TC-OBS Database | Known Issues

Known Issues for the Tropical Cyclone Observations-Based Structure (TC-OBS) Database

The TC-OBS Database uses objective methods and a large quanitity of aircraft observations in order to construct the observations-based parameter estimates. Despite extensive quality control measures, a small handful of storms have some issues. This page serves to document known quality issues with the database, and will be updated as any additional issues are brought to our attention. We hope to address each issue in a follow-on project.

Issue #1: Bad surface wind observations from the Stepped Frequency Microwave Radiometer (SFMR) cause spikes for the RMW parameter

Description of the issue

The SFMR surface winds provided in the HRD flight level data files appear to have some erroneously large values for certaint storms. While considerable effort has been made to filter these bad values out during the creation of the FLIGHT+ Dataset, it appears that a few bad SFMR surface wind speeds cause some spurious spikes in the TC-OBS intensity, RMW, and wind radii parameters for a few cyclones that occurred after 2008. Brad Klotz (HRD/CIMAS) has recently reprocessed the SFMR winds, but I will need to obtain some additional funding in order to create a new version of the FLIGHT+ Dataset and then rebuild the TC-OBS Database.

Storms affected:

Issue #2: Widely spaced observations may lead to strange oscillations for the wind radii parameters

Description of the issue

For at least one storm, it appears widely spaced observations (or some other issue) can lead to strange oscillations in the observationally-estimated wind radii.

Storms affected:

  • Hurricane Frances (2004)     Affected parameters: wind radii.

Issue #3: Quadrant wind radii (size) all have the same uncertainty bounds

Description of the issue

This issue was reported by Suz Tolwinski-Ward. It's possible that the background uncertainties are overwriting the time-dependent uncertainty information for the various quadrant wind radii.

Storms affected:

  • This issue likely affects all storms.

Additional notes

While the following is not a data quality issue, users should be aware that most cyclones prior to 2004 do not have observational refinement for the wind radii parameter. This is because the Air Force data files do not include the flight level pressure prior to 2004. Without the flight level pressure, it is not possible to reduce the high resolution flight level data to a surface equivalent wind speed and estimate the wind radii. In the future, it should be possible to reconstruct the flight level pressure from other data parameters such as pressure altitude and geopotential height. We hope to do this in a follow-on project.


This page was last updated 21 January 2016 by Jonathan Vigh.