DETERMINATION OF LOCATION-SPECIFIC WEATHER INFORMATION FOR AGRONOMIC DECISION SUPPORT

Abstract

A method performed by at least one apparatus is inter alia disclosed, said method comprising: obtaining weather model data indicative of location-specific weather information for a first set of locations (26) on a first grid (28); obtaining an area of interest (30) associated to at least one user (32); obtaining and/or determining a second set of locations (34) based on a second grid (36) within said area of interest (30); obtaining measurement data on location-specific weather information of a measurement device associated to said at least one user located at a measurement location (38) within and/or proximate to said area of interest (30); and determining, based on at least said obtained weather model data and said obtained measurement data, location-specific weather information for said second set of locations (34) based on said second grid (36).

Claims

1. A method performed by at least one apparatus, said method comprising: obtaining weather model data indicative of location-specific weather information for a first set of locations (26) on a first grid (28); obtaining an area of interest (30) associated to at least one user (32); obtaining and/or determining a second set of locations (34) based on a second grid (36) within said area of interest (30); obtaining measurement data on location-specific weather information of a measurement device associated to said at least one user located at a measurement location (38) within and/or proximate to said area of interest (30); and determining, based on at least said obtained weather model data and said obtained measure-ment data, location-specific weather information for said second set of locations (34) based on said second grid (36).

2. The method according to claim 1, wherein said second grid (36) provides a spatial resolution being higher, in particular at least one order of magnitude higher than a spatial resolution of said first grid (28).

3. The method according to claim 1, further comprising: obtaining elevation model data indicative of a surface elevation for at least one location within and/or proximate to said area of interest (30); wherein said determining of said location-specific weather information for said second set of locations (34) is further based on said elevation model data.

4. The method according to claim 1, wherein said determining of said location-specific weather information for said second set of locations (34) comprises a spatial interpolation of said weather model data on said second set of locations (34); and/or a spatial interpolation of said measurement data on said second set of locations (34).

5. The method according to claim 1, further comprising: determining location-specific bias information based on deviations of said obtained weather model data and/or said obtained measurement data; wherein said determining of said location-specific weather information for said second set of locations (34) is further based on said location-specific bias information.

6. The method according to claim 5, further comprising: determining at least one field parameter of a statistical model based at least partially on said location-specific bias information, wherein said determining of said location-specific weather information for said second set of locations (34) is further based on said statistical model using said at least one field parameter.

7. The method according to claim 5, further comprising: carrying out at least one machine learning process based at least partially on said location-specific bias information to obtain at least one field parameter of a model for said location-specific bias information, wherein said determining of said location-specific weather information for said second set of locations (34) is further based on said model for said location-specific bias information using said at least one field parameter.

8. The method according to claim 1, wherein said determining of said location-specific weather information for said second set of locations comprises a spatial process model of said weather model data and said obtained measurement data on said second set of locations (34).

9. The method according to claim 1, further comprising: obtaining and/or determining a third set of locations (40) based on a third grid (42) within said area of interest (30); obtaining surface topography data for at least one location within and/or proximate to said area of interest (30); determining, based on said obtained surface topography data and said determined location-specific weather information for said second set of locations, location-specific weather information for said third set of locations (40) based on said third grid (42).

10. The method according to claim 9, wherein said third grid (42) provides a spatial resolution being higher, in particular at least one order of magnitude higher than a spatial resolution of said second grid (36).

11. The method according claim 9, wherein said surface topography data is based on at least one of lidar data, radar data, a crop growth model, and/or a crop growth measurement.

12. The method according claim 9, wherein said surface topography data is indicative of surface elevation and/or surface roughness.

13. The method according to claim 1, wherein said location-specific weather information for said first, second and/or third set of locations (26, 34, 40) is indicative of at least one of precipitation, wind speed, wind direction, temperature, dew point temperature, air pressure, ge-opotential, relative humidity, sunshine duration, global radiation and/or solar irradiation; and wherein said location-specific weather information for said first, second and/or third set of locations (26, 34, 40) is indicative of a current weather status and/or a weather forecast.

14. The method according to claim 1, wherein said location-specific weather information for said first, second and/or third set of locations (26, 34, 40) is indicative of at least one of near surface wind profile, near surface humidity profile, water transfer of crops, and/or near surface water balance.

15. The method according to claim 1, wherein said first grid (28) comprises a spatial resolution for locations in the order of 10 km to 100 km.

16. The method according to claim 1, wherein said second grid (36) comprises a spatial resolution for locations in the order of 500 m to 2 km.

17. The method according to claim 9, wherein said third grid (42) comprises a spatial resolution for locations in the order of 5 m to 50 m.

18. The method according to claim 1, wherein said location-specific weather information for said second and/or third set of locations (34, 40) is used in a crop growth model, a crop pathogenic risk model and/or a pest model.

19. A computer program code, the computer program code when executed by a processor causing an apparatus (69) to perform the actions of the method of claim 1.

20. A computer readable storage medium (110-116) in which computer program code according to claim 19 is stored.

21-24. (canceled)

Description

BRIEF DESCRIPTION OF THE FIGURES

[0076] FIG. 1 is a flow chart illustrating an exemplary embodiment of a method according to the invention;

[0077] FIG. 2 is a schematic illustration of the first set of locations based on the first grid, the area of interest and the second set of locations based on the second grid;

[0078] FIG. 3 is a schematic illustration of a system comprising an apparatus for performing the method according to the invention;

[0079] FIG. 4 is a block diagram of an apparatus for performing the method according to the invention;

[0080] and

[0081] FIG. 5 is a schematic illustration of examples of tangible storage media according to the present invention.

DETAILED DESCRIPTION OF THE FIGURES

[0082] FIG. 1 is a flow chart illustrating an exemplary embodiment of a method 2 according to the invention. The method 2 is performed by at least one apparatus.

[0083] In step 4, weather model data indicative of location-specific weather information for a first set of locations on a first grid is obtained. For instance, the weather model data may be data generated on basis of a global weather model and may be provided by an external service provider. The first grid may be considered coarse and may be have a resolution that is large compared to relevant scales for agronomic decision support. The resolution may for example be in the order of 10 km to 100 km, in particular 10 km or 20 km, indicating the distance between nearest neighboring locations of the first set of locations. The location-specific weather information for the first set of locations is indicative of at least one of precipitation, wind speed, wind direction, temperature, dew point temperature, air pressure, geopotential, relative humidity, sunshine duration, global radiation and/or solar irradiation.

[0084] In Step 6, an area of interest associated to at least one user is obtained. The area of interest may represent arable land such as an agricultural field associated to the user managing the agricultural field.

[0085] In step 8, a second set of locations based on a second grid within said area of interest is obtained and/or determined. For instance, the user may specify certain points of interest within the area of interest, such as locations where certain types of crops are grown. A second grid may also be determined based on a given resolution, for example, the second grid comprises a spatial resolution for locations in the order of 500 m to 2 km, in particular around 1 km.

[0086] In this embodiment, the location-specific weather information is downscaled from the first grid to the second grid by means of at least two steps. In step 10, elevation model data indicative of a surface elevation for at least one location within and/or proximate to said area of interest is obtained. The surface elevation model is, for instance, a high-resolution DEM. The DEM may be representative for the shape of mountains, hills, lakes, water streams or valleys. The DEM may allow for a regionalization step 12, which for example takes into account regional effects such as mountain and shadowing effects, e.g. valley winds, and altitude effects, e.g. orographic precipitation.

[0087] Further, in step 14 measurement data on location-specific weather information of a measurement device associated to said at least one user located at a measurement location within and/or proximate to said area of interest is obtained. For example, the user may have access to at least one local weather station in the area of interest. The measurement data may in this regard indicate the ground truth, e.g. the actual local weather conditions at the measurement location, and be used to correct, verify, and/or select weather model data. In step 16, location-specific bias information based on deviations of said obtained weather model data and said obtained measurement data is obtained. The location-specific bias information may allow for a bias correction of the weather model data to obtain more precise location-specific weather information on the fine second grid.

[0088] In step 17, location-specific weather information for said second set of locations based on said second grid is determined. The determining is based on at least said obtained weather model data and said obtained measurement data (in particular by means of the location specific bias information and a bias correction in step 16), as well as on said elevation model data (by means of the regionalization step 12). The determining step 17 of said location-specific weather information for said second set of locations comprises a spatial interpolation of said weather model data on said second set of locations. In particular, at least one field parameter of a statistical model based at least partially on said location-specific bias information is determined, wherein determining step 17 is further based on the statistical model using said at least one field parameter. At least one machine learning process, e.g. involving an artificial neuronal net with a deep learning routine, based at least partially on said location-specific bias information may be utilized to obtain the at least one field parameter. As an alternative or in addition, a spatial process model of said weather model data and said obtained measurement data may be used.

[0089] The location-specific weather information for said second set of locations determined in step 17 may be used in a crop growth model, a crop pathogenic risk model and/or a pest model, for cases where the resolution of the second grid is already sufficient for such a model. To obtain location-specific weather information with an even higher and in particular hyper-local resolution, this embodiment of the method involves obtaining and/or determining a third set of locations based on a third grid within said area of interest in step 18. The third grid may for example be obtained from the user or a resolution is chosen or predetermined, for example a spatial resolution for locations in the order of 5 m to 50 m, in particular around 10 m.

[0090] In step 20, surface topography data for at least one location within and/or proximate to said area of interest is obtained. Said surface topography data is indicative of surface elevation and/or surface roughness and has in particular a higher resolution than the surface elevation data. The surface topography data is based on lidar data, radar data and a crop growth model. Lidar and radar data may provide information relating to surface elevation and surface roughness with a high resolution. A crop growth model may yield information on size and growth stage of crops, further being representative for surface elevation and surface roughness. These crop-specific effects are taken into account in step 22.

[0091] In step 24, based on said obtained surface topography data and said determined location-specific weather information for said second set of locations, location-specific weather information for said third set of locations based on said third grid is determined. Weather information with hyperlocal resolution may therefore be obtained by the method.

[0092] The location-specific weather information for said second and third set of locations may be indicative of at least some of the characteristics that are also represented by the weather model data, in particular one of precipitation, wind speed, wind direction, temperature, dew point temperature, air pressure, geopotential, relative humidity, sunshine duration, global radiation and/or solar irradiation. The location-specific weather information for said first, second and/or third set of locations may be indicative of a current weather status and/or a weather forecast.

[0093] Further, in particular the downscaling on the third grid may yield additional characteristics of the location-specific weather information. For instance, incorporating crop-specific effects in step 22 may allow for the location-specific weather information for said third set of locations to be indicative of at least one of near surface wind profile, near surface humidity profile, water transfer of crops, and/or near surface water balance.

[0094] The location-specific weather information for said second and/or third set of locations may be provided to the user for agronomic decision support. In particular, the location-specific weather information is used in a crop growth model, a crop pathogenic risk model and/or a pest model such as numerical computer models. The crop growth model is in particular related to the one used for obtaining surface topography data in step 20.

[0095] FIG. 2 is a schematic illustration of the first set of locations 26 based on the first grid 28, which represents global weather model data on a coarse grid with 10 km resolution, for example. The area of interest 30 associated to the at least one user 32 is regional and may represent arable land such as agricultural fields associated with the user. The second set of locations 34 is based on the second grid 36 and represents a regional grid with 1 km resolution, for example. The measurement device associated to the user is located at a measurement location 38.

[0096] FIG. 2 further illustrates the third set of locations 40 based on the third grid 42. The third grid 42 may for example relate to a part of the area of interest 30 such as a specific agricultural field. The third grid 42 may be a hyperlocal grid with a resolution of 10 m and can be used to characterize weather conditions on hyperlocal scales that are important for agronomic decision support.

[0097] FIG. 3 is a schematic illustration of a system 46 comprising an apparatus for performing the method according to the invention. Measurement devices are associated with the user 32, which may be a mobile device 48 such as a handheld sensor or a sensor associated with a vehicle such as a tractor, as well as a local weather station 50. Mobile device 48 and local weather station 50 are in communication with a network 52, e.g. a local network or the internet. Mobile device 48 and local weather station 50 may further communicate with field data generator 54 that uses the measurement data to derive further information on the agricultural field.

[0098] An external weather service provider 56 may for instance provide weather model data based on a coarse first grid, and also communicate with the network 52.

[0099] Agronomic decision support system 58 comprises a communication layer 60 that communicates with network 52 and may serve to obtain weather model data, an area of interest and measurement data.

[0100] Field weather generator 62 is configured as means for determining, based on at least obtained weather model data and obtained measurement data, location-specific weather information for a second set of locations based on a second grid and/or for a third set of locations based on a third grid.

[0101] Data management means 64 may have access to data base 66 that stores weather data, such as location-specific weather information on the first, second and third grid, output from the field data generator 54 as well as model data. The weather data may for instance be used to further refine statistical models and for machine learning processes used in the method to obtain the location-specific weather information on the second and third grid. The model data may comprise parameters of statistical models and machine learning processes.

[0102] Agronomic decision support system 58 further comprises a presentation/visualization layer 68 that is configured to provide output to the user, e.g. location-specific weather information on the second and third grid.

[0103] FIG. 4 is a block diagram of an apparatus 69 for performing the method according to the invention. Apparatus 69 for instance is or forms a part (e.g. as a module) of a server, and may interact with clients and in particular mobile devices.

[0104] Apparatus 69 comprises a processor 70. Processor 70 may represent a single processor or two or more processors, which are for instance at least partially coupled, for instance via a bus. Processor 70 executes a program code stored in program memory 71 (for instance program code causing apparatus 69 to perform the example embodiment of a method according to the invention, when executed on processor 70). Processor 70 further interfaces with a main memory 72 (for instance acting as a working memory) and a mass storage 74, which may for instance store measurement data, weather model data and elevation model data. The measurement data, weather model data and elevation model data may for instance be stored in a database. Further, the apparatus 69 may also be able to store a model (e.g. respective model parameters of the model), e.g. in a mass storage 74. A communication interface 75 allows for a communication with a network.

[0105] FIG. 5 illustrates examples of tangible storage media that may for instance be used to implement program memory 71 of FIG. 4. To this end, FIG. 5 displays a flash memory 110, which may for instance be soldered or bonded to a printed circuit board, a solid-state drive 111 comprising a plurality of memory chips (e.g. Flash memory chips), a magnetic hard drive 112, a Secure Digital (SD) card 113, a Universal Serial Bus (USB) memory stick 114, an optical storage medium 115 (such as for instance a CD-ROM or DVD) and a magnetic storage medium 116.

[0106] Any of the processors mentioned in this text, in particular but not limited to processors 70 of FIG. 4, could be a processor of any suitable type. Any processor may comprise but is not limited to one or more microprocessors, one or more processor(s) with accompanying digital signal processor(s), one or more processor(s) without accompanying digital signal processor(s), one or more special-purpose computer chips, one or more field-programmable gate arrays (FPGAS), one or more controllers, one or more application-specific integrated circuits (ASICS), or one or more computer(s). The relevant structure/hardware has been programmed in such a way to carry out the described function.

[0107] Moreover, any of the actions described or illustrated herein may be implemented using executable instructions in a general-purpose or special-purpose processor and stored on a computer-readable storage medium (e.g., disk, memory, or the like) to be executed by such a processor.

[0108] References to ‘computer-readable storage medium’ should be understood to encompass specialized circuits such as FPGAs, ASICs, signal processing devices, and other devices.

[0109] It will be understood that all presented embodiments are only exemplary, and that any feature presented for a particular exemplary embodiment may be used with any aspect of the invention on its own or in combination with any feature presented for the same or another particular exemplary embodiment and/or in combination with any other feature not mentioned. It will further be understood that any feature presented for an example embodiment in a particular category may also be used in a corresponding manner in an example embodiment of any other category.