Method for determining storm predictability

• US 6,340,946 B1
• Filed: 08/03/2000
• Issued: 01/22/2002
• Est. Priority Date: 08/03/2000
• Status: Expired due to Term

• Current Assignee: Massachusetts Institute of Technology, University Corporation For Atmospheric Research
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Wilson, James, Moore, Michael, Hallowell, Robert, Wolfson, Marilyn, Forman, Barbara, Henry, Sandra
• Primary Examiner(s): Gregory, Bernarr E.
• Application Number:
  US09/631,157
• Time in Patent Office: 537 Days
• Field of Search: 342/26, 342/27, 342/28, 342/175-186, 342/192-197, 342/361-366, 382/100, 382/103, 382/162, 382/165, 382/168, 382/170, 382/181, 702/1, 702/2, 702/3, 702/4, 706/928, 706/930, 706/931
• US Class Current: 342/26.00R
• CPC Class Codes: G01S 13/953 mounted on aircraft
• 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 determining the predictability of an element in a weather radar image. An image filter approximating the envelope of the organized storm radar image is applied to a pixel in a received weather radar image to generate a processed pixel value. A variability value is determined from the variation in the pixel values of the neighboring pixels which lie within the image filter. The predictability is generated from the processed pixel value and the variability. Pixels having high processed pixel values and low variabilities typically correspond to pixels within a strong organized storm and, therefore, are more predictable. Pixels having low processed pixel values and high variabilities, such as pixels representative of airmass storms, generally have lower predictabilities.

Claims

1. A method for determining predictability of a selected element within a weather radar image comprising:

receiving the selected element and neighboring elements for the weather radar image;

processing the selected element and the neighboring elements to generate a processed selected element value;

determining a variability from the selected element and the neighboring elements; and

generating the predictability of the selected element from the processed selected element value and the variability.

2. The method of claim 1 wherein the weather radar image is a reflectivity image.

3. The method of claim 1 wherein the weather radar image is derived from weather radar image data.

4. The method of claim 3 wherein the weather radar image is a vertical integrated liquid water (VIL) image.

5. The method of claim 3 wherein the weather radar image is the autocorrelation of another weather radar image.

6. The method of claim 3 wherein the weather radar image is the correlation of two other weather radar images, each of the two other weather radar images representative of different times.

7. The method of claim 1 wherein the elements are pixels.

8. The method of claim 1 wherein the step of processing comprises filtering the selected element and the neighboring elements.

9. The method of claim 8 wherein the step of filtering further comprises rotating a filter in coordinate space and filtering the selected element and the neighboring elements.

10. The method of claim 8 wherein the step of filtering further comprises iteratively rotating the filter in coordinate space and, for each rotation, filtering the selected element and the neighboring elements to generate a processed selected element value.

11. The method of claim 10 wherein the step of rotating further comprises iteratively rotating the filter through 180 degrees.

12. The method of claim 10 further comprising the step of determining a maximum from the plurality of processed selected element values.

13. The method of claim 1 wherein the variability is the standard deviation of the selected element and the neighboring elements.

14. The method of claim 1 wherein the variability is the variance of the selected element and the neighboring elements.

15. The method of claim 1 wherein the predictability is inversely proportional to the variability.

16. The method of claim 1 wherein the predictability is proportional to a power of the processed element value.

17. The method of claim 1 wherein the predictability is proportional to the square of the processed element value.

18. The method of claim 1 further comprising the step of displaying the predictability.

19. The method of claim 18 further comprising displaying the predictability on a graphical display.

20. The method of claim 19 further comprising displaying a color-coded representation of the predictability.

21. A method for determining predictability of a selected element within a weather radar image comprising:

receiving the selected element and neighboring elements for the weather radar image;

filtering the selected element and the neighboring elements to generate a set of filtered element values and a selected filtered element value;

incrementally rotating the filter in coordinate space to generate a rotated filter and, for each rotation, filtering the selected element and the neighboring elements to generate at least one additional set of filtered element values and at least one additional selected filtered element value;

determining the maximum of the selected filtered element values;

determining a variability from the sets of filtered element values that were generated from the rotated filter which generated the maximum selected filtered element value; and

generating the predictability of the selected element from the maximum selected filtered element value and the variability.

22. An apparatus for determining a characteristic of a selected weather radar image element, comprising:

an input module for receiving the element and neighboring elements within a weather radar image;

a prediction module for determining the characteristic, comprising;

a rotation module for iteratively rotating a filter in coordinate space;

a filter module for applying each rotated filter to the selected element and the neighboring elements to generate a plurality of selected filtered element values and a plurality of sets of filtered element values; and

a processor module for determining the characteristic from the plurality of selected filtered element values and the plurality of sets of filtered element values; and

an output module for displaying the characteristic.

23. The apparatus of claim 22 wherein the characteristic is a predictability value.

24. The apparatus of claim 22 wherein the element is a pixel.

25. The apparatus of claim 22 wherein the output module is a graphical display.