Method and apparatus for short-term prediction of convective weather

• US 7,062,066 B2
• Filed: 05/11/2002
• Issued: 06/13/2006
• Est. Priority Date: 02/19/2002
• Status: Expired due to Term

• Current Assignee: Massachusetts Institute of Technology, Biovitrum AB (Swedish Orphan Biovitrum AB (publ))
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Johnson, Richard, Wolfson, Marilyn, Boldi, Robert, Dupree, William, Wilson, Carol, Delanoy, Richard L., Theriault, Kim E., Forman, Barbara, Hallowell, Robert G.
• Primary Examiner(s): Mancuso, Joseph
• Assistant Examiner(s): KRONENTHAL, CRAIG W
• Application Number:
  US10/144,667
• Publication Number:
  US 20030156734A1
• Time in Patent Office: 1,494 Days
• Field of Search: 382/100, 382/159, 382/224, 382/225, 382/103, 382/107, 342/26, 342/460, 702/3
• US Class Current: 382/100
• CPC Class Codes: G01S 13/86 Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
• CPC Class Codes: G01S 13/951 ground based
• CPC Class Codes: G01W 1/10 Devices for predicting weather conditions computers per se G06; display devices G09
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

A method and apparatus for forecasting the likely occurrence of convective weather events, such as thunderstorms. An image filter is used to identify areas of interest within a meteorological image that are likely to contain convective weather. The image filter and an image difference processor identify sub-image regions within the meteorological image that are likely to experience a growth and/or decay of weather events. The classification filter classifies sub-image regions within the meteorological image into a number of predetermined storm categories. The meteorological images are filtered using matched filters, features within the filtered images are tracked, and the resulting track vectors are combined according to the storm classification. The meteorological image, interest image, growth/decay image, classification image, and combined vectors are processed to produce the short-term forecast.

Claims

1. A method for developing a short-term storm forecast comprising the steps:

(a) receiving a first and a second meteorological image at a first and a second time, respectively, each meteorological image comprising information indicative of a meteorological indicator in relation to a predetermined geographical area;

(b) processing each meteorological image to determine a sub-image region as being within at least one of a plurality of predetermined storm classifications;

(c) generating a combined tracking vector field comprising at least one tracking vector for each determined sub-image region, the combined tracking vector field indicative of relative motion; and

(d) generating a short-term forecast by advecting a meteorological image according to the combined tracking vector field.

2. The method of claim 1, wherein step (b) comprises filtering each meteorological image with at least one matched image filter.

3. The method of claim 2, wherein the matched image filter comprises an approximate ellipse.

4. The method of claim 3, wherein the approximate ellipse is representative of an elliptical geographical area having an aspect ratio greater than approximately 4 and a major axis greater than approximately 20 km.

5. The method of claim 3, wherein the approximate ellipse is representative of an elliptical geographical area having minor and major axes of 13 km and 69 km, respectively.

6. The method of claim 2, wherein the matched image filter comprises an approximate circle.

7. The method of claim 6, wherein the approximate circle is representative of a geographical area having a diameter less than approximately 20 km.

8. The method of claim 6, wherein the approximate circle is representative of a circular geographical area having a 13 km diameter.

9. The method of claim 1, wherein the predetermined storm classifications are selected from the group consisting of:

a line storm;

a large cell storm;

a small-cell storm; and

stratiform convection.

10. The method of claim 1, wherein step (b) comprises the steps:

(b-1) detecting within each meteorological image corresponding sub-image regions of such image indicative of line storms;

(b-2) determining from each meteorological image a respective variability image indicative of the variability of the meteorological indicator; and

(b-3) identifying a sub-image region of such image indicative of airmass storms responsive to the determined line-storm portions of such image and the determined variability image.

11. The method of claim 1, wherein step (c) comprises the steps:

(c-1) filtering each meteorological image to determine a large-scale feature image;

(c-2) filtering each meteorological image to determine a small-scale feature image;

(c-3) determining a large-scale feature tracking vector field based on the large-scale feature image;

(c-4) determining a small-scale feature tracking vector field based on the small-scale feature image; and

(c-5) generating the combine tracking vector field using the classified sub-image regions and the large-scale and small-scale tracking vector fields.

12. The method of claim 1, further comprising the step of filtering the combined tracking vector field to reduce spurious tracking vectors.

13. The method of claim 12, wherein the step of filtering comprises:

(a) determining a global correlation vector representative of the general movement of the meteorological indicator of the received image;

(b) removing from each line region and each stratiform region substantially all tracking vectors having a direction difference measured in relation to the global correlation vector of greater than a predetermined first angular threshold;

(c) removing from each large airmass region substantially all tracking vectors wherein the global vector is greater than a predetermined first tracking-velocity threshold and the direction difference between the tracking vector and a global correlation vector is greater than a predetermined second angular threshold;

(d) correlating the tracking vectors within each small airmass region; and

(e) replacing the small airmass region tracking vectors with the correlated tracking vector.

14. The method of claim 13, wherein the predetermined first angular threshold is approximately between 45 degrees and 90 degrees.

15. The method of claim 13, wherein the predetermined first angular threshold is approximately 70 degrees.

16. The method of claim 13,wherein the predetermined first tracking-velocity threshold is approximately 10 meters per second.

17. The method of claim 13,wherein the predetermined second angular threshold is approximately between 90 degrees and 180 degrees.

18. The method of claim 13, wherein the predetermined second angular threshold is approximately 135 degrees.

19. The method of claim 1 wherein step (d) comprises the steps:

(d-1) determining a growth/decay image indicative of the relative growth;

(d-2) trending a meteorological image using the growth/decay image; and

(d-3) advecting the trended meteorological image using the combined tracking vector field.

20. An apparatus for developing a short-term storm forecast comprising:

a receiver receiving a first and second meteorological image at a first and a second time, respectively, each meteorological image comprising information indicative of a meteorological indicator in relation to a predetermined geographical area;

a classification processor determining at least two sub-image regions, each sub-image region being associated with a different predetermined storm classifications;

a tracker determining a combined tracking vector field comprising at least one tracking vector for each determined sub-image region, the combined tracking vector field indicative of relative motion;

an image filter filtering the combined tracking vector field to reduce spurious tracking vectors; and

a forecast processor generating a short-term forecast by advecting a meteorological image according to the combined tracking vector field.

21. The apparatus of claim 20, further comprising a growth/decay processor generating a growth/decay image, wherein the forecast processor generates a short-term forecast by advecting a meteorological image according to the combined tracking vector field and the growth/decay image.

22. The apparatus of claim 20, further comprising a scale-selection filter to produce scale-specific images to be tracked.

23. An apparatus for developing a short-term storm forecast comprising:

means for receiving a first and second meteorological image at a first and a second time, respectively, each meteorological image comprising information indicative of a meteorological indicator in relation to a predetermined geographical area;

a means for processing each meteorological image to determine at least two sub-image regions, each sub-image region being associated with a different predetermined storm classifications;

means for generating a combined tracking vector field comprising at least one tracking vector for each determined sub-image region, the combined tracking vector field indicative of relative motion;

means for filtering the combined tracking vector field to reduce spurious tracking vectors; and

means for generating a short-term forecast by advecting a meteorological image according to the combined tracking vector field.

24. The apparatus of claim 23, further comprising a means for determining a growth/decay image, wherein the forecasting means generates a short-term forecast by advecting a meteorological image according to the combined tracking vector field and the growth/decay image.

25. The apparatus of claim 23 wherein the means for generating the combined tracking vector field comprises a large-scale filter means and a small-scale filter means.