processed. in addition, to validate the methods from the depth of the well surface area piezometer was used. the high standard deviation of psi method, which indicates the dispersion of results and its high noncompliance with changes in well water table level compared to sbas method, has less standard deviation than the previous method and has more compliance or correlation with well water change showing that the sbas method is more valid in this region than the psi method. the results of classifying the base object by the nearest neighbor method showed that sentinel 2 images with a total accuracy of 89% and a kappa coefficient of 0.86 have a good performance for producing land use maps. at the same time, the sbas method provides a wide and continuous coverage of the area, which makes it possible to determine the area under the subsidence affected. moreover, a comparison between the extent and pattern of subsidence occurred in this area, resulting from radar interferometry technique. the location and density of groundwater abstraction wells in this plain show that subsidence has occurred in the same areas where the density of these wells is high.