Pluviometric method for estimating a volume of precipitating water, and corresponding meteorological device

Abstract

The invention relates to a pluviometric method for estimating a volume of liquid water precipitating above a geographical position, said method being implemented by a GNSS signal-receiving and processing device comprising: an antenna able to pick up GNSS signals and positioned at the geographical position; and a computing unit able to receive and process the GNSS signals picked up by the antenna; the method comprising the following steps: a step of receiving at least one GNSS signal, referred to as a received signal, the spectrum of which is composed of at least two distinct frequencies; a step of determining a total delay at the zenith of the received signal at the geographical position; a step of determining a volume of liquid water precipitating above the geographical position from only the total delay at the zenith.

Description

[0001] DESCRIPTION

[0002] TITLE OF THE INVENTION: Rainfall method for estimating the volume of precipitating water and corresponding meteorological device

[0003] Technical field of the invention

[0004] The invention relates to the technical field of satellite rainfall and connected rainfall objects. It relates more particularly to a rainfall method for estimating a volume of precipitating liquid water above a predetermined geographical position by analyzing GNSS signals.

[0005] The invention also relates to a meteorological device implementing such a method and a rainfall estimation system implementing such a method for estimating a volume of liquid water precipitating above a geographical position.

[0006] Technological background

[0007] Estimating rainfall amounts, and particularly estimating precipitation rates, is an essential component in weather prediction. Estimating precipitation rates is crucial in predicting extreme weather scenarios such as floods (especially so-called "flash" floods), river overflows, and associated phenomena such as dam bursts or flooding of populated areas.

[0008] Therefore, the successful management of civil security in the face of these scenarios is conditioned by the launch of an early warning, as accurate as possible, making it possible to warn the populations concerned as quickly as possible and, if necessary, to implement safeguard and evacuation plans for potentially affected areas in order to keep the population living there safe.

[0009] Today, there is a rich set of rainfall techniques for estimating precipitation rates; these techniques are based on one or more of the following: networks of weather stations, weather radars; satellites for observing clouds and their evolution.

[0010] Although capable of detecting precipitation and indicating its intensity and location, the main drawbacks of the aforementioned techniques lie in their low spatial resolution, due in particular to a total or partial absence of network coverage. However, better spatial resolution and / or higher density, particularly in the order of a kilometer, are necessary to be able to accurately estimate the precipitation rate in order to make sufficiently relevant predictions, particularly around riverside villages.

[0011] Furthermore, the aforementioned technologies are difficult to access by local entities such as town halls or civil associations. This is due, on the one hand, to their high cost and technicality; and on the other hand, to the lack of adaptability and deployment of said technologies on a small scale.

[0012] There is therefore a need for a method for estimating with high precision rainfall quantities, such as the precipitation rate, over a given geographical area, in particular for areas with a surface area of ​​the order of a square kilometer.

[0013] There is also a need for a robust, reliable and low-cost rainfall estimation system that can implement the rainfall estimation method sought above. In addition, the system must be able to be deployed simply and quickly while being weather-resistant and easily repairable.

[0014] Objectives of the invention

[0015] The present invention aims to remedy all or part of the drawbacks of the prior art, in particular those set out above, by proposing a solution making it possible to determine the volume of precipitating liquid water above a geographical position from the sole measurement of a total delay at the zenith of a GNSS signal.

[0016] The invention aims to provide a pluviometric method for estimating a volume of liquid water precipitating above a geographical position.

[0017] The invention aims in particular to provide, in at least one embodiment, a meteorological device capable of receiving and processing GNSS signals and of implementing the pluviometric method for estimating a volume of precipitating liquid water above a geographical position.

[0018] The invention also aims to provide, in at least one embodiment, a hydrological alert system in a geographical area comprising watercourses and capable of alerting to risks of flooding and / or rising water levels.

[0019] Statement of the invention

[0020] To do this, the invention relates to a pluviometric method for estimating a volume of liquid water precipitating above a given geographical position.

[0021] The pluviometric method comprises the following steps: a step of reception, by an antenna, called the reception antenna, arranged at said predetermined geographical position and capable of capturing GNSS signals, of at least one GNSS signal (21), called the received signal, the spectrum of which is composed of at least two distinct frequencies; a step of determination by a calculation unit capable of receiving and processing GNSS signals, of a total delay at the zenith of each signal received by said reception antenna at said geographical position; a step of determination, by said calculation unit, of a volume of precipitating liquid water above said predetermined geographical position from only said total delay at the zenith determined in the previous step and atmospheric data relating to said predetermined geographical position.

[0022] A pluviometric method according to the invention therefore makes it possible to estimate the volume of precipitating liquid water contained in the atmospheric column above the given geographical position from the sole total delay at the zenith of the received GNSS signal.

[0023] If desired, a plurality of GNSS signals may be received by the receiving antenna in order to improve the accuracy of the estimation of the volume of precipitating liquid water above said column of said geographical position. The plurality of GNSS signals may in particular come from a plurality of distinct satellites, in particular satellites belonging to distinct satellite constellations such as GPS, Galileo, GLONASS, Bei-Dou.

[0024] These different constellations emit signals in well-defined frequency bands. The invention can use signals whose spectrum includes frequencies in any of said bands, in particular the GPS LI, L2 and L5 bands; and the Galileo E1, E6, E5a, E5b bands.

[0025] It is customary to use L1 = 1575.42 MHz and L2 = 1277.60 MHz as the two carrier frequencies used to transmit information coded in binary form.

[0026] The accuracy in estimating the volume of precipitating liquid water contained in the fluid column above the given geographical position is a function of different parameters, one of them is the accuracy in determining the total zenith delay which, in turn, depends on the precise knowledge of the geographical location of the receiving antenna.

[0027] In one embodiment of the invention, the calculation unit determining a volume of precipitating liquid water above said predetermined geographical position from only said total delay at the zenith is a remote calculation unit, in particular forming part of a remote computer server.

[0028] This feature allows the computing unit to be configured according to specific computing power requirements. In particular, in order to carry out the mathematical calculations necessary to determine the said volume of precipitating liquid water.

[0029] In one embodiment of the invention, the pluviometric method for estimating a volume of precipitating liquid water above a given geographical position comprises a preliminary step of positioning the receiving antenna by an absolute positioning method.

[0030] By doing so, the accuracy of determining the total zenith delay, and consequently that of estimating the volume of precipitating liquid water, is improved.

[0031] In a cumulative embodiment of the invention, the absolute positioning method is a precise point positioning method.

[0032] The precise point positioning method allows for precise positioning using a single receiver and does not depend on the availability of data from reference stations near the user. In addition, this method ensures centimeter-level positioning accuracy of the receiving antenna.

[0033] Alternatively, the absolute positioning method is a satellite augmentation system SBAS method or a PPP-RTK precise point positioning method.

[0034] In one embodiment of the invention, the step of receiving at least one GNSS signal further comprises the following preprocessing sub-steps: a sub-step of reception timestamping, by the calculation unit, of at least one received signal; a sub-step of decoding, by the calculation unit, the transmission timestamp of said received signal; a sub-step of determining and correcting the delays and / or disturbances generated by the crossing of the ionosphere of said received signal from said at least two distinct frequencies composing the spectrum of said received signal; a sub-step of determining, by the calculation unit, at least one total oblique delay due to at least one received signal from said transmission and reception timestamps of said at least one received signal.

[0035] Thus, the sub-steps of decoding the transmission timestamp and the signal reception timestamp make it possible to determine the signal travel time between the satellite at the time of transmission of said signal and the receiving antenna.

[0036] Since the ionosphere is a dispersive medium, it propagates the different frequencies composing the signal emitted by the satellite at different speeds, thus causing a delay in the signal, called ionospheric delay. This delay can be corrected using mathematical models such as the "ionosphere free", "Klobuchar" and "NeQuick-G" models.

[0037] In the case of the "ionosphere-free" model, first-order ionospheric effects on both code and carrier phase measurements depend on the inverse square of the signal frequency. Therefore, dual-frequency receivers can eliminate ionospheric effects by a linear combination of code or carrier measurements.

[0038] The passage of the GNSS signal through the troposphere causes a signal delay, called tropospheric delay. This delay results from a change in the refractive index of the propagation medium due to air humidity and variations in pressure and temperature. Tropospheric delay is thus strongly correlated with the season, local time, and latitude. The troposphere is composed of neutral gases and is not a dispersive medium for electromagnetic signals, which implies that atmospheric delay affects GNSS frequencies in the same way.

[0039] The main species in the near atmosphere, nitrogen and oxygen, correspond to the dry part of the troposphere. These effects contribute 90% of the total delay and are slow dynamic, i.e., about 1% variation over a few hours, and are therefore fairly well modeled.

[0040] The humid part of the troposphere, i.e. water vapor, is responsible for the remaining 10% of the delay; the main reason for this delay is due to the dipole moment of the water vapor molecules.

[0041] Thus, the tropospheric delay is written as the sum of these two contributions.

[0042] In the context of the present invention, "total slant delay" means the total delay that the GNSS radio signal experiences due to the neutral atmosphere along the path between a satellite and a ground receiving antenna. In one embodiment of the invention, the value of a total zenith delay of the received signal at the geographical position is determined from said previously obtained total slant delay of the received signal.

[0043] According to this characteristic, the total oblique delay is transformed into a total zenith delay allowing the determination of the volume of liquid water precipitating on the vertical column above the geographical position of interest.

[0044] Advantageously, this transformation is performed using a roll-off function describing the dependence of the delay on the elevation angle, which makes it possible to project an oblique delay into a zenith delay as a function of the local elevation of the line of sight.

[0045] Examples of such drawdown functions are those known as the "Mariani", "Davis" and "Herring" drawdown functions. The "Mariani" drawdown function is given by the following continued fraction:

[0046] [Math. 1]

[0047] Due to the nature of the total delay at the zenith of the received signal, it provides information on the mass of water in liquid and gaseous form in the column above the geographical position.

[0048] In one embodiment of the invention, the step of determining a volume of precipitating liquid water above said geographical position comprises the following sub-steps: a sub-step of determining a dry component of said total delay at the zenith of the received signal; a sub-step of determining the volume of precipitating liquid water above the geographical position from said total delay at the zenith of the received signal and the dry component of said total delay at the zenith of the received signal; By proceeding in this way, the invention seeks to differentiate the so-called "dry" and "wet" contributions constituting said total delay at the zenith of the received signal.

[0049] If desired, the sub-step of determining a dry component of the total delay at the zenith of the received signal, denoted ZHD, can be obtained by a hydrostatic physical model, non-limiting examples of such models are the “Saastamoinen” model, the “Hopfield” model or the model known as “MOPS”.

[0050] The mathematical expressions of the “Hopfield” and “Saastamoinen” models are given respectively by the following equations:

[0051] [Math. 2] rm Kl 1

[0052] ZHD = 0.62291 —— - + 0.0023081 hPal T

[0053] [Math. 3]

[0054] 0.022767 Ug-1 P

[0055] ZHD = - - -

[0056] 1 - 0.00266 cos( ) - 0.00028 [-M h

[0057] L / C772 J

[0058] In the preceding mathematical expressions, P is the pressure surface, T the temperature, h the ellipsoidal height and < > the latitude.

[0059] In one embodiment of the invention, the sub-step of determining the volume of precipitating liquid water above the geographic position from said total zenith delay is performed by a machine learning algorithm.

[0060] This advantageous embodiment provides a tool for quickly determining the volume of precipitating liquid water from the sole value of the total delay at the zenith. This feature is particularly interesting in decision-making associated with the volume of precipitation, such cases being encountered for example in the assessment of flood risks and / or risks of overflowing watercourses.

[0061] Furthermore, the generalization power of machine learning algorithms makes it possible to process the set of possible cases according to the combinations of meteorological parameters. In one embodiment of the invention, the machine learning algorithm is a scalar predictor taking as input a total zenith delay value of a GNSS signal; said scalar predictor being trained on a set of data relating to total zenith delay values ​​of GNSS signals, geographical position and meteorological data.

[0062] By doing so, the invention makes it possible to establish the dependency links between the different meteorological parameters, the geographical position and the total delay values ​​at the zenith. Thus, it is possible to estimate the volume of precipitating liquid water from the sole value of the total delay at the zenith.

[0063] According to an advantageous embodiment, said scalar predictor is a graph neural network, one of the vertices of which is made up of the architecture of the “MetNet” network; this part makes it possible to take into account the different meteorological conditions. The invention uses the transfer learning technique by supplementing the graph with:

[0064] 1. an input vertex consisting of a dense-layer perceptron-type neural network taking into account the total zenith delay; and by

[0065] 2. an output vertex consisting of a single neuron and an activation function returning a real value corresponding to the volume of precipitating liquid water sought.

[0066] In an alternative embodiment of the invention, the sub-step of determining the volume of precipitating liquid water above the geographical position from said total zenith delay is performed by a physical model of the gaseous water portion of the troposphere.

[0067] This advantageous characteristic alternatively allows the volume of precipitating liquid water to be deduced by subtracting the dry part and the wet part in the form of vapor from the total delay at the zenith.

[0068] The invention also relates to a meteorological device capable of receiving and processing GNSS signals and capable of implementing the pluviometric method for estimating a volume of precipitating liquid water above a geographical position according to the invention.

[0069] The meteorological device makes it possible to measure the volume of precipitating water above the said geographical position.

[0070] In one embodiment of the invention, said meteorological device comprises means for communicating with a remote meteorological server, including in particular meteorological data such as temperature, humidity, wind speed and direction, precipitation, atmospheric pressure, as well as short and long term forecasts of thunderstorms and other meteorological phenomena.

[0071] Advantageously, the weather data is regularly updated and allows the weather device to have accurate and up-to-date weather information. In addition, the weather device can also receive real-time weather alerts in case of extreme weather conditions.

[0072] Advantageously, the meteorological devices may be meteorological devices for monitoring watercourses when these are positioned at the level of a watercourse as described in the applicant's patent document WO2023174922 describing a device dedicated to measuring the height of the water surface and the surface velocity of hydrological systems by remote sensing.

[0073] According to this advantageous characteristic, the determination of the volume of precipitating water above the location of said rain gauge device makes it possible to determine the river contribution directly integrated into the watercourse at said geographical location.

[0074] The invention also relates to a meteorological system for estimating a volume of precipitating liquid water above a geographical area comprising a plurality of meteorological devices, according to the invention, distributed within said geographical area.

[0075] The plurality of devices makes it possible to cover an area larger than a single geographical position and to estimate by weighting and / or extrapolation a volume of precipitating liquid water in said geographical area.

[0076] In one embodiment of the invention, a meteorological device among the plurality of meteorological devices is a master device responsible for centralizing the measurements carried out by all of the meteorological devices distributed in said geographical area.

[0077] According to this advantageous characteristic, it is possible to make all the meteorological devices cooperate in order to centralize the information on the volumes of precipitating liquid water calculated by each of said devices.

[0078] In one embodiment of the invention, each meteorological device can be designated, at any desired time, as master with respect to the complementary set of meteorological devices.

[0079] According to this advantageous feature, it is possible to operate the system even when a failure is detected in one of the meteorological devices and especially the master device.

[0080] Advantageously, such meteorological devices for monitoring watercourses may include means for measuring the level of water flowing in a watercourse as well as means for measuring the flow rate of said watercourse, in particular through LIDAR scanners, as described in the applicant's patent.

[0081] The invention also relates to a hydrological alert system in a geographical area having watercourses, said alert system comprising: a plurality of meteorological devices for monitoring watercourses deployed at each watercourse in said geographical area, each of said meteorological monitoring devices comprising a meteorological device according to the invention and a device for measuring the level of said watercourses; a remote meteorological monitoring server capable of communicating with the plurality of meteorological monitoring devices; characterized in that said alert system is capable of issuing a flood alert, for any one of said watercourses, based on the level of said watercourses and / or an estimate of a volume of precipitating liquid water in said geographical area obtained by the method according to the invention and a predetermined threshold, called the alert threshold.

[0082] According to this advantageous feature, the presence of a plurality of meteorological river monitoring devices deployed at each river allows continuous and real-time monitoring of meteorological and hydrological conditions. These meteorological monitoring devices provide accurate and up-to-date data on meteorological conditions and river levels, which allows for a reliable assessment of flood risk.

[0083] In addition, the use of a remote weather monitoring server allows for efficient and rapid communication between the server and the meteorological monitoring devices. This real-time communication facilitates the collection, analysis, and interpretation of meteorological and hydrological data, thus enabling the warning system to issue accurate and relevant flood warnings. The flood warning is triggered not only from the level of the rivers, but also from an estimate of the volume of liquid water precipitating in the geographical area, which reinforces the reliability and accuracy of the warning system.

[0084] The alert threshold typically takes the form of a volume of water in cubic meters, but it can also be defined in terms of height in meters, in terms of a flow rate in cubic meters per second, etc.

[0085] In one embodiment of the invention, the computing unit is a remote computing unit, in particular a computing unit of a remote computer running in a computer server.

[0086] List of Figures

[0087] Other aims, characteristics and advantages of the invention will appear on reading the following description given solely for non-limiting purposes and which refers to the appended figures in which:

[0088] [Fig. 1] is a block diagram of a pluviometric method for estimating a volume of precipitating liquid water according to one embodiment of the invention.

[0089] [Fig. 2] is a schematic view of a device for receiving and processing GNSS signals capable of implementing the method of Fig. 1 according to one embodiment of the invention.

[0090] [Fig. 3] is a schematic view of the geometric configurations defined by the propagation of GNSS signals from satellites to a meteorological monitoring device capable of implementing the method of Fig. 1 according to one embodiment of the invention.

[0091] [Fig. 4] is a schematic view of a meteorological monitoring device according to one embodiment of the invention.

[0092] [Fig. 5] is a schematic view of a hydrological warning system in a geographic area comprising a plurality of meteorological monitoring devices according to one embodiment of the invention.

[0093] Detailed description of an embodiment of the invention

[0094] In the figures, scales and proportions are not strictly respected, for the purposes of illustration and clarity.

[0095] Furthermore, identical, similar or analogous elements are designated by the same references in all figures.

[0096] Fig. 1 schematically and partially represents a block diagram of a pluviometric method for estimating a volume of precipitating liquid water V above a given geographical position G. The method begins with a preliminary step EO of positioning by a method of precise point positioning of a reception antenna 11, called reception antenna, capable of capturing GNSS signals 21 arranged at said predetermined geographical position G.

[0097] Following the preliminary step EO of positioning the receiving antenna 11, the method comprises a step E1 of receiving, by said antenna 11, at least one GNSS signal 21, called the received signal, the spectrum of which is composed of at least two distinct frequencies, in this case frequencies L1 = 1575.42 MHz and L2 = 1277.60 MHz.

[0098] The step E1 of receiving the at least one GNSS signal 21 further comprises the following preprocessing sub-steps: a sub-step E11 of reception timestamp, by a calculation unit 12 capable of receiving and processing GNSS signals of each signal received 21 by said reception antenna 11 at said geographical position G; a sub-step E1.2 of decoding, by the calculation unit 12, the transmission timestamp of said at least one received signal.

[0099] Sub-steps E11 and E1.2 make it possible to determine the travel time of the signal 21 between the satellite 20 at the time of transmission of said signal and the receiving antenna.

[0100] The step E1 of receiving the at least one GNSS signal 21 further comprises the following preprocessing sub-steps: a sub-step E1.3 of determining and correcting, by the “Klobuchar” method, the delays and / or disturbances generated by the crossing of the ionosphere of said received signal 21 from said at least two distinct frequencies making up the spectrum of said received signal 21; a sub-step E1.4 of determining, by the calculation unit 12, at least one total oblique delay of the at least one received signal 21 from said transmission and reception timestamps of said at least one received signal 21.

[0101] Subsequently, the method comprises a step E2 of determination by the calculation unit 12 of a total delay at the zenith T of said received signal 21 at said geographical position G. The total delay at the zenith T is determined from the total oblique delay of the received signal 21 using a “Mariani” fold-down function.

[0102] Finally, the method ends with a step E3 of determining, by the calculation unit 12, a volume V of precipitating liquid water above said geographical position G from only said total delay at the zenith T determined in step E2 and atmospheric data relating to said predetermined geographical position G. To do this, the calculation unit executes a machine learning algorithm in the form of a scalar predictor taking as input the total delay value at the zenith T of a GNSS signal 21; said scalar predictor being trained on a set of data relating to total delay values ​​at the zenith T of GNSS signals 21, the geographical position G and historical meteorological data.

[0103] Fig. 2 schematically represents a device 10 for receiving and processing GNSS signals 21 transmitted by a satellite 20; the device 10 is capable of implementing the method of Fig. 1 according to one embodiment of the invention. The device 10 comprises an antenna 11 for receiving GNSS signals 21 and a calculation unit 12.

[0104] Antenna 11 is positioned at geographic position G using a precise point positioning method.

[0105] Fig. 3 represents a schematic view of the geometric configurations defined by the propagation of GNSS signals 21 from satellites 20 to a meteorological monitoring device positioned at point G and capable of implementing the method of Fig. 1 according to one embodiment of the invention.

[0106] Also shown are the ionosphere 100 and the troposphere 200.

[0107] The center of the Earth is represented by the point O and the elevation angle ex of the satellite 20 is the angle formed by the tangent space T to the Earth at the point G. The direction of the zenith Z at G is also represented. Fig. 4 represents a schematic view of a meteorological device 10 for monitoring, receiving and processing GNSS signals according to an embodiment of the invention. The device 10 comprises an antenna 11 for receiving GNSS signals, a computing unit 12 capable of analyzing said GNSS signals.

[0108] The device 10 further comprises a storage memory 15 capable of recording the measurements of the volume of precipitating liquid water over time, as well as communication means 17 capable of communicating with external devices such as remote servers.

[0109] The device 10 is also provided with a rechargeable electrical power source (not shown) as well as means for supplying electricity and recharging said electrical power source in the form of solar panels 16.

[0110] Finally, the device 10 is equipped with a camera allowing the recording and determination of the surface speed of a watercourse; and a sensor 13 of the level of a watercourse, in the form of a LIDAR scanner 13.

[0111] Fig. 5 shows a schematic view of a hydrological warning system in a geographical area comprising a plurality of meteorological monitoring devices 10 according to one embodiment of the invention.

[0112] The plurality of meteorological monitoring devices 10 are positioned at a watercourse 50 and regularly measure the level of the watercourse as well as determine the value of the volume of liquid water precipitating above the location of each of said devices by the method of Fig. 1.

[0113] The devices 10 maintain communications with a remote server 60 containing regularly updated meteorological data. Said remote server 60 comprising a computing unit (not shown) and a communication unit (not shown) allowing said server to communicate with the plurality of meteorological monitoring devices 10; the computing unit of the remote server 60 is capable of executing the method of determining a volume of precipitating liquid water above each geographical position given by the geographical location of said meteorological warning devices.

[0114] The devices 10 are capable of transmitting an alert to a device 40 for transmitting flood and / or flood alerts at the level of the geographical area covered by the plurality of devices 10 based on an estimate of a volume V of liquid water precipitating in said geographical area and comparing said volume to a predetermined threshold.

[0115] The invention is not limited to the embodiments described. In particular, it will be possible to use several meteorological monitoring devices on different waterways.

[0116] It will also be possible to envisage the use of new satellite constellations and / or new frequency bands.

Claims

CLAIMS 1. A rain gauge method for estimating a volume (V) of precipitating liquid water above a predetermined geographical position (G) comprising the following steps: a step (El) of receiving, by an antenna (11), called the receiving antenna, arranged at said predetermined geographical position (G) and capable of capturing GNSS signals (21), at least one GNSS signal (21), called the received signal, the spectrum of which is composed of at least two distinct frequencies; a step (E2) of determining, by a calculation unit (12) capable of receiving and processing GNSS signals (21), a total delay at the zenith (T) of each signal received (21) by said receiving antenna (11) at said geographical position (G);a step (E3) of determining, by said calculation unit (12), a volume (V) of precipitating liquid water above said predetermined geographical position (G) from only said total delay at the zenith (T) determined in the previous step and atmospheric data relating to said predetermined geographical position (G).; 2. Method according to claim 1, characterized in that it comprises a preliminary step (EO) of positioning the receiving antenna (11) by an absolute positioning method.

3. Method according to claim 2, characterized in that the absolute positioning method is a precise point positioning method.

4. Method according to one of claims 1 to 3, characterized in that the step (El) of receiving at least one GNSS signal further comprises the following pre-processing sub-steps: a sub-step (El.l) of reception timestamp, by the calculation unit (12), of each received signal (21); a sub-step (E1.2) of decoding, by the calculation unit (12), the transmission timestamp of each received signal (21); a sub-step (El.3) of determining and correcting the delays and / or disturbances generated by the crossing of the ionosphere (100) of each received signal (21) from said distinct frequencies composing the spectrum of each received signal (21); a sub-step (El.4) of determining, by the calculation unit (12), at least one total oblique delay of each received signal (21) from said transmission and reception timestamps of this received signal (21).

5. Method according to claim 4, characterized in that the step (E2) of determining a total delay at the zenith (T) of each received signal (21) at the predetermined geographical position (G) is carried out from said total oblique delay of the received signal (21).

6. Method according to one of claims 1 to 5, characterized in that the step (E3) of determining a volume (V) of precipitating liquid water above said geographical position (G) comprises the following sub-steps: a sub-step (E3.1) of determining a dry component of said total delay at the zenith (T) of the received signal (21); a sub-step (E3.2) of determining the volume (V) of precipitating liquid water above the geographical position (G) from said total delay at the zenith (T) of the received signal (21) and the dry component of said total delay at the zenith (T) of the received signal (21).

7. Method according to claim 6, characterized in that the sub-step (E3.2) of determining the volume of precipitating liquid water above the geographical position (G) from said total delay at the zenith (T) is carried out by a machine learning algorithm.

8. Method according to claim 7, characterized in that the machine learning algorithm is a scalar predictor taking as input a total zenith delay value (T) of a GNSS signal (21); said scalar predictor being trained on a set of data relating to total zenith delay values ​​(T) of GNSS signals (21), the geographic position (G) and meteorological data.

9. Meteorological device (10) comprising a receiving antenna (11) capable of capturing GNSS signals (21) arranged at a predetermined geographical position (G) and a calculation unit (12) capable of processing the GNSS signals (21) received by said antenna (11), said meteorological device (10) being characterized in that it implements the pluviometric method for estimating a volume (V) of precipitating liquid water above a geographical position (G) according to one of claims 1 to 8.

10. A meteorological system for estimating a volume of precipitating liquid water over a geographic area comprising a plurality of meteorological devices (10) according to claim 9 distributed within said geographic area.

11. Hydrological warning system in a geographical area comprising watercourses (50), said warning system comprising: a plurality of meteorological devices (10) for monitoring watercourses (50) deployed at each watercourse (50) of said geographical area according to claim 9; a remote meteorological monitoring server (60) capable of communicating with the plurality of meteorological monitoring devices (10); characterized in that said warning system is capable of issuing a flood warning, for any one of said watercourses, from an estimate of a volume (V) of precipitating liquid water in said geographical area obtained by the method according to one of claims 1 to 8, when said estimated volume of precipitating liquid water is greater than a predetermined threshold.

Images