Archived News

17 June 2008

The IMaRS facility is once again receiving MODIS Direct-Broadcast data from both the Aqua and Terra satellites. The Terra MODIS Direct-Broadcasts resumed on 13 June, and the transmissions have been sucessfully received at IMaRS for the last several days. Our thanks to those at NASA who determined what the problem was and restored Terra's X-band Direct-Broadcast transmissions.

12 June 2008

The IMaRS facility continues to receive MODIS Direct-Broadcast data from the Aqua satellite, but the Terra satellite's Direct-Broadcast transmissions are still suspended until problems with Terra's power system are resolved. It may be a week or longer before there is a resumption of Terra Direct-Broadcasts.
The AQUADirect-Broadcast transmissions continue to be monitored, and the processing of the data continues as usual. The Look-Up-Tables used by IMaRS for processing the AQUA data have been updated to version 5.0.35.3a.

02 June 2008

The IMaRS facility continues to receive MODIS Direct-Broadcast data from the Aqua satellite, but the Terra satellite's Direct-Broadcast transmissions have been temporarily suspended and the Terra imagery isn't appearing on the IMaRS MODIS website. Terra's X-band Direct-Broadcasts were suspended on June 1st due to a power problem within the satellite. The transmissions will not resume until the Direct Broadcast transmitter is reconfigured (this may take several weeks).
The Direct-Broadcast transmissions continue to be monitored, and the processing of the AQUA Direct-Broadcast data continues as usual.

20 May 2008

Today, the MODIS Look-Up-Tables on IMaRS for the Terra sensor were updated to version 5.0.40.5a, and the Look-Up-Tables for the Aqua sensor were updated to version V5.0.35.2b.

22 April 2008

Today, the MODIS Look-Up-Tables on IMaRS for the Terra sensor were updated to version 5.0.40.4a, and the Look-Up-Tables for the Aqua sensor were updated to version V5.0.35.2a.
Several new production areas have been added to the IMaRS Direct-Broadcast MODIS web site recently. High Resolution (250m) color imagery is now available for areas near Cape Cod (ccod); Long Island, NY (longIsland); and the leeward Islands of the Lesser Antilles (leeIslesLA).

18 March 2008

Several new production areas have been added to the IMaRS Direct-Broadcast MODIS web site recently. High Resolution (250m) color imagery is now available for areas near Cape Cod (ccod); Long Island, NY (longIsland); and the leeward Islands of the Lesser Antilles (leeIslesLA). Links to these areas are listed on the IMaRS MODIS home page (probably the page you are viewing). In the near future there may be some modification of these areas and I anticipate adding an additional area with 1km MODIS imagery for the North West Atlantic region.
Today, the MODIS Look-Up-Tables for the Terra sensor were updated on IMaRS to version 5.0.40.3a.

27 February 2008

NASA reports that the Terra satellite was instructed to conduct an Inclination Maneuver from approximately 1400 to 1719z today. While the IMaRS antenna system was able to track and record data from the 1600 pass of Terra, the data could not be completely processed to level3 or remapped to the standard projections for display on the IMaRS web pages.

25 February 2008

There was a delay in the complete processing of 3 Terra passes from yesterday and this morning because of a 'bad' file used for geolocation. This morning the file was replaced and processing delays should be returning to their normal duration.

14 February 2008

A computer hardware problem this morning resulted in processing delays for some of today's MODIS data. The most vexing part of the problem appears to have past, and the IMaRS processing system is operating again.

04 February 2008

It was reported that on this (the IMaRS MODIS Direct Broadcast) web page, the links to data from 2004 and 2005 were not always displayed. The problem has been fixed and the links to the MODIS imagery from 2001 to the present should now be displayed correctly.

25 January 2008 am

One of the MODIS data processing components of IMaRS system has failed this morning. Data from the MODIS passes is being received and archived, but the system will not be able to resume MODIS processing until the problem is repaired. I hope that the repair will be put in place today.

25 January 2008 pm

The problem that the IMaRS MODIS processing system had this morning has been corrected. The posting of MODIS data received at IMaRS this morning was delayed, but is now being posted to the web page.

07 January 2008

The update of the IMaRS MODIS-processing software is not yet complete. The quality of the IMaRS Direct-Broadcast imagery has improved, and a further improvement (especially for the Terra chlorophyll imagery) is anticipated with the completion of IMaRS updates. The remaining updates should not produce as dramatic a change as that which resulted from December's update.

28 December 2007

Yesterday, the IMaRS MODIS-processing software was updated (Thank you Judd), and initial tests show a significant improvement in the quality of the IMaRS Direct-Broadcast imagery, especially the Terra chlorophyll imagery. The majority (and hopefully, most troubling part) of the update should be complete. While there are still additional changes which should be made over the next week, these should not cause as dramatic a change in the IMaRS as yesterday's update. Since yesterday afternoon, the IMaRS direct-broadcast data available on the webpages should have been processed using the new software.

27 December 2007

The IMaRS MODIS-processing software is being updated which should improve the quality of the IMaRS MODIS imagery. The majority of the update was completed today and is now being tested on the data from this afternoon's MODIS passes. However, the update is not yet complete, and not all of the MODIS data is now using the new software. If all goes well, then after tomorrow, the direct-broadcast data received after tomorrow should be processed with the updated software.

18 December 2007
This morning there was a reception problem at IMaRS with the ~1600 Terra pass, so this pass will not be displayed on the IMaRS MODIS webpage. I believe the problem has been corrected and that the rest of today's passes should be sucessfully received and processed.

13 December 2007

The recent data-feed problem with the Aqua satellite has been corrected by NASA. However, this morning a problem arose aboard the Terra satellite and NASA is working to correct that problem. There will not be daytime Terra imagery on today's webpage.
In order to improve the quality of the IMaRS MODIS imagery, the IMaRS MODIS-processing software is being updated. There have been some problems encountered during this update, and this has delayed the posting of some of the MODIS imagery. The update is not finished, but it is hoped that significant delays in making the MODIS data available can be avoided.

03 December 2007

Yesterday, December 2nd, a problem arose with the Aqua satellite's Data Feed. NASA is working to identify and correct the problem, and updates can be read from the System Status link below. Though full MODIS passes were not received at IMaRS, the data received was processed and displayed.

08 November 2007

This morning the X-band antenna's suffered a hardware failure and subsequent shutdown. The system was unable to receive today's ~15:10 Terra pass because the system was inoperable at that time. A temporary repair has been made which has allowed the system to restart and function (more or less) normally, and MODIS data reception has resumed.

17 October 2007

You may have noticed that the Terra passes of 15:36 on October 10th and 16:37 on October 16th are displayed with a swath image, but without individual mapped-area imagery. This is because the Terra spacecraft performed some positioning maneuvers on October 10th and October 16th, which resulted in a temporary offset in spacecraft location/orientation information. Consequently, while the swath imagery from the 1536 pass on the 10th and the 1637 pass on the 16th was processed and displayed, the geo-located imagery was not produced for the usual IMARS Modis 'product' areas.

Since the last antenna hardware repair and replacement operation on October 4th, the frequency and severity of the intermittent reception problems appears to have decreased. There have been previous relatively uneventful periods without resolution of the source of the difficulties, so we will continue to closely monitor the system.

Though I try to post timely announcements concerning the status of the IMARS MODIS Direct-Broadcast site; events which effect the MODIS system in general, rather than this site in particular, may not be included in this news and annoucement section.
NASA provides this site with announcements about the status of the MODIS system: NASA MODIS Rapid Response System Status

3 October 2007

The problems that were causing delays in processing MODIS imagery from Friday to Monday have been resolved.
The MODIS X-band antenna system is still having intermittent reception problems. These problems have continued despite the replacement and testing of multiple components during the last several weeks. Efforts to identify and fix the reception and hardware problems continue.

30 September 2007

The MODIS processing of several of Saturday and Sunday's passes halted several times, and as a result, there were delays in making the MODIS data available. The cause of these problems is not readily apparent to me, and the processing has continued uneventfully after being manually restarted. I will try to determine what is amiss, but the availability of the MODIS data imagery may be delayed if these problems continue to occur. Wish these glitches would learn to respect a weekend.
Additionally, the MODIS X-band antenna system is still having intermittent reception problems. These problems have continued despite the replacement and testing of multiple components during the last several weeks. Efforts to identify and fix the problem continue.

25 September 2007 pm:

The MODIS X-band antenna system is still having intermittent reception problems. These problems have continued despite the replacement and testing of multiple components during the last several weeks. We are working with SeaSpace to try and identify and fix the problem.
We again apologize for any inconvience, and thank you for your patience.

31 August 2007 pm:

One of the components of the MODIS X-band antenna system was replaced this morning, however, the replacement was not completed before the 1450 Terra overflight so we did not receive data from that pass. While the new receiver component functioned as expected for the subsequent (1625) Terra broadcast, it did not work with the data from this afternoon's Aqua satellite passes.
We have now re-installed the original component which should allow the reception of most of the Aqua direct-broadcast data.
We apologize for any inconvience and thank you for your patience.

31 August 2007:

One of the components of the MODIS X-band antenna system was replaced this morning. Because the replacement was not completed before the Terra overflight at ~14:50, the data from this pass was not captured. Data from the subsequent Terra overflight was received and is being processed. If the new receiver component is working well, then the data from the Aqua overpass at ~1805 should also be received and processed and available on the IMaRS website by 1930. Thank you for your patience.

17 August 2007:

The Terra satellite has resumed its direct-broadcast transmissions, and the data from both MODIS s atellites is again being received and processed here at IMaRS.

09 August 2007:

The component of the antenna system that failed yesterday was replaced this morning, and the antenna now seems to be working again.

While this antenna problem may have been corrected, we still will be receiving data only from the Aqua satellite until the Terra satellite resumes broadcasting. The Terra satellite stopped broadcasting on August 3rd and NASA forecasts that it may take several weeks for the problem on the satellite to be analyzed and corrected.

08 August 2007:

The USF antenna is not working properly at this time. We are investigating the cause and will try to resume reception of the MODIS data as soon as possible.

When the antenna problem is resolved, we still may only be receiving data from the Aqua satellite. The Terra satellite stopped broadcasting on August 3rd and the word from NASA is that it may take several weeks for the problem on the satellite to be analyzed and corrected.

04 August 2007:

Only MODIS data from the Aqua satellite is being received and posted at this time. The Terra satellite stopped broadcasting on August 3rd and the word from NASA is that it may be two weeks before the problem on the satellite is analyzed and corrected. We continiue to receive and process the broadcasts from the Aqua satellite.

22 July 2007:

The air conditioning unit in the antenna shack has been replaced and seems to be working fine. Hence, the probability that the antenna system will need to be shut down during these hot summer days has been dramatically reduced.
The X-band antenna system has been receiving data from both MODIS satellites since the upgrade of the antenna computer system. We continue to work on resolving the remaining antenna anomolies.

20 July 2007:

afternoon update:

Replacement of the air conditioning unit in the antenna shack was started today. However, a series of thunderstorms meandered about Tampa Bay this afternoon, preventing the installation of the new AC unit. We hope that the installation can be completed tomorrow, and that we can avoid a further disruption to the MODIS data reception and processing. But, the possibility remains that the antenna system will need to be shut down until the temperature control can be restored.

20 July 2007:

The air conditioning unit in the antenna shack is being replaced today. We'll be checking the temperature throughout the day as the replacement is made. The AC service people will be trying to get the replacement made relatively quickly. If it gets too hot in the shack, then the antenna system will be shut down until the operational temperature returns to normal. Though the antenna system is still not working flawlessly, it has been receiving data from both MODIS satellites since the upgrade of the antenna computer system. We continue to work on resolving the remaining antenna anomolies.

Apologies for the inconvenience during these repairs and upgrades.

16 July 2007:

Several days and multiple MODIS overflights have passed since the replacements of several components in the X-band antenna system computer. The transmitted data from both the Terra and Aqua satellites has been received without the data quality problems that prevented the processing and display of the direct-broadcast MODIS data. Hopefully these problems have now been resolved and any remaining problems will be minor.

09 July 2007:

The MODIS X-band antenna system is starting to come back up again. The system's hard drive has been replaced and the antenna software has been upgraded. We hope that not only will this allow the MODIS data to be received again, but also that some of the intermittent problems in reception and data quality will be eliminated. There are still a few more issues to be resolved before the MODIS antenna and processing systems are operating as they should be.

We apologize for the recent gaps in the MODIS data coverage and look forward to the resumption of 'normal' operations. I do not have an estimate on when the MODIS data reception and processing capability will be restored and repaired to a satisfactory level, but we are continuing to work toward that goal.
Thank you for your patience.

06 July 2007:

The MODIS X-band antenna system is still not working correctly. While it is encouraging that the system did sucessfully receive data from one of the MODIS-Aqua passes on the 3rd, the data was from both Terra and Aqua passes on the afternoon of the July 5th and morning of the 6th was not sucessfully processed by the antenna system. During the next several days we will be upgrading several components of the antenna's computer system as part of the efforts to return to normal MODIS data processing.
We apologize for the absence of current MODIS data on the USF IMaRS site. Our efforts to identify and rectify the causes of the recent data reception problems are continuing. While I hope to resume MODIS data reception soon, I do not have an estimate for the restoration of normal MODIS operations.

11 June 2007:

The X-band antenna antenna system which receives the MODIS Direct-Broadcast (DB) data here at USF has had some intermittent problems lately. As a result, you may see several dropped lines or large gaps in some of the MODIS pass images. Several components have been replaced in an effort to isolate and eliminate the problem. The problem is intermittent, so while the last several passes have been collected without any large gaps, the problem is still being investigated.

The region adjacent to Puerto Rico and the Virgin Islands has been added to the USF data products. The 'short name' of the area (such as used in the "Data Products by Coverage Area") is pr_vi. Only the The high-res (~250m) color imagery and HDF files are being generated and posted now, and there may be a bit of fine-tuning over the next several days.

... and a belated introduction: I am David English, and I've started working part-time here at with Frank Muller-Karger and Brock Murch at IMaRS to try and keep the DB MODIS system running smoothly. My background, though diverse, has consisted mostly of oceanographic rather than computer science. While I'm certainly not a programming wizard, I'll try to help Brock maintain the considerable MODIS processing system that he and Judd have built. The fact that it is still running, even after my initial explorations of it, is a testament to the quality of Judd's work in creating a stable and robust processing system at USF.
Though my primary task is to keep the DB MODIS data available, I approach this opportunity with some end-user sympathies. Several years ago I worked with Mary Jane Perry and Karl Banse analyzing CZCS data at the University of Washington. I continue to work (as I have for the last several years) with Ken Carder using near-surface remote-sensing and in-water optical measurements. I hope to continue examining ocean color from small-scale observations in the water, to large-scale spatial and temporal observations from satellite. If you hae any questions for me, I'll try to offer my perspective (or admit ignorance).

8 June 2006:

I just swapped in our repaired receiver, and things are looking fine so far. This should be the end of the data disruptions that we have been seeing over the past few weeks.

Also, I finally turned off the RSMAS ocean code processing for good, so you will no longer be seeing the MODOCL2*.hdf and MOD28L2*.hdf swath L2 products, nor any of the mapped products derived from them.

I've ramped up the SeaDAS processing to replace all of our products that were previously produced by the RSMAS ocean code. I hope you get used to the new products real fast, as there's no going back at this point!

Finally, (and I mean finally), tommorrow is my last day working here at USF IMaRS. I'm leaving to work for a company in Dallas, TX, but will still be checking in on things now and again from there until a suitable replacement is found for my position. For questions and comments in the mean time, please email Brock Murch .
It has been a great five years working with this group at IMaRS, and working with all of our great MODIS data users on so many interesting projects, I can't even count. I will be monitoring my current email address for probably a long while, so drop me a line to keep in touch.

Sincerely,
Judd Taylor

9 May 2006:

We are currently experiencing antenna problems with our X-band system. Unfortunately, it looks like the receiver is failing, and a repair or replacement for that part may take a while. This is causing the complete loss of data reception from the Terra and Aqua satellites, upon which the MODIS instruments reside. We are working to resolve this problem as soon as possible.

In the mean time, I'm working on ingesting data from the DB antenna at NASA's Goddard Space Flight Center, which should provide comperable coverage for several of our areas of interest. I'll post an update when I've got the backup stream up and running.

Update:

SeaSpace is sending us a loaner receiver, so we might start getting data again as early as tomorrow!

31 January 2006:

We've managed to survive another hurricane season, and have been continuing to work on major backend upgrades to our systems. There's no visible change to our systems yet, but a lot of things are moving around behind the scenes.

There are a couple of things I want to notify everyone about regarding some upcoming changes, and some nagging issues we're dealing with at this time.

First, the EPTOMS mission has ended on Dec. 31, 2005, and the Ocean Color group has not yet implemented the new OMI datastream. For now our reprocessing of ocean color datasets is hung up on this issue. We're working with the Ocean Color group to determine how to best deal with the shifting requirements for ancillary ozone data. Hopefully this will be resolved very shortly.

Our primary MODIS file server is filling up with out mapped products. Currently, this server hosts our level 0, level 1B and level 2 MODIS datasets. Very shortly, we will be splitting either the mapped products, or the lower level datasets to another server to deal with this issue. Externally, I'm going to try to keep things the same as much as possible, however, there may be some changes in how you can access the datasets. This will most likely just mean FTP'ing data from a different server (whichever dataset we decide to move to a newer machine), instead of modis.marine.usf.edu. More details will be provided as things develop. Stay tuned updates.

We plan on migrating completely over to SeaDAS from the RSMAS ocean code in the next couple of months. Nearly all of the products that are available from the RSMAS products are now available from SeaDAS products. The new SeaDAS products are already available as level 2 datasets, and as mapped products for the florida and fullpass areas. The rest of the areas are forthcoming, and after they are all available, we will stop producing the RSMAS products completely.

Update: 8 Feb 2006
We've decided to move all of the level 3 products over to the cyclops file server. This is going to be done very soon; hopefully before the end of the week.

There will be no apparent change in the web pages (see this if you want to check). The FTP access will be the same, except you will need to log in to ftp://cyclops.marine.usf.edu/modis instead of into ftp://modis.marine.usf.edu/modis . I'm anticipating only a short transition time (no more than a day at most), where the web pages will be out of date with the existing data. The data production will not be interrupted, however, so you can still get to the L3 data products via FTP.

Also, during this transition, I will be turning off the code that updates the links on the old level 3 directory structure. Please switch to the new directory structure immediately (it's described in the archived news section below).

Again, this change only affects the location of the level 3 (mapped products) datasets; everything else will be the same as usual.

After the change is complete, we will start storing more of the MODIS level1b and level2 products on the modis file server, and less on cyclops. There's currently about 1.3TB of level1b and level2 data on cyclops, and only about 230GB of data on modis. There will eventually be around 800GB of level1b and level2 data on modis, and the excess will still be on cyclops.

Questions/Comments.

27 September 2005:

Hurricane Rita Update
Unfortunately, we haven't been as lucky with getting good, cloud-free imagery as quickly as we were able to get for Hurricane Katrina. However, there is finally some good imagery available, showing very extensive flooding of the LA coastline.

Please see our new USF MODIS Hurricane Rita Image Gallery for the best imagery we have available at this time.

Hurricane Katrina Update
New improved CRGB stretches I've gone through 18 of the best images of the Katrina flooding yet available from MODIS and reprocessed them with an improved CRGB stretch, showing more detail of the flooding an urban areas. Check out the updated gallery below to see the new images and animations.

USF MODIS Katrina Image Gallery

Please check back to this site for updates and new images as soon as they are available.

Questions/Comments.

8 September 2005:

Hurricane Katrina Update
The image links below show data from NASA's MODIS instrument, received and processed at the University of South Florida's Institute for Marine Remote Sensing (IMaRS).

In the CRGB color scheme, red normally indicates dry land, while blue indicates water.

The first image shows the same area before Katrina hit, showing red throughout New Orleans. The second image was captured today (at 16:47:35 GMT), showing large areas of New Orleans and the adjacent Gulf Coast inundated with water.

http://imars.usf.edu/~hu/scratch/katrina/FLOODING.GIF

Shockwave Flash Animation

New Orleans Close-up GIF Animation

NEW: USF MODIS Katrina Image Gallery
Please check here to see the latest imagery as it is manually corrected. All subsequent updates will be on that page.

Before Katrina (Saturday, 27 Aug 2005):

CRGB Image:
Full mriver Region:
RGB Image:

After Katrina (Tuesday, 30 Aug 2005):

CRGB Image:
Full mriver Region:
RGB Image:

Anyone has permission to use these images as long as you credit USF and IMaRS.

If these images may be of use to local management authorities in your area, please pass them on.

Please check back to this site for updates and new images as soon as they are available.

10 August 2005:

Important News About Level 3 Directory Structure Changes:
We are finally ready to go ahead with the migration to the new L3 directory structure for all of our mapped modis products. We're planning to move everything over on Monday, August 15th.

As we move to the new structure, symbolic links will be made to the old structure, for a period of a couple of months. This will mean you will still be able to get to the files in their old locations, except via FTP, which won't follow symbolic links for security reasons. Make sure any software you may have for downloading our data is changed to use the new structure, especially if it grabs the data via FTP.

The new layout of the directory structure is as follows (using HTTP access, as an example):

http://modis.marine.usf.edu/level3/ANTENNA/SITE/YEAR/DOY/SRES/TRES/STATUS/MODIS.*

Where:

ANTENNA: The collection location's 4 digit HRPT code. This is always 'husf' right now.
SITE: The site name for the mapped area. All product from this site share the projection and geographic boundaries. ('cbay', 'florida', etc...)
YEAR: The 4 digit year.
DOY: Julian Day of Year.
SRES: The spatial resolution of the data products. ('4km', '1km', '250m', etc...)
TRES: The temporal resolution of the data products. All products generated from a single pass of data are 'pass', and other codes are used for composited imagery, like '1day', etc... (coming soon!).
STATUS: The processing status of the data products. We first process data products with whatever ancillary data is available at the time (which are 'intermediate' products), and then reprocess them when the get the correct ancillary data (at which point the products are 'final' status).

The change has necessitated a couple of other minor changes to the existing products:

First, the fullpass ERGB products are now named like:
MODIS.2005222.183725.fullpass.ergb.png
Instead of something like:
MODIS.2005222.183725.fullpass.nLw.rgb.png

Second, a couple of sites had sub-products that didn't match the geographical limits of the parent sites. These sub-products are now a part of their own sites. The MODIS.*.scar.hires.* files are now MODIS.*.scar-hires.*, and the MODIS.*.yucatan.rgb.* files are now MODIS.*.yucatan-hires.rgb.*.

Until we make the switch, you can browse the new directory structure (which now consists of symlinks back to the old structure) at: http://modis.marine.usf.edu/level3

These changes have also brought about a change in both the daily web pages and the weekly web pages. We'll be making the switch at the same time we move the file to the new directory structure.

Again, you can browse the new websites before we make the switch here.

If you have any questions or comments, let me know.

14 June 2005:

We suffered a cascading power failure on campus today, and were unable to capture the early MODIS/Terra pass. We're back up now, and there should be no further interruptions in the data stream, barring further major power problems...

If you've noticed there hasn't been as much MODIS/Aqua data lately, it is because there are currently multiple failures on the NASA ground network receiving stations that normally grab the hi-rate MODIS/Aqua data. NASA is using an antenna at Goddard to capture the hi-rate MODIS/Aqua data. During those periods, the satellite does not transmit any Direct Broadcast data.

We are moving feverishly towards implementing a stream of MODIS data processed with the SeaDAS software, since there are currently no more updates of the ocean calibration tables for the RSMAS ocean processing software that we currently use. While the SeaDAS products are young, they are improving all the time, and we hope to make them available for our DB data very soon.

If you have any questions or comments, let me know.

23 February 2005:

Things have been running smoothly for quite a while now, and any changes to the web pages have been minimal since the last update. We've done some major upgrades to the back-end systems here, and in the near future we're going to use the improvements to start producing higher quality and more useful interfaces to our MODIS datasets, so keep checking back!

We've recently introduced "Coastal RGB" composite images to the product lineup for most of the 250m areas that we already have. These images have been optimized to bring out the contrast of the features visible in the coastal waters, and to show a clear boundary between the water and land. In these images, we've used the MODIS bands 2, 1, and 4 as R, G, and B, respectively. Band 4 has been "sharpened" using a ratio against band 1, to effectively increase it's resolution to 250m to match bands 1 and 2. This type of product is still preliminary, and we'd like to hear if you find it useful, or have any suggestions on any improvements.

If you have any questions or comments, let me know.

17 August 2004:

We have weathered Hurricane Charley and are now back to normal operations. Our site is located directly on Tampa Bay and we were required to evacuate early Thursday morning. To minimize the risk and length of the downtime anticipated had we been directly hit, we decided to evacuate with some of the more critical components of our antenna system. As you probably already know, Hurricane Charley turned towards Charlotte Harbor at the last minute and the Tampa Bay area was spared.

In other news, I've recently upgraded the modis mapping software to the latest version. This new version integrates many new improvements and the data quality is much better. The mapping accuracy has been increased significantly, and I've added an RGB sharpening scheme along with better scan-wise interpolation functions to create stunning 250m RGB images. The geographic polygon intersection code has also been greatly improved, which fixes some bugs as well as eliminates images that have no data (which were occasionally produced with previous versions).

Since so much of the code base for the mapping software has changed, there may still be some bugs lurking around. If you notice anything unusual, please let me know so I can correct the problems.

If you have any questions or comments, let me know.

21 May 2004:

There have been a few changes since the last update. The most significant is the addition of MODIS nighttime SST products. I've been getting some questions about missing RGB and ocean color products, but those are just the new night passes (it is not possible to produce RGB or ocean color products from nighttime data).

We've also started generating level 0 AMSR-E data from the Aqua spacecraft. These files are available on our FTP server alongside the Aqua/GBAD and Aqua/MODIS L0 datasets. This should help the general AMSR-E community by providing the data in near-real time. However, there are no plans to produce any higher level AMSR-E data products here. Look for those at Remote Sensing Systems.

To streamline our internal operations, we are hoping to implement a data product database this summer. This will increase our productivity, as well as allow for better web interface to our data. As part of this project, we will also be migrating our data products to a new file structure; and for MODIS level 3 products, a new file naming convention. If you foresee any difficulties using our data products due to these upcoming changes, please contact me.

Over the last few months, we've been working with the SEACOOS project. We are providing MODIS and AVHRR data for the integrated data portal, as well as aiding in the development. The results so far are very impressive, check it out.

If you have any questions/comments, just let me know.

18 February 2004:

NASA has announced that the MODIS ocean processing will now be created by the SeaDAS group. It looks like we will eventaully switch to this software for our data processing, but there is still much to be determined. For the transition period (at least), we will continue to process our MODIS ocean products with the software produced by the RSMAS group at the University of Miami.

Also, today an error occured with the SFE formater on Terra, so there is data from Terra/MODIS data until NASA can fix the problem.

If you have any questions/comments, just let me know.

Update: 19 February 2004

NASA has just reported that Terra/MODIS is back up!

4 December 2003:

You may have noticed that Terra/DB has been down since Sunday, November 30. There appears to be a problem with the DAS channed FIFO chip aboard the Terra spacecraft that is corrupting the data sent out over the direct broadcast antenna. We're working with the investigation team to assist in anyway we can to resolve this issue. However, if other measures fail to recover this chip, then a switch to the B-side electronics will be necessary. As you may recall about 2 years ago Terra switched to B-side, which revealed that the B-side modis is in worse shape than the A-side, which we are currently running on. If you value DB data from Terra in any way, please fill out this survey and email it to Angie Kelly.

If you have any questions/comments, just let me know.

7 October 2003:

As of 6 October 2003, our antenna is back up. We've installed more lightning protection, and a better building ground. We're working on keeping a spare LNA around as well. These measures should be adaquate to prevent any more downtime in the future.

We've finally been able to get the bit of code that makes the OISST reference images for the SST products to work on linux. This will allow us to get the HDF versions of the OISST without having to wait weeks to get them. This should improve the quality of our SST products quite a bit.

For the mapped products, we've gone through and regenerated all of the background coastlines and color tables to a new IMaRS standard across all of our products including AVHRR and SeaWiFS. We've also been able to ensure that the stretches we apply to the data before it goes into the images is consistant across all of our SST and CHL products.

I've just finished re-evaluating the quality filters for our MODIS products, with emphasis on only applying filters that do not also throw away useful data. You'll notice on the SST and CHL that there are more bad pixels included (especially around clouds), but that there is much more data overall. We decided to go this route because most of our applications for MODIS data are qualitative, rather than quantitative.

If you have any questions/comments, just let me know.

9 September 2003:

Wow! It's been a long time since the last update. Here's what's new:

I just switched the MODIS quality control web pages to be the main enrty point for our MODIS data products. These pages are a lot cooler than the old list of swath passes, so go ahead, look around, you should get used to it pretty quick. You can still get to the old site here.

I've also made a lot of improvements to the pages, including new links to the L3 HDF data products, as well as links to our DODS server for those files. With DODS you can directly query the HDF files on the web without downloading them. The web pages that are organized by coverage area are now seperated into each week's images, rather than all of them at once, and there is now a neat little calendar interface to get to the week you want.

The L3 data product quality (mainly the 250m products) has also improved recently. There were previously a couple of geolocation problems, which have now been solved. The first was a misregistration issue with the ephemeris that was uploaded to the Aqua spacecraft, which has been corrected by NASA. The other problem caused "squiglies" in some of our 250m products, which has also been fixed.

On a more tragic note, our SeaSpace X-Band antenna for reception of MODIS data is down. We suspect a nearby lightning strike fried our Low Noise Amplifier (again). We went through 2 LNA's in 3 days. We are currently looking towards a more long term solution to this problem.

In the mean time, we're pulling L0 data from the antenna at GSFC's Direct Readout Laboratory. This is the same data that we receive, except that GSFC's antenna does not receive data as far south as ours. For this reason, some of our more southern coverage areas will not be covered as often as with our antenna. I was able to start ingesting the GSFC data right as our antenna went down, so there are no gaps. Other than the coverage differences, our L3 products will be the same as those from our antenna.

29 May 2003:

We've made a lot of progress since the last update, but I've been holding off on updating the web news until our new web page is up. It looks like it will be a couple of weeks before it's finished.

I've implemented the newest version of the RSMAS ocean code, but I'm not sure what version number to call it. I'm using 4.4.3, since the last code was 4.4.2, but those type of version numbers are set by the MODAPS processing system at the DAAC and is beyond the control of the software authors at RSMAS. This is the same version of the code that will be used in the reprocessing at the DAAC slated to begin in August. For further version information, see the pcf files or email me.

I'd like to note, however, that the reference SST data we're using for Aqua chl-a data processing is not what the DAAC uses. The DAAC is using the OISST 1deg weekly average SST since Aqua SST data is still not yet fully validated. Using this OISST data for our Direct Broadcast data was not a possibility. We have decided to use the Aqua SST data based on assurances from Dr. Bob Evans at RSMAS and our own Dr. Chuanmin Hu that the quality of the chl-a output will be useful (and of higher quality than the data processed with the OISST inputs). For further calibration information, email me.

MODIS Level 3 Mapped Products are up!
I've got my map_modis processing software in full production now, and images are being updated on a regular basis now. PNG images are available for most of the product areas, as well as HDF files for our estuary product areas. Here are some links to directories containing the products (no pretty web interface yet):

Legacy Products (analog to our SeaWiFS products):
Estuary Products:
Misc. Products:

If you need to know more about any of the products, don't hesitate to email me for information; I just haven't had time to write full descriptions of the areas on the web.

Update 6 June 2003:
Check out the new quality control web pages here!

The QC pages are mainly intended for internal purposes, but until our cool browse site is finished in a couple of weeks, it's the best way to browse through all of our MODIS DB data products. I highly suggest you check it out.

If you have any questions/comments, just let me know.

26 March 2003:

I've just finished beta testing the newest version of the DAAC L1B processing software, version 1.2.2. The calibrations should now be a little better, and our L1B products are now made with the latest LUTs in use at the DAAC. It shouldn't be too long until it's available to the other DB sites.

I haven't seen any updates to the L2 LUTs, but I hear a new one should be available soon. As soon as it is, we'll be using it.

I'm now working on reconfiguring our processing streams to include our Aqua/MODIS data. Because of this, we will soon stop processing data collected from the GSFC DB antenna. This will allow us to keep more MODIS data online (while most of the GSFC DB data is available elsewhere). During this reconfiguration cycle, we will also begin producing L3 mapped products in HDF and PNG formats on a regular basis.

And someday, over the rainbow, we should have a great new web site for our MODIS products. It looks like our web programmer will be working feverishly on a new interface for our SeaWiFS products before one is available for our MODIS products.

Also, look for access to our data through DODS (Distributed Oceanographic Data System), which will be on-line soon.

If you have any questions/comments, just let me know.

06 March 2003:

Things have been running smoothly since the last update, and I'm glad there isn't any bad news to report here! The antenna and out processing systems have been working flawlessly.

Some of the users of our L2 ocean data products may be wondering about which versions of the calibration LUTs we are currently using. You can get this information from the *.hdf.met files that accompany the data files. We are using the latest LUTs that we can get our hands on at this point, but unfortunately they're not the best ones out there. Until a better way is found to get the L1B LUTs and code updated more frequently, our calibrations will lag behind what the DAACs are using. There is an ongoing discussion in the MODIS ocean community about this issue.

Our production of MODIS L3 mapped ocean (and otherwise) parameters is nearing production mode. Take a look at some of the test SST products for the Gulf of Mexico here. Both *.png images and HDF (not HDF-EOS) files are available. (disclaimer: these are still just tests...).

22 January 2003:

If you haven't noticed, our antenna is back up! It turned out to be a burnt out Low Noise Amplifier (LNA). We're still looking into the cause. In any case, MODIS data received here is now available within 1-2 hours of collection (usually).

The GSFC Direct Readout Laboratory (where we get the GSFC data, below) has put in Gigabit Ethernet on their server, and transfers from GSFC to here are now very fast. You should notice the GSFC data below now becomes available several hours earlier than before the upgrade.

If you have any questions/comments, just let me know.

09 January 2003:

Unfortunately, we are having trouble with our new SeaSpace antenna. It appears to be a hardware problem up on the antenna feed, so the guys at SeaSpace will be out here to fix it early next week. I'll keep you posted. In the mean time, we are still processing MODIS data from the GSFC DRL's antenna.

Since I didn't hear anything from users regarding features desired for my mapping software, I can only assume that either nobody reads what I write here, or nobody uses this web page (although I have server logs to prove otherwise...). I've added many features since the last update, and there are more test images available here.

11 December 2002:

You may have noticed our new antenna is now operational! In addition to our new antenna, our processing systems have been changing as well. We are still using the collection 3 L1B software from the DAAC, and are awaiting an update to collection 4. However, I have recently implemented the collection 4 MODIS ocean software, and the quality of our ocean products should increase dramatically. I'm expecting a similar increase in quality when we switch over to the collection 4 L1B software.

We are still not yet processing any Aqua MODIS data through the standard NASA processing chain, but it is high on my TODO list. In the mean time, we are using SeaSpace's Terascan software to make images from Aqua MODIS. These images are available here.

As you may know, I'm currently writing software to make (correctly) reprojected images from MODIS data. You can see some of my test images here. I'm hoping that this software will be useful to other users of MODIS data, as well as IMaRS. If you have any feature requests or other needs (projections, user interfaces, etc...), let me know. The program, called map_modis, is written in the Perl Data Language (PDL) and c++. It is several times faster than any other mapping software for MODIS data that I know of (IDL/envi, simap, ms2gt, and maybe HDFLook, but it's pretty darn fast...), and is built on open source tools. The licensing is still up in the air, but I'm open to suggestions...

24 October 2002:

Much has changed since the last update! We will soon be installing a new 2.4m x-band antenna from SeaSpace to resume collection of modis data. If anyone is interested in purchasing our old 4m antenna (for parts), please email me.

Also, we have switched over to a new server, as well as newer versions of our modis processing software. We are currently running the GSFC DAAC L1B software version 1.2.0 for Direct Broadcast. See the *.hdf.met files for more version information. An even newer version will be available soon, and we upgrade to that asap. Our ocean products have not migrated to the latest version (released 4sept2002); we are waiting for the impending L1B updated code to recompile the RSMAS software.

Given the calibration problems with Aqua/Modis, we are delaying the production of Aqua data until a later date.

13 August 2002:

Starting August 8, 2002, forward processing will stop temporarily for the GSFC data. We need this time to work on long-term fixes for our storage system. This will only be temporary. Thank you, Judd Taylor.

30 May 2002:

Our web site is back up after a disk failure. We lost 2 disks on a RAID5 array with no spares (!). This is usually an unrevoverable condition, but since I am the uberengineer, I was able to get everything back up with no losses. We will soon be putting images of our level2 ocean products on the web in near real time. You can take a peek at some of our test images here.

Also, we are still working on fixing our antenna. We are experiencing 2 problems, one with the LNA/Downconverter (waiting for parts to arrive), and another with the Y encoder cable. Hopefully, we will be up and running again very soon.

15 April 2002:

Since our antenna is down for the time being, we have resumed processing of the GSFC DB data to level2. Our L2 products are not yet available in a browse image format, but we are working on that now. We are also producing special images for the blackwater and mms discharge regions in florida. For the blackwater region, we will be reprocessing all of our data from January 1, 2002.

5 March 2002:

We have encountered a problem with the antenna and it is down until further notice. On the bright side, it appears that it may be a small issue concerning the encoder coupling. Assuming this, we hope to have it resolved soon.

25 February 2002:

We have now successfully built the University of Miami RSMAS MODIS Ocean Code. We will soon begin fully automated production of all 48 MODIS Ocean products in near real-time. Aside from the quality of the ancillary data used to create the products, these will be identical to the products available at the GSFC DAAC.

Follow the links below to see images produced from the test data.
MODCOL: 350KB 1.0MB
MODSST: 370KB 2.1MB

Also, as a preview of the mapped products that we will be producing, take a look at this test run through simap, the Simple Modis Mapper program: MOD02HKM.jpg.

Ok, after a very productive day, let's put everything together:
MODOCL2A.A2002049.1516.CZCS_pigment.png
MODOCL2A.A2002049.1516.K_490.png
MODOCL2A.A2002049.1516.calcite_conc.png
MODOCL2A.A2002049.1516.chlor_MODIS.png
MODOCL2A.A2002049.1516.chlor_fluor_base.png
MODOCL2A.A2002049.1516.chlor_fluor_effic.png
MODOCL2A.A2002049.1516.chlor_fluor_ht.png
MODOCL2A.A2002049.1516.cocco_conc_detach.png
MODOCL2A.A2002049.1516.cocco_pigmnt_conc.png
MODOCL2A.A2002049.1516.phycoeryth_conc.png
MODOCL2A.A2002049.1516.phycou_conc.png
MODOCL2A.A2002049.1516.pigment_c1_total.png
MODOCL2A.A2002049.1516.susp_solids_conc.png

The problem with our antenna now only shows up intermittently, and we are able to receive data from modis for all of almost every pass. We are still working on this problem to increase that to 100%.

28 January 2002:

After nearly two months of successful MODIS data acquisitions, we have now hit another snag. Some of the equipment appears to have been damaged during a recent power outage at the antenna hut. We are only getting the stronger part of the signal from MODIS, which is still being processed. We are frantically wokring to fix the problem.

On a lighter note, our processing system is nearly complete. The processing load is now being balanced among the ten linux computers on our cluster. We are seeing moderate gains with this improvement, but expect to see drastic results when we improve the network communications between the nodes and the fileserver.

Also, in case you are wondering, we have stopped processing the Goddard data. When we get the MODIS ocean code running, we may process ocean products from their data.

19 December 2001:

Our X-band antenna is now working! We are still not in full production mode yet, but we are getting close. We are in the process of calibrating the antenna, and automating the level 0 processing. Come by often to see our new and exciting progress!

Also, look for our new web site coming soon!