利用两次新疆于田MS7.3地震对 图像信息方法可预测性的检验

以青藏构造块体为研究区域， 采用以往利用图像信息方法预测M7.0地震得到的较好的计算参数(网格尺度为1平方度， 预测时间窗口为8年)， 系统计算了青藏构造块体1993年以来的“地震热点”图像， 特别研究了两次于田M_S7.3地震前后的图像信息演化过程， 并通过ROC方法和R值评分方法检验， 对图像信息方法在该地区的预测效能进行评价. 结果表明: ① 2008年3月21日于田M_S7.3地震和2014年2月12日于田M_S7.3地震均发生在地震热点上， 且发生在平均发震概率随时间变化的峰值点年份. ② 在回溯性检验的时间段内青藏构造块体共发生9次M≥7.0地震， 其中6次地震前震中附近均有稳定的地震热点出现， 包括两次于田M_S7.3地震. 1996年喀喇昆仑山口M_S7.1地震及2008年汶川M_S8.0地震只在个别预测时间窗内出现地震热点， 并没有出现稳定的地震热点演化过程， 而1996年丽江地震发生前震中附近则没有出现地震热点. 该结果与前人研究结果存在差异， 值得进一步研究. ③ ROC检验和R值评分结果均显示图像信息方法在本文采用的模型参数下对大地震具有较好的中长期预测效果. 本文还根据目前存在的地震热点给出了青藏高原块体2014年以后M7.0地震发生概率高的3个可能地区. 

Predictability test for pattern information method by two MS7.3 Yutian,Xinjiang,earthquakes

In this paper， the Qinghai-Tibet tectonic block was chosen to be the study region， and calculating parameters with grid of 1°×1° were employed according to the previous studies by pattern informatics (PI) method for M7.0 earthquake prediction. Hotspot diagrams of each prediction window since 1993 have been obtained， and the relationship between the hotspots and the M>7.0 earthquakes during the prediction windows were studied. The predictability of PI method was tested by verification of receiver-operating characteristic (ROC) curve and R score. The results show that: ① Both of the Yutian M_S7.3 earthquakes in 2008 and in 2014 occurred in the region with the hotspots in the year with the highest average possibility. ② Nine earthquakes of M≥7.0 occurred in the retrospective studied period since 1993. Six of the nine earthquakes could be predicted by the successive obvious hotspots imagine， including the two Yutian M_S7.3 earthquakes; two of the nine earthquakes might be predicted because hotspots only occurred in a single predicted window without successive evolution process， including the Karakorum Pass M_S7.1 earthquake in 1996 and the Wenchuan M_S8.1 earthquake in 2008; one of the nine earthquakes， i.e.， Lijiang， Yunnan， M_S7.0 earthquake in 1996， could not be predicted because no hotspots appeared before the earthquake. The results associated with Wenchuan M_S8.0 earthquake in this paper are different from those of the previous studies， which may be caused by the differentce in selected study region and calculating parameters. ③ Test results of the predictability of PI method by ROC and R score show that positive prospect of PI method could be expected for long-term earthquake prediction. According to the calculating results based on the parameters from this study， three regions with high possibility for potential M7.0 earthquake risk in the future were given.