Abstract
A methodology to perform large-scale trending of radar data, is discussed. Trends are calculated in Lagrangian space, based on motion vectors at forecast time, using a weighted linear fit of a mean value within a given-radius circle over a given period. It was observed that large-scale trending of a storm line is effective. The calculated trends are very sensitive to inaccuracies and changes in the motion vectors used. It was observed that decrease in forecast area coverage and bias between the trended and nontrended forecast results is greatest at low probability levels.
| Original language | English |
|---|---|
| Pages | 873-878 |
| Number of pages | 6 |
| State | Published - 2004 |
| Event | 11th Conference on Aviation, Range, and Aerospace Meterology - Hyannis, MA, United States Duration: Oct 4 2004 → Oct 8 2004 |
Conference
| Conference | 11th Conference on Aviation, Range, and Aerospace Meterology |
|---|---|
| Country/Territory | United States |
| City | Hyannis, MA |
| Period | 10/4/04 → 10/8/04 |
Fingerprint
Dive into the research topics of 'Large-scale trending of radar data'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver