Journal of Geomatics Science and Technology

fa بررسی تغییرات TEC پیش از وقوع تعدادی از زمین لرزه های بزرگ رخ داده در سال 2010، با استفاده از آزمون آماری تی دو-هتلینگ Investigation TEC Variations before the Several Large Earthquakes that Occurred in 2010, Using the T2-Hotelling Test ژئودزی و هیدروگرافی Geo&Hydro پژوهشي Research <div style="text-align: justify;">زمینلرزهها سالانه خسارتهای جانی و مالی عظیمی به مردم جهان وارد میکنند. از اینرو دانشمندان همواره به دنبال یافتن راهکاری برای شناسایی زمان و مکان این پدیده، پیش از وقوع آن هستند. این در حالی است که فعالیتهای لرزهای زمین باعث ایجاد تغییرات ناهنجاری در پارامترهای یونسفری، پیش از رخداد زمینلرزههای بزرگ میشوند. همین امر سبب شده است که امروزه بررسی تغییرات یونسفری به یکی از روشهای مهم پیشبینی زمینلرزه تبدیل شود. در این مقاله برای شناسایی تغییرات یونسفری-لرزهای از دادههایTEC  حاصل از نقشههای جهانی یونسفر (GIM) استفاده شده است. روش آماری مورد استفاده برای کشف تغییرات ناهنجار یونسفری، آزمون تیدو-هتلینگ چندمتغیره میباشد. در این مقاله برای اولین بار از این روش به منظور بررسی ارتباط تغییرات یونسفری و زمینلرزه استفاده شده است. جهت بررسی کارایی این آزمون در کشف تغییرات ناهنجار یونسفری-لرزهای 12 زمینلرزه با بزرگای گشتاوری بزرگتر و یا مساوی 6 که در سال 2010 رخ داده، مورد مطالعه قرار گرفتهاند. مطابق با نتایج حاصل، این آزمون آماری موفق به شناسایی آنامولی یونسفری-لرزهای در 9 مورد از زمینلرزههای مورد مطالعه شده است. این آنامولیها عموماً در اطراف مرکز زمینلرزهها مشاهده شده و اغلب یک هفته پیش از زلزلهها قابل شناسایی هستند. همچنین به طور کلی نتایج نشان میدهد انجام آزمون تیدو-هتلینگ در سطح اطمینان 99% نسبت به سطح اطمینان 95% در این مطالعه کارآمدتر بوده است.</div> <div style="text-align: justify;">The fact of matter is that, aannually, earthquakes create huge losses in life and property to people all around world. On average, about one earthquake with magnitude of <img alt="" height="16" src="file:///C:UsersDaViDAppDataLocalTempmsohtmlclip11clip_image002.png" width="32" > and fifteen earthquakes with magnitude of <img alt="" height="16" src="file:///C:UsersDaViDAppDataLocalTempmsohtmlclip11clip_image004.png" width="33" > occur in the world. So, scientists are always looking for a way to identify the time and place of the event before its occurrence in order to reduce damages and injuries. Earthquake is seismic geophysical phenomena which including irregular, nonlinear and complex processes that's why there is no simple approach to predict its parameters. However, this is a known fact that before large earthquakes occurrence, seismic activity of the Earth can cause abnormal variations in ionospheric parameters. Hence nowadays, study on ionospheric variations has become one of the most important ways to predict earthquakes. Nowadays, as regards development spatial and temporal resolution of TEC data (that is extracted from GPS and other satellites) seismo-ionospheric studies are rapidly improving. In this paper, in order to identify seismo-ionospheric variations, the TEC data has extracted from global ionospheric maps (GIM) associated with the CODE center. The statistical method that has utilised to detect abnormal ionospheric variations is multivariate T2-Hotelling test for two samples. T2-Hotelling test is a very powerful test for detect the unusual changes of mean of two samples. The first sample which TEC reference values has extracted from it,includ data of the 44 to 15 days before the earthquake, then days has removed with high level of solar and geomagnetic activities (<img alt="" height="18" src="file:///C:UsersDaViDAppDataLocalTempmsohtmlclip11clip_image006.png" width="38" > or <img alt="" height="16" src="file:///C:UsersDaViDAppDataLocalTempmsohtmlclip11clip_image008.png" width="54" >). Second sample has formed from four days, finally for investigating this point that whether the difference between the mean of reference sample and second sample’s mean is significant or not, T2-Hotelling test has used. If the test rejectes, it means that difference is abnormal, otherwise it is usual. Abnormal difference may be through the earthquake. In this paper, this method has used for investigating the relation between the ionospheric abnormal variations and earthquakes, for the first time. In order to evaluate the ability of the test to detect seismo-ionospheric variation, 12 earthquakes with the magnitude greater than or equal 6 that occurred in 2010 has studied. Whereas solar activities in the year 2010 were low and several major earthquakes occured in this year, so that this year has selected for study. Given that the behavior of the ionosphere at diverse latitudes is very different, earthquakes have chosen with various latitudes.Also earthquakes have selected from areas with good IPP coverage. According to the results, this statistical test can detect seismo-ionospheric anomaly for 9 cases of studied earthquakes. These anomalies have generally observed around the center of earthquakes and these are often identifiable a week before the earthquakes (for Haiti, Drake, Honshu, Mexico and Ecuador earthquakes). Another important point is that anomaly of Ecuador earthquake, despite its great depth (206.7 km) is well identified. Results generally indicate that T2-Hoteling test at confidence level of 99%, has been more effective in this study.  </div> زمین لرزه, یونسفر, آنامولی, TEC, تی دو-هتلینگ Earthquake, Ionosphere, Anomaly, TEC, T2-Hotelling 21 32 http://jgst.issgeac.ir/browse.php?a_code=A-10-590-1&slc_lang=fa&sid=1 Z. Sadeghi زهرا صادقی z.sadeghi@mail.kntu.ac.ir 10031947532846005845 10031947532846005845 Yes Faculty of Geodesy & Geomatics Engineering, K. N. Toosi University of Technology دانشگاه صنعتی خواجه نصیرالدین طوسی M. Mashhadi Hossainali مسعود مشهدی حسینعلی hossainali@kntu.ac.ir 10031947532846005846 10031947532846005846 No Faculty of Geodesy & Geomatics Engineering, K. N. Toosi University of Technology دانشگاه صنعتی خواجه نصیرالدین طوسی