Due to the considerable loss of life and damages caused by powerful Earthquakes, along with the strengthening of structures and disaster management systems, many researches are being conducted associated with the possibility of creating Earthquake warning systems. However, no reliable scientific report on successful Earthquake prediction has been published so far. Although there are a couple of published scientific papers indicating the detection of unusual changes in physical and chemical parameters obtained from ground stations, but due to the limitations of in-situ measurement in terms of number, location, time and cost, there was no noticeable progress in Earthquake warning. After the advent of remote sensing satellites, the number of statistical studies related to Earthquake precursors increased significantly, and most of them emphasized the existence of abnormal behaviors in physical and chemical precursors around the time (about 1–30 days before) and locations of powerful Earthquakes. In this study, an attempt has been made to address the advances and challenges in the way of Earthquake prediction using satellite data and provide a perspective of the future of these researches. It should be noted that the increase in the number of satellites designed and launched specially for Earthquake studies, the variety of available Earthquake precursors (multi-precursor analysis), the development of classic and intelligent anomaly detection and predictor algorithms (multi-method analysis), the provision of intelligent systems for the integration of various precursors (fusion and decision systems), the creation of cloud storing and processing data services (Google Earth Engine, Giovanni, etc.) and the development of intelligent user interfaces for public, have made researchers more hopeful for the appearance of low-uncertainty Earthquake warning systems in the near future.
Read full abstract