Seismic-ionospheric disturbances in ionospheric TEC and plasma parameters associated with the 14 July 2019 Mw 7.2 Laiwui earthquake detected by the GPS and CSES

In this study, with cross-valid analysis of total electron content (TEC) data of the global ionospheric map (GIM) from GPS and plasma parameters data recorded by China Seismo-Electromagnetic Satellite (CSES), signatures of seismic-ionospheric perturbations related to the 14 July 2019 Mw7.2 Laiwui earthquake were detected. After distinguishing the solar and geomagnetic activities, three positive temporal anomalies were found around the epicenter 1 day, 3 days and 8 days before the earthquake (14 July 2019) along with a negative anomaly 6 days after the earthquake, which also agrees well with the TEC spatial variations in latitude-longitude-time (LLT) maps. To further confirm the anomalies, the ionospheric plasma parameters (electron, O+and He+densities) recorded by the Langmuir probe (LAP) and Plasma Analyzer Package (PAP) onboard CSES were analyzed by using the moving mean method (MMM), which also presented remarkable enhancements along the orbits around the epicenter on day 2, day 4 and day 7 before the earthquake. To make the investigations more convincing, the disturbed orbits were compared with their corresponding four nearest revisiting orbits, whose results indeed indicate the existence of plasma parameters anomalies associated with the Laiwui earthquake. All these results illustrated that the GPS and CSES observed unusual ionospheric perturbations are highly associated with the Mw 7.2 Laiwui earthquake, which also strongly indicates the existence of pre-seismic ionospheric anomalies over the earthquake region


Introduction
Electromagnetic phenomena possibly associated with natural disasters (earthquake, tsunami and volcanic activities) have been extensively investigated in recent years, and seismic related anomalies are the most important ones. Although the physical mechanism about the seismic ionospheric anomalies is still unclear, a significant number of observational studies suggest that there is indeed a connection between the two phenomena. In general, the seismic ionospheric disturbance mainly includes the "earthquake precursor" effect of ionospheric TEC and plasma parameters.
There are currently two major types of methods for the measurement of seismic associated ionospheric anomalies: the ground-based stations and space-based satellites.
The total electron content (TEC) derived from measurements of local ground-based GPS receivers was first employed by Liu et al. (2001) to study ionospheric electron density variations during the 1999 7.6 ChiChi earthquake and he found that the GPS TEC around the epicenter dramatically decrease in the afternoon period a day, 3 days, and 4 days before the earthquake. After that,

Seismic Event Information
Indonesia is one of the most seismically active regions in the world, with comparatively much higher probability of seismic events occurrences. The frequent occurrence of earthquakes in this area provides an excellent chance and condition to study the phenomenon of seismic-ionospheric anomalies. Consequently, in this paper To investigate the TEC variations, the GIM data provided by NASA Jet Propulsion Laboratory (JPL) were adopted to this study. The GIM is constructed into 5°× 2.5° (Longitude, Latitude) grid with time resolution of 2 hour. GIM data are generated using data from 150 GPS sites of the IGS and other institutions. In our study, the TEC data based on the date and geographic location of Laiwui earthquake from 75 days before to 10 days after (30 April to 24 July) the main shock occurred.

China Seismo-Electromagnetic Satellite Data
The China Seismo-Electromagnetic Satellite (CSES), which is also named as The scientific payload of the CSES is composed of several instruments that provide a nearly continuous survey of ionospheric plasma, waves, and energetic particles. In this study, the electron density and electron temperature data derived from LAP (Langmuir Probe), ion density ( + , + ) and ion temperature data derived from PAP (Plasma Analyzer Package) were applied to this research. Besides, also as shown in Figure 1 there were about 100 flight orbits (most are revisited orbits) above the earthquake region from one month before to 10 days after the earthquake (14 June to 24 July, 2019), which provided a significant amount of scientific observation data to our study.

TEC anomalies
The moving median and inter-quartile scope of data are used to shape the upper and lower bounds so that the seismic-anomalies could be separated from the background . In addition, to calculate the statistical parameters, the length of the period was selected as about 55 days in order to avoid affects by the seasonal variations. The upper and lower bound of the mentioned range can be calculated using the following equations (1)-(4): and  are the 30-day   TEC moving median value, TEC inter-quartile range, TEC upper bound, TEC lower bound, TEC observed value, differential of TEC and threshold of the anomaly, respectively. Here we set the k = 2.0 considering the magnitude of the main shock to select the anomalies interval. Over and above, while the absolute value of is larger than the value (| | ≥ ), the behavior of the pertinent TEC value will be noted as anomalous.

Plasma parameters perturbations
In this study, GIM TEC anomalies derived from GPS satellites 45 days before to 10 days after the earthquake have already been analyzed. To further confirm the observed TEC anomalies, a cross-valid examination is conducted with the application of the observation data from CSES.
As introduced in Section. 2.3, the data recorded by payloads LAP and PAP on CSES are adopted to study the ionospheric plasma parameters perturbations above the earthquake preparation zone during the period of 30 days before (14 June 2019) to 5 days after (19 July 2019) the Laiwui earthquake. We examine the percentage deviation of the plasma parameters recorded by CSES within 30 days before and 10 days after the earthquake via moving mean method. Deviation of the plasma parameters can be calculated using the following formulas: and are the CSES observed values for each plasma parameter, while and are perceived as the background values, which are the corresponding moving means from previous 30 days orbit data (data cell is sampled by 2 ° in latitude and 4° in longitude). Indeed, unusual perturbations in different ionospheric plasma parameters are detected prior to 14 July Laiwui earthquake. Figure   5 (a) to (e) displays the percentage deviation of electron density (Ne), electron temperature ( ), + density ( + ), ion temperature ( ) and + density ( + ) respectively. However, due to the measurement limitation of the PAP instrument, there is little valid data for + density above the earthquake area and the measurement of + density is also not persistent for some certain orbits.
From the TEC anomalies analysis, the TEC anomalies were detected on 8 days (6 July), 3 days (11 July) and 1 day (13 July) prior to the earthquake. A further cross-valid analysis is conducted during these periods. As shown in Figure 6(a), the electron density increased significantly on day 4 (10 July) and day 2 (12 July) before the earthquake, the maximum value increased by approximately 135.32% and 115.69% respectively when it approached the epicenter. While, on day 3 (11 July) and day1 (13 July) before the earthquake, the maximum only increased by about 16.80% and 39.58%. Similarly, as shown in Figure 6(c) the main component + density also increased dramatically on day 4 (10 July) and day 2 (12 July) before the seismic event when it approached the epicenter, the maximum value increased by 160.10% and 153.74% respectively. While, the + density remained relatively stable 3 days before and 1 day before with a slight increase about 10.63% and 21.73%. Although the observation data of + density is not persistent for some orbits, as shown in Figure 6(e), the variation tendency of + density can still be observed from the data recorded by CSES. + density profoundly increased when flying above the epicenter on 4 days before the earthquake with maximum increased by 154.76%. The electron temperature and the ion temperature, however, remained relatively stable during the observation period, with a comparatively slight increase no more than 60% for all orbits. Also, Figure 6(a) to (e) displays the percentage deviation of the same plasma parameters, while the observation period is from 4 July to 8 July. Since a magnetic storm occurred on 8 July, as shown in Figure 2, it is difficult to distinguish whether the anomalies on 8 July is caused by the magnetic storm or the earthquake. As displayed in Figure 6(a), the electron density dramatically increased on 5 July and 7 July when approaching the epicenter with a maximum increase by approximately 129.29% and 151.17% respectively. While the electron density on 4 July and 6 July remained relatively stable, with the greatest deviation percentage no more than 80%. + density increased significantly by the order of 112.61% and 197.77% on 5 July and 7 July, while the adjacent orbits remained relatively stable. Besides, + density also increased simultaneously with the + density on 7 July, with a maximum increase by 186.29% Furthermore, the variation of the electron and ion temperature still remained relatively stable (deviation no more than 50%) without significant perturbations during the observation period. Furthermore, it also should be noted that CSES usually flew above the earthquake preparation zone twice a day, with a descending orbit and an ascending orbit respectively. The obvious disturbances were observed by CSES when it was flying above the region with descending orbits. However, we do not find any similar variations in plasma parameters before the occurrence of earthquake during ascending orbits at about 02:00 LT (17:00 UT).

Discussion and Conclusion
In this study, seismic-ionospheric anomalies associated with the 7. Nevertheless, further discussions are required for some of the results. A cross-valid analysis of TEC anomalies was conducted using the LAP and PAP data of CSES, and the results are indeed consistent with the former one. There were great enhancements in plasma parameters (electron density, ion density, etc.) prior to the earthquake, however, there was also a difference between the two results. For example, the most significant anomalies of the electron density were observed on day 4 (10 July) and day 2 (12 July) before the earthquake, while the TEC anomalies were exactly one day after the CSES observations. Besides, a negative anomaly in TEC was also detected 6 days after the earthquake, but similar disturbances were not detected by CSES. There might be several reasons accounting for these discrepancies, on the one hand, as shown in Figure 4b, the Kp index increased at 4:00-10:00 UT on 10 July, while the electron density also increased simultaneously during the same period, so the disturbances on 10 July may be related to the geomagnetic activity. On the other hand, the discrepancies may be mainly attributed to the difference of the two datasets. To be specific, CSES is a spacecraft exploring the topside ionosphere at an altitude about 507 km with in-situ observations, while the GPS-TEC is calculated under the assumption of the ionospheric single layer.
Besides, some of the TEC anomalies were observed outside the preparation zone., as shown in Figure 3 and Figure 4. This can be attributed to lithosphere-atmosphereionosphere coupling (LAIC) process, the Dobrovolsky formula is an ideal equation without the consideration of the LAIC process. The earthquake-related anomalies induced by the LAIC mechanism works complicatedly, whose wave channels mainly is At daytime, the ionosphere received much more radiation from the sun resulting in more ionized particles, which significantly increases the density of the ionospheric electrons and ions. While the densities of the electrons and ions are much lower at night time. Due to the relatively lower electron and ion densities, the variations of the ionospheric plasma parameters will be much more difficult to be detected by CSES.
Thus, the plasma parameters observed by CSES remained relatively stable during night time prior to the earthquake. Also, perturbations in electron density occurred more often than those of electron temperature from the observation results of CSES, which illustrate that electron density is much more sensitive to seismic activity than electron temperature, this is also consistent with the statical investigation conducted by Liu et al. (2014). Besides, it also should be noted that the PAP instrument of CSES is slightly contaminated, with lower absolute value in observation data, therefore the data of PAP can only be applied to the relative deviation analysis.
In conclusion, during these periods, the measurements of GPS and CSES yield similar tendencies, the temporal and spatial anomalies of the TEC and ionospheric plasma perturbations detected by CSES over the epicenter did indicate significant positive seismic-ionospheric anomalies. Based on the results presented, we can also safely draw the conclusion that CSES data are reliable for the study of seismic events.
Also, the localization and synchronization of the longtime anomalies around the occurrence of earthquake suggest that these perturbations are highly associated with 7.2 Laiwui earthquake, but further investigations are required in the future to obtain a more accurate knowledge of the perturbation process.