The Global Oscillation Network Group ( GONG ) Project is a community-based activity to develop and operate a six-site helioseismic observing network for at least three years, to do the basic data reduction and provide the data and software tools to the community, and to coordinate analysis of the rich data set that should result. The Project is currently looking forward to deployment of its first sites in 1994 and a fully operational network and data management and analysis center in 1995. GONG data will be available to any qualified investigator whose proposal has been accepted, however active membership in a GONG Scientific Team will allow early access to the data and the collaborative scientific analysis that the Teams have already initiated. The GONG Newsletter provides status reports on all aspects of the Project and related helioseismic science.
This year's annual GONG Meeting - Helio- and Astero-Seismology from the Earth and Space: GONG '94 - took place in Los Angeles May 16 - 20, hosted by UCLA's Roger Ulrich and USC's Ed Rhodes and Werner Däppen. It was extremely well attended, and the proceedings will be published by the Astronomical Society of the Pacific. The 145 participants had the opportunity to enjoy Mt. Wilson in the snow, after which Ed Rhodes was rumored to have indicated that he was eager to host the annual meeting again sometime after 2014. One of the high points of the meeting was an impromptu, after dinner recollection of the discovery of the "five-minute" oscillations by Bob Noyes, and Franz-Ludwig Deubner's confession of how he first encountered this intriguing phenomenon. The participants all signed the very eye-catching poster which Roger Ulrich's son had designed for the meeting, and Arvind Bhatnagar presented it to Bob Leighton, whose seredipitous discovery "enabled" the whole discipline of helioseismology, after the meeting.
In late June, GONG organized a workshop at the University of Sydney, Australia to encourage research interest in helioseismology in Australia in association with our establishing a GONG site at Learmonth, Western Australia. The workshop was sponsored by the IPS Radio and Space Services ( our host organization at Learmonth ) and the School of Mathematics and Statistics of the University of Sydney, with support from the NSF and the Australian government ( DITRD ). The 35 participants from seven Australian universities were subjected to a hard week's lectures and discussion, but there are signs that they survived and the meeting was a real success.
During the workshop, Jim Kennedy, David Cole ( Director of the IPS Radio and Space Services ) and John Leibacher had the opportunity to present an overview of GONG's activities to the U.S. Ambassador to Australia, Edward Perkins, who had recently visited the Learmonth Solar Observatory.
The first official ground breaking for GONG took place during the third week in June at the IPS Radio and Space Services' Learmonth Solar Observatory. John Leibacher, Jim Kennedy, and Frank Hill were in attendance representing the Project. The neck of the champagne bottle with lopped off with the blade of a shovel and work was allowed to "officially" get under way. Since then, Asset Services has proceeded with the grading of the site and the foundation excavations. Arden Petri, the Project's mechanical designer, has traveled to the site to perform the tasks of the Project's "Advance Team", the first of three teams involved in the deployment of our instruments. He witnessed the concrete pour, and supervised the placement of the critical instrument mounting fixtures. If the contractor remains on schedule, this site will be ready to receive an instrument in September, well in advance of the February target date.
The first site to receive a station ( excluding Tucson, of course ) will actually be El Teide in the Spanish Canary Islands. That site will also be ready well in advance of the arrival of the station, slated for December. The permitting and licensing process is now well in hand, and we expect to be able to select a contractor by the end of August, or early September. The current schedule predicts that our Advance Team will be visiting the Canaries in the latter part of September, and site preparation will be complete by mid-October.
Jim Kennedy visited Mauna Loa in mid-July to oversee the "leveling of the lava" at and around the GONG site. With sleet and rain spinning off of a hurricane hovering less than two hundred miles away, he witnessed a classic "cut-and-fill" job as the machinery prepared the site in conjunction with the land preparation for NOAA's new lidar facility.
The Project has obtained the services of an engineering firm to help us prepare the bid packages for site preparation at Big Bear Lake. Core samples have already been obtained along the causeway to allow a detailed specification for the dredge-and-fill operation required to make a pad for the instrument. Things are also proceeding smoothly on the permitting front, although one critical permit application cannot be submitted until the preliminary engineering study has been completed, and drawings prepared. Nevertheless, it now appears unlikely, that these obstacles that will keep us from beginning work this Fall.
In India we have now established that the Udaipur Solar Observatory itself will be our general contractor. Work on their new campus on which our instrument will be located is now under way and progressing well. A package of updated prints was recently released to them to facilitate their planning.
In Chile, our colleagues at CTIO are in the process of putting together detailed cost estimates for our site preparations. Here again, the observatory staff will provide us with general contracting services. Since NSO and CTIO are part of the same observatory system, the act of "letting the contract" is here just a relatively simple transfer of funds within the organization.
The GONG Project was pleased to host Cristina Soares, from the IAC in Tenerife, as the most recent GONG scientific site visitor. Cristina spent three months with us in Tucson, and reduced a set of Taiwanese Oscillation Network ( TON ) data. The Project is happy to congratulate Jesús Patrón Recio on the completion of his thesis entitled "Tridimensional Distribution of Horizontal Velocity Flows under the Solar Surface". Jesús has returned to Tenerife and will defend his work in September.
The electronics group has now finished building and checking out the cables in the field shelters. With that huge task behind us, our efforts are now concentrated on component and systems certification. All of the production data-system boards have been certified and are being installed in their chassis for system tests. The balance of the instrument-control boards are at various stages in the production process. At the time of this writing, the last four boards are about to be released for full production. Populated and tested versions of these last boards should be available to us by mid August when testing of the first three integrated systems is scheduled to get under way.
In the last Newsletter we noted that we had experienced an "anomalously low yield" in our production dual memory boards. At that time, we had reason to believe that the problem was likely bad components rather than any sort of fundamental board-manufacturing problem. In the end our worst fears were realized when we discovered serious flaws in the production boards. After considerable consultation with the vendor, and eventually a plant visit, we were able to identify the problem and have 50 more memory boards made quickly. We are pleased ( and relieved ) to report that the new lot has recently arrived and all of the boards passed the certification tests.
The two real-time programmers assigned to the instrument group have basically completed the data acquisition software. This code is now under strict configuration control; the majority of changes being made are bug fixes, or relate to minor items such as the comment fields of the data header-block parameters. This has provided a "resource wedge" for the component certification activities discussed above. With the help of Jean Goodrich, on temporary loan to us from Jim Pintar's DMAC operation, certification code has been written to test all of the electronic boards which are directly accessible via the VME bus.
We have also made a start on the development of the instrument's user interface. Shane Walker, a student at the University of Arizona, has been hired to help us implement the graphical user interface for the instrument control software. This interface will be the "window on the GONG instrument" for our hosts at the field sites, allowing the status of the instrument and data systems to be evaluated at a glance. The display will also show the most recent intensity and magnetogram image, a running "strip chart" of the mean velocity signal, and log message window. A second "strip chart" window can be opened on demand to display selected instrumental parameters as a function of time. An IRAF window will also be available for other custom reductions and displays, such as velocity and modulation images. Good progress has been made on this work, and we hope to run a beta-test version in early August.
On the optics front, assembly of Lyot Filters has resumed. Two units are already available ( one of which is in service at the prototype ), and Jack Harvey and Roberta Toussaint are committed to having three more assemblies finished and tuned by mid August when field integration begins. The last looming concern with optics is the 4.5 Å interference prefilters: we have still only received two of the ion-assisted filters from the vendor. Another production run is in progress, but attempts to move the passband during annealing destroyed one filter and caused uniformity problems in another. We are continuing to work closely with the vendor. Our backup plan calls for going into integration with conventional ( temperature sensitive ) interference filters.
Much of the month of June, as well as the first few days of July, was taken up with an extended observing run at the prototype. Typically the prototype instrument's time has been divided between software development and engineering checkout, punctuated every ten days to two weeks by a four-day observing run. With development winding down, and with the long, clear spring days in Tucson during June, we suspended that schedule and let the instrument run. As the accompanying chart shows the run lasted 24 days, 11 June through 4 July. Of these, we were able to obtain data worth reducing on 20 days. We had two days of instrumental down time. The first of these was due to a software problem: eleven days into the run, the instrument was no longer able to acquire the Sun without manual intervention. This was traced to an error in the time used to make ephemeris calculations. An inaccurate clock was inadvertently being read, rather than the precision GPS-derived time that provides the time base for the rest of the instrument. The second day of down-time was self inflicted: we chose to use a clear day to reinstall and align the magnetogram modulator and associated electronics, which had been unavailable when the run began.
The last major algorithm task is the installation of an a-coefficient expansion into the determination of the frequencies in some regions of the spectrum. This matter was discussed in LA, where it was decided that it was pointless to attempt to fit every peak in the region of the spectrum where the modes are essentially local rather than global. Current observations indicate that this transition takes place around l = 190, so the current plan is to fit every peak below that point, and to fit a-coefficients to entire ridges above. There will be a region of overlap in which both methods are used. In addition, year-long time series and power spectra will be produced for modes with l < 20, where the lifetimes are long.
As the baseline development work winds down, resources will become available to allow the Project to address issues that had to be skipped over to assure functionality of the minimum pipeline processing. Items include temporal filters, the low-frequency analysis, spatial aliasing, non-spherical harmonic decomposition, and feature tracking. The choice of directions will be made with the input of the DMAC User's Committee ( DUC ).
The membership of the DUC changed at the LA meeting. Tuck Stebbins, Roger Ulrich, and Jørgen Christensen-Dalsgaard rotated off the committee as their two-year terms ended. They provided an invaluable service to the Project. Tuck is especially to be commended for his yeoman work as the Chair of the DUC, and he was awarded one of the coveted Hero Of GONG ( HOG ) awards in LA. Stepping into Tuck's shoes is Dave Hathaway, who has agreed to chair the DUC for the next 2 years. Also joining for two-year terms are Sylvain Korzennik, and Mike Thompson. Tim Brown and Todd Hoeksema will continue for the last year of their three-year terms. The DUC met again on July 26 in Boulder, in conjunction with a mini-workshop on inversions.
Last fall, the Data Storage and Distribution System (DSDS) embarked on a major project to upgrade its computer systems and software. This effort was completed as planned in June. The DSDS now consists of two SUN SPARC10's running SOLARIS 2.3 using ORACLE's database management system with a new design for the file catalog. The users' machine was upgraded to a DECstation5000. The number of possible simultaneous users has been increased from two to eight. The anonymous 'ftp' disk area and space available for network distributions of data products was also increased significantly. The database on the SPARC10's that supports the cartridge volume and file catalogs was converted from Ingres to ORACLE. The redesign of the file catalog provided a significant increase in performance beyond that derived from the workstation upgrade.
At this point the DSDS is operational. Future DSDS software activities will be maintenance and enhancements. One of these enhancements will provide a mechanism for reporting the errors that may occur when the data products are produced, communicating this information to the scientific community, and providing a systematic solution for managing the reprocessing that will replace the affected data products.
The Project anticipates that during the next year as the community increases its use of the DSDS for obtaining data products additional enhancements and modifications may be required.
During the previous quarter, the DMAC calibrated and produced site-day l- spectra and 4-minute averages for 13 prototype data days: March 29, April 24, 30, May 27, 31, and June 11-16, 18, 22. In addition, there are eleven days of raw data beginning on June 23 that are currently being reduced. The interval from June 11 through July 4 was a continuous observing run during which the prototype operated routinely. It was cloudy from June 17 through June 21; however, raw data was recorded each day. Also during the quarter, the Field Tape Reader processed nine raw data cartridges from the prototype instrument that contained 37 site-days. The off-site copy and storage facility copied ten cartridges containing data products.
The development of the data reduction pipeline is proceeding. Those involved spend their time writing software, reducing data, and diagnosing various problems which range from software bugs through functional problems with the reduction algorithms to problems with the raw data.
To explore the spatial frequency range of the instrument, an l- spectrum to l_max_ = 500 was produced. Two features are clearly evident: the ridges are visible well beyond the nominal range used by the Project (l_max_= 250) and spatial aliasing can be seen beginning at l = 250. (The first traces of spatial aliasing can be seen at l = 200 with a very sensitive examination of the spectra.)
The Project also made some progress in understanding the low-l noise in the l- spectra. It was discovered that at least for l = 0, the noise could be suppressed by discarding mode coefficient time samples whose magnitude was greater than 3.5 times the 'rms' of the time series. Subsequent investigation revealed that most of the anomalous samples can be associated with guider faults (signals from the instrument that the guider was not functioning for some period of time during the one-minute recording interval). Combining anecdotal evidence from observers at the prototype site led to the conclusion that birds which seem to congregate in large numbers in the fields near the prototype cause most of the guider faults by occasionally flying between the Sun and the light feed. The guider trips when the intensity falls below a threshold. This was confirmed by video tape of the signal sent to the video monitor in the prototype. This tape also included a remarkable series of frames showing part of an airplane as it approached the runway at a nearby airport.
GONG is now making project information available to the World Wide Web (WWW) via an Http server on the DSDS Users' Machine, helios.tuc.noao.edu ( 184.108.40.206 ) . The GONG "Homepage" may be accessed through hyperlinks in the NOAO Homepage, the NSO Homepage, or directly at http://helios.tuc.noao.edu/greek/gonghome.html.
The GONG Homepage provides links to introductory documents describing the instrument, GONG sites, the DMAC and the DSDS. There are additional documents available containing up-to-date project status, information and a browser such as NCSA Mosaic, however access through Lynx seems to work just fine for users without access to an imaging display.
This good budget news has also allowed us to move forward to fill two open positions presently on our books. We are currently recruiting a Senior Associate in Research to work closely with Jack Harvey with activities relating to assembling, testing, and evaluating the six field instruments, and performing quality assurance checks on the incoming data and diagnosing instrumental problems in the field during the operation of the network. The second position we hope to fill is that of an electrical engineer with a strong background in programming to assist with integration and deployment of the field instruments. This second position ultimately derives from the position vacated by our friend and colleague Warren Ball, who's untimely stroke deprived us of his skills and experience back in April 1993. Since then, we have been forced to keep these funds in reserve in the event that the worst budget predictions had been realized. The research associate job is currently posted and we are accepting applications.
With all of the usual budgetary caveats, the schedule for the deployment remains unchanged with the Big Bear instrument being brought on-line in Tucson early this Fall, and the Tenerife and Learmonth stations being shipped late this year, and becoming operational early in 1995. The Mauna Loa, Udaipur, and Cerro Tololo would be deployed during the Spring and finally the Big Bear would be relocated from Tucson to BBSO. We really are in the home stretch!