|
|
Boulder, Colorado - August 11-12, 2005
|
Comments and Questions for CLASS
- Include an option to gzip the files. That would reduce the
network traffic by 50% at the cost of some cpu. (The NASA DAAC's provide this.)
- To improve documentation I have downloaded a couple of days of GOES data. I am able to read the data fine, but I have not found documentation on the line prefixes or the navigation headers.
- I like the fact that you provide the ftp address to retrieve
the data upon order. I can then create a batch script to get the
data and schedule it for late in the day. That is a big improvement
over the NASA DAAC's.
- I am not sure how one would set up a system to retrieve data every day automatically. That might be a burden on the delivery system, but, I am sure there are users who would like that facility.
- There are some purposes (e.g., in assimilative modeling), that could be running with real-time data on a much faster cadence than daily (e.g., 5 minutes), and that would want an automatic data feed from CLASS at that cadence.
- We can select or deselect all images. It would be nice to be able to select or deselect a selected block of images using the standard shift-select paradigm.
- It is not at all clear what the "order query" function (button) is supposed to do. Is it supposed to return the status or numbers of my queries or the whole systems. If it is my queries, then false values are being returned.
- It would be really nice to see the search box overlayed on the thumbnail images that are returned (at least for AVHRR searches). As it stands now, the user has to use too much imagination to go from the boxes displayed on the map to the actual image.
- This is more of a wish item. It would be fantastic if the AVHRR granules had a percentage cloud cover associated with them. It would be even nicer if the results could be filtered using this.
- The search screen is a separate window from the search result screen. If the search window is inadvertently closed (as I have done many times) and the user has not saved the search parameters, it is not possible to retrieve the search window with the current parameters. Or at least I don't know how to do it.
- The file names of the granules that are placed on the ftp site for downloading have no helpful information to the user. Helpful information would be satellite number, date and time. This is actually a pretty important issue when a lot of data is ordered which spans a long time period.
- Need calibration when requesting NetCDF. The CLASS system can supply GOES imager data in the netcdf format. This is good. But, it seems that there are no options for the UNITS of the data. So, a user is left with "raw" values that need to be divided by 32 and then calibrated. Yet, I don't see how a user is supposed to know this. Are there plans to offer RAD or TBB units as well? Until then, I don't see this data as being useful to any user that needs quantitative units like radiance or brightness temperatures.
- What are the plans with file names? As you may know, I requested about 18 months ago that a gif file end with a gif extension and that the netcdf end with a nc extension. I had hoped that this was to be in the recent June updates. Also, I don't think the file name differentiates between imager and sounder data. Right now the jpg image ends sometimes with .L6968551! I note that you now supply some meta data. Given this is an ASCII file, do you plan to amend a "txt" extension?
- It says on the web page that you recently had an upgrade on June 23rd, see http://www.class.noaa.gov/nsaa/products/welcome. Is it your plans to post a summary of the changes? (And/or what is coming in the next version?)
- I find it confusing that in the first email after ordering data that a "tar" file is referred to. But, I think it's only an Data set name: goes12.2005.145.114514.tar? Also, any "comments" that I added (say to remember the file type) are only in the first email (verification) and not the second (notification) email.
- I note that one can request various data formats in an order. Can one request 2 types? For example, maybe a gif and netcdf? That way an user can check they are reading the data properly.
- While I understand that most of the GOES users deal with the imager and not the sounder, it looks like one still can only get separate for each band mcidas areas. I'm not sure of the status of this request. Almost all the quantitative applications from the GOES Sounder need one multi-band image. Any plans to allow for a multiple band (imager and sounder) option?
- Who officially saves the important, but non-operational GOES data? For example, the month or so of GOES-12 data as it moved to replace GOES-8? Or the GOES-11 data from this summer or the GOES-N data during its check out?
- The beauty of GOES is the ability to loop images. As you know, they are literally 100s of web sites where one can loop realtime GOES images. Wouldn't it be great of CLASS could build a customized loop for the user with historical data! There are many options for formats (animated gif, avi, etc), my suggestion would be for a web-based java applet (developed by Tom W.) to be used http://www.ssec.wisc.edu/data/geo/east/latest_east_ir4_conus4j.html. The user could download the images if they want after looking at the loop. This way someone could look at a loop from GOES for a storm that they were affected by the previous week on the camping trip, etc.
- When one asks for a WV image as a gif or jpg, it's not enhanced, but all washed out. Why not put a default WV enhancement on?
http://www.ssec.wisc.edu/data/geo/east/latest_east_wv_conus.jpg
- There should be a gray bar with units (TEMP) labeled option when getting gif/jpg files.
- There are other ideas (thumbnails, correlated searches, etc), but I'm not sure if it's worth the time to list them all until some of the core issues (non-calibrated data, only single bands, strange file extension names) are addressed.
|
| |
|