Method and system for monitoring particle precipitation in the magnetosphere - Patents.com

Abstract

A method (100) for monitoring magnetospheric particle precipitation, comprising the steps of: detecting charged magnetospheric particles by at least one particle detector (10) mounted on at least one satellite vehicle (1) in orbit (2) and associating (101) the detected particles with respective detection data; processing the detection data to associate (102) respective estimates or measurements of kinetic energy with the detected magnetospheric particles; and generating a first count value N correlating to the number of charged particles detected in a period of time, the first count value N being associated with a relatively high estimate or measurement of kinetic energy falling within a first energy range. H and obtaining (104) a second count value N correlating to the number of charged particles detected during said time period associated with a relatively low estimate or measurement of kinetic energy within a second energy range. L (105) acquiring a first count value N H The second count value N L determining that impulsive events of charged particle precipitation in the magnetosphere (MPP events) have occurred during said time period and comparing said fluctuations with a threshold; assigning said MPP events to geomagnetic longitudes and times at which they occurred (109); defining one or more groups of MPP events, each group including MPP events that occurred at the same geomagnetic longitude or in a time range of relatively close geomagnetic longitudes (110); and identifying (111) a group of MPP events as indicative of terrestrially-originating activity, e.g., pre-earthquake or seismic activity, based on the number of MPP events included in the group and / or based on associated fluctuations found in the detecting step (106).

Description

[Technical field]

[0001] The present invention relates to the technical field of satellite systems for monitoring particles in space, and in particular for acquiring and processing data on particle activity. In particular, the present invention relates to a method and system for monitoring particle precipitation in the magnetosphere, which may be implemented using particle detectors. Such a method and such a monitoring system may be used, for example, to identify possible areas of pre-earthquake activity and to provide warnings for potential activity related to earthquakes. The monitoring system and method may be used generally to monitor electromagnetic perturbations of the Earth's magnetosphere induced by terrestrial phenomena. [Background technology]

[0002] Demonstrating how to correlate earthquakes with physical signals detectable by current technology, with high and reliable statistical significance, is a difficult problem to solve. Previous attempts in the past decades have involved several research groups in many countries affected by strong seismic activity, including groups in Italy, Greece, the United States, Japan and China. Methods used have included ground-based measurements and, more recently, measurements with space-based instruments.

[0003] Several cases of anomalous emissions of electromagnetic signals with peculiar characteristics have been reported, coinciding in time and space with energetic earthquakes, i.e., earthquakes of great intensity in particular. Such results, although interesting, are limited to occasional evidential determinations between earthquakes and ground or space measurements. The reasons for this are numerous: lack of systematic studies of such effects over sufficiently long periods (several years), problems with reproducibility of results, high background noise affecting the measurement process, weak overall statistical evidence, etc. Correlation attempts obtained so far have not been supported by highly statistically significant post-test determinations of the probability of occurrence of such results.

[0004] Since the late 1980s, research has been carried out into the possible correlation between peculiar magnetospheric particle events and earthquakes. The first work on this topic was by Galper AM et al. (1989), who described, for example, -Aleksandrin, S.Yu.;Galper, AM;Grishantzeva, LA;Koldashov, SV;Maslennikov, LV;Murashov, AM;Picozza, P.;Sgrigna, V.;Voronov, SA, "Energy charged particle bursts in near-Earth space as earthquake precursors", Annales Geophysicae, 21, 597-602(2003); -Galper, AM et al., "Energy charged particle fluxes in the radiation belts and their relationship to the Earth's seismic activity," Cosmic Research, 27, 789-792 (1989).

[0005] In these studies, attempts were made to develop methods for obtaining correlations between seismic events and electromagnetic perturbations. The method of Galper et al. is based on the following: - a coupled model of the magnetosphere and lithosphere based on the propagation of electromagnetic waves along the magnetic L-shell (defined as the magnetic field lines identified by the distance in units of ground line where they intersect with the equatorial plane of the Earth's magnetosphere) and the interaction of these waves with particles trapped in the L-shell, a phenomenon that leads to particle precipitation; - selection of earthquakes with magnitude (MMS scale) M ≥ 5 (this is a rather low value, leading to a correlation between magnetospheric particle precipitation and earthquakes of low statistical significance); - Identification of particle precipitations based on increases or bursts of energetic particles produced by wave-particle resonance. These particle precipitations are observed by satellites equipped with appropriate instruments for detecting magnetospheric particles; -Temporal correlation between magnetospheric particle precipitation and seismic events based on the time difference between magnetospheric particle precipitation and seismic events occurring in the same L-shell within a narrow range.

[0006] However, due to the chosen methodology and the chosen population of earthquakes (magnitudes too low), the results of the above studies are insignificant from a statistical point of view. In recent years, no substantial improvements have been observed since these first attempts.

[0007] A general description of the above methods is presented, for example, by Pulinets & Boyarchuk in Pulinets, S. & Boyarchuk, K. "Ionospheric precursors of earthquakes", Springer (2004). In this publication, the section devoted to the detection of magnetospheric particle precipitation is based on the aforementioned article by Galper et al.

[0008] An informed description of electromagnetic waves that may be related to earthquakes is given in Hayakawa M., Earthquake Prediction by Radio Wave Techniques, (Singapore, John Wiley & Sons, 2015). In this and other publications: -Hayakawa, M., Yoshino, T. & Morgounov, VA, "On the possible influence of seismic activity on the propagation of magnetospheric whistlers at low latitudes", Phys. Earth and Planets. Interior, 77, 97-108 (1993); -Hayakawa, M., Hobara, Y., Ohta, K., & Hattori, K., "Very low frequency magnetic disturbances associated with earthquakes", Earthquake Science, 24(6), 523-534(2011); -Ohta, K., Izutsu, J., Schekotov, A. & Hayakawa, M., "ULF / ELF electromagnetic emissions prior to the 11 March 2011 earthquake in Japan", Radio Science, 48, 589-596 (2013); Circumstantial evidence is presented of low frequency electromagnetic emissions observed in association with large magnitude earthquakes. These ground-based observations focus on the detection of ULF and VLF waves using ground-based instruments, and the emissions are likely consistent with powerful earthquakes. See also publications. -Currie, JL & Waters, JL, "On the use of geomagnetic indicators of earthquake precursors and ULF waves as earthquake precursors", J. Geophys. Res. Astrophysics, 119, 992-1003 (2014); -Park, SA, Johnston, MJS, Madden, TR, Morgan, FD & Morrison, HF, "Electromagnetic precursors of earthquakes in the ULF band: a review of observations and mechanisms", Rev. Geophys., 31, 117-132 (1993).

[0009] Furthermore, some cases of magnetospheric anomalies regarding the so-called total electron content (TEC), taking into account their spatial and temporal proximity to some earthquakes, are also discussed, for example, in the abovementioned publication: Hayakawa, M., Earthquake Prediction by Radio Techniques, (Singapore, John Wiley & Sons).

[0010] However, none of the aforementioned studies presented a systematic study, and a global statistical analysis of possible correlations between ULF / VLF signals and / or TEC anomalies and earthquakes was not discussed, even if candidates could be identified therefrom.

[0011] The DEMETER satellite group (operating between 2004 and 2010) published observations of wave perturbations in the ionosphere in spatial and temporal proximity to earthquakes of very high magnitude, see for example publications. -Onishi, T., Parrot, M. & Berthelier, J.-J., "DEMETER mission, recent investigations of ionospheric influences related to artificial activities and seismic phenomena", Comptes Rendus Physique, 12, 160-170 (2011); -Parrot, M., Berthelier, J.~J., Lebreton, J.~P., Sauvaud, J.~A., Santolik, O., Blecki, J., "Examples of anomalous ionospheric observations in earthquake regions by the DEMETER satellite", Physics and Chemistry of the Earth, 31, 486-495 (2006).

[0012] These observations from space (from an altitude of about 700 km) show interesting detections of VLF and LF waves. However, global analysis of DEMETER data in close temporal proximity to the earthquakes does not support the existence of a correlation with high statistical significance between the emission of these waves and earthquakes.

[0013] In fact, so far no satisfactory demonstration (based on rigorous and complete statistical analysis) of a systematic correlation between magnetospheric particle events and earthquakes has been provided. The existence of such a correlation would mean that earthquakes emit electromagnetic and plasma waves that can propagate along magnetic field lines and interact by resonance with electrons and positrons (secondary albedo particles) of the Earth's inner magnetosphere. Only circumstantial evidence has been presented in the literature mentioned above, and so far evidence of a possible correlation has been entirely discussed with insufficient statistical analysis.

[0014] One of the main problems with the Galper et al. method is related to the use of the L-shell to identify geographical areas that may be affected by seismic events. Conventionally, the areas identified in the aforementioned method consist of two nearly horizontal "strips" of longitude, one above and one below the magnetic equator, both extending around the entire circumference of the Earth, corresponding to the intersection of the magnetic field lines corresponding to the specified L-shell with the Earth's surface. This feature creates a large uncertainty about the geographical areas where seismic events may occur.

[0015] Other major limitations of Galper et al.'s method are the low statistical significance and the large number of spurious correlations (false positives) caused by the selection of magnetospheric events to correlate with earthquakes of a relatively low threshold magnitude (M = 5).

[0016] With regard to proposed magnetospheric anomalies associated with earthquakes (such as phenomena affecting the TEC), this evidence has always been presented as events of an "after the fact" nature, without a full discussion of the global context in which such anomalies usually become apparent, and without mentioning the lack of discussion of "false positive" events. Similar conclusions can be drawn for the results of the DEMETER satellite, which aimed to highlight ionospheric perturbations and / or anomalous electromagnetic signals in the lower magnetosphere.

[0017] It can therefore be concluded that the current magnetospheric measurements of particle bursts associated with earthquakes via the magnetic L-shell, TEC-type anomalies, and ionospheric perturbations are, at best, circumstantial evidence for a possible correlation with earthquakes. Investigations based on these assumptions or the methods proposed so far are therefore not based on sufficient statistical evidence.

[0018] There is therefore still a felt need to develop a monitoring method that makes it possible to completely or at least partially overcome the drawbacks and limitations of the above-mentioned prior art methods. It is therefore a general object of the present specification to provide a monitoring method that makes it possible to meet the aforementioned need.

[0019] This general object is achieved by a monitoring method as generally defined in claim 1. Preferred and advantageous embodiments of said monitoring method are defined in the attached dependent claims.

[0020] The present invention will be better understood from the following detailed description of particular embodiments, given by way of example and not limiting in any way as a result, with reference to the accompanying drawings, which are briefly described in the following paragraphs. [Brief description of the drawings]

[0021] [Figure 1] 1 shows a schematic diagram of one non-limiting embodiment of a system adapted and configured to implement a method for monitoring particle precipitation in the magnetosphere according to the present invention; [Diagram 2] 2 illustrates a possible embodiment of a particle detector that may be employed in the system of FIG. 1; [Diagram 3] 1 shows a non-limiting embodiment of a flow chart of a method for monitoring particle precipitation in the magnetosphere according to the present invention. [Figure 4] The acquired data on particle sedimentation is shown in a two-dimensional time-longitude diagram. [Figure 5a] 4 shows a timing histogram of data acquired with a monitoring method according to the present invention. [Figure 5b] The figure in Fig. 5a shows a timing histogram of historical data relating to earthquake events for the same period. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] Figure 1 shows a non-limiting embodiment of a monitoring system 1, 2 adapted and configured to implement a method 100 for monitoring particle precipitation in the magnetosphere according to the present invention. Figure 3 shows a flow chart of a possible exemplary and non-limiting embodiment of said monitoring method 100.

[0023] With reference to Figure 1, the monitoring system 1, 2 comprises at least one satellite vehicle 1, or more simply a satellite 1 or a space segment including a satellite 1, movable in an orbit 3 relative to the planet Earth 4. On board the satellite vehicle 1, at least one particle detector 10 is provided, adapted and configured to detect charged electromagnetic particles, preferably charged particles having a kinetic energy of less than 1 GeV.

[0024] According to an advantageous embodiment, said space segment comprises a plurality of satellite vehicles 1, ie a constellation of satellite vehicles 1, each of which is equipped with a respective particle detector 10.

[0025] The monitoring system 1, 2 comprises at least one ground segment including at least one ground station 2 adapted and configured to be operatively connected, directly or indirectly, with the satellite vehicle 1 in order to receive the telemetry data transmitted by the satellite vehicle 1. The ground station 2 comprises, in a manner known per se, a satellite data receiving device and one or more processors adapted and configured to process the received data to generate processed data. Said operative connection may be direct or may be made by other satellite vehicles 1 or by space vehicles or stations in general. The ground station may comprise several stations that are geographically distributed and operatively connected to each other. The monitoring system 1, 2 preferably comprises a display system (not shown), such as, for example, at least one display adapted and configured to display the processed data. The display system is, for example, housed in the ground station 2 or in another station or processing center operatively connected to the ground station 2.

[0026] According to a particularly advantageous embodiment, the orbit 3 of the satellite vehicle 1 is an equatorial orbit, preferably an equatorial low Earth orbit (LEO). Since the charged particles of interest in the monitoring method 100 according to the invention are charged particles that resonate with magnetospheric plasma waves (the latter being a physical phenomenon that occurs efficiently only with particles such as electrons and positrons), some advantages can be obtained by choosing an equatorial LEO orbit with respect to a polar orbit. In particular, - In equatorial LEO orbits, the flow of particles in the energy range between 0.01 GeV and 0.10 GeV is dominated by electrons and positrons (which appear as secondary albedo particles with respect to primary cosmic rays). The flow is 10 3 ~10 4 (m 2 s sr GeV) -1 It is. - An equatorial orbit allows one to avoid backgrounds caused by the massive flow of solar particles and by the flow of cosmic protons in the kinetic energy range below 1 GeV. - Equatorial orbits are best for acquiring signals based on geomagnetic longitude. Conversely, polar orbits are not suitable for this purpose, as they acquire signals based primarily on the L-shell, not on longitude.

[0027] According to a particularly preferred but non-limiting embodiment, the satellite vehicle 1 is a satellite called AGILE, for example as described in the article "The AGILE Mission, Astron. & Astrophys", 502, 995-1013 (2009) by Tavani, M. et al.

[0028] In fact, the AGILE satellite is in unique conditions to study the phenomenon of particle precipitation in the inner magnetosphere, since its orbit is not subject to solar storms or external magnetic perturbations. The AGILE satellite currently orbits the Earth in an equatorial orbit at an altitude of about 500 km. On the AGILE satellite, a particle detector 10 is installed, which is in particular an imaging gamma detector (GRID) which, in addition to detecting cosmic gamma rays with energies above 20 MeV, is also able to efficiently detect charged electromagnetospheric particles. This is precisely described, for example, in the article "AGILE as a particle detector" by Argan, A., Piano, G., Tavani, M. & Trois, A., Journal Geophys. Res., 121, 3223-3239 (2016). This latter article is fully incorporated herein by reference for describing the structure and operation of a non-limiting example of a particle detector 10 suitable for use in the monitoring method 100 according to the invention, in particular with regard to the structure and operation of the particle detector GRID.

[0029] For the AGILE satellite, the flow of charged particles in an orbit of about 95 minutes is dominated by electrons and positrons with kinetic energies less than 1 GeV. The charged particle detection rate by the particle detector GRID (after passing through the anti-coincidence system of the detector mentioned above) is relatively low (a few Hz). Orbit 3 of the AGILE satellite is not affected by temporary increases in the number of charged particles induced by solar activity or magnetic perturbations, which strongly affect other satellites in inclined or polar orbits.

[0030] In the case of the AGILE satellite, the acceptance of the particle detector grid to detect particles is approximately 50 cm, with the instrument response depending on the angle of incidence of the particle with respect to the axis of the particle detector grid. 2 sr. The detection of cosmic gamma photons therefore coexists with the efficient acquisition of data of charged particle events. These latter events trigger the on-board data acquisition system and are preferably transmitted to the ground station 2 together with precise data on the time of acquisition, the local magnetic field, and information on the orbital position.

[0031] Referring now to FIG. 3, a monitoring method 100 includes step 101 of detecting charged electromagnetic particles ("Part_Detect") by a particle detector 10 mounted on an orbital satellite vehicle 1 and associating the detected charged particles with respective detection data.

[0032] As already explained, a possible example of the particle detector 10 is the particle detector GRID on board the AGILE satellite, in such case e.g. a gamma ray imaging detector detecting charged electromagnetospheric particles with kinetic energies less than 1 GeV. In alternative embodiments, the particle detector 10 may be e.g. a simpler and less bulky particle detector, for example configured to detect and track only charged particles.

[0033] According to a typical embodiment, the particle detector 10 comprises a particle imaging tracker and / or an imaging calorimeter.

[0034] According to a particularly advantageous embodiment, the particle detector 10 comprises a multi-layer structure 11 or a segmented structure 11, for example comprising a plurality of solid planar sub-detectors, for example made of silicon-tungsten, arranged to form a stack of sub-detectors aligned along the axis z. Each of the aforementioned sub-detectors makes it possible to detect, for example per charged particle, the coordinates x, y on the impact surface and, preferably, the amount of charge deposited on the sub-detector by the charged particle. The aforementioned multi-layer structure thus makes it possible to track in 3D the path of the charged particle in the particle detector 10. A charged particle having a relatively large amount of kinetic energy may traverse the entire multi-layer structure 11 and affect all the sub-detectors, also according to a direction, while a charged particle having a relatively small amount of kinetic energy may stop after a collision with one or more of the sub-detectors. In the particular example shown in FIG. 2, corresponding to the particle detector GRID on board the AGILE satellite, the particle detector 10 comprises a multi-layer structure 11 and a calorimeter 12, preferably of, for example, caesium iodide, the multi-layer structure 11 being located above the calorimeter 12. The particle detector 10 may, for example, comprise an anti-incidence screen 13 .

[0035] For example, the detection data associated with each charged particle in step 101 of the monitoring method 100 may include data relating to the path, e.g., in 3D, of the charged particle within the particle detector 10, and / or the amount of charge deposited by the particle within the detector 10, and / or the time at which detection of the charged particle occurred, and / or the geomagnetic longitude at which detection of the particle occurred.

[0036] The monitoring method 100 further comprises a step 102 ("E_Assign") of processing the detection data to associate, i.e. assign, the detected magnetospheric particles to a respective estimate or measurement of their kinetic energy. An estimate or measurement of kinetic energy means a quantitative type of data that corresponds to or correlates with the kinetic energy of the charged particle. This estimate or measurement can be obtained according to various methodologies known to the person skilled in the art, for example by counting the number of sub-detectors crossed by the charged particle, and / or by analyzing the geometric path of the charged particle in the multi-layer structure (e.g. the deviation of such a path), and / or by measuring the amount of charge deposited by the particle, for example on each of the sub-detectors involved, and / or by using the data provided by the calorimeter 12 (if provided).

[0037] According to an advantageous embodiment, the monitoring method 100 further comprises a step 103 of determining the pitch of the arrival direction of the charged magnetospheric particles relative to the local magnetic field and selecting (for further processing) particles with a pitch that is included in one or more pitch ranges. This makes it possible to discard detection events that are absolutely or statistically not important for further processing. For example, charged particles whose arrival direction is perpendicular or nearly perpendicular to the local magnetic field can be discarded, since charged particles characterized by these arrival directions are statistically not particles undergoing a precipitation phenomenon.

[0038] The monitoring method 100 includes: a first count value N correlating to the number of charged particles detected during a period of time; H Step 104 ("N H _Obtain" associated with a relatively high estimate or measurement of kinetic energy that falls within a first energy range; a second count value N correlating to the number of particles detected during said time period; L Step 105 ("N L"_Obtain" associated with a relatively low estimate or measurement of the kinetic energy that falls within a second energy range; Further includes:

[0039] The first and second energy ranges may be defined as high and low energy channels, respectively.

[0040] According to a particularly advantageous embodiment, the first and second energy ranges are determined so that under normal conditions the first count value N H and a second count value N L are selected to be equal or nearly equal to one another.

[0041] Conveniently, the first energy range includes energies above 60 MeV and the second energy range includes energies below 40 MeV. Advantageously, the second energy range includes energies above 5 MeV, for example above 10 MeV and below 40 MeV.

[0042] According to a possible embodiment, the two energy ranges are adjacent to each other and have in common an extreme value equal to or equal to about 50 MeV, for example, the first energy range includes energies lower than (or lower / equal to) about 55 MeV and the second energy range includes energies higher than (or equal to) 55 MeV.

[0043] For purposes of this specification, the word "about," when in relation to a numerical value, is intended to indicate both the exact numerical value as well as a variation above and below that numerical value of up to 10%.

[0044] Referring again to FIG. 3, the monitoring method 100 includes: -First count value N H The second count value N L and a step 106 of detecting the relative variation of ("Delta_Detect"); - step 107 ("MPP_Determ") of comparing said change with a threshold to determine that an impulsive event of particle precipitation in the magnetosphere, or "MPP event", has occurred during said time period; Further includes: From a practical point of view, an MPP event is an event that indicates the impulsive precipitation of charged particles due to resonance phenomena between charged particles and radiation, e.g., plasma waves, propagating in the Earth's magnetosphere, said radiation may be caused by natural and / or artificial phenomena affecting, e.g., the Earth's atmosphere and / or lithosphere.

[0045] According to a particularly advantageous embodiment, said threshold value is established by relating said historical data of variations to historical data relating to events of earthquakes or pre-earthquake activity, including geolocation and intensity information of said events.

[0046] The aforementioned MPP events constitute events that represent impulsive precipitation of charged magnetospheric particles, with respect to the precipitation of charged particles generally detectable in the absence of perturbations.

[0047] The above time range may be a wide range or a narrow range, for example, on the order of 1 second or on the order of 10 seconds.

[0048] According to an advantageous embodiment, the step 106 of detecting the relative variation includes detecting a first count value N H and the second count value N L and vice versa. Thus, step 107 of determining that an MPP event has occurred ("MPP_Determ") includes verifying whether said HR ratio is above, below or equal to said threshold value.

[0049] For example, HR=N H / N L If N is 0, the threshold above which it may be determined that an MPP event has occurred is equal to or greater than 10, e.g., equal to about 15, e.g., equal to 14. H Value and N LIt has been shown to be advantageous to select the energy range in which values ​​are counted so that such values ​​will, under normal conditions, be equal or nearly equal to one another, and therefore the HR ratio will normally be equal to or nearly equal to 1.

[0050] According to an advantageous embodiment, the monitoring method 100 further comprises a step 108 of removing and / or identifying spurious events ("SE_Discard"), correlating said variations detected in the detection step 106 with indices of magnetospheric and solar activity. Indeed, it is known that several types of events, such as, for example, solar storms, can cause anomalous precipitation of charged magnetospheric particles.

[0051] The monitoring method 100 includes: - a step 109 ("LT-Assign") of assigning to each MPP event a geomagnetic longitude and a time when the MPP event occurred; - defining one or more groups of MPP events or clusters of MPP events ("CL_Def"), step 110, each group including MPP events occurring in the same geomagnetic longitude or in a time range of relatively close geomagnetic longitudes; Further includes:

[0052] For example, the aforementioned time range may be on the order of a few weeks, such as equal to 3 to 5 weeks, such as equal to 4 weeks.

[0053] For example, relatively close geomagnetic longitudes are those that fall within a range of longitudes having an amplitude equal to a maximum of 20°, such as longitudes equal to or approximately equal to 15° or 10°.

[0054] Following the step 110 of defining one or more groups of MPP events, the monitoring method 100 includes a step 111 ("PPA_Ident") of identifying a group or cluster of MPP events as indicative of a natural or man-made terrestrial activity, such as pre-earthquake or seismic activity, for example based on the number of MPP events contained in the group and / or based on the associated variations found in the detection step 106. Such a number is equal to or greater than 1, preferably equal to or greater than 2. For example, if a very high HR is associated with an MPP event, it may be determined that the cluster is a degenerate cluster and contains only one MPP event. For MPP events with a lower HR, for example equal to 10, it may be established that a group of MPP events must contain a minimum number of events equal to two, or three or four, to indicate pre-earthquake or seismic activity. It is noted that the aforementioned activities are not limited to seismic or pre-earthquake activities. This is because other natural activities (for example of a meteorological nature, such as terrestrial gamma ray flashes) or man-made activities that may affect particle precipitation may also be included.

[0055] Preferably, the monitoring method 100 further comprises a step 112 of generating an earthquake or pre-earthquake alert ("SA_GEN"), said alert being for example displayed on a screen and / or sent by a data message, for example a text message or email.

[0056] For example, in Figure 4, MPP events with calculated HR (Hardness Ratio) values ​​of 14 or greater are mapped onto a two-dimensional longitude-time diagram. These MPP events in fact experienced a large depletion of low-energy charged particle precipitation with respect to high-energy charged particle precipitation. In Figure 4, six groups of MPP events or clusters identified in step 111 of the monitoring method 100 are each marked with a dashed circle.

[0057] It should be noted that the steps of the monitoring method 100 described above may be performed in a different order than in the diagram of Fig. 3. For example, the counting steps 104 and 105 may be performed in reverse or simultaneously. Furthermore, while it is necessary to perform step 101 of detecting particles on board the satellite vehicle 1, storing the data on board the satellite vehicle 1 and transmitting the data acquired in step 101 to the ground station, subsequent processing steps may be performed on the ground, but the execution of some of said steps on board the satellite vehicle 1 may be advantageous to discard particles and / or MPP events already on board the satellite that are not important for subsequent processing steps, in order to reduce the bandwidth resources required for data transmission to the ground. Experimental Results

[0058] Below, we describe the experimental results of data acquired by the AGILE satellite, processed using a processing system based on specially developed software.

[0059] The focus has been on detecting statistically significant impulsive reductions in particles detected in the low energy channel relative to those detected in the high energy channel by the particle detector GRID on the AGILE satellite. "Hardness" is defined as the Hardness Ratio (HR), defined as the number of high energy events divided by the number of low energy events, both determined during the same time range δt. Impulsive MPP events are defined by their associated hardness HR during the time range δt. HR(N) is the hardness value for MPP events with HR values ​​equal to or greater than N. The ratio of particle numbers measured in the two energy ranges (and therefore the HR(N) value) is subject to Poisson statistics and other effects that may affect their presence.

[0060] The particle data flow provided by the AGILE satellite along an equatorial orbit continuously provides HR values ​​according to time T and geomagnetic longitude LG. Given the detection capabilities of AGILE, a suitable binning for analyzing the longitude and time characteristics of HR is given by a time bin of 1 day and a geomagnetic longitude bin of 1 degree. As an example, Figure 4 shows the longitude and time distribution of MPP events measured at HR(14) during the data acquisition of the present analysis (16 April 2015 - 30 November 2017). It is very important to understand that the measurements of MPP events in this case are not influenced by magnetospheric effects due to solar events or atmospheric flashes. Therefore, the lack of local and global uniformity seen in the distribution of MPP events in Figure 4 is not caused by external solar or meteorological effects or by the characteristics of the particle detector on board the AGILE satellite.

[0061] The developed processing system allows to correlate the distribution of detected MPP events with earthquake occurrence, i.e. seismic events. The data required for such earthquake analysis (time, geographic location, magnitude and physical parameters of earthquake events) are obtained from the public archives of the USGS. To minimize the influence of random statistical fluctuations, the analysis considers only earthquakes with a magnitude of M>6 (MMS scale), with a special emphasis on earthquakes with a magnitude of M>7.

[0062] The processing system made it possible to verify the presence of a "global" time correlation between MPP events and earthquakes by considering the summed quantities over 100-day time bins. Such a choice is motivated by the need to first check a possible correlation between the apparently random values ​​of MPP events and a quantity proportional to the total seismic energy of the most energetic earthquakes. For this purpose, the temporal behavior of two amplitudes was compared. The first amplitude is a quantity proportional to the negative value of the logarithm of the joint probability according to Poisson statistics that there is an MPP event in a 100-day bin. The second amplitude is a quantity proportional to the negative value of the logarithm of the joint probability of having an earthquake in the same time bin, assuming an occurrence probability of earthquakes that is anticorrelated with the energy of the earthquake. These quantities were summed up over the entire range of longitude (excluding values ​​for the South Atlantic Anomaly, SAA) and within 100-day time bins, and the cumulative energy of MPP events (HR) and earthquakes (Ψ N ) are plotted separately as histograms of the normalized values ​​of HR(14) and magnitude M>6 for the Northern Hemisphere earthquakes. Figures 5a and 5b show the results of such an analysis for earthquakes with HR(14) and magnitude M>6 in the Northern Hemisphere. The similarity of the two distributions of MPP events and earthquakes is evident at a glance, since their maxima and minima lie within the same time range. A joint statistical analysis of the two distributions aimed at testing the hypothesis of correlation between the two distributions gives a value of the Kolmogorov-Smirnov coefficient equal to 0.97 and a Pearson coefficient equal to 0.928.

[0063] As a next step, geomagnetic longitude L G Considering the fact that a value of can be attributed to each MPP event, the geolocation of MPP events was considered. At this point, MPP events can be mapped onto a two-dimensional longitude-time diagram, and it is possible to identify groups of MPPs (clusters) that contain statistically significant MPPs.

[0064] If some of the MPP events are directly correlated with electromagnetic signals generated in the earthquake-affected regions and channeled along the north-south magnetic meridian, a correlation between prominent MPP events and high-intensity earthquakes can be expected not only in terms of time but also in terms of geomagnetic longitude. Such electromagnetic signals propagate along magnetic field lines (as expected from VLF / ELF and Wisler waves). Given the statistics of MPP events and the coverage of the AGILE satellites per unit of longitude and time, a possible correlation between MPP events and earthquakes can be expected with the Dobrovolski radius R D (This is the radius of the area affected by the earthquake phenomenon; R D are studied on daily and weekly scales in the longitude range given by (depends on the magnitude of the seismic event). For each earthquake, the range ΔL'=2R D A two-dimensional "active region" (AR) is determined, centered on the location of a seismic event, defined in geomagnetic longitude by and time span ΔT'. This latter quantity is the distribution of time delays between the MPPs of HR(15) and earthquakes (EQ) in the geomagnetic northern hemisphere with magnitudes M ≥ 6.4 and M > 7, ΔT = T MPP -T EQ (7-day bins), which occurs within a region within ΔL'. Both distributions show an excess of peak events in the first negative time bin (implying that a subset of MPP events predicts the seismic event) and an asymmetric amplitude of the distributions around the peak. From the observed distributions, the time range defining the active region is taken from the set of three bins preceding the seismic event and the following bins. Thus, the total time range defining the active region is ΔT' = 4 weeks, 3 weeks before EQ and 1 week after EQ. To determine the statistical significance of such a procedure, a full simulation was performed (see section "Statistical Considerations" below).

[0065] After defining the regions of seismic events, areas of activity were assigned to MPP clusters in the 2D map and identified as areas of possible pre-seismic activity (PPA).

[0066] For future events, in case of significant clustering above a threshold of MPP events, the developed processing system generates a pre-earthquake warning by specifying the geomagnetic longitude range, time range, and occurrence probability of clusters of observed MPP events.

[0067] The processing system may also create MPP event maps obtained for various threshold parameters and make them available for electronic transmission (computer and smart phone via a dedicated page / application). statistical analysis

[0068] MPP events with large values ​​of hardness ratio (HR) and active regions of earthquakes with large magnitudes are correlated not only in time but also in geomagnetic longitude. Here we consider the two-dimensional distribution of MPPs with HR(15) and active regions of earthquakes with magnitude M>7. A first estimate of the probability of random occurrence of similarity between the two distributions is the binomial probability p = 2 × 10 that observed MPP events overlap in time and longitude of active regions. -8 These values ​​are confirmed by thorough simulations, as shown below.

[0069] The statistical significance of the obtained results was determined by performing a full simulation of the HR signal, reproducing the random fluctuations caused by Poisson statistics affecting the MPP events. Three conditions were imposed to select the cases of interest for the randomly generated distribution. The first condition (C1) is based on the global temporal correlations shown in Figures 5a and 5b. The simulated distribution of MPP events must meet the requirement of reproducing correlations similar to those shown in Figures 5a and 5b. The second condition (C2) is imposed by the two-dimensional distribution. The total number of HR(15) located within the active area of ​​M>7 seismic events must be equal to or greater than 17 (determined by observations). The third condition (C3) is prescribed by imposing that the number of active areas with at least one HR(15), obtained from observations, is equal to or greater than 10 out of a total of 13. The first estimate of the pretrial probability is the product of the three conditional probabilities C1-C2-C3, P tot-pre =P(C1,C2,C3)=P(C1)P(C2|C1)P(C3|C1,C2).

[0070] Based on the values ​​of the Pearson coefficients in Figures 5A and 5B, the estimated P(C1) = 7.3 × 10 -5 has already been obtained. (Unconditional) probability P(C2) = 2 × 10 -8 We also obtained a binomial estimate of . Furthermore, P(C3|C1,C2) = 0.22. Therefore, the product of these probabilities is P(C1)P(C2)P(C3|C1,C2) = 3.2 × 10 -13 It is.

[0071] However, because the conditional probability P(C2|C1) is difficult to determine, only a complete simulation using a sufficiently large number of iterations can determine the total prior probability P tot-pre Then, we can obtain the final value of 6.267×10 12A complete simulation was performed with a total number of iterations equal to . Such a simulation would generate 25 cases of simulated configurations that fulfill the conditions C1-C2-C3 for the HR(15) and EQ cases in the geomagnetic north hemisphere with M>7. Thus, the result of this simulation is that the pretrial probability for this case is P tot-pre =3.99×10 -12 It is.

[0072] The final probability is obtained by considering the number of trials performed. A conservative estimate of the number of trials is N trial =2.6×10 4 Therefore, the final total probability after trials that the distribution found will occur randomly is P tot-pre and N trial is given by the product of P tot-post [HR(15),M>7]=1.04×10 -7 This value corresponds to a statistical significance greater than 5 sigma in units of the equivalent standard deviation of a two-sided Gaussian distribution.

[0073] HR(15) and earthquakes of M≧6.4 (6.058×10 12 Independent simulations of the cases (repetitions) are performed for conditions C1-C2-C3 and value P tot-pre =1.39×10 -11 Generate 84 cases that satisfy P tot-post [HR(15),M≧6.4]=3.6×10 -7 Such values ​​are just below 5-sigma significance in units of the equivalent standard deviation of a two-sided Gaussian distribution. conclusion

[0074] From the above, it is clear that the above described monitoring method 100 and monitoring systems 1, 2 make it possible to fully achieve the preset objectives in terms of overcoming the shortcomings of the prior art. By analysing the variations in the precipitation of charged particles in the relatively high energy channels and in the relatively low energy channels, respectively, if such variations are deemed significant, a high correlation can be obtained between MPP events and earthquake events of relatively large magnitude (M>6), and therefore an earthquake warning or pre-earthquake warning can be generated.

[0075] Having established with unprecedented statistical significance the existence of a correlation between MPP events and energetic earthquakes in the Northern Hemisphere of the geomagnetic field, it is possible to attribute the cause of the phenomenon of impulsive precipitation of charged particles observed by the described monitoring method to wave-particle resonances in the range of ELF type electromagnetic waves. This interpretation is consistent with what is expected from the propagation of electromagnetic waves in the lithosphere, which must pass through rocks and salt water.

[0076] The embodiments and structural details may be varied widely with respect to the above description disclosed as non-limiting examples, without prejudice to the principles of the invention and without departing from the scope of the invention as defined in the appended claims.

Claims

1. A method (100) for monitoring magnetospheric particle precipitation, comprising: - detecting (101) charged electromagnetic particles by at least one particle detector (10) mounted on at least one satellite vehicle (1) in orbit (2) and correlating the detected particles with respective detection data; - processing the detection data to associate respective estimates or measurements of kinetic energy with the detected magnetospheric particles (102); a first count value N correlating to the number of charged particles detected in a period of time associated with a relatively high estimate or measurement of kinetic energy falling within a first energy range; Ｈ and obtaining (104) a second count value N correlating to the number of charged particles detected during said time period, associated with a relatively low estimate or measurement of kinetic energy falling within a second energy range; Ｌ and a step (105) of obtaining - First count value N Ｈ The second count value N Ｌ detecting (106) a relative variation of - determining (110) that an impulsive event of precipitation of charged particles in the magnetosphere (MPP event) has occurred during said period and comparing said variations with a threshold value; - assigning (109) to the MPP event a geomagnetic longitude and a time at which the MPP event occurred; - defining one or more groups of MPP events, each group including MPP events occurring in the same geomagnetic longitude or in a time range of relatively close geomagnetic longitudes (110); - identifying (111) a group of MPP events as indicative of activity of terrestrial origin, for example pre-seismic or seismic activity, based on the number of MPP events contained in the group and / or based on the associated variations found in the detection step (106); The method includes:

2. The method comprises: The method (100) of claim 1, comprising the step of generating (112) a pre-earthquake or seismic activity alert.

3. The step of detecting the relative variation (106) includes detecting the second count value N Ｌ The first count value N Ｈ 3. The method (100) according to claim 1 or 2, wherein the step of determining (109) comprises an operation of calculating a HR ratio (Hardness Ratio) of the hardness of the material, or vice versa, and wherein the step of determining (109) comprises an operation of verifying whether the HR ratio is higher, lower or equal to the threshold value.

4. The method (100) according to any one of claims 1 to 3, wherein the at least one satellite vehicle (1) is in an equatorial orbit (3), preferably in an equatorial low earth orbit (LEO).

5. The method comprises: The method (100) of any one of claims 1 to 4, comprising the step of removing and / or identifying spurious events (108) correlating the detected variations with indices of magnetospheric and solar activity.

6. The method (100) according to any one of claims 1 to 5, wherein the charged particle detector (10) allows for the detection of charged magnetospheric particles in distinction from gamma rays.

7. The method (100) of any one of claims 1 to 6, wherein the particle detector (10) comprises a particle imaging tracker and / or an imaging calorimeter.

8. The first energy range and the second energy range are, under normal conditions, Ｈ and the second count value N Ｌ 8. The method (100) of any one of claims 1 to 7, wherein are selected to be equal to one another or approximately equal to one another.

9. The method (100) of any one of claims 1 to 8, wherein the charged electrosphere particles are predominantly electrons and / or positrons.

10. The method (100) of any one of claims 1 to 9, wherein the particle detector (10) is adapted and configured to detect and track particles having energies below 1 GeV.

11. The method comprises:

11. A method (100) according to any one of claims 1 to 10, comprising the steps of storing the detection data on board the satellite vehicle (1) and transmitting, directly or indirectly, at least a portion of the data to a ground station (2) operatively connectable to the satellite vehicle (1).

12. The method comprises:

12. The method (100) of claim 11, comprising at least partially processing said sensed data on board said satellite vehicle, and said transmitting step comprises transmitting said processed data.

13. 13. The method (100) of any one of claims 1 to 12, wherein the threshold is set by correlating the historical data of the variations with historical data related to earthquake events including geolocation and intensity information of earthquake events.

14. The method comprises:

14. The method (100) of any one of claims 1 to 13, further comprising the steps of determining (103) the pitch of charged electromagnetospheric particles relative to a local magnetic field and selecting particles having a pitch that is included in one or more pitch ranges.

Images