云南地区GNSS应变率场异常识别方法及地震预测效能评估

(1.云南省地震局,云南 昆明 650201; 2.云南省地质环境监测院,云南 昆明 650216; 3.自然资源部云南省高原山地地质灾害预报预警与生态保护修复重点实验室,云南 昆明 650216)

GNSS; 应变率场; 地震预测; R值评分; 异常指标; 云南地区

Identification Method of GNSS Strain Rate Field Anomaly and Evaluation of Earthquake Prediction Efficiency in Yunnan
WANG Lingli1,3,HONG Min1,NIU Tian1,LI Qin2,3,YANG Xinjun1,Yu Shixian1

(1.Yunnan Earthquake Agency,Kunming 650201,Yunnan,China;2.Yunnan Institute of Geo-environment Monitoring,Kunming 650216,Yunnan,China;3.Yunnan Key Laboratory of Geohazard Forecast and Geo-ecological Restorationin Plateau Mountainous Area,Kunming 650216,Yunnan,China)

GNSS; strain rate field; earthquake prediction; R-value scoring; abnormal indicators; Yunnan area

DOI: 10.20015/j.cnki.ISSN1000-0666.2025.0006

备注

利用1999—2020年云南及邻区近300个GNSS测站的观测数据解算获取的速度场为约束,采用克里金插值方法分时段估计了1999—2007年,2009—2014年,2015—2020年三期区域应变率场; 通过回溯各个观测时段之后3年内MS≥5.0地震事件,分析区域地壳形变特征与地震事件发生地点之间的相关性,结果表明,绝大部分地震都发生在面应变高梯度带的张压转换区和最大剪应变率沿断层方向的高值区或其边缘。基于上述应变率场异常特征,提出格网地震危险因子算法,建立地震危险区识别模型,通过估计格网最大剪应变率和面应变率风险区划因子,定量提取异常区地震危险指标,结果显示采用数值模型识别出的异常区与地震事件具有较好地对应关系; 进一步采用R值评分的方式对应变率场异常区模型识别方法进行地震预测效能量化评估与分析,结果显示3期应变率场预测结果均通过R值评分检测。
In this paper,the velocity field is calculated by using the observation data from nearly 300 GNSS stations in Yunnan and its adjacent areas in the period from 1999 to 2020.Based on the velocity field,three stages(1999-2007,2009-2014,and 2015-2020)of the regional strain rate fields are estimated by Kriging interpolation method.By backtracking the earthquake events(MS≥5.0)in the following 3 years of each observation stage,the correlation between the characteristics of the regional crustal deformation and the location of the earthquake events is analyzed,and some abnormal criteria of strain rate field identifying the risk locations of the strong earthquakes are summarized.That is,most of the earthquake cases occur in the tension-pressure transition zone of the high gradient zone of surface strain,and the high value zone or the edge of the maximum shear strain along the fault.Based on the abnormal characteristics of strain rate field,this paper proposes a grid algorithm of the seismic risk factor,and establishes a model of the seismic-risk area identification.By estimating the recognition factors of the risk zones of the maximum shear strain and the surface strain with the grid algorithm,the seismic risk indicators of the abnormal areas are quantitatively extracted.The abnormal areas identified by the numerical model corresponds well with the seismic events.In order to provide an objective evaluation basis for the subsequent use of this model to track earthquake conditions and to determine earthquake risk locations,this paper adopts the R-value scoring method to evaluate the efficiency of this proposed model.The results show that the prediction of the strain rate field in the three stages passes the R-value scoring.