The transition from the metastable state to the meta-instability stage indicates that the seismic fault has entered an irreversible deformation process and will lead to an inevitable instability(Ma Jin et al., 2014). Therefore, identifying the meta-instability stage is helpful for the judgment of short-term earthquake precursor anomalies. Under laboratory conditions, the meta-instability stage can be visually identified through stress-time curves, thus potentially predicting the occurrence of the laboratory earthquake. However, there are significant differences between field conditions and laboratory environments. Firstly, the underground medium and structural conditions in the real earthquake source region are unclear and far more complex than laboratory specimens. Secondly, the distribution of the sensors, sensor density, as well as measurement accuracy are limited by various conditions, making it impossible to construct an ideal observation environment covering the entire region. Thirdly, the loading stress cannot be directly measured, and the current actual stress state of the study area is unknown, which is the most difficult problem to solve. Therefore, under the guidance of meta-instability experiments and theories, it is a beneficial attempt to conduct retrospective studies on typical earthquake cases with relatively good observation conditions in the past, analyze the spatial-temporal evolution of different physical fields at different stages before the earthquakes, compare the observed phenomena with the characteristics and change processes of meta-instability stages obtained from experiments or theoretical research. Its final goal is to find possible characteristics or indirect criteria for meta-instability stages under field observation conditions.
Therefore, taking the Ludian MS6.5 earthquake as an example and under the guidance of the meta-instability experiments and theories, the paper comprehensively analyzes the relationship between the spatial-temporal evolution of precursory anomalies and the meta-instability process based on the seismic activity and the geophysical observation data prior the earthquake, and combined with numerical simulation results of the earthquake nucleation. The Ludian MS6.5 earthquake occurred on August 3, 2014 in northeastern Yunnan Province, China. The observation conditions in this region were relatively good, with 26 seismometers within a 300-km radius of the epicenter, which were able to basically monitor earthquakes with completeness magnitude ML≥1.5 and locating accuracy of less than 20km. There were 79 fixed geophysical observation stations, including 11 within 100km, 32 between 101~200km, and 36 between 201~300km. The observation terms covered 43 deformation observations(22 tilt observations, 18 borehole strain observations, and 3 gravity observations), 187 underground fluid observations(90 water physical observations such as water level and temperature, 43 material compositions measurements including radon, mercury and so on, 26 gas measurements such as CO2, and 28 ion measurements including bicarbonate, calcium, and magnesium), and 52 electromagnetic observations(36 geomagnetic observations, 16 resistivity and electromagnetic wave observations). There were a large number of credible medium- and short-term precursor anomalies before the Ludian MS6.4 earthquake, a total of 48 precursor anomalies were identified. Among of them, there were 8 seismic anomalies and 40 geophysical anomalies, accounting for approximately 15% of all measurement items. Among these 40 geophysical anomalies, 31 were proposed before the earthquake, and most of them were investigated and verified on-site with reliable changes.(Wu, et al., 2019).
Based on this abundant precursor abnormally data before the Ludian MS6.4 earthquake and further systematic analysis, a typical earthquake case and relevant observational facts have been provided which can support the viewpoint that during the meta-instability stage, the earthquake nucleation occurred in the epicenter region and the synergy process evolved continuously in surrounding area of the epicenter. The results show that based on large-scale strong earthquake activities and the observation data of the mobile gravity, it can be determined that the concerned area was already in a high-stress state before the Luding earthquake. At that time, the stress level in the large area including the epicenter of the Ludian earthquake was relatively high, and the northeastern Yunnan region and its nearby areas where the Ludian earthquake occurred were already in a critical stress state where strong earthquakes could occur at any time. Under the premise of determining a high-stress state, according to the precursors of seismic activities and geophysical observation precursor anomalies, it can be roughly determined that the meta-instability process of the Ludian earthquake may have begun seven or eight months before the mainshock. The most prominent phenomenon or judgment index is the transition of the fault stress state from accumulation to release, characterized by the active of small earthquakes near the epicenter, as well as the synergistic phenomenon of fault deformation characterized by the significant increase in the number of geophysical observation anomalies, which is related to the expansion process of the core weakening zone in the late period of the earthquake nucleation. After that, until the occurrence of the mainshock, two times should be paying attention to. Firstly, four to five months before the mainshock, the spatial distribution range of the geophysical observation anomalies expands significantly from the epicenter area to the periphery region, indicating accelerated synergistic deformation of the fault. Secondly, after two months before the mainshock, the small earthquake activities near the epicenter began to weaken, and the micro-earthquake activities and the geophysical anomalies showed a migration and contraction towards the epicenter, which is associated with the contraction process of the core weakening zone during the final stage of the earthquake nucleation. The concept of seismic meta-instability proposed from the perspective of stress changes in seismic fault has an explicit physical implication, and the meta-instability stage is associated with the earthquake nucleation process(He et al., 2023). The basic premise for the meta-instability theory to play a role in short-term earthquake prediction lies in how to apply laboratory research results to natural earthquakes, understand whether the regional or fault stress state tends to or enters a meta-instability state through field observations, and further utilize it for practical earthquake prediction.
