News and information relating to the Total Ozone Mapping Spectrometer. To have additional newsworthy items added, please contact Dave Larko. Items in TOMS NEWS SPACE are newer than 12 months. Items dating back farther can be found at the News Archive.

We are still having problems with the TOMS calibration, but the instrument now appears to be stabilizing. Interventions in the calibration were put in on August 1, 2001, and again on December 21, 2001, in order to keep the error within bounds. The error we have seen so far is only about 5% at worst. The data still cannot be used for trends, but they should be useful for many purposes. The good news is that the rapid, wavelength dependent changes have stopped and the instrument appears to be stabilizing at about 25% throughput (which is not a problem - our signal to noise is still very good). While there are no guarantees what the instrument will do in the future, We are cautiously optimistic that we will be able to reprocess and get pretty good accuracy. NASA plans to operate the EP TOMS instrument through 2004 to get overlap with the OMI instrument on Aura. So don't give up on us - we are trying hard.

Since August 1 of 2001 we have been applying a correction for the cross track bias in TOMS - an error such that ozone measured looking to the left of the orbital track was consistently lower than ozone measured looking to the right. This error appears to be caused by changes in the optical properties of the front scan mirror. We thought that the instrument was stabilizing, but recent instrument changes are again introducing an error of several dobson units into the TOMS data.

TOMS is performing well for purposes that don't require the highest accuracy - ozone hole monitoring, and aerosol detection for example - but current TOMS data should not be used for trend analysis. Data through 1999 are of very high quality; data in 2000 may be used with caution.

With the current state of the instrument we can no longer produce high accuracy data on a near real time basis. We must move to a system of delayed processing. Near real time data of limited accuracy will continue to be produced. Then six months to a year later the data will be re-processed with a much better calibration. Users need to be aware of this distinction.

We will do our best to continue to provide accurate TOMS data until the launch of the advanced ozone mapper OMI on AURA in early 2004.

We have decided to change the EP/TOMS operational processing configuration in order to apply a correction for the cross-track bias errors that have grown large over the past year or so. These data are being made available at the TOMS web site on an operational basis. Data starting August 1, 2001 will be reprocessed with this configuration and made available as well to provide a consistent data set over the 2001 ozone hole season. The data previous to August are the old data, so a discontinuity will be seen at the July 31 / August 1, 2001 boundary. The reprocessed ozone images will also use our new color scale. The cross-track bias correction will remove almost all of the cross-track bias, but will not remove residual long-term changes in ozone due to related instrument degradation. The current EP/TOMS data should not be used for studies of long-term changes in ozone beyond the summer of 2000.

In a few months we hope to finalize the calibration and cross track correction for Earth Probe TOMS. At that time we will re-process the entire EP TOMS data set and make it available online as a consistent data set.

We regret to announce that the QuikTOMS satellite was lost during launch on Friday (see official press release below). We on the Ozone Processing Team will now concentrate on our efforts to keep Earth Probe operating and in good calibration. Our main problem is that there has been significant instrument degradation, which makes it a challange to meet our goal of 1% long term accuracy. The orbit is stable and the satellite seems to be performing well. The instrument is now over 5 years old, but with a bit of luck we hope to keep it operating for several more years.

(Dulles, VA 21 September 2001) - Orbital Sciences Corporation (NYSE: ORB) announced today that the launch of its Taurus rocket, which was carrying the OrbView-4 satellite for ORBIMAGE and the QuikTOMS satellite for NASA, did not achieve the mission's intended orbit. Approximately one and a half minutes after the Taurus rocket lifted off from Vandenberg Air Force Base in California, an in-flight anomaly was observed during the rocket's second stage powered flight. As a result, Orbital believes the rocket released the satellites at a lower altitude and velocity than intended and that they did not achieve a stable orbit.

We have done a test reprocessing of the Earth Probe TOMS data implementing a correction for the scan angle dependance. The corrected data look quite good so far, but we are still evaluating the accuracy of the data to see if we need to do one more iteration on the correction term. When we are satisfied with the correction (possibly in September) we will announce that the new data are ready and replace all the online EP data beginning with July 1996.

TOMS sees forest fires in Washington State.

TOMS data anomalies: in the last year Earth Probe TOMS has begun to experience two problems.

1) a drop in throughput of the instrument of about 50%, and

2) a cross track bias such that ozone measured when looking to the far left of the orbit track is 2 to 3% lower than ozone measured when looking to the far right of the orbit track. This bias is visible in the TOMS images near the equator, where the ozone distribution is most uniform.

Both changes appear to result from some real change in the front optics of the instrument, probably the scan mirror, that is not completely understood. The drop in throughput is only a minor problem and is dealt with in the processing. The scan bias is a more difficult problem. We are now (early 2001) beginning to see data dropouts at the far right scan positions.

We believe that the TOMS data continue to be usable for most purposes. Average ozone appears to be stable, but individual measurements can have significant error if they are made near the edge of each orbit track. TOMS data past mid 2000 should not be used for trend analysis until the problem is better understood. We plan to carefully characterize the effect on our measured radiances and reprocess the last year or so of data, possibly this summer.

Leap year glitch - there was a problem in the software in the ground control center such that the leap year was not being handled correctly. Data for December 31st were being assigned to January 1st. Leap years in the past were handled correctly by this software, so it not clear why it failed this time and it is taking longer to track down the problem than expected.

Earth Probe is fine and we have not lost any data. We hope to have everything back on line tomorrow (the 4th).

Solar Eclipse. Datum over North America were not processed in the vicinity of the partial solar eclipse.

Discovery On-Line EarthAlert annouces a major Saharan Dust Event.

Shortly after takeoff from Abidjan, a Kenya Airways Airbus A310 jet carrying 169 passengers and 10 crew crashed in the sea Sunday night. The flight orginiated in Nairobi and was scheduled to stop over in Lagos. Harmattan winds coming down from the Sahara to the north had made skies over Lagos unusually hazy on Sunday and the airport stopped incoming flights. After a three-hour layover, Kenya Airways Flight 431 took off at 9:08 p.m. and crashed into the Atlantic Ocean off the Ivory Coast, one minute later. The noontime TOMS data for 1/30 and 1/31 shows the dust cloud over the KA 431 crash site. The data shows that the dust was quite heavy. Authorites have declined to comment on the whether the plane's engines could have been effected by dust and sand. The NRL Marine Meteorology Division has a thorough analysis of this case on the NRL MMD Aerosol Webpage.

Starting COB on Thursday, 30 December 1999, the computers that do the processing for TOMS data to derive ozone will be shut down to comply with NASA's Year 2000 policy. Science processing will resume on Monday, 3 January 2000, at which point all the data collected over the weekend will be processed.

The spacecraft and instrument are doing fine. This shutdown will not lead to any loss of data in the TOMS record (unless civilization completely collapses, but then we will have bigger things to worry about than ozone anyway).

EP-TOMS has observed the lowest value of ozone ever seen in the Northern Hemisphere since satellites first began ozone measurements in 1978.

Again this year, EPTOMS will be shut down on November 17 for approximately 24 hours due to the Leonid meteor shower. It will return to normal operations on November 18.

A new TOMS instrument, to be called QuikTOMS, has been approved for launch in August of the year 2000. The current instrument, Earth Probe TOMS, was launched in July of 1996. It was designed as a two year mission but is now entering its fourth year of operation. In order to assure continuity of the ozone data record for the detection of long term ozone trends a new TOMS instrument is needed. The last of the three new TOMS instruments (the Earth Probe and ADEOS instruments have already flown) - Flight Model 5 - is built and in storage.

QuikTOMS is being procured under the Goddard RSDO (Rapid Spacecraft Development Office) program, following the procedure under which QuikSCAT was developed and launched. By using a commercial spacecraft and launch vehicle instead of developing a new system, a large savings in cost and rapid development can be achieved.

The Orbital Sciences Corporation proposal to build, integrate, and launch QuikTOMS was accepted on July 27, 1999. The spacecraft used will be a MicroStar, which has flown successfully 31 times. The instrument will be launched on an OSC Taurus a four-stage, ground-launched vehicle that can deliver satellites of up to 3,000 pounds into low-Earth orbit. QuikTOMS will share the launch with an Orbcomm satellite. The launch is currently scheduled for August 15, 2000. QuikTOMS will be placed in an 800 km sun synchronous orbit with a 10:30 am equator crossing time.

Earth Probe TOMS is now back in full operation. The initial data appear to be good, but there might be small ozone errors in the first day of data (January 3) as the instrument warmed up. We will be watching the instrument for any signs of a calibration shift as a result of the failover.
The problems that caused the spacecraft emergency are now fairly well understood. It does appear that a single event upset (SEU) in the clock channel produced the 81 degree pitch error that led to the failover. During a failover, the spacecraft points towards the sun to ensure power for the solar panels and spins up for stability. Before spacecraft spin-up it waits ten minutes for the momentum wheels to spin down. Unfortunately the bearing friction in the momentum wheels was much less than at launch and at the end of the ten minute waiting period the wheels were still spinning at 600 rpm. When the thrusters began to spin the spacecraft, gyroscopic forces caused it to wobble. This led to continuous firing of the thrusters as it tried to compensate for the wobble, largely exhausting the onboard fuel. (Initially we had thought that the wobble was caused by a bad thruster.) Software to recover from any future failover using magnetic torquing has now been uploaded. If nothing else fails, we know of no reason why TOMS should not be able to operate for several more years.

The Earth Probe spacecraft emergency was terminated at 18:22z on January 2, and TOMS was restored to science mode at 00:20z on January 3, 1999. Recovery was done entirely using magnetic torquing because the on-board fuel supply was exhausted during the failover. The control software has been re-written to do all attitude control using the magnetic torque rods. The orbit is slightly more eccentric than before the failover, but it remains sun synchronous and should be very stable for the next few years. [back-to-the-top]

The Earth Probe control team began to de-spin the spacecraft this morning using magnetic torqing. This is a slow process and will take a couple of days. So far everything is going exactly as expected. Next the spacecraft will need to transition to Earth pointing and some software will be uploaded for operation under magnetic control. If everything continues to go well, we could be back in Science Mode next week.

Analysis of the Earth Probe failover is continuing. The failover appears to have been caused by an SEU (single event upset) in the clock channel that produced a spurious 81 degree pitch error. The current theory is that one of the attitude thrusters was also bad, which led to continuous firing as the computer tried to spin stabilize the spacecraft in safehold, exhausting the on-board fuel supply. If nothing else is wrong, we should be able to bring Earth Probe back into science mode using magnetic torque control. This will require re-writing the control software. We hope to complete recovery by mid-January. Wish us luck.

Earth Probe TOMS went into safehold at 15:11 UT on Sunday, December 13, 1998 and is currently in sun-pointing safe mode. The situation is under study, but it appears that the normal recovery procedure will not be possible and an alternate recovery procedure will have to be used. We expect that TOMS will be out of operation for some time. [back-to-the-top]