2015 Site Updates

Some links and images are no longer available in our older site updates. Additionally, some information is no longer applicable.

For the latest site updates, click here.

Historic HDOB messages in the recon system have been rewritten and now have the storm slice feature available for them.

The next major step is to have Google Maps mapping available for all messages. Our previous recon system had simple Google Maps mapping. We have not yet written the code to bring that to the current system. Since the Google Earth browser plugin will be discontinued at the end of this year, we hope to have some Google Maps mapping available by then, eventually having all recon data available like it is currently in the Google Earth plugin, only it would be in 2D in Google Maps rather than in the 3D environment of the Google Earth plugin. (Data viewable in Google Earth, the program, will of course still be available.)

We hope to have all the Google Maps mapping completed by early 2016 at the latest.

I had to make updates to the recon system after the NOAA P-3 radar data starting becoming available in other file formats. Those updates are now live, however they are untested in real time. I have temporarily turned off the live mirror we had for radar data. The recon system should add radar data automatically into the recon system's archive, including into Google Earth files. If it can't find where the data should be added in real time, it will instead add the data here where it will not be associated with any mission data. If there are significant issues I will restore the temporary live mirror we had.

The recon system is now checking in real time for airborne Doppler radar data from NOAA P-3's. It will appear in Google Earth mapping. This has yet to be tested in real time with incoming imagery, so there are likely issues that will arise. The temporary live mirror we have setup will remain for now. Skew-t images came in for a few missions earlier this year and our recon system does not handle that data as of now. We will add some radar imagery from 2014 at a later date. The administrative system to add radar data has not yet been constructed, so adding the data must be done manually which takes time.

90L's name did not appear in the model system correctly. There was an unexpected number of commas in the best track file which confused our system. The system is now coded to allow the extra comma.

There is now a new look to AXBTs in our Google Earth recon system. The actual SST temperature, floored to the nearest whole number, will be displayed. That icon will also contain a color coded to that number. If the SST is not available, the default AXBT icon will be displayed. All older AXBT mapping has been updated with the new icons, both in the normal mission archive and the AXBT archive.

A few minor bug fixes have been applied to the live aircraft pages.

All the links on the pages on our site were checked to make sure they still work. Our Google Earth overlay file was reviewed as well.

An early version of the live aircraft pages are now available. There are further improvements that will eventually be made.

This morning, across all of our Tropical Globe sites, we made significant updates to our cascading style sheets (CSS). We have a lot more customizing to do. The changes to the CSS will allow us to more easily distribute our recon system and model system in the future with just the CSS needed to style each system, rather than including unnecessary CSS that styles other aspects of our site. If you notice anything that doesn't look right or doesn't work right, please let us know.

The live aircraft pages will continue to be developed offline this week. As parts of this feature are completed they will be added to the operational recon system.

Due to the March 30th update there was an issue with a non-tasked mission that ran past midnight Zulu on April 10th. After offline testing, the problem that occurred should now be resolved for future missions. This was the first live test of a non-tasked mission running past midnight Zulu since that section of code was rewritten.

I am continuing to work on the design of the live aircraft pages, as well as how product pages will work.

The issue where mission details on the front page were unavailable for a period of about 10 minutes around midnight Zulu continued to be a problem. Over the past few months a significant amount of testing was needed after a lot of code was rewritten on the processing side of the recon system. The system was updated with the new code today. The code that was rewritten included sections that handle moving non tasked data to a new directory when a non tasked mission extends past midnight Zulu and also missions that cross into another basin. These are very complicated sections of code and therefore the system may not update properly until it is tested more thoroughly in the live operational system.

I am currently working offline on the visual design of the live aircraft pages. I hope to have those done by the start of the hurricane season.

There was an issue with data being added more often than needed from the NHC's recon archive. A side effect was displaying older missions from the day on the front page and in the Google Earth file. I think this has now been fixed. There has also been an issue where mission details on the front page are unavailable for a period of about 10 minutes around midnight Zulu for some missions. (or until the next available mission data for that mission is available) This issue might be related to the other issue or may have been fixed by another change made. I will need to keep an eye on it carefully to see if the problem is still occurring. Meanwhile, work continues on further updates to the recon system. I hope for these additional updates, which are being constructed and tested offline, will be available by the start of the hurricane season.