云南地区形变观测资料短临异常指标提取

提出了面应变综合预测指标法，目的是将全球卫星定位系统（global position system，GPS）观测结果应用于地震短临预测。基于云南省GPS连续观测资料，利用GPS高精度数据处理软件（GAMIT/GLOBK）处理后获得位移场。对位移场进行网格化插值获得均匀分布于56个网格的位移场，基于每个网格位移场时间序列求取其面应变时间序列，并对每个网格面应变时间序列的映震能力进行评价、综合，提取出适合于云南地区的地震短临异常识别指标。实践表明，综合预测指标对云南地区2011年以来发生的20次5级以上地震进行3个月短临预测，其中准确预测次数为19次，漏报数为1次，虚报5次，认为该指标对地震发震时间、震中均具有较好的指示意义，可用于云南地区未来的地震短临预测实践。

Extract Short Impending Abnormal Indicator from Deformation Data in Yunnan Province

This paper presents a comprehensive predictor indicator for surface strain, and aims to achieve short-term and impending earthquake prediction based on GPS (global positioning system) observation data. We obtain the velocity field of the Yunnan area from continuous GPS observation data with the GAMIT/GLOBK(High precision GPS data processing software) software. The Yunnan region was divided into 56 grids by interpolation and created a uniform displacement field. We obtain the surface strain time series of each grid. By evaluating and synthesizing the earthquake response capacity of each grid, a short temporary anomaly identification indicator was extracted. Since 2011, there were 20 earthquakes with Magnitude greater than five, occuring in Yunnan, 19 times of them were accurately predicted, with one omission and five false detections, Tested in practice with GPS observations recorded in Yunnan area, this indicator can not only predict the time but also the epicenter of earthquakes that happened in Yunnan area. It can be used to forecast future earthquakes in Yunnan region.