Next: Support Capabilities Workshop
Previous: NSO Telescope/Instrument Combinations
Table of Contents - Search this issue - NOAO Newsletter Home Page

NOAO Newsletter - Global Oscillation Network Group - June 1997 - Number 50

Global Oscillation Network Group

The GONG Project is a community-based activity to operate a six-site helioseismic observing network, to do the basic data reduction, provide the data and software tools to the community, and to coordinate analysis of the rich data set. GONG data are available to any qualified investigator whose proposal has been accepted; however, membership in a GONG Scientific Team encourages early access to the data and the collaborative scientific analysis that the Teams are undertaking. Information on the Project and the scientific investigations, as well as access to the data, is available on our WWW at:


The GONG instrument continues to run with a high degree of reliability at all six sites around the world. Since the beginning of the calendar year we managed to collect all but about 1.5% of the possible images, and as usual, many of these potential gaps were filled by adjacent sites observing simultaneously.


Caption:Comparison of Intensity and Velocity Power Spectra, for l = 150, showing surprising differences. While asymmetry at lower frequencies had been noticed previously, it had not been seen in a single data set and was "suspect." The coincidence of the resonant peaks at lower peaks gives us confidence that the processing is similar for the two variables, but the increasing difference of the peaks of the intensity and velocity resonances with frequency was totally unanticipated. While there may be a small effect arising from the different distribution of the signals from the center of the solar disk to the limb, the principal differences probably result from the increasing transparency of the atmosphere to sound waves with increasing frequency, and the different dependence of temperature and velocity fluctuations on wave form. The spectra have been smoothed with a wavelet denoising technique.

Two thirds of the images lost this quarter resulted from a lightning storm at El Teide where the ensuing power-down and power-up resulted in the failure of a critical power supply. With the help of the excellent staff at the observatory the problem was diagnosed quickly, but shipping delays kept the instrument down longer than we might have liked. We are in the process of addressing this problem by supplying all of the sites with a full set of power supplies to be stored with the instrument.

The next most damaging source of down time this winter has been dew and frost on the turret optics. This has caused delayed startup of the instrument at Big Bear and El Teide on clear mornings following winter storms. Some experimentation during a dew event in the midst of the recent maintenance trip to the Canary site taught us that the optics can be cleared relatively quickly by diverting all or most of the available "clean air" from the optical table to the turret. (During normal operation the flow is split equally between these two demands.) We have also installed an experimental air-drying system in the air line at the Big Bear site. While the system worked well in Tucson during a few wet mornings earlier this winter, results at Big Bear have so far been disappointing.

Although costing little in the way of actual down time, tape-drive issues were a major source of frustration at several sites recently. The problems were eventually traced to a bad batch of head-cleaning tapes. The manufacturer admitted that oil-contaminated cleaning cartridges had been distributed, causing performance to degrade with each cleaning. In our case, several brand new drives were rendered useless in just one cleaning. Repeated cleaning with a "new-and- improved" type of cleaning cartridge brought the recalcitrant drives back to their duty. If you have had similar problems with Exabyte drives, call for details!

We have welcomed Humberto Villegas to the operations group this winter, replacing Sang Nguyen as an electronic technician. Sang, who constructed most of the electronic chassis in our instruments, has moved to NOAO's engineering and technical services group. Humberto is a fine addition to the group, and has already attended one preventive maintenance trip to Tenerife. Later this spring he will travel with the India team to look in on that instrument.

Data Management and Analysis

During the past quarter, month-long (36-day) time series and power spectra were produced for GONG months 15 and 16 (ending 961202) with fill factors of 0.91 and 0.94. The fill factors for these two months were significantly higher than for the previous two months (0.73 and 0.76). The fill factors for GONG months 13 and 14 were unusually low because of global weather patterns (including monsoons at Udaipur and typhoons at Learmonth), preventive maintenance visits to several sites, and some instrument problems. These factors did not adversely affect GONG months 15 and 16. The month 16 fill factor (0.94) matches the previous high from GONG month 11.

As part of the preparation for a reprocessing campaign to regenerate p-mode power spectra from calibrated velocity images, the project temporarily halted the routine production of month-long p-mode power spectra and reprocessed GONG month 16 with the software and processing parameters that the project intends to use during the reprocessing campaign. The evaluation of the results and consultation with the project's scientific community through the DMAC Users' Committee is expected to be completed within in the next few weeks.

Resources that would normally have been used for routine p-mode reduction were redeployed to produce month-long time series and power spectra from the intensity and modulation images from GONG month 16. The intercomparison of these and the same products from the velocity images is underway.

Data Algorithm Developments

The project has been implementing some substantial changes to the processing of the data upstream of the spectral fitting. The new reprocessing implements a new spatial map and apodization in the spherical harmonic decomposition; a temporally varying longitude zero point in the remapping to compensate for the changes in the synodic to sidereal correction caused by the Earth's elliptical orbit; the normalization of the site-day time series moduli to partially correct for different instrumental velocity scale factors; and the restriction of the gap filling to two-point gaps only. These changes are ready to go, and will be routinely used after approval by the GONG DMAC Users Committee.

In the area of spectral fitting, we have installed an iterative background removal, which has reduced the number of noise spikes identified as modes. An asymmetrical line profile model is also under development. The leakage matrix calculation is undergoing improvements, being ported from IDL to Fortran to improve its speed, incorporating pixel integration, and calculations for intensity and modulation. We are also about to begin work on an alternative peak finder tailored for low-degree only. A wavelet denoising and multi-taper package is up and running thanks to Rudi Komm and Yeming Gu, and is now being applied to month 16 l-v spectra.

With a bit of effort, we have merged together one day of GONG network data with one day of MDI/SOI data from SOHO. The resulting spectrum shows a noise level reduction greater than the square root of 2 factor expected from simple averaging, and substantially lower than either of the two individual spectra. This result is probably due to the power of combining coherent signals and suggests that it may be very profitable to routinely merge together GONG and SOI data.

We are extremely pleased to welcome Stuart Jefferies to the GONG/SOI data team. Stuart brings a wealth of experience in sophisticated spectral fitting techniques to the Project. On the down side, Yeming Gu has left the project to work on advanced image processing for supermarket scanners at NCR. We wish him the best of luck distinguishing apples from oranges. Long-term visitors are currently Frederic Baudin and David Freilly-Fraillon. Frederic is working on applying homomorphic deconvolution to GONG and SOI spectra, and studying the details of mode excitation. David is compiling statistics on mode amplitudes and lifetimes.

Recently, the project analyzed one month of GONG data using all three available observables---Doppler velocity (V), total intensity (I), and modulation (M, a proxy for equivalent width). The objective was to evaluate the scientific uses of M and I, and to discard the calibrated data products if they contained little information. While we expected to observe the well-known difference in line asymmetry, we were unprepared for the completely surprising and substantial frequency shift between V and I! As shown in the figure, there is a relative shift in the apparent peaks of tens of microHz between V and I in the region of 4.5-5.5 mHz. There is no apparent shift between V and M. Further comparison of the V, M, and I spectra showed the presence of the f and p1 peaks in V (but not in M or I); the expected asymmetry difference between V and I, and the similar asymmetry of the peaks observed in V and M.

We are striving to understand these effects. There are qualitative differences between the observation of V, M, and I. The velocity is a vector field that is primarily directed normal to the solar surface, and observed projected on the line of sight. The intensity and modulation are scalar fields, with rather different center-to-limb variations in magnitude. These differences affect the apparent amplitude of the modes, particularly the distribution of artifacts known as spatial leaks that arise from our inability to observe the entire solar surface. This differing spatial leak structure may play a role since the frequency range is precisely where the short mode lifetimes cause the spatial leaks to blend together. This effect is being modeled.

Alternatively, the physics of the wave propagation in the solar atmosphere suggests that the waves are evanescent, and implies a roughly 90o phase difference between V and I. This, coupled with the location of the modes driving close to the solar atmosphere, the different effective heights of formation for V, M, and I, and an acoustic cutoff frequency around 5 mHz suggests that the comparison of the spectra in the three observables may yield new insights into mode physics. Finally, the fitting of the mode parameters may also be improved by considering all three observables simultaneously.

Looking Forward

The project has received a three-year travel grant from the NSF International Division to support enhanced scientific collaboration between U.S. and Indian science team members.

GONG has just received the details of its FY 1997 operations budget. The good news is that we have the $2.0M requested. This was not unexpected, but it is always good to have it finally written down.

NOAO is committed to operating the GONG network for a full eleven-year solar cycle, with the usual caveats about the availability of future funding. The project is in the process of renewing three of the site MOU's that expire during the current year. Verbal agreements are in place for all three sites and much of the paperwork is well under way.

Design and planning efforts continue for the proposed larger-format camera for the GONG instrument, an acceptable version of which is currently in house. A conceptual design review is scheduled for June with participation by the GONG community. A proposal is being prepared for submission to the NSF for funding of the construction of these cameras.

John Leibacher

Next: Support Capabilities Workshop
Previous: NSO Telescope/Instrument Combinations
Table of Contents - Search this issue - NOAO Newsletter Home Page

NOAO is operated by the Association of Universities for Research in Astronomy (AURA), Inc. under cooperative agreement with the National Science Foundation