摘要
大数据是当今的热点,是各种信息技术和互联网发展到现今阶段的一种表象或特征,正逐步渗入我们的日常生活和工作.大数据技术的战略意义不在于掌握庞大的数据信息,而在于对数据的深度挖掘.通过丰富的时空动态数据分析,大数据为我们提供了理解事物间相互作用的巨大可能性.因此,大数据蕴含着巨大生产力,将对全球人类生活、科技、经济、文化及政治发展带来深远影响,许多国家已经开始对大数据应用研究做出了重大部署.大数据的核心是积累数据、分析数据、应用数据.大数据案例表明,通过提高人们自己的特殊技能和洞察力,大数据分析具有预测事物发展趋势、改变传统观念和发现新事物的功能,并有助于我们从错误信息中挖掘有价值信息.大数据改变着人类探索世界的传统观念和方法,尤其是通过了解模型的优点和局限性,大数据可以使数据产生知识.地震观测数据是大数据.地震是地球的诸多现象之一,地球又是一个多圈层的系统.随着人类获取数据、使用数据能力的重大突破和进展,数据当中隐藏的规律和趋势将不断被挖掘利用,给探索地球各种现象和减轻自然灾害,特别是地震机理,带来新的途径.1)大数据让地震预测不再热衷于寻找因果关系,大数据时代预测将以密集观测和多样本分析为基础,极有可能发现哪些地震前兆与地震有真正的关系,因此,在大数据的背景下,相关关系会促进地震预测水平,提高地震预测的可靠性;2)大数据促进部门间、地区间、国际间地震数据融合,加速数据实时分析,提升短临预测价值;3)大数据时代更需要高密度综合观测,让我们看到更多以前无法被关注到的细节,提高我们的洞察力;4)大数据改变地震监测预报方式方法,以往的有些数据模型、地震参数计算方法、前兆异常认识需重新修正,从而获得更精准的答案.然而,大数据战略思维在地震行业还未有得到充分应用,缺少有效汇集、存储海量数据的大数据技术,来实现数据集中分析和深度挖掘.我们需要为地震监测预报大数据实现做好准备:1)决策管理层推动大数据平台建设,培养数据分析科学家;2)整合所有观测数据,实现数据共享,并加强不同行业间的数据交换和新技术应用;3)加密现有的地震监测网,拓展数据资源;4)挖掘与地震有关的现象,研究高密度观测下地震参数计算方法,真正实现大数据价值挖掘;5)创建大数据下地震监测预报新理论.总之,大数据时代会给人类社会、经济、生活方式、创新思维带来一系列变革.地震监测也一样随着大数据时代会有新的变革,会改变现有地震监测预报思维模式和方法,进而推动地震科学的创新.对此,决策层应做好顶层设计.
Big data is a hot point at present.And it is a kind of representation or features of the development of a variety of information technology and Internet at the present stage.It is gradually infiltrated into our daily life and work.The strategic significance of big data technology is not the master of huge data information rather than data mining.Big data offer enormous possibilities for understanding interactions between things,with rich spatial and temporal dynamics,and for detecting complex interactions and nonlinear among variables.Therefore,big data contains a huge productivity and will have a far-reaching effect on the global human life,science and technology,economic,cultural and political development.Many countries have begun to make a major deployment of large data application.The core of big data is accumulating data,analyzing data,and applying data.The cases of big data show that big data analysis can make a prediction of developing trend,and change traditional ideas and discover something new,and help us mining valuable information from error information by improving people＇s the special skills and insight themselves.Big data is changing the traditional concepts and methods of human＇s exploring the world,especially through understanding the advantages and limitations of the models,big data can make the data generated knowledge.Seismic observation data is big one.Earthquake is one of the many phenomena of earth itself,and the earth also is a system of multiple spheres.Along with a major breakthrough of capacity and progress of the human＇s using the data and getting data,laws and trends hidden in data will continually be mined. This will bring new ways to exploring various phenomena on the earth and preventing and mitigating natural disasters,especially studying earthquake mechanism.First of all,at big data age,earthquake prediction is no longer interested in finding the causal relationship.The prediction will be based on intensive observation and much more sample analysis,the real relationship between somewhat precursors and earthquakes will be very likely found.Therefore,In the context of big data,cor-relativity relationship will promote the level of earthquake prediction and improve the reliability of earthquake prediction.The second,big data promote seismic data fusion stored and used by various departments in different regions and countries.And it will accelerate real-time data analysisto enhance the value of short impending prediction;Thirdly,the age of big data we also need more the high density and integrated observation to let us see more details which were not been previously noticed,for improving our insight.The fourth,big data will change the methods of earthquake monitoring and prediction.The past some data model,calculation methods of seismic parameters and recognition of precursor abnormalities need to be re-amended,so as to obtain more precise answers.The strategic thinking of big data were not be fully run in earthquake monitoring and prediction,however,the lack of the technology of effective data collection and mass data storage to realize large-scale scientific data analysis and data mining.We need to prepare for application of big data.Firstly,the decision-making management should promote platform construction of big data and train scientists of data analysis;The second,all of the observed data need to come together and realize the data sharing.At the same time,we need enhance the data exchange and new technology application between different areas. The third, we need encrypt earthquake monitoring network and expand data resources to get rich data.The Fourth,related phenomena about earthquake should be excavated.We should study the new ways of calculating seismic parameters under the conditions of high density observation to dig the something value hidden in big data.The fifth,a new theory of big data for earthquake monitoring and prediction should be created.In short,the age of big data will bring us a series of changes in human society,such as economy,life style and innovative thinking.Some new change will take place in the age of big data,too.And the existing mode of thinking and methods of earthquake monitoring and prediction will be changed.These will promote the innovation of the earthquake science.In this regard,the decision layer should do the top design well.
出处
《地球物理学进展》
CSCD
北大核心
2015年第4期1561-1568,共8页
Progress in Geophysics
基金
"科技支撑项目--地震科学数据共享"项目(503130108)资助
关键词
大数据
地震监测预报
数据挖掘
相关分析
决策
big data
earthquake prediction
dala mining
correlation analysis
decision
作者简介
张晁军，男，1965年生，河北省邢台人，博士，高级工程师，主要从事地震学与信息学科和新技术方法应用研究．（E-mail：zhangchaojun@seis．ac．cn）
