DE102004000025B4 - System and method for detecting, transmitting and evaluating electromagnetic radiation data and information - Google Patents

Abstract

Use of a measuring station (20) with an antenna body (30), which is designed for the measurement of low-frequency magnetic fields, for a lightning location network (100) with a plurality of spatially separated measuring stations (20 or 20 *) as a pseudo-3D network for the identification of Discharges emitted in high altitude storm clouds, wherein: - the antenna body (30) has a mechanically self-supporting shape with at least two simple, circularly arranged circular coils without moving and / or weather-sensitive components designed for outdoor installation; - Wherein the magnetic induction (B (t)) from the x and y components (Bx (t), By (t)) of the flash field is detected by means of the two coils, - wherein the primary circuit of the antenna body (30) from the secondary circuit the antenna body (30) is galvanically isolated, and - wherein the magnetic induction (B (t)) is measured directly as a function of time, the measuring station (20) having at least ns a time measuring device (38) for determining the respective time course of electromagnetic radiation of the low-frequency magnetic fields of detected signals (S or S *), wherein the electromagnetic radiation from at least one pulse source of natural origin, namely from at least one atmospheric discharge (P ) and where the height (H) of the pulse source, namely the emission height, and the directionality (C), namely the spatial directional course, of the pulse output or pulse emission effected by the pulse source is located by taking the deviation of the arrival time of the signal (S) at the measuring station closest to the pulse source, from the arrival time of the signal to ...

Description

The present invention relates to a use of a measuring station with an antenna body, which is designed for the measurement of low-frequency magnetic fields, for a lightning location network with a plurality of spatially separated measuring stations as a pseudo-3D network for identifying discharges that are emitted in storm clouds from high altitude. Furthermore, the present invention relates to a use of an antenna body for the measuring station described above.

Lightning tracking systems have been in use around the world for more than ten years, allowing localization of ground flares with increasing local accuracy, which is less than one kilometer on modern networks. Detection efficiency is reported for lightning greater than five kiloamps, usually over ninety percent, but weaker lightning has no statistically reliable data.

Traditionally, the operational networks are designed to report predominantly ground flashes (= CG or cloud-ground) and cloud-cloud flashes (= I [ntra] C [loud] within a storm cloud or C [loud-] C [loud] between Thunderclouds) are suppressed as much as possible. Recently, however, the trend is increasing to also detect discharges in the clouds for purposes of meteorological use.

State of the art

A distinction of CG flashes and IC / CC flashes is possible with special methods, in particular with three-dimensional methods (so-called 3D methods), whereby measurements in the V [ery] H [igh] F [requenzy] range and can be combined in the V [ery] L [ow] F [requency] range [cf. Kawasaki, Z.-1. et al. (1994), SAFIR operation and evaluation of its performance, Geophys. Res. Lett. 21 (12), pages 1133 to 1166; Thery, C. (2001), Evaluation of LPATS data using VHF interferometric observations of lightning flashes during the EULINOX experiment, Atmospheric research 56, pages 397 to 409].

In the article by Jacobson et al. (2000), "FORTE radio-frequency observations of lightning strokes detected by the National Lightning Detection Network", J. Geophys. Res. 105, pages 15653 to 15662, a three-dimensional method is described, for example, in which a three-dimensional bearing initially takes place with VHF radio frequencies and then temporal coincidence, the associated VLF signal from an existing two-dimensional VLF network (so-called NLDN) is searched ,

Furthermore, Smith et al. (1999) in the article "A distinct class of isolated intracloud lightning discharges and their associated radio emission, J. Geophys. Res. 104, pages 4189 to 4212, a three-dimensional method, but providing a pure VHF bearing without any VLF reference.

A three-dimensional process with good resolution of discharge channels is described by Krehbiel et al. (1999) in the publication "Three-dimensional lightning mapping observations during MEAPRS in Central Oklahoma, 11th Int. Conf. on atmosphere. Electricity, NASNCP, Alabama, June 1999, pages 376 to 379. This method, however, again has no VLF reference and relies purely on VHF bearing in the clouds.

Already very early in the document by Taylor (1978), A VHF technique for space-time mapping of lightning discharge processes, J. Geophys. Res. 83, pages 3575 to 3583, the possibility of direct spatial targeting of high emission sites by means of transit time measurements shown.

In this known method by means of two closely adjacent stations, each requiring two sensors in a vertical arrangement (among other ground level and at a height of about fifteen meters) transit time differences measured in the nanosecond range and hereby determines the elevation angle whose intersection by triangulation the Emission locations.

• – Richard and Auffray (1985), VHF-UHF interferometric measurements, applications to lightning discharge mapping, Radio Science 20, Seiten 171 bis 192;
• – Rhodes et al. (1994), ”Observations of lightning phenomena using radio interferometry”, J. Geophys. Res. 99, Seiten 13059 bis 13082;
• – Laroche et al. (1996), ”3D structure of lightning discharge within storms”, 10th Int. Conf. Atmosph. Electricity, Osaka, 10. Juni 1996 bis 14. Juni 1996, Seiten 330 bis 332;
• – Onuki et al. (1996), Imaging of lightning channel in three dimensions using interferometer, 10th Int. Conf. Atmosph. Electricity, Osaka, 10. Juni 1996 bis 14. Juni 1996, Seiten 325 bis 332; und
• – Richard and Lojou (1996), ”Assessment of application of storm cell electrical activity monitoring to intense precipitation forecast, 10th Int. Conf. Atmosph. Electricity, Osaka, 10. Juni 1996 bis 14. Juni 1996, Seiten 284 bis 287.

Further three-dimensional methods are described in the following documents:

• Richard and Auffray (1985), VHF-UHF interferometric measurements, applications to lightning discharge mapping, Radio Science 20, pages 171-192;
• Rhodes et al. (1994), "Observations of lightning phenomena using radio interferometry", J. Geophys. Res. 99, pages 13059 to 13082;
• - Laroche et al. (1996), "3D structure of lightning discharge within storms", 10th Int. Conf. Atmosphere. Electricity, Osaka, June 10, 1996 to June 14, 1996, pages 330 to 332;
• - Onuki et al. (1996), Imaging of lightning channel in three dimensions using interferometer, 10th Int. Conf. Atmosphere. Electricity, Osaka, June 10, 1996 to June 14, 1996, pages 325 to 332; and
• - Richard and Lojou (1996), "Assessment of application of storm cell electrical activity monitoring to intense precipitation forecast, 10th Int. Conf. Atmosphere. Electricity, Osaka, June 10, 1996 to June 14, 1996, pages 284 to 287.

However, global positioning techniques use less sophisticated two-dimensional systems (called 2D systems) that make CG lightning versus IC / CC lightning differentiation more difficult. Although some manufacturers report impulse-form-dependent reliable differentiation procedures, overall the literature on this problem is extremely contradictory.

In addition, the document shows US 5,771,020 A a flash detection system for detecting and locating an initial discharge of an initial first lightning strike of a flash. An initial discharge of a flash generates a pulse which can be used to precisely detect the flash, and in particular the location of the initial flash discharge. To detect and determine according to an embodiment of the US 5,771,020 A at least three sensors detect the location of the first pulses from initial flash discharges by using the time difference of information arrival of the pulses at each of the three sensors.

From the publication WO 98/58279 A1 Furthermore, an electromagnetic measuring system for meteorology is known, which contains a plurality of measuring stations, which are arranged spatially distributed in a given spatial area.

In addition, CUMMINS, Kenneth L .; MURPHY, Martin J .; TUEL, Jeffrey V. in the essay: Lightning detection methods and meteorological applications. (Published in: IV International Symposium on Military Meteorology, 26-28 September, 2000, 13 p. - ISBN 8391356523) proposes to use three detection technologies and localization techniques (VLF, LF and VHF) for different lightning discharges using networks from radiation-detecting (in particular electro-magnetic radiation from a lightning discharge) sensors.

In addition, CUMMINS, Kenneth L. [et al.], In the article: A combined TOA / MDF technology upgrade of the U. S. national lightning detection network. (Published in: Journal of Geophysical research: Atmospheres, Vol 103, 1998, No. D8, pp 9035-9044 - ISSN 0148-0227) a flash detection system with a network consisting of over 100 radiation-detecting (in particular electromagnetic radiation from a lightning discharge ) Sensors. The authors discuss in this essay in particular the design and implementation of a signal-transit time-based or direction-bearing-based sensor network and its acquired data with regard to potential for improvement.

In the publication DE 41 33 209 A1 discloses a method for detecting the data of electromagnetic radiation in the atmosphere, which are of natural origin and thereby meteorological processes arise, which spread by measurement technically detectable up to a few exchange kilometer distance and are preferably assigned to the frequency range from 1 kHz to 100 kHz.

In addition, PARMIGIANI, Roberto; PARMIGIANI, Claudio; SELERI, Gabriele: VLF antenna "BIKELOOP". URL: http://www.vlf.it/bikeloop/bykeloop.htm a formed by two orthogonal coils aligned antenna body, which is adapted to VLF (very low frequency) waves from natural and / or non-natural origin to receive.

In addition, in Limann, Otto: wireless technology without ballast. 14th edition 1978, Franzis-Verkag GmbH, Munich; Pages 210-212 Examples of capacitive and inductive antenna coupling in AM antennas and the use of external and ferrite antennas in the MW and LW range disclosed.

task

Object of the invention

Based on the disadvantages and inadequacies set out above, as well as in appreciation of the prior art outlined, the present invention has the object, a Using a measuring station with an antenna body for a lightning location network of the type mentioned above and a use of the antenna body for the measuring station of the type mentioned so that an accurate characterization of the pulse source, for example, a reliable distinction between cloud-bottom flashes (C [loud ] G [round]) and cloud-cloud-flashing (= I [ntra] C [loud] within a cloud or C [loud] -C [loud] between clouds).

Presentation of the invention: solution, advantages

This object is achieved according to the teaching of the present invention by using a measuring station having the features specified in claim 1 and by using the antenna body for the measuring station with the features specified in claim 7. Advantageous embodiments and expedient developments of the present invention are characterized in the respective subclaims.

Thus, the gist of the present invention resides in the three-dimensional bearing of impulse emissions of natural and / or unnatural origin, particularly lightning, charge discharges, charge emissions or the like, in V [ery] L [ow] F [requency] - networks; in particular, no use of frequencies in the radio or V ery high range requency range is required.

To distinguish from conventional, already known Peilverfahren applies to the present invention that the bearing is not close to only one or two of the pulse source, in particular lightning near, sensors in a direct manner, that is not done by locally targeted in height, but by using Deviations of the signal arrival times at the respective closest to the pulse source, in particular lightning next sensor stations compared to other, not the pulse source nearest, in particular not lightning next, sensor stations.

Consequently, the present system, as well as the present method, deals with a newly conceived pulse localization system, in particular a flash detection system, which is based in principle on the proven two-dimensional method (so-called 2D flash bearing) in the V [ery] L [ow] F [requency] range based.

It should be noted that in many countries 2D lightning detection systems are in operation. Examples include the North American network NALDN in the USA [cf. Orville, R.E. et al. (2002), The North American Lightning Detection Network (NALDN) - First results: 1998-2000, Monthly Weather Review 130 (8), pages 2098 to 21091] and the European Union system EUCLID mentioned.

The latter EUCLID system originated from an amalgamation of the German-operated system BLiDS (with US sensor technology and network software) and the Austrian ALDIS network [cf. Diendorfer, G. et al. (1994), "Results of a performance analysis of the Austrian Lightning Location Network ALDIS. In: 22nd Int. Conf. on Lightning Protection, Sept. 19-23, Budapest, Hungary] and its extension to many other surrounding countries. These networks operate in the VLF range and use the classical methods of direction finding (so-called D [irection] F [inding]) and / or the signal transit times (T [ime] O [f] A [rrival]). The VLF radiation emitted by a pulse source, in particular by a flash, can be received at a sufficient amount of strength at a plurality of sensor stations. To determine the source location, the directions of incidence (DF) measured at the individual stations and / or the propagation time differences (TOA) can then be used. If the signals are now detected at only two sensors, the directions provide a location that can be optimized by incorporating the signal times. For signal acquisition at more than two stations, the T [ime] O [f] A [rrival] method is preferable for accuracy because relatively easily achievable accuracies of about one microsecond already result in better than one kilometer locations.

Accordingly, in the schematic diagram according to 5 the critical propagation delay dT = T _P -T _H due to I [ntra] C [loud] discharges (compared to C [loud] G [round] strikes).

The device of signal processing and pulse processing is thus advantageously designed so that significantly higher time accuracies (better than a microsecond, that is less than a microsecond) can be achieved than in conventional systems. This accuracy is essential for determining event locations of pulses, in particular lightning, according to the transit time method.

By means of the advantageously used analysis method, an accurate time determination (better than one microsecond, ie less than one microsecond) is possible even with irregular pulse shapes, in particular for signals which are detected at different sensors and belong to the same pulse event, in particular to the same lightning event. This can be achieved in, inter alia, pattern and shape recognition algorithms.

For example, in the network center, by means of special algorithms, in particular by means of algorithmic pattern recognition, from the numerous signals arriving from the individual sensors, those are found which belong to the same pulse event.

For this purpose, the peak patterns received from the various individual sensors are "superimposed" in digitized form in order to be able to unambiguously identify the initial peak in this way. By means of this "matching" method, a time accurate and therefore good Peilergebnis can be achieved, which in turn is essential for the height bearing. As a result, the high sensitivity of the event can be achieved and the high number of signals that can be primarily detected can actually be used further and also utilized expediently with regard to the signal properties.

• – zum einen einen sehr starken Blitz mehrfach melden, weil die Einzelsignale der Sensoren nicht richtig gruppiert werden, und
• – zum anderen schwache Signale wegen nicht identischer Impulsformen nicht als zusammengehörig erkennen.

Filtering out signals that belong to the same lightning event is not trivial when there is a high volume of data, because time overlaps occur. However, if a wrong signal is contained in a signal group, lightning detection can not take place successfully. In conventional systems this problem is not solved; This is also evident from the fact that conventional lightning detection systems

• - On the one hand report a very strong flash multiple times, because the individual signals of the sensors are not properly grouped, and
• - On the other hand, weak signals because of non-identical pulse shapes do not recognize as belonging together.

Furthermore, conventional flash detection systems can cause misalignments with significant deviations in location.

If more than three measured data are available, the three unknown quantities to be determined (= the length, the width, the pulse output time, in particular the flash time) can be calculated using standard minimization methods. As far as systematic errors also remain on the order of about one microsecond, the DF accuracy mentioned can also be achieved sustainably in practical continuous operation and verified for example by means of impacts in measuring towers.

Since the system as well as the method a particularly high detection efficiency, even in the field of weak pulses, especially weak flashes achieved, and no measures to suppress IC discharges, in particular IC flashes must be applied, the new network provides significantly more discharge events than conventional comparison systems. For this reason, the question of the origin of the discharges arises in a very special way.

For this purpose, according to the teaching of the present invention, the network (in particular the flash detection network) is expanded by a three-dimensional mode (so-called 3D mode) so that emission levels of pulses, in particular of discharges, can be determined. Thus, according to the invention, a novel and decisive contribution to the identification of IC / CC discharges, in particular of IC / CC flashes, is proposed.

• – weil nur eine Station impulsquellennah, insbesondere blitznah, sein muss, sind große Sensorabstände möglich, das heißt es ist kein dichtes teures Netz erforderlich;
• – aufgrund der Nutzung G[lobal]P[ositioning]S[ystem]-gesteuerter Zeitmessungen können auch noch größere Entfernungen von bis zu etwa einhundert Kilometern zur Blitz-Station als ”der Impulsquelle nah” bzw. als ”blitznah” gelten und signifikante, das heißt für die Höhenpeilung nutzbare Zeitabweichungen produzieren;
• – die Verwendung von V[ery]L[ow]F[requency]-Messnetzen (= Bereich um zehn Kilohertz) sichert eine große Reichweite der der Impulsquelle nicht nächstgelegenen Messstationen, zum Beispiel der nicht blitznächsten Blitzstationen (,die den Blitz auch noch erfassen müssen), das heißt etablierte VLF-Netze können eingesetzt werden, sofern die Zeitmessungen im Mikrosekundenbereich genau sind;
• – das System wie auch das Verfahren kann auf der Hardware herkömmlicher 2D-Blitzortungsverfahren aufsetzen; eine Verwendung ”echter” 3D-Techniken im Radio- oder VHF-Bereich ist insofern nicht erforderlich, als Ziel des Systems wie auch des Verfahrens nicht die räumlich präzise Auflösung von Entladungskanälen, zum Beispiel von Blitzkanälen, im 100 Meter-Bereich, sondern die Bereitstellung einer Datenbasis für eine Entscheidung hinsichtlich der Frage ist, ob das detektierte Ereignis als I[ntra]C[loud]-Impuls, insbesondere als I[ntra]C[loud]-Blitz, einzustufen ist;
• – die Existenz von Impulsabgaben bzw. Impulsaussendungen, insbesondere von Blitzemissionen, aus Kilometer-Höhen lässt sich theoriefrei und ohne Annahme justierbarer Parameter aus einem Vergleich der an einzelnen (,jeweils blitznächsten) Sensoren und an Gruppen von (nicht blitznächsten) Sensoren gemessenen Verteilungen von Zeitabweichungen zweier Impulsabgabezeiten (im Netz durch Gesamtpeilung ermittelte Impulszeit minus Differenz von Ankunftszeit und Laufzeit an der jeweiligen der Impulsquelle nächstgelegenen Station), insbesondere zweier Blitzzeiten (im Netz durch Gesamtpeilung ermittelte Blitzzeit minus Differenz von Ankunftszeit und Laufzeit an der jeweiligen blitznächsten Station), erkennen;
• – die Emissionshöhe lässt sich mit Algorithmen, die mit denen der Ortspeilung zumindest vergleichbar sind, durch Hinzufügen der Höhenvariablen berechnen;
• – das Verfahren kann in andere existierende Impulsortungssysteme, insbesondere Blitzortungssysteme, integriert werden, sofern die notwendige Zeitgenauigkeit erzielt wird;
• – das Verfahren kann auch bei intensitätsschwachen Impulsen, insbesondere bei intensitätsschwachen Blitzen, eingesetzt werden, sofern der Impuls, insbesondere der Blitz, an drei, bei Nutzung von Einfallsrichtungen zur Peilung auch an nur zwei Sensoren gemessen wird;
• – die Ergebnisse des Verfahrens können dazu benutzt werden, Fragen einer Impulsform-Diskriminierung von Wolke-Wolke-Entladungen (IC), insbesondere von Wolke-Wolke-Blitzen, gegenüber Wolke-Erde-Entladungen (CG), insbesondere Wolke-Erde-Blitzen, zu klären; und
• – das System wie auch das Verfahren kann zusammen mit der Impulsform-Analyse zur Diskriminierung eingesetzt werden, um den (kleinen) Prozentsatz strittiger Fälle zu lösen, der weder mit dem einen Verfahren noch mit dem anderen Verfahren klar entscheidbar ist.

Such a procedure offers a number of technical as well as economic advantages:

• - Because only one station impulsquillennah close, in particular lightning-fast, must be, large sensor distances are possible, that is, it is not a dense expensive network required;
• - due to the use of G [lobal] P [ositioning] S [ystem] -controlled time measurements, even greater distances of up to about one hundred kilometers to the flash station can be considered "near the pulse source" or "close to the flash" and significant, that is to produce usable time deviations for the height bearing;
• - the use of V [ery] L [ow] F [requency] measurement networks (= range around ten kilohertz) ensures a long range of the measuring stations not closest to the pulse source, for example the non-lightning flash stations (which still detect the flash) must), that is established VLF networks can be used, provided that the time measurements in the microsecond range are accurate;
• The system as well as the method can be based on the hardware of conventional 2D flash detection methods; It is not necessary to use "real" 3D techniques in the radio or VHF range, as the object of the system as well as the method is not the spatially precise resolution of discharge channels, for example of lightning channels, in the 100 meter range, but the provision of a database for a decision as to whether the detected event as I [ntra] C [loud] pulse, especially as I [ntra] C [loud] flash;
• The existence of momentum emissions, in particular of lightning emissions, from kilometer altitudes can be calculated theory-free and without assuming adjustable parameters from a comparison of the time-deviations measured at individual (, respectively closest to the lightning) sensors and at groups of sensors (not the closest to the lightning) two pulse delivery times (in the network by total bearing determined pulse time minus the difference of time of arrival and transit time at the respective closest to the pulse source station), in particular two flash times (in the network determined by total bearing flash time minus difference of arrival time and duration at each lightning next station) recognize;
• - The emission level can be calculated with algorithms that are at least comparable to those of the location bearing, by adding the height variables;
• The method can be integrated into other existing pulse localization systems, in particular flash detection systems, provided the necessary time accuracy is achieved;
• - The method can also be used in low-intensity pulses, especially in low-intensity flashes, provided that the pulse, in particular the flash, is measured at three, when using directions of incidence for bearing also on only two sensors;
• The results of the method can be used to address issues of pulse shape discrimination of cloud-to-cloud discharges (ICs), in particular cloud-to-cloud flashes, to cloud-to-earth discharges (CG), in particular cloud-to-earth flashing, to clarify; and
• - The system as well as the procedure can be used together with the Pulse Shape Analysis for Discrimination to solve the (small) percentage of contentious cases that is not clearly decidable with either method or with the other method.

In an advantageous embodiment of the present invention, the sensor operates passively and without power supply. This saves error sources compared to known systems. In addition, when working passively and without power, the sensor can process much higher data rates than previously described or existing prior art systems.

• – meteorologisch und wolkenphysikalisch wichtige Erkenntnisse über das Auftreten von IC-Blitzen,
• – eine höchst sichere Unterscheidung von CG-Blitzen versus IC-Blitzen,
• – eine bessere Identifizierung von CG-Blitzen, was zu wichtigen Folgedaten in Form von zuverlässigeren Blitzdichte-Karten führt (Verbesserung des Blitzschutzes, was zum Beispiel für die Automobilindustrie, für das Baugewerbe, für Freizeitveranstalter, für Reiseveranstalter und/oder für die Versicherungswirtschaft interessant ist) oder
• – eine Erfassung von wesentlich mehr Blitzen, als herkömmliche Netze melden, was in Zusammenhang mit CG-Identifizierungen zu deutlich höheren Boden-Blitzdichten als bisher angenommen führt.

In addition to the technical and economic advantages set out above, the present invention also offers a number of application advantages, such as

• - Meteorological and cloud physical important knowledge about the occurrence of IC flashes,
• A most reliable distinction of CG flashes versus IC flashes,
• - A better identification of CG flashes, which leads to important follow-up data in the form of more reliable flash density cards (improvement of lightning protection, which is interesting for example for the automotive industry, for the construction industry, for leisure operators, for tour operators and / or for the insurance industry ) or
• - Detecting significantly more flashes than conventional grids report, resulting in connection with CG identifications to significantly higher ground-lightning densities than previously thought.

• – dem Erfordernis einer genauen Zeitbestimmung auch bei intensitätsschwachen, verrauschten, komplex geformten und/oder übersteuerten Signalen, um eine möglichst quantitative Erfassung und Unterscheidung von Impulsen (I[ntra]C[loud] versus C[loud]G[round]), insbesondere von Blitzen, zu gewährleisten,
• – der Notwendigkeit einer effizienten Erkennung nicht impulskorrelierter, insbesondere nicht blitzkorrelierter, Störsignale,
• – dem Erfordernis einer optimalen Einrichtung des Zeitmanagements aller Stations- und Netzkomponenten oder
• – der Notwendigkeit einer Benutzung komplexer Algorithmen zur möglichst zuverlässigen

Furthermore, the present invention provides a contribution to a number of difficulties, such as

• The requirement of precise time determination even with weak-intensity, noisy, complex-shaped and / or overdriven signals in order to obtain the most quantitative possible detection and differentiation of pulses (I [ntra] C [loud] versus C [loud] G [round]), in particular of lightning, to ensure
• The need for efficient detection of non-pulse-correlated, in particular non-lightning-correlated, interference signals,
• The requirement of an optimal time management of all station and network components or
• - the need to use complex algorithms for the most reliable

Impulsabgabeunterscheidung, in particular discharge discrimination, of real and statistical and systematic errors due to random height estimates.

The requirement of time determination thus includes, according to an advantageous development of the present invention, a particularly accurate time marking of the signals arriving at the individual sensors. This time marking takes place for signals of the same pulse, in particular the same flash, at the different sensor stations as similar as possible.

On the basis of this precise time marking, it can be determined from the marking times whether a sensor signal near the pulse source, in particular on a sensor close to the lightning, is measured with respect to the other sensors, a delayed signal time; if such a delay of the signal time is present, this indicates an extended running distance and thus an origin from a great height.

Since this time difference to be exploited becomes progressively smaller as the distance between the pulse source location, in particular the flash location, and the sensor under consideration increases, efficient height estimation requires the measurement and use of very small time differences of at most about one microsecond. In an advantageous embodiment of the invention, therefore, the markings of the individual signals are not inaccurate than this time difference.

• (i) Zunächst könnte unterstellt werden, dass verzögerte Ankunftszeiten aufgrund von Systemfehlern rein statistisch verteilt auftreten und keine physikalische Bedeutung haben. Da die in Rede stehenden Verzögerungen je nach Distanz zur Impulsquelle, insbesondere zum Blitz, meist mehrere Mikrosekunden betragen, müssten relativ große Zeitfehler unterstellt werden, die aufgrund der nachweislich genauen Peilungen auszuschließen sind. Ferner würde man dann erwarten, dass Fehler mit kleineren Verzögerungszeiten häufiger auftreten und die resultierende Höhenverteilung entgegen den tatsächlichen Ergebnissen zu geringen Höhen der Impulsquelle, insbesondere zu niedrigen Blitzhöhen, nahe Null stark zunimmt. Eine Untersuchung der Zeitabstände zwischen ermittelter Impulsabgabe- oder -aussendungszeit, insbesondere ermittelter Blitzzeit, und den aus den Einzelstationen gemäß Ankunftszeit und Laufzeit zum Ort der Impulsquelle, insbesondere zum Blitzort, resultierenden Impulsabgabe- oder -aussendungszeiten, insbesondere Blitzzeiten, ergeben erfindungsgemäß keinerlei systematische Abweichungen der Zeitmeldungen einer Station relativ zur Gesamtheit der Stationen.
• (ii) Signalverzögerungen an einer Sensorstation können durch elektronische Effekte zustande kommen undan den verschiedenen Stationen unterschiedlich stark ausgeprägt sein. Wäre dies der Fall, würde die 3D-Peilung in praktisch allen Fällen zu fiktiven, stationsbezogen unterschiedlichen Impulsquellenhöhen, insbesondere Blitzhöhen, führen und nicht nur bei einem Bruchteil der Signalgruppen. Zudem wäre dieser Fehler in der Prüfung gemäß vorstehendem Punkt (i) zu entdecken.
• (iii) Falsche Ankunftszeiten können durch fehlerhafte Auswertung der eintreffenden Signale verursacht sein; insbesondere bei Impulsen mit mehreren signifikanten Extremwerten oder bei nicht aufgelösten Doppelpeak-Strukturen könnte eine falsche Zeitmarkierung entstehen. Bei noch gut aufgelösten Peaks würde ein solcher Fehler jedoch in der Regel zum gegenteiligen Effekt führen (--> zu frühe Ankunftszeit anstatt zu später Ankunftszeit), weil die impulsnächste, insbesondere blitznächste, Station, das stärkste Signal empfängt und daher kaum zu erwarten ist, dass der erste relevante Signalpeak unterdrückt und statt dessen der zweite (”verzögerte”) Peak zur Zeitmarkierung erfasst wird. Bei schlecht aufgelösten Doppelstrukturen könnte allerdings bei der impulsnächsten, insbesondere blitznächsten, Station das Maximum im zeitlich späteren Teil liegen, während bei weiter entfernten Stationen infolge von Dämpfungs- und Ausbreitungseffekten die Struktur nicht mehr aufgelöst und ein früheres Maximum in der Mitte des Gesamtpeaks gefunden wird. Die Signalauswertung gemäß der vorliegenden Erfindung berücksichtigt und löst derartige Probleme.
• (iv) Die Sensoren können insbesondere bei nahen Impulsquellen, insbesondere bei nahen Blitzen, übersteuern und daher keine oder verfälschte Zeitmarkierungen und Einfallsrichtungen melden. Es wird über vorhandene Ortungsnetze häufig berichtet, dass zur Vermeidung solcher Komplikationen der Impulsquelle nahe Stationen, insbesondere blitznahe Stationen, bei der Auswertung nicht berücksichtigt werden. Für eine 3D-Peilung würden dann die interessantesten Informationen verloren gehen. Die Empfangsstationen gemäß der vorliegenden Erfindung sind daher auf hohe Intensitätsauflösung mit großem Dynamikbereich ausgelegt – ohne bei schwächsten Signalen unempfindlich zu werden –, und die Impulsbehandlung kann auch gesättigte Signale in zuverlässiger Weise verarbeiten. Dies ist dadurch erleichtert, dass das Feld B(t) und nicht die Ableitung dB/dt gemessen wird, so dass Integrationsprozeduren entfallen.
• (v) Es ist wohlbekannt, dass lokale ”site-errors” zur Verfälschung der gemessenen Einfallsrichtungen führen; weniger ist über die damit verbundenen Zeitfehler berichtet worden. Sollten diese Zeitfehler den Bereich von etwa einer Mikrosekunde übersteigen, was durchaus vorstellbar erscheint, würden sich zwar nicht zwingend gravierende Fehler in der 2D-Lokalisierung des Netzes, jedoch fiktive Emissionshöhen in der dreidimensionalen Ortung ergeben. Ein solcher Effekt würde dann bei nahezu allen Signalen auftreten und dadurch feststellbar sein (vgl. vorstehender Punkt (i)).

Furthermore, the present invention makes it possible to discuss new questions concerning VLF emission from storm clouds. Thus, for the possibility of altitude estimation essential to the invention that the delay of the pulse arrival time, in particular the flash arrival time, has its cause in the longer term to the emission source and is not caused by equipment or other systematic errors. In the following, the five most important possible effects that can influence the altitude determination are discussed:

• (i) First, it could be assumed that delayed arrival times due to system errors occur purely statistically distributed and have no physical significance. Since the delays in question, depending on the distance to the pulse source, in particular to the flash, usually amount to several microseconds, relatively large time errors would have to be assumed, which can be ruled out due to the verifiable accurate bearings. Further, one would then expect errors with smaller delay times to occur more frequently, and the resulting altitude distribution greatly increases, contrary to the actual results, to low heights of the pulse source, especially to low flash heights, near zero. An investigation of the time intervals between determined Impulsabgabe- or -aussendungszeit, in particular determined flash time, and from the individual stations according to arrival time and duration to the location of the pulse source, in particular the flash point resulting Impulsabgabe- or -aussendungszeiten, in particular flash times, according to the invention result in any systematic deviations of Time messages of a station relative to the totality of the stations.
• (ii) Signal delays at a sensor station may be due to electronic effects and be different in severity at the various stations. If this were the case, the 3D bearing would in almost all cases lead to fictitious, station-related different pulse source heights, in particular flash heights, and not just a fraction of the signal groups. In addition, this error would have to be detected in the test according to (i) above.
• (iii) Incorrect arrival times can be caused by incorrect evaluation of the incoming signals; in particular with pulses with several significant extreme values or with unresolved double peak structures, a wrong time marking could arise. In the case of well-resolved peaks, however, such an error would generally lead to the opposite effect (-> too early arrival time instead of too late arrival time), because the next impulse, especially lightning-fast, station receives the strongest signal and is therefore hardly to be expected, that the first relevant signal peak is suppressed and instead the second ("delayed") peak is detected for timing. In the case of poorly resolved double structures, however, the maximum at the impulse-next, especially lightning-next, station could lie in the temporally later part, while at more distant stations due to damping and propagation effects the structure is no longer resolved and an earlier maximum in the middle of the total peak is found. The signal evaluation according to the present invention takes into account and solves such problems.
• (iv) The sensors can overdrive especially at near pulse sources, especially near lightning, and therefore report no or falsified timing marks and incident directions. It is frequently reported on existing localization networks that in order to avoid such complications of the pulse source near stations, in particular lightning near stations, are not taken into account in the evaluation. For a 3D bearing then the most interesting information would be lost. The receiving stations according to the present invention are therefore designed for high intensity resolution with a large dynamic range - without becoming insensitive to the weakest signals - and the impulse treatment can also process saturated signals in a reliable manner. This is facilitated by measuring the field B (t) and not the derivative dB / dt, thus eliminating integration procedures.
• (v) It is well known that local "site-errors" lead to falsification of the measured directions of incidence; less has been reported about the associated timing errors. Should this time error the In the range of about one microsecond, which seems quite conceivable, not necessarily serious errors would result in the 2D localization of the network, but fictitious emission levels in the three-dimensional location. Such an effect would then occur in almost all signals and thus be detectable (see the above point (i)).

In view of the above findings, it can be assumed in the present invention that the 3D bearings presented here will signal real propagation delays caused by emission in the storm clouds rather than near the ground.

The inventively determined heights of the pulse sources, in particular flash heights, are also well compatible with the geometry of typical storm clouds. Since it is generally and for good reasons assumed that CG discharges, in particular CG flashes, generate the main emission in the VLF range close to the ground, it is obvious that the 3D bearings are negative and positive IC discharges , in particular IC flashes, which have a strong vertical discharge direction.

Rather horizontally oriented IC or CC discharges, for example IC or CC flashes, should hardly be able to be received at several stations with amplitudes according to a 1 / R dependence because of the less favorable radiation characteristic for the measuring technique used according to the invention.

The present invention also includes, especially for linear pulse sources, the option of separating pulse sources, for example IC (or CC) flashes, into vertical and horizontal discharge channels, in particular flash channels, that is the possibility of discriminating IC ( or CC-) discharges, in particular of IC (or CC) flashes, which are predominantly vertical or horizontal.

This is achieved according to the invention by checking the amplitudes A measured at a sufficient number of sensor stations with respect to their distance R: is the distance-dependent amplitude A (R) compatible with a 1 / R law and is the pulse event, in particular the lightning event, by means of 3D method is classified as I [ntra] C [loud] or I [nter] C [loud] type, then it is a predominantly vertical flash.

However, strong deviations from the 1 / R law can be detected and corrected by taking into account the radiation characteristics according to two angles, for example the elevation angle and the angle between the momentum / emission axis, in particular the flash axis, and the direction to the sensor , then it is a predominantly horizontal discharge channel, in particular lightning channel.

The amplitudes A (R) dependent on the distance R may be vaporized by variable ground conductivity, which may interfere with a 1 / R dependence. In the present system, as well as the present method, this fact is accounted for by jointly analyzing groups of pulse sources, for example, flashes, from narrow spatial areas according to the present system as well as the present method. With this optional technical measure, it can be determined whether with the same or very similar running distance the damping effects occur consistently or only with individual (horizontal) pulses, in particular flashes.

This latter group analysis may conveniently be done online even if, for example, the particular thunderstorm within the chosen reporting period for three-dimensional momentum locations, especially lightning locations, provides (for example one minute) enough flashes from the same area. Otherwise, this additional information about horizontal pulses, in particular about horizontal flashes, in any case offline, so in subsequent analysis, can be determined.

Of interest in connection with the present invention is the question to what extent upward-directed earth-cloud discharges (G [round] C [loud]), for example earth-cloud-lightning, emanate at the lower cloud boundary. In this respect, it is useful to check the findings obtained according to the invention by means of real 3D systems [cf. Defer, E. et al. (2000), Simultaneous observations of CG activity from NALDN and ONERA-ITF interferometric mapper during the STERAO-A experiment. In: Int. Lightning Detection Conf., Nov. 7-8, Tucson, Ariz. (Global Atmospherics Inc.)].

A refinement of the pseudo-3D system or the pseudo-3D method for determining heights of pulse sources, for example of lightning emission levels, in the VLF range with respect to even more accurate time resolutions than are realized in the first phase of the network, and the combination with pulse shape Analyzes allow more reliable and near-quantitative (on-line) discrimination of CG pulses, particularly CG flashes, and IC pulses, especially IC flashes, with minimal overhead.

Basically, the use of the 2D lightning location network as a pseudo-3D network, mediated by the teachings of the present invention, allows the identification of discharges emitted in high altitude storm clouds.

In an advantageous embodiment of the present invention, the sensor actually measures the magnetic induction B (t) directly as a function of time. In doing so, the induction law is used - just as physically required. The use of the law of induction leads in conventional application to a measuring voltage which is proportional to the derivative dB / dt of the magnetic induction B (t) after time t.

However, an expedient embodiment of the present invention includes an electrical circuit that actually leads to B (t) in the selected frequency range. An important side-effect is that it does not require subsequent elements to integrate dB / dt into B (t) (if necessary, such subsequent elements would in turn lead to additional inaccuracies because it would also be integrated over the unavoidable noise). The system is also designed to provide extremely high signal sensitivity at the same time.

Prerequisite for the application of the system as well as the method is even with complex pulse shapes, especially lightning pulse shapes, to about one microsecond accurate determination of the signal arrival times at the individual sensor stations, which can be achieved with optimal use of GPS receivers and intelligent pulse analysis is.

It is thus possible to subject a majority of the detectable in a localization network pulses, especially lightning, in networks with lower station distances even the majority of all pulses, especially lightning, a 3D analysis and very directly - without any theoretical assumption and without any adjustable parameters - determine whether the emission was close to the ground.

This is an essential contribution to a comprehensible reliable identification of CG pulses, in particular of CG flashes, achieved even in spatially large networks, without significant additional effort to operate. The advantages of "true" 3D systems remain entirely unaffected, because these have more fundamental physical processes and smaller-scale discharge channel resolutions, for example, flash channel resolutions, as the goal, as is the case here with the pure lightning location.

As already stated in the presentation of the present invention, the location of pulse sources, in particular the lightning location, can be determined by means of the V [ery] L [ow] F [requency] measurements either by the direction-finding method (so-called "D [irection] F [inding] ") and / or by means of the transit time difference method (so-called" T [ime] O [f] A [rrival] method ").

In traditional direction finding (DF = direction finding), the direction of incidence of the electromagnetic radiation, in particular the direction of incidence of the flash signal, is measured at the sensor stations and the pulse source is localized by means of intersection methods, in particular the flash location. In this conventional method, inaccuracies occur, which are referred to as a local error or "site error".

In the past, methods have been published to correct these errors retrospectively and thus to increase the locating accuracy. This is usually done by locating without a specific sensor; This results from the location of the pulse source, in particular from the flash location, and from the sensor location a target angle. This target angle is compared with the actually measured angle, and an angle correction is derived from the difference. By cyclically repeating the process, an optimization can take place. In this case (more precise) T [ime] O [f] A [rrival] -local determinations can also be used.

• – unterschiedliche Laufzeiten der Signale in der Elektronik der einzelnen Sensorstandorte und/oder
• – umgebungsbedingte Zeitverzögerungen, wie sie zum Beispiel durch Unterschiede in der Leitfähigkeit und damit in der Signal-Ausbreitungsgeschwindigkeit oder durch Reflexionen und Überlagerungen infolge nicht-ebener Geländestrukturen auftreten können.

The angle errors to be corrected are caused by errors in the antenna structure as well as by influences such as reflections or conductivity gradients of the environment. The correction methods compensate for the total error, in particular if not a single angular correction but a correction function is determined as a function of the angles of incidence. In the T [ime] O [f] A [rrival] method, the two aforementioned errors are also possible accordingly, namely

• - Different maturities of the signals in the electronics of the individual sensor locations and / or
• Ambient time delays, such as those that may occur due to differences in conductivity and thus signal propagation speed, or reflections and overlaps due to non-planar terrain structures.

So far, no correction methods for the T [ime] O [f] A [rrival] method are known which correspond to the D [irection] F [inding] correction.

Due to the problem described, the system of the type mentioned and the method of the type mentioned is to be further developed so that errors in the determination of the respective time course, in particular the respective arrival time and / or the respective signal propagation time, the detected signals are minimized.

• – zum Lokalisieren der Höhe, insbesondere der Emissionshöhe bzw. der Sendehöhe, der Impulsquelle und/oder
• – zum Bestimmen der Direktionalität, insbesondere des räumlichen Richtungsverlaufs, der Impulsabgabe/-aussendung natürlichen und/oder nichtnatürlichen Ursprungs, insbesondere des Blitzes, der Ladungsabgabe, der Ladungsaussendung oder dergleichen, erfolgt.

This problem is solved by a system and by a method in which at least one of the measuring stations or sensor (station) s is adjusted and / or calibrated, this adjustment and / or this oak instead of or in addition

• For locating the altitude, in particular the emission level or the transmission altitude, the pulse source and / or
• - For determining the directionality, in particular the spatial direction course, the impulse delivery / -auss natural and / or unnatural origin, in particular the flash, the charge delivery, the charge emission or the like, takes place.

This calibration correction and / or adjustment method can be used in particular for optimizing and / or correcting the T [ime] O [f] A [rrival] technique, since it offers a significant advantage with regard to the achievement of improved time stamps or time markings. By means of this calibration and / or adjustment method, both 2D spatial bearings and 3D height bearings of lightning discharges can be performed with improved accuracy.

In an advantageous embodiment, first the position, in particular the location and / or the height, of the pulse source is localized by excluding the to be adjusted and / or calibrated measuring station from the determined respective time profile, in particular from the determined respective arrival time, the respective Difference in the transit time of pulses of the same pulse source to the respective measuring station is determined.

Furthermore, the respective time profile, in particular the respective arrival time, of the signals originating from the localized pulse source and detected at the measuring station to be adjusted and / or calibrated is calculated.

Based on this, the difference between the calculated respective time profile, in particular the calculated respective arrival time, and the ascertained respective time profile, in particular of the determined respective arrival time, of the signals originating from the localized pulse source and detected at the measuring station to be calibrated and / or calibrated determined and processed in a statistically meaningful way. If necessary, at least one time correction term, in particular for later localizations and / or for later localizations, is determined on the basis of the determined respective difference and the measuring station to be adjusted and / or calibrated is adjusted and / or calibrated by means of this determined time correction term.

The present adjustment and / or calibration method is based on essential to the fact that the signal analysis in the central station, the detection of the distribution of time deviations between measured signal arrival time and due to the successful bearing of electromagnetic radiation, in particular the flash bearing, calculated arrival time in a statistically meaningful manner from the form of the determined distribution to check the proper technical time management of the signal propagation times and readjust if necessary. After carrying out this test and adjustment procedure, an optimal use of the transit time differences for height bearing is guaranteed.

• – der Höhe der Impulsquelle, insbesondere der Emissionshöhe bzw. der Sendehöhe, und/oder
• – der Direktionalität, insbesondere des räumlichen Richtungsverlaufs, der durch die Impulsquelle bewirkten Impulsabgabe oder Impulsaussendung; insbesondere kann System wie auch das vorliegende Verfahren bei der genauen sowie zuverlässigen Abgrenzung von Impulsabgaben oder Impulsaussendungen, zum Beispiel von Entladungen, innerhalb einer Gewitterwolke (sogenannte I[ntra]C[loud]-Blitze) und/oder zwischen mindestens zwei Gewitterwolken (sogenannte C[loud]C[loud]-Blitze), gegenüber Impulsabgaben oder Impulsaussendungen, zum Beispiel Entladungen, zwischen Wolke und Erde (sogenannte C[loud]G[round]-Blitze) eingesetzt werden, was wiederum eine im Vergleich zum Stand der Technik zuverlässige Erstellung von Impulsdichtekarten, zum Beispiel von Blitzdichtekarten, ermöglicht.

Finally, the present system, as well as the present method, relates to the use of at least one system according to the type set out above and / or a method according to the above-described type for localization

• The height of the pulse source, in particular the emission level or the send height, and / or
• The directionality, in particular the spatial direction, of the pulse output or pulse emission caused by the pulse source; In particular, the system as well as the present method can be used for the accurate and reliable demarcation of pulse outputs, for example discharges, within a storm cloud (so-called I [ntra] C [loud] flashes) and / or between at least two storm clouds (so-called C [loud] C [loud] flashes) are used against pulse outputs or pulse transmissions, for example discharges, between cloud and earth (so-called C [loud] G [round] flashes), which in turn is a reliable one compared to the prior art Creation of impulse density maps, for example of flash density maps.

The system of the kind set forth above and / or the method of the kind set forth above may be used to calibrate and / or adjust at least one measuring or sensing station in lightning detection networks.

• – zur Vorhersage und zur Lokalisation, insbesondere zur Früherkennung, von klimatologischen und/oder meteorologischen Vorgängen, wie etwa von Blitzschlag, von (Extrem-)Niederschlag, von Gewittern, von Hagel, von Orkanen, von (Stark-)Regen, von Sturm, von Unwettern, von Wind oder von anderen besonderen klimatologischen und/oder meteorologischen Ereignissen,
• – zur dreidimensionalen Ortung von Impulsquellen, zum Beispiel von Blitzen, unter Einbeziehung schwacher Signale zum Optimieren der Vorhersage und Lokalisation, insbesondere der Früherkennung, von Gewitterzellen,
• – zur Erkennung von Signalherden zur Vorhersage und Lokalisation, insbesondere zur Früherkennung, von Gewitterzellen,
• – zur Analyse von Gruppen der Daten und Informationen zur Charakterisierung meteorologischer Klassen,
• – zur Extrapolation von räumlichen Zug bahnen erkannter Gewitterherde und/oder
• – zur Gewinnung biometeorologischer Informationenverwendet werden.

Regardless or in connection therewith, the system according to the kind set out above and / or the method according to the kind set out above may be implemented in the real time or in the short term

• - for prediction and localization, in particular for early detection, of climatological and / or meteorological processes, such as lightning, extreme precipitation, thunderstorms, hail, hurricanes, (heavy) rain, storm, from severe weather, wind or other special climatic and / or meteorological events,
• For the three-dimensional location of pulse sources, for example of lightning, using weak signals to optimize the prediction and localization, in particular the early detection, of thunderstorm cells,
• For detecting signal sources for prediction and localization, in particular for the early detection of thunderstorm cells,
• - for the analysis of groups of data and information for the characterization of meteorological classes,
• - For extrapolation of spatial train tracks detected thunderstorms and / or
• - be used to obtain biometeorological information.

embodiment

Brief description of the drawings

As already discussed above, there are various possibilities for embodying and developing the teaching of the present invention in an advantageous manner. Reference is made on the one hand to the independent claims 1 and 7 subordinate dependent claims, on the other hand, further embodiments, features and advantages of the present invention hereinafter with reference to the by 1 to 8D illustrated embodiment explained in more detail.

It shows:

1 a schematic representation of an embodiment of an inventive use of a measuring station with an antenna body for a lightning location network;

2A in a schematic representation of an embodiment of a measuring station 1 ;

2 B in perspective view an embodiment of an antenna unit, the measuring station from 2A assigned;

3 a schematic representation of an embodiment of a central station 1 ;

4 in topographical representation of an embodiment of a geographical distribution of the measuring stations and the central station 1 ;

5 schematically the principle of the present method, where H is the emission level of V [ery] L [ow] F [requency] flash radiation;

6 in diagrammatic representation, the signals occurring within a longitude range plotted on the right axis and within a latitude range plotted on the vertical axis of atmospheric discharges evaluated by the present method;

7A in a diagrammatic representation, determined by the present method, plotted on the vertical axis time delay of the signal arrival times as a function of the plotted on the right axis observation distance for different emission levels;

7B a diagrammatic representation of a first example of a discharge height distribution determined by the present method (= number of discharges noted on the right axis, plotted against the measured on the vertical axis, measured in kilometers emission level);

7C a diagrammatic representation of a second example of a discharge height distribution determined by the present method (= number of discharges recorded on the right axis, plotted against the measured on the vertical axis, measured in kilometers emission level);

7D a diagrammatic representation of a third example of a discharge height distribution determined by the present method (= number of discharges noted on the right axis, plotted against the measured on the vertical axis, measured in kilometers emission level); and

8th in a diagrammatic representation, the distribution (= number) of the vertical axis, determined by the present method, on the right axis plotted time deviations (in μs or microseconds) of the arrival signals (negative times mean time delay).

The same or similar embodiments, elements or features are in the 1 to 8th provided with identical reference numerals.

Best way to carry out the invention

• – sowohl zur Charakterisierung von nichtnatürlicher Strahlung, zum Beispiel
• – von mittels eines Senders K eines Flugzeugs M gesendeten Signalen oder
• – von Mobilfunksignalen,
• – als auch zur Charakterisierung von natürlicher Strahlung, zum Beispiel von Blitzentladungen P,

eingesetzt werden kann.In 1 is a system 100 represented, that

• - both for the characterization of non-natural radiation, for example
• From signals sent by means of a transmitter K of an aircraft M or
• - of mobile radio signals,
• - as well as for the characterization of natural radiation, for example of lightning discharges P,

can be used.

Since the principles underlying the measurement of such pulse outputs or charge discharges (charge discharges) for the electromagnetic radiation originating from pulse sources of unnatural origin and for the electromagnetic radiation originating from pulse sources of natural origin are the same or at least analogous, the system becomes 100 described below by way of example when used in the flash measurement.

That on the basis of 1 to 8th illustrated, in particular as a Sferics / Blitzmessungsanordnung trained (= somewhat three-dimensional flash detection network 100 ) System has several, namely up to one hundred individual measuring stations 20 . 20 * (in 1 are exemplarily four such measuring stations 20 . 20 * shown) and a central device or central station 10 on.

As the representation of 1 and 2A is removable, is every measuring station 20 . 20 * an antenna or an antenna body 30 (see. 2 B ), which is designed for the measurement of low-frequency magnetic fields.

This is off 2 B it can be seen that this antenna body 30 is designed in a mechanically self-supporting form without moving and / or weather-sensitive components for outdoor installation. The primary circuit of the antenna body 30 is from the secondary circuit of the antenna body 30 galvanically isolated; The electromagnetic fields can be coupled out in a broadband and low-noise manner and can therefore be measured in real time.

The in the housing of the measuring station 20 respectively. 20 * arranged station electronics 40 initially has an amplifier 42 for low-noise amplification of the antenna body 30 coming signals S and S * on. The amplifier 42 Downstream is a filter stage 44 , That by means of the filter unit 44 filtered signal S "is then in the unit 46 subjected to a signal digitization.

By means of the low-noise amplifier 42 and the filter stage 44 a smoothing and optimization method of the signals S and S * can be performed, which represents a significant advantage due to the problems presented below.

In the case of signals with a large amplitude and with a simple, undisturbed form, the determination of a characteristic arrival time is relatively easy to accomplish. Conventional systems according to the prior art relate in their application only to such cases smooth waveforms and seek only the maximum of the respective signal.

• – klein,
• – mit eigenen, oft komplexen und unregelmäßigen Strukturen behaftet und
• – von Rauschanteilen überlagert.

The in practice at the sensors 20 respectively. 20 * however, incoming signals S and S * are mostly

• - small,
• - Afflicted with its own, often complex and irregular structures and
• - superimposed by noise.

This makes it more difficult to obtain a clear timestamp, especially if the signal S or S * is weak and is close to the lower detection threshold.

In particular, then, searching for the (single) absolute signal maximum in different sensors conventionally leads to inaccuracies which, regardless of the use of the most accurate time normals, such as the G [lobal] P [ositioning] S [ystem] method, usually clearly above one Microsecond lie. In rare cases, the difficulty or inaccuracy of a conventional maximum finding ultimately leads to the rejection of the entire signal.

To eliminate the aforementioned problems, the system 100 according to 1 combined with 2A the signal amplitudes, which are predominantly, but not exclusively, around a maximum, are subjected to a smoothing and optimization process with which irregularities in the signal curve can be compensated.

• – in der Art der Anwendung auf die speziellen Verhältnisse des Systems 100 adaptiert und
• – mathematisch auf kurze Rechenzeit optimiert.

The corresponding algorithms are based essentially on standard methods, however they are

• - in the way of application on special conditions of system 100 adapted and
• - Optimized mathematically for short computing time.

This makes it possible to mark even weak, irregular and noisy signals with sufficient time accuracy.

Furthermore, the system 100 according to 1 combined with 2A not only the largest peak of a signal S or S * subjected to the smoothing and optimization method set forth above; rather, the entire signal structure above the noise threshold is analyzed according to this smoothing and optimization method. Thus, a plurality of time or structure information can be available for each individual signal S or S * as a function of the actual total pulse shape actually present.

With regard to the station electronics 40 It should be noted that between the digitizing unit 46 and one for storing the of the station electronics 40 processed data and information D or D * provided storage unit 50 a P ersonal C [computer] 22 is switched. In the central evaluation software of the P [ersonal] C [omputer] s 22 the structures of the time information determined according to the described smoothing and optimization method are matched with one another.

In this context, it should be considered that conventionally only one time mark is available for the sensor signal, so that in the prior art an (unwanted) inclusion of simultaneous interference signals can occur.

To avoid this disadvantage, the analysis properties of the central evaluation with regard to the smoothing and optimization method are designed so that the possibly existing multiple information in the form of time structures in the individual sensor signals, that is within a signal group belonging to the same flash, due to the accurate time recording and marking are sufficiently compatible with each other.

• – die Genauigkeit der Signalzeit-Markierung erhöht und
• – die Wahrscheinlichkeit eines (unerwünschten) Einbezugs eines Störsignals, zum Beispiel technischen Ursprungs mit anderen Zeitstrukturen, erheblich verringert, so dass die sich anschließende Analyse zu keiner Fehlpeilung führt.

By the system preferred use of multiple time stamps and / or multiple time structures is

• - Increases the accuracy of the signal time marker and
• - Significantly reduces the probability of (unwanted) inclusion of an interference signal, for example of technical origin with other time structures, so that the subsequent analysis does not lead to a mis-bearing.

Starting from the memory arrangement 50 Finally, you can have a basically bi-directional connection 60 a transmission of selected parameters of the signals S and S * to the central station 10 be made.

In 3 is the structure of the central station 10 shown. The Sferics online recording takes place in different amplitude and time ranges on two separate recording systems 40 . 40 ' , Both recording systems 40 . 40 ' is a highly accurate G [lobal] P [ositioning] S [ystem] time base 38 with an accuracy of about three hundred nanoseconds. All other individual components of the local network are GPS time via a central data server 70 made available.

The trained as station electronics first recording system 40 (See also 2A ) is used to collect so-called standard sferics and is not only part of the central station 10 but also of all measuring stations 20 respectively. 20 * , The recording of the field amplitudes takes place in each case for a time interval of 512 microseconds in an amplitude range of the magnetic field strength to six nanotesla.

If the specific trigger condition for a component of the signal S or S * is fulfilled, the GPS event time is frozen in a hardware register. After completion of the digitization, the magnetic field vector is calculated from the ratio of the time-dependent magnetic field profiles and its time-dependent amount, that is, the current Sferics position, on a screen or monitor 72 shown.

With an uncertainty of 180 degrees, the magnetic field vector indicates the direction of incidence of the signal S or S *. If the time course of the electric field amplitude is also available, the direction of incidence can be determined relatively accurately. This direction of incidence is combined with the trigger time and the time course of the magnetic field amount on the central data server 70 stored. In addition, a F [ast] F ourir T ransformation 74 the signal S or S * with online representation.

• – für ein Zeitintervall von 512 Mikrosekunden sowie
• – für ein Zeitintervall von 655 Millisekunden

in einem Feldstärkebereich bis sechzig Nanotesla aufzeichnet.For purposes of lightning research using measurement arrangements in the V ery] [ow] F [requency] range (so-called applied sferics research) is at the central station 10 a second recording system 40 ' installed, the amplitudes respectively

• - for a time interval of 512 microseconds as well
• - for a time interval of 655 milliseconds

in a field strength range up to sixty nanotesla records.

With this second recording system 40 ' Consequently, both the first events of a possible sequence of partial flashes over a period of 512 microseconds and the entire temporal structure of partial pulses within a flash P can be detected. The recording of the strong signals S and S * originating from the near field therefore takes place simultaneously in two different time windows with different temporal resolution: in addition to the standard time window of 512 microseconds (compare first recording unit 40 ), which in general records the signal S or S * generated by the first partial flash (= so-called "first return stroke") in high resolution, the time sequence of the individual strokes is recorded in a 655 millisecond time window (temporal interpolation points of sixteen kilobits ).

The sensitivity can be reduced because of the generally strong signals S and S *, so that in the second recording unit 40 ' a measurement range up to about sixty nanotesla results. The trigger time, the direction of incidence and both digitized waveforms are in turn centrally on the server 70 saved.

The second recording system 40 ' at the central station 10 therefore allows to observe events in two different time windows with different resolution. Thus, both the time sequence of the strokes within a lightning event as well as high-resolution the Sferics caused by individual strokes can be recorded.

By means of the communication unit 76 All required data and information D or D * from the external measuring stations 20 respectively. 20 * retrieved and on the central server 70 be filed. For certain standard data is done fully automatically at fixed times; special data records can be transferred manually.

In addition, by means of the communication unit 76 also serves external data requirements; For this a Fileil [ransfer] p [rotocol] server and a mailbox are available. All measuring stations 20 respectively. 20 * are via modem 76m and dial-up line 76w to the central station 10 tethered.

• – ein Meteosat-Empfänger 16, der stets die aktuellen Satellitenbilder anzeigt und archiviert, sowie
• – die Mailbox 18 eines Wetterdienstes, die per Wählleitung den Zugriff auf Bodenwetterkarten, auf Radarwetterkarten oder auf dergleichen ermöglicht.

In order to be able to examine the obtained Sferics data for correlations with meteorological parameters, weather information Q is required. Essentially two sources are available, namely

• - a Meteosat receiver 16 who always displays and archives the current satellite imagery, as well
• - the mailbox 18 a weather service that allows dial-up access to ground weather maps, on radar weather maps or the like.

Satellite images, the ground weather maps, the radar weather maps and the like are thus displayed online by means of a screen or monitor (= reference numerals 78 in 3 ). How finally the representation according to 3 is removable (in 3 exemplarily three) jobs 80 for data analysis to the central server 70 [= File Server, GPS Time Server (Netware Server)] connected.

• – die von den jeweiligen Messstationen 20 bzw. 20* übermittelten Daten und Informationen D bzw. D*, insbesondere auch in hohen Raten und/oder insbesondere auch ohne zeitlichen Verlust, auszuwerten und
• – abgestufte und/oder vollautomatische Warnungen vor klimatologischen und/oder meteorologischen Vorgängen, insbesondere vor Gewittern, abzugeben.

The invention essential functions of the central station 10 consist in

• - that of the respective measuring stations 20 respectively. 20 * transmitted data and information D or D *, especially in high rates and / or in particular without loss of time to evaluate, and
• - deliver graded and / or fully automatic warnings of climatological and / or meteorological processes, in particular thunderstorms.

In the lightning detection system 100 We have succeeded in making all essential components particularly efficient, emphasizing simplicity and cost minimization. The essential characteristics are briefly listed:
The sensors for the V [ery] L [ow] F [requency] radiation of flashes are two simple, crossed coils, with which the magnetic flux B (t) from the two components B _x (t) and B _y (t) of the flash field is measured (see. 2 B ).

A sufficiently faithful representation of the temporal course of the incoming impulses is given in the range of almost ten kilohertz to about four hundred kilohertz. From the ratio of the magnetic field components B _x and B _y , the direction of incidence of the lightning radiation can be determined. The VLF sensor works passively and does not require power.

For time stamping, a commercially available, separately mounted G [lobal] P [ositioning] S [ystem] receiver is used to accurately determine the signal arrival time to within about one microsecond; This accuracy is important for the functionality and efficiency of the bearings.

The sensor data runs in a commercially available P ersonal computer with a specially developed module card for digitization and preprocessing.

The sampling rate for signal digitization is set to one megahertz. Incoming signals are recorded continuously and analyzed in parallel, so that no loss occurs even at extremely high rates. The maximum processing rate is over a thousand signals per second.

By continuously sampling and using a parallel processing loop, the sensor measures 20 . 20 * without dead time. This allows the sensor 20 . 20 * process much higher data rates than conventional systems of the prior art. In conventional systems, after a signal detection for a certain (processing) time can not respond to other signals, so at high signal rates data loss occurs.

• – Signale können zeitlich früher gemessen werden, wodurch eine zuverlässige Gewitter-Frühwarnung ermöglicht wird; und
• – die Form von Gewitterzellen wird wesentlich besser sichtbar, weil aufgrund der hohen Empfindlichkeit um etwa eine Größenordnung (= um etwa Faktor zehn) mehr Signale erfasst werden als in Systemen gemäß dem Stand der Technik. Damit werden die Zellenkonturen besser sichtbar, und eine neuartige meteorologische Nutzung der Daten wird ermöglicht. Insbesondere eignen sich diese Daten wegen ihrer großen Anzahl zur Einspeisung in andere meteorologische Anwendungen, wie zum Beispiel für eine Veredlung und Vorhersagenutzung mittels der M[odel]O[utput]S[tatistics]-Technik.

Because the system 100 Compared to the prior art, it is possible to process significantly higher data rates, two successive properties are possible:

• - Signals can be measured earlier in time, allowing a reliable lightning early warning; and
• - The shape of thunderstorm cells is much more visible, because due to the high sensitivity by about an order of magnitude (= by about a factor of ten) more signals are detected than in systems according to the state of the art. This makes the cell contours more visible and allows a novel meteorological use of the data. In particular, because of their large number, these data are suitable for feeding into other meteorological applications, such as for example for refining and forecasting using the M [odel] O [utput] S tatistics technique.

For each signal, a compact data packet is formed and transmitted via telephone line to a central office; Due to the bandwidth of the transmission bandwidth, only about 100 signals per second can actually be transmitted, which proves to be sufficient even in the case of strong thunderstorms, so that buffer functions are rarely needed.

The entire digitized pulses B _x (t) and B _y (t) are archived at each station and retrieved at low-activity times, ie usually at night, in order to be usable for research and development purposes.

The sensors 20 . 20 * (see. 2A ) are at, for example, six locations in southern Germany (cf. 4 ) mounted at a mean distance of about 115 kilometers, run in continuous operation and send data to the control center 10 (see. 3 ).

The central computer is also a common computer [omputer] equipped with software for data organization and flash bearing and designed for continuous operation. Even larger networks can be operated according to the invention with a suitably equipped P [ersonal] C [computer] due to the efficiently designed operating modes and software solutions, so that no new and / or complex additional measures are required for the application of the present invention.

• – die Emissionshöhe H und/oder
• – die Direktionalität C, das heißt der räumliche Richtungsverlauf

einer Blitzentladung P lokalisiert werden kann, wenn die Abweichung der Ankunftszeit des Signals S an der der Blitzentladung P nächstgelegenen Messstation 20 von der Ankunftszeit des entsprechenden Signals S* an zum Beispiel drei der gleichen Blitzentladung P nicht nächstgelegenen Messstationen 20* bestimmt werden kann (vgl. 1), sind als praktisches Beispiel gemessene Blitze im Überwachungsgebiet mit Längen von 9,5 Grad bis 13,3 Grad und mit Breiten von 47 Grad bis 49,8 Grad in 6 dargestellt.Based on the basis 5 illustrated principles of the present invention, according to which

• - the emission level H and / or
• - the directionality C, that is the spatial direction course

a lightning discharge P can be located if the deviation of the arrival time of the signal S at the nearest of the lightning discharge P measuring station 20 from the arrival time of the corresponding signal S * to, for example, three of the same lightning discharge P measuring stations not nearest 20 * can be determined (cf. 1 ), as a practical example, measured flashes in the surveillance area with lengths of 9.5 degrees to 13.3 degrees and with widths of 47 degrees to 49.8 degrees in 6 shown.

It's just over 14,000 flashes, including partial flashes that are sent to at least three sensors 20 . 20 * were recorded. Bearings with only two sensors 20 . 20 * are not included and would increase the total number of events even more; However, the inclusion of these data requires, for reasons of avoiding incorrect bearings beforehand, a correction of the bearing directions with regard to the so-called "site error", which can only take place after an overall evaluation of seasonal data.

The in the 6 to 8th presented data of the network 100 are consistent in their sense and show an obviously meteorologically meaningful structure. Now the question arises as to how the division into C [loud] G [round] flashes and I [ntra] C [loud] flashes can be done.

The literature on discrimination of CG and IC flashes is contradictory and will not be discussed here. In any case, this question can still be regarded as open, whereby more clarity can be expected from still ongoing waveform analyzes with improved time resolution.

Consequently, the 2D network is expanded, namely in the form of the use of this 2D network as a pseudo-3D network, whereby a precise detection of the signal times, in particular the signal propagation times, at the individual sensor stations 20 . 20 * is realized.

• – für eine Emissionshöhe H von fünf Kilometern (= unterste Kurve),
• – für eine Emissionshöhe H von 7,5 Kilometern (= zweitunterste Kurve),
• – für eine Emissionshöhe H von zehn Kilometern (= mittlere Kurve),
• – für eine Emissionshöhe H von 12,5 Kilometern (= zweitoberste Kurve) und
• – für eine Emissionshöhe H von fünfzehn Kilometern (= oberste Kurve).

In this way, the possibility opens up to perform three-dimensional bearings at sufficiently close to the station lightning. The diagram shows 7A while the 7B . 7C and 7D set the appropriate sensitivity for altitude measurements:
So is in 7A the expected delay of the arrival times of the signals S, S * (in microseconds: us = μsec) as a function of the (observation) distance (in kilometers) for different emission levels H, namely

• - for an emission height H of five kilometers (= lowest curve),
• - for an emission height H of 7.5 kilometers (= second lowest curve),
• For an emission height H of ten kilometers (= mean curve),
• - for an emission height H of 12.5 kilometers (= second highest curve) and
• - for an emission height H of fifteen kilometers (= top curve).

7B shows the distribution (= number plotted on the right-hand axis) of the high-axis, measured in kilometers emission levels H of 951 flashes, which are targeted within a radius of up to about sixty kilometers around three measuring stations. It shows an emission maximum at a height H of just under ten kilometers.

At least three other measuring stations have detected the flashes. In total, 6,929 lightning strokes were taken at four or more stations during the same period; In addition, there are about 7,800 bearings with three measuring stations that allow 3D bearings only with the additional use of bearing directions.

• – n einer Höhe H zwischen sechs und sieben Kilometern (vgl. 7C) bzw.
• – in einer Höhe H von etwa zehn Kilometern (vgl. 7D).

In the 7C and 7D is in each case a corresponding lightning height distribution, that is to say a distribution of the emission height H of three-dimensionally-targeted lightnings during a thunderstorm at other times than in 7B shown. This results in an emission maximum

• - n a height H between six and seven kilometers (cf. 7C ) respectively.
• - At an altitude H of about ten kilometers (cf. 7D ).

The data shown does not include ground flashes.

According to the present invention, assuming a time resolution of the flash arrival times of about one microsecond, a height bearing at least in the distance range of up to about one hundred kilometers is possible.

The results of the evaluations based on this are in the 7A to 7D exemplified. It turns out that the focus of the V [ery] L [ow] F [requency] emission of discharges is in many cases not close to the ground, but at a higher altitude within the storm clouds.

The present method and its results are now to be examined more closely in order to better assess the relevance and reliability. If the method is illustrated by a concrete example, which is typical for the height estimates presented here, then an unusually strong flash can be used, followed 251 milliseconds later by a single weaker discharge at the same location.

Table 1 lists the essential data of the DF method for this event, namely the peak data of the main flash with the subsequent flash after 251 milliseconds: length width Type main flash 3D Bearing 11.1832 47.1962 CC 2D bearing 11.1889 47.1965 - follow Blitz 3D Bearing 11.1846 47.1974 CC 2D bearing 11.1909 47.1979 -

The specified deviations apply relative to the former bearing; The intrinsic quality of the bearing results obtained with data from six stations according to the present invention improves quite considerably in the transition from the two-dimensional bearing to the three-dimensional bearing.

If the lightning location takes place without consideration of emission levels, the result is an acceptable location, but the bearing error is higher than would be expected on average.

An inspection of the flash times recorded at all six stations of the network proves that the time of arrival at the station in question is atypical by almost seven microseconds compared to the time expected from the whole network and against the corresponding time errors of the other five stations.

Now, according to the invention, the 3D effect switched on and executed a new bearing with the flash height as an additional parameter to be optimized, resulting in a much more coherent overall result. The newly introduced and optimized emission level is 15.7 kilometers with a statistical error of 3.5 kilometers. If the completely independent follow-up stream is analyzed as well, you get a nearly identical flash point with an emission height of 16.4 +/- 3.5 kilometers.

The described procedure is applied to all data considered and provides comparable results. To make the 3D bearing as reliable as possible, it is possible to use only those cases in which discharges are measured at at least four sensor stations.

According to the invention, however, it is also possible to carry out a 3D bearing with sufficient accuracy by means of weaker signals detected at only three stations when the directions of incidence in the optimization procedure are added.

Table 2 shows the number of people around a measuring station, for example the central station 10 , three-dimensionally-targeted flashes (= detected signals) and compares them to the total number in the network (R denotes the radius of circumference, Ns the number of stations used for two-dimensional bearings, under "H>0" those with at least four measuring stations 20 respectively. 20 * three-dimensionally-targeted signals S and S * with high emission levels H are listed; Consider an exemplary surveillance area with longitudes between 9.5 degrees and 13.3 degrees and latitudes between 47 degrees and 49.8 degrees): R (km) Ns> = 3 Ns> = 4 H> O 20 365 173 36 30 1459 718 202 40 5254 2788 540 50 9088 4851 679 60 13508 7098 711 total area 60327 36909 2114

A finding can be seen insofar as only a significant proportion of higher-lying emission locations can be assigned to the signals registered in the network.

With the present system 100 as well as with this system 100 underlying method can also be a delimitation or separation of cloud flashes (= discharges P within a cloud W and / or between at least two clouds W) against earthflashes (= discharges P between cloud W and earth E) are made.

So is in 8th the symmetrical curve around the zero time point (= no temporal deviation) can be identified as a bottom flash or earth flash curve, that is, in 8th Symmetrical curve shows the 4,450 non-height detectable bottom flashes (C [loud] G [round] x 0.26).

• – für den Distanzbereich bis zwanzig Kilometer (= Wolkeblitzkurve mit dem niedrigsten Peak in 8),
• – für den Distanzbereich zwischen zwanzig und vierzig Kilometern (= Wolkeblitzkurve mit dem drittniedrigsten Peak in 8),
• – für den Distanzbereich zwischen vierzig und sechzig Kilometern (= Wolkeblitzkurve mit dem viertniedrigsten Peak in 8) und
• – für den Distanzbereich über sechzig Kilometer (= Wolkeblitzkurve mit dem zweitniedrigsten Peak in 8).

In the 8th The other four curves shown are from the cloud flashes, the negative times signifying a time delay; these are plotted for four different distances between flash and sensor station, namely

• - for the distance range up to twenty kilometers (= cloud flash curve with the lowest peak in 8th )
• - for the distance range between twenty and forty kilometers (= cloud flash curve with the third lowest peak in 8th )
• - for the distance range between forty and sixty kilometers (= cloud flash curve with the fourth lowest peak in 8th ) and
• - for the distance range over sixty kilometers (= cloud flash curve with the second lowest peak in 8th ).

Overall is in 8th a distribution of 2,247 determined time deviations of the next closest station at a total of 6,697 lightning strikes.

All in all, the system as well as the flash detection method not only has an increased efficiency, but also offers the possibility to perform a three-dimensional distinction of discharges within a cloud and cloud-to-ground discharges:
For the past twenty years, flash-localization systems have been developed for commercial use primarily in the US, but also in France and Japan; this development culminated in the construction of the North American Lightning Detection Network (N [orth] A [merican] L [ightning] D [etection] N [etwork]) in 1989. Since then, various upgrades have been carried out, and the technique has been expanded into many Countries introduced.

• – werden Effizienzen im Hinblick auf Blitz-Nachweise berichtet, die etwa neunzig Prozent betragen,
• – sind die erreichten Lokalisierungsgenauigkeiten besser als ein Kilometer und
• – wird die IC-CG-Unterscheidung durch Wellenform-Kriterien als nahezu perfekt bezeichnet.

According to the commonly used, discussed at the beginning of the prior art

• Efficiencies are reported with regard to lightning evidence, which is about ninety percent,
• - the localization accuracies achieved are better than a kilometer and
• The IC-CG distinction is called nearly perfect by waveform criteria.

Ground flash densities are calculated from operating results and are intended to be important data for lightning protection technologies.

For more advanced networks, conventionally, the possibility of accurate flash localization has been checked primarily by measurements involving flashes in radio towers, yet the detection efficiency and the method of distinguishing IC discharge from CG discharge remains a questionable matter; initially, networks were developed which were only intended to indicate potential damage-causing CG discharges on the ground, while IC discharges were suppressed by methods not described in detail.

Although it is reported that the rise time and peak zero crossing time of the detected VLF pulses can be used; however, the relevant distinguishing parameters have been repeatedly changed in the past. In recent years, it has finally been realized that IC discharges are at least metereologically relevant, and thus modern networks also indicate the detected IC flash events.

Accordingly, a three-dimensional option for the determination of emission levels as well as for an on-line differentiation of C [loud] G [round] and G [round] C [loud] discharges in demarcation or in comparison to I [ntra] opens up. C [loud] or C [loud] C [loud] discharges:
The new, permanently operable and, for example, in South-East Germany covering an area of approximately 300 kilometers over 400 kilometers of lightning localization network makes it possible to analyze the recorded lightning activity, in particular with regard to the occurrence of severe thunderstorms and a comparison with data established lightning localization networks pull.

• – Verglichen mit konventionellen Netzwerken gemäß dem Stand der Technik zeigt das vorliegende Netzwerk bis zu einer Größenordnung mehr Entladungen an; die meisten zusätzlichen Signale weisen eine niedrigere Amplitude auf.
• – Für die Unterscheidung von Entladungen innerhalb einer Wolke (I[ntra]C[loud]) gegenüber Wolke-Boden-Entladungen (C[loud]G[round]) wird eine neue dreidimensionale Technik angewendet; dies bedeutet unter anderem, dass die konventionelle, mittels Wellenform-Kriterien bewerkstelligte Unterscheidung von IC- und CG-Entladungen zu modifizieren ist, nicht zuletzt weil die meisten lokalisierten Ereignisse Bodenblitze waren.
• – Verglichen mit dem Stand der Technik sind auch die Bodenblitzdichten einer Revision in Richtung höherer Anzahlen zu unterziehen.

The most notable results can be summarized as follows:

• Compared with conventional prior art networks, the present network displays up to an order of magnitude more discharges; most additional signals have a lower amplitude.
• - For the differentiation of discharges within a cloud (I [ntra] C [loud]) against cloud-to-ground discharges (C [loud] G [round]), a new three-dimensional technique is applied; among other things, this means that the conventional waveform-based discrimination of IC and CG discharges needs to be modified, not least because most of the localized events were bottom flashes.
• Compared with the prior art, the bottom flash densities are also subject to a revision in the direction of higher numbers.

The network according to the present invention includes, for example, six sensor stations and is designed to measure as much of the arising thunderstorms as possible in V [ery] L [ow] F [requency] activity; Highly efficient and advanced data processing is implemented in the present invention to maximize the metrologically relevant output and to identify potential opportunities for better data and information related to early warning purposes.

As a result, by means of the present invention, lightning discharges can be collected in much larger quantities than expected. A comparison with the data of conventional networks shows a surplus of factor three to ten, so that a much better detection of thunderstorm cells and thunderstorm contours becomes possible. These results raise the question of where the additional hits come from and give rise to a more detailed investigation of the procedures for IC-CG distinctions.

Since the prior art method of analyzing waveform criteria is highly complex, the more immediate method of 3D localization is used, which requires neither assumptions nor any parameters to be adjusted.

In a manner which is surprising to a person skilled in the art, this technique, which has hitherto been used in no other network, proves to be very successful and permits the identification of IC flashes in approximately 85 percent of all cases. The principle of this method is shown below:
If a sensor sufficiently close to a discharge senses an IC event (as opposed to horizontal propagation in the event of a CC thunderstorm), that sensor receives the VLF signal with a time delay dT = T _P -T _H (see FIG. 5 ). An emissions level of ten kilometers, for example, causes a delay of about three microseconds at a sensor distance of fifty kilometers.

Since the accuracy of GPS-based measurements of arrival times is usually on the order of one microsecond or better, 3D discrimination of IC flashes is possible as long as the average distance of the network sensors from the discharge event is not too great.

The applicability of the 3D method to a given network can be determined by examining the distribution of the time delay P (dT) obtained from the respective sensor closest to the localized flash. In this context, localization implies that the discharge must be detected at a sufficient number of sensor stations.

In the absence of IC discharges, P (dT) is a highly symmetric distribution, while the existence of detectable IC discharges causes a shift in P (dT) distribution to negative times (corresponding to a time delay) (cf. 8th ).

Given the latter facts, the emission level H can be introduced into the localization algorithm as an additional parameter. The experimental distributions p (dT) according to the present invention are indeed asymmetric and give typical results. The deduced emission levels of a few kilometers appear reasonable, can not result from CG discharges and therefore probably represent IC discharges.

In a refinement essential to the invention, network simulations are used to make the results more resilient in accordance with the present invention:
Thus, calculated arrival times are statistically deformed until the accuracy of the entire localization for the network matches actual results, and thus lies between half a kilometer in the center and one kilometer near the network boundary.

If no IC discharges are accepted, the artificial time shifts give about fifteen percent "false" IC identifications, while assuming one hundred percent IC discharges with the assumed height distribution, 85 percent is detected.

An examination of the actual data sample cases shows that fewer than thirty percent of the events can be attributed to IC discharges. As a consequence, it has to be stated that most of the additionally recorded events are CG discharges.

• – Eine neue Technik für IC-CG-Unterscheidungen, die im Prinzip jedes moderne Lokalisierungsnetzwerk bewerkstelligen kann und die neuen Input in Bezug auf die Wolkenphysikforschung gibt, wird zur Verfügung gestellt.
• – Die Ergebnisse gemäß der vorliegenden Erfindung können genutzt werden, um die vorstehend erwähnten, an sich bekannten Wellenform-Kriterien zu überprüfen und zu verbessern; es steht zu erwarten, dass eine Kombination der beiden Verfahren eine sehr zuverlässige IC-CG-Unterscheidung ergibt.
• – Das Auftreten einer großen Anzahl zusätzlicher Ereignisse erfordert eine Aktualisierung der Nachweiseffizienzen von Netzwerken.
• – Der erfindungsgemäß erhaltbare Anstieg von CG-Entladungen erfordert eine Aktualisierung der Karten von Bodenblitzdichten.

Various implications of the results according to the present invention require further attention:

• - A new technique for IC-CG distinctions, which in principle can accomplish any modern localization network and gives new input in terms of cloud physics research, is provided.
• The results according to the present invention can be used to check and improve the above-mentioned waveform criteria known per se; It is expected that a combination of the two methods will yield a very reliable IC-CG distinction.
• - The occurrence of a large number of additional events requires updating the detection efficiencies of networks.
• The increase of CG discharges obtainable according to the invention requires an update of the maps of bottom flash densities.

With the results of the present invention, the waveform criteria can be checked for a much larger data basis than before; Also according to the invention an optimization of the detection efficiency, a comparison with VHF data of other highly specialized 3D research systems and an investigation of the processes in relation to IC discharges is ensured.

Finally, the application of the signal analysis according to the invention allows the use of weak and irregularly shaped flash pulses and allows in all such cases a sufficiently accurate time recording. This has proven to significantly improve the detection efficiency of the lightning measurement network and, in addition to a reliable two-dimensional spatial bearing, also to a clear height bearing of lightning emissions in the VLF range.

In addition to this two-dimensional spatial bearing and / or to this three-dimensional height bearing or directionality bearing according to the present invention, at least one of the measuring stations 20 respectively. 20 * Adjusted and / or calibrated, whereby an improved time tracking is achieved.

• a) die zweidimensionale Blitzortung wird ohne Einbezug der zu eichenden Station 20 bzw. 20* ausgeführt;
• b) es wird die Laufzeit vom Blitzort zur betrachteten Station 20 bzw. 20* ermittelt;
• c) die so ermittelte Signal-Ankunftszeit wird mit der tatsächlich gemessenen Ankunftszeit verglichen;
• d) es wird eine Statistik über die unter Schritt c) ermittelten Zeitdifferenzen erstellt;
• e) bei korrekter Eichung sollte die Differenz den Mittelwert Null aufweisen und um Null schwanken;
• f) es stellt sich im Allgemeinen heraus, dass besagte Differenz einen Trend entweder zu positiven Werten oder zu negativen Werten aufweist; diese Differenz ist der Zeit-Korrekturterm;
• g) es ist jedoch darauf zu achten und durch geeignete Maßnahmen zu berücksichtigen, dass eine Signalverzögerung nicht durch Wolke-Wolke-Blitze verursacht wird (vgl. unten) bzw. als solche erkannt wird;
• h) bei folgenden Ortungen ist die gemittelte Differenz aus Schritt f) als Zeit-Korrektur anzubringen;
• i) die besagte Korrektur kann ebenfalls in Abhängigkeit von der Richtung zum Blitzort als Korrekturfunktion ermittelt werden, um gegebenenfalls nicht-isotrope Umgebungseffekte einzubeziehen.

Such a calibration (correction) method advantageously has the following steps:

• a) the two-dimensional lightning location is without involving the station to be calibrated 20 respectively. 20 * performed;
• b) it will be the duration of the flash to the station under consideration 20 respectively. 20 * determined;
• c) the thus determined signal arrival time is compared with the actually measured arrival time;
• d) a statistic is generated on the time differences determined in step c);
• e) if correctly calibrated, the difference should be zero and fluctuate around zero;
• f) it generally turns out that said difference has a trend towards either positive values or negative values; this difference is the time correction term;
• (g) however, care should be taken and appropriate measures taken to ensure that signal delay is not caused by cloud-cloud flashes (see below) or recognized as such;
• h) in the case of subsequent location, the averaged difference from step f) is to be applied as the time correction;
• i) said correction can also be determined as a function of the direction of the flash location as a correction function, to include possibly non-isotropic environmental effects.

• – wird die zweidimensionale Ortspeilung verbessert und/oder
• – wird bei Einbezug blitznaher Sensorstation(en) die dreidimensionale Höhenpeilung hinsichtlich der Wolke-Erde-Blitz-Diskriminierung sicherer, weil noch kleinere Zeitverzögerungen als signifikant zu werten sind; die Höhenpeilung wird quantitativ verbessert, weil die ermittelten Höhen aufgrund genauerer Zeitdifferenzen ebenfalls automatisch genauer werden.

As a result of the calibration correction or (post-) adjustment

• The two-dimensional spatial bearing is improved and / or
• - When incorporating lightning-fast sensor station (s), the three-dimensional height bearing becomes safer with regard to cloud-earth-lightning discrimination, because even smaller time delays are considered to be significant; the height bearing is improved quantitatively, because the determined heights also automatically become more accurate due to more precise time differences.

LIST OF REFERENCE NUMBERS

100

system

10

Central station or central office

16, 18

Interface of the central station 10 , especially

16

Meteosat receiver

18

Mailbox of a weather service

20

a pulse source spatially nearest measuring station or sensor (station)

20 *

a pulse source spatially not nearest measuring station or sensor (station)

22

Data processing system, in particular P [ersonal] C [computer]

30

Antenna, in particular broadband antenna body

38

Time measuring device, in particular G [lobal] P [ositioning] S [ystem] clock

40

Station electronics of the central station 10 as well as the measuring station 20 . 20 * , in particular first Sferics online recording system of the central station 10

40 '

second station electronics of the central station 10 , in particular second Sferics online recording system of the central station 10

42

Amplifier unit of the station electronics 40

42 '

Amplifier unit of the central station 10

44

Filter unit of the station electronics 40

44 '

Filter unit of the central station 10

46

A [nalog] / D [igital] converter unit of the station electronics 40

50

Storage unit of the measuring station 20 . 20 *

60

Connection between central station 10 and measuring station 20 . 20 *

70

Server unit of the central station 10 , in particular central data server and / or time server

72

the server unit 70 associated display unit, for example, screen or monitor, in particular for displaying the current Sferics situation

74

Ourour transformation of the signal S, S * with online representation

76

Communication center of the central station 10

76m

Modem of the communication center 76

76w

Dial-up line of the communication center 76

78

Online display of satellite images, of ground weather maps, of radar weather maps and the like by means of at least one display unit, in particular by means of at least one screen or monitor

80

Workplace for analysis of data and information D, D *

A

Amplitude of the signal S, S *

C

Directionality, in particular spatial direction course, the impulse delivery or impulse emission

d

Distance of the measuring stations 20 . 20 * to each other

dT

Propagation delay, where dT = T _P -T _H

D

from the measuring station or sensor (station) spatially closest to a pulse source 20 upcoming data and information

D *

from the measuring station or sensor not stationed spatially next to a pulse source (station) 20 * upcoming data and information

e

Earth, especially soil

H

Height of the pulse source, in particular emission level or send height

K

transmitter

M

plane

P

atmospheric discharge, especially lightning

Q

additional climatological and / or meteorological data and information source

R

Removal of the pulse source from the measuring station 20 . 20 *

S

the measuring station or sensor (station) spatially closest to a pulse source 20 associated signal

S *

the measuring station or sensor not stationed spatially closest to a pulse source (station) 20 * associated signal

S '

by means of the amplifier unit 42 amplified signal

S ''

by means of the filter unit 44 filtered signal

T _H

adjusted for the effect of height H impulse delivery time or pulse emission time, in particular by effect of the height H cleaned discharge time

T _P

Pulse output time or pulse transmission time, in particular discharge time

W

cloud

Claims (8)

Use of a measuring station ( 20 ) with an antenna body ( 30 ) designed for the measurement of low-frequency magnetic fields for a lightning localization network ( 100 ) with several spatially separated measuring stations ( 20 respectively. 20 * ) as a pseudo-3D network for identifying discharges emitted in high-altitude storm clouds, wherein: - the antenna body ( 30 ) has a mechanically self-supporting form with at least two simple, circularly arranged circular coils without moving and / or weather-sensitive components, which is designed for setting up outdoors, - wherein by means of the two coils, the magnetic induction (B (t)) from the x - and y-components (B _x (t), B _y (t)) of the flash field is detected, - wherein the primary circuit of the antenna body ( 30 ) from the secondary circuit of the antenna body ( 30 ) is galvanically isolated, and - wherein the magnetic induction (B (t)) is measured directly as a function of time, the measuring station ( 20 ) at least one time measuring device ( 38 ) for determining the respective chronological course of electromagnetic radiation of the low-frequency magnetic fields of detected signals (S or S *), wherein the electromagnetic radiation originates from at least one pulse source of natural origin, namely at least one atmospheric discharge (P), and wherein the height (H) of the pulse source, namely the emission height, and the directionality (C), namely the spatial direction, the pulse output or pulse emission effected by the pulse source localized by the deviation of the arrival time of the signal (S) at the closest measuring station of the pulse source from the arrival time of the signal at at least one of the same pulse station ( 20 * ), wherein the directionality (C) of the pulse output is identified as being substantially vertical when the amplitude (A) of the signal (S or S *) is reciprocal to the distance (R) of the pulse source of the respective measurement station, and identified as being substantially horizontal when the amplitude (A) of the signal (S or S *) deviates from the reciprocal proportionality, and this deviation by taking into account - the elevation angle and - the angle between the momentum output axis and direction to the respective one Measuring station is correctable. Using the measuring station ( 20 . 20 * ) with the antenna body ( 30 ) according to claim 1, wherein measuring station ( 20 . 20 * ) one in the housing of the measuring station ( 20 . 20 * ) arranged station electronics ( 40 ), which preferably comprises an amplifier ( 42 ) for low-noise amplification of the antenna body ( 30 ) (S, S *), the station electronics ( 40 ) preferably further comprises a filter stage ( 44 ), a digitizing unit ( 46 ) and a storage unit ( 50 ) having. Using the measuring station ( 20 . 20 * ) with the antenna body ( 30 ) according to claim 2, wherein the filter stage ( 44 ) the amplifier ( 42 ), wherein preferably by means of the filter stage ( 44 ) filtered signal (S '') in the digitizing unit ( 46 ) is subjected to a signal digitization, and wherein by means of the low-noise amplifier ( 42 ) and the filter stage ( 44 ) a smoothing and optimization method of the signals (S, S *) can be performed. Using the measuring station ( 20 . 20 * ) with the antenna body ( 30 ) according to claim 2 or 3, wherein between the digitizing unit ( 46 ) and the storage unit ( 50 ) a P ersonal C [computer] ( 22 ) is switched. Using the measuring station ( 20 . 20 * ) with the antenna body ( 30 ) according to claim 2, wherein the station electronics ( 40 ) a highly accurate G [lobal] P [ositioning] S [ystem] time base ( 38 ) is available with an accuracy of about three hundred nanoseconds, or a commercially available, separately mounted G [lobal] P [ositioning] S [ystem] receiver is available. Using the measuring station ( 20 . 20 * ) with the antenna body ( 30 ) according to one of claims 1 to 5, wherein the measuring station ( 40 ), in particular as V [ery] L [ow] F [requency] measuring station trained, passive and operates without power. Use of an antenna body ( 30 ), designed for the measurement of low-frequency magnetic fields, for a measuring station ( 20 . 20 * ) according to one of claims 1 to 6, wherein - the antenna body ( 30 ) has three coils, which are arranged at right angles to each other, for the location of pulse sources, in particular for lightning location, preferably by means of V [ery] L [ow] F [requency] measurements either by means of the direction-finding method (so-called "D [irection] F [inding] ") and / or by means of the transit time difference method (so-called" T [ime] O [f] A [rrival] method "). Use of the antenna body ( 30 ) according to claim 7, wherein - the antenna body ( 30 ) has three circularly formed coils, wherein the coils each define a plane, and - preferably the planes of the coils are arranged at right angles to each other.

Images