生物气溶胶施放源项参数反演是生物气溶胶袭击危害评估的反问题,对危害评估及应急响应具有重要指导意义。本文基于贝叶斯推理方法,利用生物传感器检测数据和正向大气扩散模型,构造似然函数,采用结合Metropolis-Hasting算法的马尔可夫链...生物气溶胶施放源项参数反演是生物气溶胶袭击危害评估的反问题,对危害评估及应急响应具有重要指导意义。本文基于贝叶斯推理方法,利用生物传感器检测数据和正向大气扩散模型,构造似然函数,采用结合Metropolis-Hasting算法的马尔可夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)抽样,对施放源位置、高度、施放剂量进行反演。统计分析表明,反演结果和初始源项参数设置吻合非常好,证明了方法的有效性。展开更多
On April 20, 2013, a magnitude M s 7.0 earthquake occurred in Lushan, Sichuan Province, China, and caused heavy casualties and economic losses. Based on the local broadband waveforms in Sichuan and adjacent provinces ...On April 20, 2013, a magnitude M s 7.0 earthquake occurred in Lushan, Sichuan Province, China, and caused heavy casualties and economic losses. Based on the local broadband waveforms in Sichuan and adjacent provinces regional networks and teleseismic broadband records from IRIS stations, the focal mechanism and the focal depth are determined by the CAP (Cut And Paste) and its upgraded methods, CAPtele and CAPjoint, respectively. The results show that the focal mechanisms and depth from different methods are steady, and the best double couple solution derived from the joint inversion is 210°, 44°, and 91° for strike, dip, and rake angles respectively for one nodal plane and 29°, 46°, and 89° for another with 16 km focal depth and M w 6.66 moment magnitude. In order to verify the reliability of the results, a number of tests are performed based on local seismograms with different velocity models. They indicate that there is about 10 degree's fluctuation in focal mechanisms and about 2 km variation in focal depth with a complex velocity structure. Furthermore, inverted by re-sampling the teleseismic waveforms on the basis of epicentral distance, the solutions are consistent with each other, which manifests that the teleseismic records are effective for constraining source parameters of the Lushan earthquake.展开更多
文摘生物气溶胶施放源项参数反演是生物气溶胶袭击危害评估的反问题,对危害评估及应急响应具有重要指导意义。本文基于贝叶斯推理方法,利用生物传感器检测数据和正向大气扩散模型,构造似然函数,采用结合Metropolis-Hasting算法的马尔可夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)抽样,对施放源位置、高度、施放剂量进行反演。统计分析表明,反演结果和初始源项参数设置吻合非常好,证明了方法的有效性。
基金supported by National Natural Science Foundation of China (Grant Nos.41174086,41074052 and 41021003)Special Project Seismic Commonweal research (Grant No.201308013)Key Development Program of Chinese Academy of Science (Grant No.KZZD-EW-TZ-05)
文摘On April 20, 2013, a magnitude M s 7.0 earthquake occurred in Lushan, Sichuan Province, China, and caused heavy casualties and economic losses. Based on the local broadband waveforms in Sichuan and adjacent provinces regional networks and teleseismic broadband records from IRIS stations, the focal mechanism and the focal depth are determined by the CAP (Cut And Paste) and its upgraded methods, CAPtele and CAPjoint, respectively. The results show that the focal mechanisms and depth from different methods are steady, and the best double couple solution derived from the joint inversion is 210°, 44°, and 91° for strike, dip, and rake angles respectively for one nodal plane and 29°, 46°, and 89° for another with 16 km focal depth and M w 6.66 moment magnitude. In order to verify the reliability of the results, a number of tests are performed based on local seismograms with different velocity models. They indicate that there is about 10 degree's fluctuation in focal mechanisms and about 2 km variation in focal depth with a complex velocity structure. Furthermore, inverted by re-sampling the teleseismic waveforms on the basis of epicentral distance, the solutions are consistent with each other, which manifests that the teleseismic records are effective for constraining source parameters of the Lushan earthquake.
