AN MY D IO NO NO O S R P T H S E A R E L

A C

N E

O N

I

T

T

E

A

R N

A R Y EC R IB TO O OBSERVA March 2001, Number 32 Photo: David Parker, 1997/Science Photo Library

Observing Proposal Reminder: The next proposal deadline is June 1, 2001. Please make a note to get your proposals for observations using the Arecibo Observatory facilities submitted by that date. Details can be found at our web site http://www.naic.edu/vscience/proposal/proposal.htm.

Resetting the Arecibo Primary Reflector Surface Paul Goldsmith lthough not strictly considered part Aof the Arecibo Upgrade project, the surface of the 305 m telescope plays a critical role in the overall system per- formance, particularly at the higher fre-

INDEX Resetting the Primary Reflector ..... 1 Radio Astronomy Highlights ...... 4 Observing with the Upgraded Arecibo Telescope ...... 5 Space and Atmospheric Sciences .. 12 Planetary Science ...... 16 Computer Department News ...... 16 Employee of the Year 2000 ...... 17 A School Science Project ...... 17 Visit from Congressional Staffers . 19 Colloquia since the last Newsletter 19 Comings and Goings ...... 19 Figure 1: An image of the errors in the main reflector surface processed from Lynn Baker’s photogram- metry data by Germán Cortés. Blue indicates positive devation from an ideal surface and red/yellow means negative deviation. The unweighted rms is about 15 mm. (Courtesy Germán Cortés).

The NAIC is operated by Cornell University under a Cooperative Agreement with the National Science Foundation. quencies that is one goal of the Upgrade. of 0.25 K/Jy at 10.67 GHz. And this was combine the angles to a given target from The surface was last surveyed and ad- before the contractors started redoing the three or more viewing positions, you can justed about 15 years ago, and a lot has surface for us! The mini-Gregorian il- solve for the three dimensional location happened since. The main cables sup- luminated only a small fraction of the of the target. This technique has been porting the surface are held to the ground total surface, but the derived surface rms refined and turned into a commercially by about 2000 cables which connect to from those measurements was 3.3 mm— available combination of hardware and concrete blocks sitting on the ground be- not too terrible, but higher than one software by a company called Geodetic neath the reflector. Soil motions thus can would like for efficient operation. Services Inc. NAIC has been working directly impact the shape of the reflec- with the president of GSI, Mr. John Previous campaigns to set the prima- tor surface. Brown, since 1994, and last year we fi- ry reflector were based on optical sur- nally were able to get an order in for the Based on monitoring by surveying, veys with theodolites. In this procedure, special equipment needed. A somewhat José Maldonado (NAIC) had indications the location of targets located above the different version of this same approach that these motions had been significant, main (North-South) cables was deter- was used to measure the secondary and particularly in the southeast quadrant of mined by triangulation based on mea- tertiary reflectors—the main difference the reflector. That part of the natural surements made from several points is for those relatively small reflectors, a “sinkhole” in which the reflector was around the reflector rim. The locations CCD camera was used. built had been filled in extensively with of these selected points could be mea- dirt and construction material from oth- sured and adjusted to an accuracy of ap- For the measurement of the primary, er parts of the bowl and elsewhere. This proximately 1 mm rms. However, the we have to use a large-format film cam- area was less stable than the rest of the panels are only 1/4 the size of the spac- era. Part of the reason why is evident if ground, and it was not surprising that it ing of the main cables, and each of them you compare the number of pixels in a 6 should be more subject to gradual move- can be adjusted. In the approach used inch by 8 inch piece of film, versus even ment. Puerto Rico is in a fairly active earlier, the positions of panels between the biggest “megapixel” CCD. My crude seismic zone and there are tremors that the main cables were “interpolated” be- estimate is that we get at least a “gigapix- produce small motions—in particular, of tween the measurements of the widely- el” format with the film camera. This is this not very well compacted portion of spaced targets on the main cables. It was necessary if you want to measure a tar- the ground under the reflector. thought that the overall rms was on the get 500 m away to an accuracy of 1 mm. order of 2.5 mm, only slightly less than In addition to the subsidence, the What happens in practice is that the implied by the X-band measurements upgrade work itself was quite traumatic camera is taken up to the top of one of mentioned above. for the reflector surface. There were the towers. It is accompanied by several many panels damaged by items dropped To perform really well, one needs the intrepid NAIC staff members, typically by construction crew working on the overall rms surface error to be less than Lynn Baker, Felipe Soberal, and some- platform and the feed arm. Also, there 1/20 wavelength, which translates to 3 times others. From the tower top, they was one large cable that was dropped, mm rms at 10 GHz. The panels them- take a number of photographs of the dish and when this hit the dish surface, it de- selves are thought to have an error of surface—several photographs are neces- stroyed over 100 panels, and broke some approximately 1 mm rms, and the sec- sary to cover the entire area, and they of the cables that support the dish sur- ondary and tertiary reflectors contribute also take photographs with the camera face. These panels and cables have long smaller errors. So it would be desirable rotated by 90 degrees to be able to iso- been repaired, but the process may cer- to get the primary surface adjustment late any distortion in the camera’s imag- tainly have contributed to a deterioration error below 2 mm rms. It was judged ing system, and take views from two of the accuracy of the reflector surface. impractical to reach this level using the different positions on each tower top as technique employed previously. In as- well. The illumination is provided by a Since one of the major goals of the sessing options, we decided to adopt powerful strobe lamp, which together upgrade was to raise the upper frequen- optical photogrammetry. with the retroreflective properties of the cy limit of operation to 8 GHz or higher target, guarantees that the targets stand (wavelengths less than 4 cm and hope- For this approach, reflective targets out with good contrast relative to the fully as short as 3 cm) it was evident that are put on the panels; these targets are 3 general dish surface. It also means that we would be pushing the accuracy of the inch diameter disks of retroreflective the effective exposure time is very short, primary reflector. Some very limited material. Using a special camera, pho- minimizing any mechanical vibrations, tests using the mini-Gregorian carried tographs of the dish are taken from the etc. Getting all the required equipment out by Phil Perillat (NAIC) in April 1991 top of each of the towers. You can imag- to the tower tops is no mean feat, and we showed that things were not terrific in ine each photograph as yielding the an- have to admire those who carried out this this frequency range, with a sensitivity gular coordinates of the target. If you difficult and sometimes hazardous work.

March 2001, Number 32 2 NAIC/AO Newsletter After the photographs are taken, the Cortés—NAIC) is shown in Figure 1. installed by José Maldonado’s team, and film is developed, and then each image Blue means high and red/yellow means they have not yet been adjusted. The is digitized using a special scanner that low. The big surprise is that the un- largest errors outside the center are on is located in lab adjacent to Tony Ace- weighted rms is about 15 mm! This is the order of 100 mm! This is even big- vedo’s (NAIC) office. This scanner is a worse than had been determined in 1991. ger than José Maldonado had expected, close relative to plate measuring ma- The obvious conclusion is that all the and shows how much that part of the dish chines used by astronomers; it measures upgrade work (plus the passage of the surface had sunk. the centroid of each of the “spots” pro- intervening 10 years!) severely degrad- While the first round of photogram- duced by the targets to an error of no ed the surface accuracy. It is difficult to metry was going on, José Maldonado more than a few microns. These cen- compare this photogrammetric rms di- and his crew were undertaking to refur- troid positions are entered into a data file rectly with that derived radiometrically, bish a lot of the panel support hardware on the PC controlling the scanner. because the Gregorian system does not that had corroded since installation in illuminate the entire surface, and large After all the photographs are 1974. Several thousand panel supports scale errors of the illuminated region scanned, the data files are combined us- needed to be replaced, and many more (linear gradients and quadratic errors) are ing a special program developed by GSI, to be cleaned up and greased so that ad- taken out by calibration runs, appearing which outputs the location of each tar- justment of the individual panels would as pointing and focus offsets, respective- get in three coordinates. Next, Lynn fits be possible. This work is still ongoing, ly. However, there is no doubt that we a sphere to the data set, and derives the and should be completed in April 2001. have adequate explanation for the rela- errors for each target relative to the best- tively poor performance we have seen That work was interrupted by the ar- fit sphere. The software also gives the at 5 GHz, and also for the variability of rival of the results of the first round of uncertainty in each position; this depends gain as a function of source declination photogrammetry indicating the presence on where on the dish the target is locat- and hour angle. Note in particular the of very severe large-scale errors. We ed, and in how many photographs the large errors seen in the “fill area”. The immediately started a project to adjust target appears. We have been impressed large errors seen in the panels right at the 2000 or so tieback cables to get the that the formal uncertainty in position the center of the dish are not surprising surface closer to the desired spherical when we have a full set of photographs as those are the “new” panels recently shape. This work was completed in a is about 0.6 mm rms. Thus, the system appears to really have the capability to measure the whole surface to the re- quired accuracy, but then it will be up to us to adjust the panels to achieve our goal. During the Fall of 2000, about 2,000 targets were placed on the primary sur- face of the antenna. Most of these were located above the points where the “tie- back cables” (which connect the surface to the concrete anchors on the ground below, mentioned above) are located. Some extra targets were put in dense patches to fully sample the panel-to-pan- el setting errors. It was a real struggle to get the necessary data, as that was one of the rainiest Fall periods in recent memory, but this was finally accom- plished. The usual learning curve for de- veloping, scanning, and reducing data was ascended, and we obtained the first set of post-upgrade surface measurement data. Figure 2: This image is from the second set of photogrammetry tests completed in January by Lynn An image of the errors (processed Baker and Felipe Soberal. The improvement is immediately evident when compared with Figure 1, from Lynn Baker’s data by Germán and is quantified by the reduction in the rms surface error from ~15 mm to just over 5 mm.

March 2001, Number 32 3 NAIC/AO Newsletter remarkably short time, and in January again that these are very preliminary. As indicated above, many people Lynn went again to Arecibo, and with However, I am confident that the photo- have been working very hard on the sur- Felipe obtained a second set of data, grammetry is giving us the right answers, face adjustment project. Lynn Baker, which is shown in Figure 2. The im- and that we can do better yet. Don Campbell (NAIC), José Maldona- provement is immediately evident (the do, Mike Nolan, Phil Perillat, and Felipe So —what happens next? After we color tables are the same), and is quanti- Soberal have been extensively involved, complete the ongoing second round of fied by the reduction in the rms surface and they have been supported by many tieback cable adjustments, Lynn and error from ~15 mm to just over 5 mm. others at Arecibo and also in the NAIC Felipe will do the photogrammetry for a The situation is really somewhat better. Maple Avenue laboratory. Mr. John third time. We also will be scheduling If one allows for fact that some of the Brown of GSI has been extremely help- additional telescope time to define the adjustments were done in the wrong di- ful in getting us up to speed with the pho- antenna performance more completely. rection (something that always happens togrammetry system at Arecibo. These This is all a prelude for Phase II, in which in a campaign like this, despite the best people are the ones who deserve credit we adjust the position of each panel in- efforts of the field crews), and one ex- for getting Arecibo working through the dividually. The first step here is to get cludes them, the rms is down to about entire cm wavelength range. approximately 39,000 targets out on the 3.5 mm. Again, this does not allow for antenna surface. The targets themselves any low-order terms, and includes the are currently on order and should arrive entire antenna surface (except the cen- Radio Astronomy Highlights within a month or so. By that time, all tral panels which still were not adjust- Chris Salter of the panel support hardware should be ed). Part of the remaining error may be refurbished, and the nontrivial task of due to the fact that some of the adjust- putting those targets on the antenna sur- ments were so large that there were in- Pulsar Scintillations face will be accomplished. Then, the teractions between adjustment points and he Oberlin/Cornell collaboration really demanding job of doing the pho- possibly even nonlinearities in the rela- lead by Dan Stinebring (Oberlin) togrammetry, but measuring 39,000 rath- T tionship between tieback cables and sur- continues to investigate the high-Q “par- er than 2,000 targets will begin. This is face position. abolic arcs” that they have been seen in conceptually not different, but in prac- pulsar secondary spectra (power spectra Naturally, although this already rep- tice the amount of time and effort to scan of the dynamic spectra). These arcs, resents a huge improvement, we are not the photographs will increase greatly, which are the transform domain equiva- satisfied, and a second set of required simply due to the increased number of lent of the criss-cross patterns that have adjustments has been generated. As this targets. often been noted in pulsar dynamic spec- is written, over half of the tieback cable Adjustment of the individual panels tra since the early 1970s, will be famil- turnbuckles have already been adjusted can then begin, and this too may require iar to faithful readers of these pages. In for the second time. The central panels several iterations. Thus, this project is fact, these arcs made their debut as will also be included in this round. So likely to go on for another year. In addi- “wisps” in the Spring 1999 NAIC-AO we can really hope to get the large-scale tion to the surface adjustment itself, we Newsletter (No. 27) after the group made rms down to a few mm. Unfortunately, will be installing the tertiary actuators intensive observations during January, this has all happened so quickly that we and computer control system, which will 1999. It will interest some readers that have not had time to schedule the re- be necessary to make the small focus and that article caught the attention of none quired telescope time to see what has pointing corrections necessary for oper- other than Ronald Bracewell, who had happened to the antenna performance. In ation at the shortest wavelengths. Bill some interesting suggestions to make some limited time creatively obtained by Sisk (NAIC) has been working with this concerning further analysis of the pat- John Harmon (and thanks to those who system extensively and it is almost ready terns. gave up their scheduled observations!) to go, but installation needs to be syn- Phil Perillat and Mike Nolan (NAIC) did In addition to roughly biweekly ob- chronized with a couple of other nasty carry out some measurements. There are servations — mostly performed remote- tasks including shimming the elevation indications (but these must be considered ly — that the group makes to monitor rails. It does seem that efficient opera- preliminary) that the L-band gain may time variability of the phenomenon to- tion at 5 GHz is now within our grasp, be up by about 10%, that the S-band gain ward half a dozen strong, nearby pulsars, and 10 GHz is not too far off. I hope is up from 5.5 to 7.0 K/Jy, and that at C- they are continuing to explore the effect that in the next newsletter we can give band (5 GHz) we have a single beam in archival data, much of it taken at you some detailed results of antenna with 4 - 5 K/Jy consistently. I am going Arecibo by Jim Cordes (Cornell) during measurements at the higher frequencies, out on a limb to even put these results in the 1980s. The most remarkable result and before long, some scientific results print, but I know that they are what ev- to come out of the analysis of this earlier as well. eryone wants to hear about. I caution data is how little the arc pattern changes

March 2001, Number 32 4 NAIC/AO Newsletter Observing with the Upgraded Arecibo Telescope: Methods and Recent Results AAS Special Session, Thursday, 05 June 2001 Pasadena, CA Arecibo Observatory will be having a special session at the upcoming American Astronomical Society meeting in Pasadena, entitled “Observing with the Upgraded Arecibo Telescope: Methods and Recent Results”. This is motivated by the significant number of improvements to the Arecibo telescope which have been made within last seven years— including improvements in telescope sensitivity, correlator response, observing methods, and data reduction techniques. The purpose of this session will be to familiarize the community with these changes and to demonstrate the observing possibilities with the upgraded telescope. It will start with a general introduction to the capabilities of the Arecibo system, including telescope sensitivity and performance, available receivers, observing procedures (both for on-site and remote observing), and data reduction options. This will be followed by a variety of shorter talks chosen to demonstrate the variety of observing possibilities at Arecibo, from Zeeman studies through asteroid imaging. There will also be a contributed poster session to complement this special session. If you are interested in giving a talk or poster at this session, please send a preliminary title to [email protected] by 7 March, 2001. For questions, please contact Karen O’Neil ([email protected]).

for a particular pulsar over more than (IMBP) systems, where the companion SIGPROC converts raw fast-sampled twenty years! The basic curvature of the star is thought to be a CO white dwarf. data into a standard filter-bank format arc, which they believe is fixed by the IMBPs differ from the classical millisec- for subsequent processing. Folding and location of a dominant scattering screen ond pulsar low-mass white dwarf sys- dedispersion of the filter-bank data is along the line of sight, as well as the tilt tems in that the radio pulsars in IMBPs also incorporated into the SIGPROC (with respect to the pulsar velocity vec- have longer spin periods and higher in- package, the idea being that the general tor) of “filamentary” structure in the ferred surface magnetic field strengths, user can produce data products quickly plane of the screen, stays roughly con- as well as significantly larger (but still for use with other software packages. stant over two decades. Work continues essentially circular) orbital eccentricities. Applications of SIGPROC to date in- on making this statement quantitative, as Despite the necessarily poor statistics at clude searches, polarimetry, scintillation well as inferring an axial ratio for the this stage, there are also suggestions that and single-pulse studies (see Fig. 3 for secondary image that they believe is re- the scale height of IMBPs above and an example of single pulses from the mil- sponsible for the parabolic arc when it below the Galactic plane is a factor of 2- lisecond pulsar, J1713+0747). In addi- interferes with direct rays from the main 4 times smaller than for the millisecond tion to WAPP data, SIGPROC can read (compact) pulsar image. Further details pulsars. These various different proper- in data from the Penn State Pulsar Ma- of this work can be found in Stinebring ties point to a different origin for the chine (PSPM), and the filter-bank at the et al. (ApJ, 549, L97). IMBPs, distinct from either the low or Ooty Radio Telescope (ORT). Work con- high-mass X-ray binary systems which tinues on extending the range of pulsar are thought to produce millisecond pul- machines catered for to include the Pulsar Timing sars and double neutron star binary sys- AOFTM and the Berkeley Pulsar Pro- As part of a program to time some pul- tems respectively. To date, the Parkes cessors. SIGPROC has been extensive- sars from the Parkes Multibeam survey survey has found 5 such systems, and ly tested and documented and is available that lie in the Arecibo sky, Ingrid Stairs full details are given in Camilo et al., for download from the Arecibo Pulsar (NRAO), Fernando Camilo (Columbia), (ApJ, 548, L187). page, (http://www.naic.edu/~pulsar). and other members of the multibeam survey collaboration have been making Pulsar Signal Processing regular Arecibo observations of the 71- Polarimetry Software ms binary pulsar J1904+0412 using the Motivated primarily by the need to ana- In parallel with the SIGPROC develop- PSPM. The orbital parameters of this lyze fast-sampled data that is now pour- ment effort, Jim Cordes has been adapt- 14.9-day circular-orbit binary system ing out of the Wide-band Arecibo Pulsar ing existing software from the old NAIC/ imply that the mass of the companion Processor (WAPP), Dunc Lorimer Berkeley 40-MHz correlator to reduce star is at least 0.2 M
. The pulsar is a (NAIC) has just finished the initial re- polarimetry data taken with the WAPP. member of the growing number of so- lease of the SIGPROC pulsar signal pro- The package takes folded correlation called intermediate-mass binary pulsar cessing package, (suggestions for a more functions from SIGPROC as input to catchy acronym are most welcome!).

March 2001, Number 32 5 NAIC/AO Newsletter produce fully calibrated Stokes profiles. Flux calibration uses a pulsed noise di- ode calibrated against extragalactic Single pulses from PSR J1713+0747 sources of known flux densities. The software can correct for cross coupling in the feed system, and it can derotate the polarization ellipse versus frequen- cy to account for Faraday rotation across the band and for the differential time delay between the two polarization chan- nels. Alternatively, the polarization an- gle rotation can be fitted to determine the Faraday Rotation Measure. The soft- ware will be released for general use shortly, pending the finalizing of a num- ber of scripts to streamline the data re- duction process. A sample output profile from the package is shown in Fig. 4.

Pulsar Absorption Study of Very Figure 3: Signal-to-noise ratios of single pulses from the 4.57-ms pulsar J1713+0747 observed at L-band Small Scale Structure in the ISM using the WAPP. Inset: a 100-ms time series showing the single pulses. The signal-to-noise distribution of the For many years there has been theoreti- pulses is very similar to those of long-period pulsars which indicates that the emission process is the same in cal and observational support for struc- millisecond pulsars. (Courtesy Dunc Lorimer) ture in the ISM on scales from ~1 pc to 1 kpc (Dickey & Lockman, 1990, B1929+10 MJD 51853.863 1.4750 GHz 100 MHz ARA&A, 28, 215), while structures on 60 scales smaller than 1 pc were expected to make only a tiny contribution. How- 40 ever, more recent observations have re- vealed structure in the atomic, molecular 20 and ionized gas that is orders of magni- tude smaller (5 –100 AU) and higher in 4–5 -3 0 density (~10 cm ) than previously suspected. This tiny-scale structure has

-20 been found in all directions in which it has been searched for and in all of the different phases of the ISM, suggesting 150 that it is more likely to be a general prop- erty of the ISM than the effect of some local phenomenon. However, its origin 100 and existence still remain a puzzle. Fur- thermore, its presence poses serious

50 questions for the models of the ISM, as the over-density of these features sug- gests they are not in pressure equilibri- 0 um with other components of the ISM. -100 0 100 Pulse phase (deg) One of the principal techniques for studying very small scale features in the Figure 4: Polarization profiles for PSR B1929+10 observed with the WAPP. The top panel shows the total cold atomic medium is based on multi- intensity, linearly polarized flux density, and circular polarization (negative-going curve). The integration epoch HI absorption measurements to- time was about 20 s and the circular polarization has not been corrected for cross coupling. The software package allows correction if the relevant parameters are known. The non-randomness of the position angle wards pulsars. As pulsars move quickly (lower panel) is real, owing to the close alignment of the spin and magnetic axes in this pulsar. (Courtesy Jim through the ISM, our line-of-sight at dif- Cordes) ferent epochs samples different interven-

March 2001, Number 32 6 NAIC/AO Newsletter stars over 15 yr in a summer student project. IRAS 18455+0448, which has previously been discussed in AO News- letter No. 30 as having an exponentially decaying maser, is no longer detectable. U Equ (alias IRAS 20547+0247), the second instance, is shown in Fig. 6 at the three observed epochs: it has faded from ~0.5 Jy to less than 6 mJy. Finally, IRAS 15060+0947, which was first de- tected in May 1985 with 143- & 370- mJy peaks, had a peak intensity of 15 mJy last October. A normal amplitude variation in intensity for 1612-MHz masers around a pulsation cycle is by a factor of 2 – 3. The birth of an OH/IR star is more difficult to document, as a prior non-de- tection may have been caused by inter- ference, or by encountering the star during a low intensity phase of its cycle: Murray has to read his old mildewed tapes to reach closure on this. Never- theless, the number of births presently matches the number of deaths, as expect- ed for a steady state population. So, in Figure 5: The HI emission and absorption spectra towards PSR B2016+28 as recorded recorded using the addition to V1511 Cyg (see AO News- Caltech Baseband Recorder (Courtesy Snezana Stanimirovic) letters Nos. 28 and 30), Murray has re- cently observed 1612-MHz masers from ing clouds. Comparison of spectra tak- serving sessions have been completed so IRAS 18432+1343 (250 mJy, see Fig. 7) en at different epochs gives information far, with two more to come in summer and from IRAS 18280+0521 (100 mJy). about the size of intervening clouds, their 2001. The data processing is under way These masers are both strong enough that density and pressure, as well as the frac- and it is hoped to compare spectra from even if they had been a factor of three tion of the cold ISM gas they comprise. different observing runs in the near fu- weaker when first surveyed, they should ture. As an example, the emission and have been detected during their previ- Joel Weisberg (Carleton), Snezana absorption spectra towards PSR ous Arecibo search, made in May 1987 Stanimirovic (NAIC), Stuart Anderson, B2016+28 are displayed in Fig. 5. These and May 1988 respectively. Murray Rick Jenet (Caltech), Simon Johnston were obtained with about 30 min of in- Lewis and Dieter Engels (Hamburg, Ger- (Sydney), Kiriaki Xilouris (UVA), and tegration. many) detected water masers from both Carleton undergraduate students, Abby stars at Effelsberg in 1994. These births Hedden, Katie Devine, Amanda Kir- stand apart from another 20 objects with schner, & Kenji Spielman have under- Studies of OH/IR Stars newly-recognized weak (~20 mJy) ma- taken multi-epoch HI-absorption Arecibo observations are making it ever sers, which are primarily detected at this Arecibo observations towards six bright more obvious that the 1612-MHz emis- time due to increased sensitivity follow- pulsars. The main aim of this project is sion phase of many OH/IR stars is very ing simultaneous observations of the to probe smaller ISM clouds with high- brief — frequently little more than a few 1612-, 1665- and 1667-MHz lines. er velocity resolution than has been hundred years. The most recent evidence achieved by previous studies (e.g. Frail, comes from accumulating statistics on Murray has also been active in de- Weisberg, Cordes & Mathers, 1994, ApJ, their births and deaths, where Murray termining better positions for Arecibo 436, 144). The data are recorded with Lewis (NAIC) now has three instances OH/IR stars. Most OH/IR stars in the the Caltech Baseband Recorder (CBR), of each. The death of an OH/IR star (loss Arecibo sky have IRAS positions with a and data processing is done remotely of all masers) is the easiest phase to reli- typical precision of order 10 arcsec. using a supercomputer and a robot-based ably document. Moreover, these cases Now these can often be improved by tape storage system at Caltech. Two ob- derive from the reobservation of just 250 identifying their near-IR counterparts

March 2001, Number 32 7 NAIC/AO Newsletter from the 2MASS survey, where the typ- 560 ical precision is ~0.2 arcsec. This would IRAS 20547+0247 still be impossible for most objects in the 480 Galactic Plane without first identifying 1612 MHz 400 the Midcourse Spacecraft Experiment 8 May 1988 320 (MSX) counterpart, this having an inter- mediate precision of ~2 arcsec. Derek 240 Kopon, an undergraduate summer stu- Flux (mJy) 160 dent from Cornell, together with Mur- ray & Yervant Terzian (Cornell), have 80 thus succeeded in identifying the 0 2MASS counterpart from 114 fields about Arecibo OH/IR stars using extant public data. Fig. 8 is a plot of the result- 40 24 april 1999 ing K-band magnitudes against the ratio of the IRAS 25- & 12-µm flux densi- 20 ties: the thicker the shell, the fainter the magnitude. Two objects appear discrep- 0 ant: the reddest is the ambiguous object IRAS 05506+2414 (BC Tau), situated at Flux (mJy) the far right of the figure, which is ei- -20 ther a proto-planetary nebula or a YSO. The bluer is IRAS 19453+1917, which -40 25 dec 2000 has a secure identification with a much fainter K-band magnitude than its IRAS -110 -100 -90 -80 -70 -60 -50 -1 color would suggest, even though it is LSR Velocity ( km s ) still much brighter than the catalog threshold of ~14.7. Perhaps, speculative- Figure 6: 1612-MHz spectra at three observed epochs for U Equ (alias IRAS 20547+0247). The upper panel ly, this is the signature of a rapidly-thick- shows the observation of 8 May, 1988, while those of 24 April, 1999, and 25 December, 2000, are displayed in ening circumstellar shell. the lower panel. (Courtesy Murray Lewis)

An HI Survey of Dark Clouds 280 Previous l21-cm observations toward a selected sample of dark clouds by Di Li 240 IRAS 18432+1343 & Paul Goldsmith (NAIC & Cornell) 1612 MHz yielded interesting and somewhat con- 200 troversial results (see AO Newsletter No. 01 Jan 2001 29). These observers detected narrow, 160 deep absorption features which are most likely associated with dark clouds. The 120 large amount of cold atomic gas inferred Flux (mJy) from the absorption optical depths re- 80 quire a modification of the standard H2 40 formation rate. Recent progress in grain modeling, lab simulation, and grain sur- 0 face chemistry corroborate this line of

thought. Di & Paul have run a H2 forma- -200 204060 tion simulation based on the most recent LSR Velocity ( km s-1 ) chemical rates. This demonstrates a slow down of H2 production at low HI densi- ties, and can thus explain their observa- tions well. Figure 7: The 1612-MHz spectrum of IRAS 18432+1343 taken on 01 Jan, 2001. (Courtesy Murray Lewis)

March 2001, Number 32 8 NAIC/AO Newsletter 2MASS identifications for OH/IR Stars In their HI spectra, absorption fea- 16 tures unrelated to OH emission are of- ten present. However, any direction with 14 molecular emission will also have cold 12 HI in the beam. This suggests a more

10 comprehensive view of the atomic ISM, K s with temperature fluctuations both asso- 8 ciated with, and independent of, molec- 6 ular cooling. The H2 formation simulation of Di & Paul illustrates that 4 it is feasible to maintain a significant 2 atomic component inside dark clouds. -0.8 -0.6 -0.4 -0.2 0.0 0.2 More studies involving models of pho- (25-12) µm ton dissociative regions (PDRs) are be- ing conducted to address quantitatively Figure 8: 2MASS K-band magnitudes plotted against the ratio of the IRAS 25- & the issue of HI in the cloud envelope. 12-µm flux densities for 114 Arecibo OH/IR stars (Courtesy Murray Lewis)

HI in High Latitude Clouds Two aspects of the controversy still clear. Overall inspection attests to the The widths of almost all radio spectral remain. First, the location of the self-ab- prevalence of HI self-absorption associ- lines from molecular clouds indicate that sorption. A recent Canadian HI survey ated with dark clouds. They also sur- a substantial fraction of the molecular using the DRAO-ST detects large num- veyed two OFF positions at distances of gas is undergoing suprathermal motion. bers of cloud-like structures in absorp- 10 and 20 arcmin for the 29 targets iden- Given this basic observation, the diffi- tion, of which only 50% have known CO tified above. Seventeen of these show culty is in explaining how these suprath- associations. The existence of cold HI significant reduction in HI absorption ermal motions can be maintained over condensation in the solar neighborhood and OH emission in the OFF positions molecular cloud lifetimes. In the past, has been proposed. Second, is the atom- and therefore identify the HI self-absorp- ideas centering on long-lived MHD tur- ic component associated with molecu- tion as clearly associated with the dark bulence and a spectrum of Alfven waves lar clouds exists in the envelope or is clouds. L1544 is shown in Fig. 9. distributed inside the clouds? In the light of their analysis and these competing views, Di & Paul have extended their HI project into a survey toward optically selected dark clouds. From Feb 1-3, 2001, they observed 36 sources covering all starless cores in the Arecibo sky with 0 hr < RA < 6 hr, and with angular sizes of at least 2 arcmin. The observing strategy was to utilize the high sensitivity and spectral resolution of the upgraded Gregorian system. HI and OH (1665 & 1667 MHz) were ob- served simultaneously with down to ~0.1 kms-1 channel width. Only ON scans were taken to avoid the uncertainty in- troduced by the variation in the back- ground HI emission. Out of 36 sources observed, 29 have narrow absorption features with corresponding OH emis- sion lines in velocity space. Four of the remaining seven have the OH lines on the slope of the λ21-cm emission, and Figure 9: The HI self-absorption (black) and OH 1667-MHz emission (red, amplified 15-fold) are plotted in uncorrected antenna temperatures for the source L1544. The offsets are labeled on the individual plots as (ra, the effect of absorption is visible, but not dec) in arcmin. (Courtesy Di Li)

March 2001, Number 32 9 NAIC/AO Newsletter the λ21-cm HI line in the environs of an isolated, nearby, non star-forming, high- latitude molecular cloud (MBM16). Without the complication of galactic ro- tation or long sight-lines through the disk, the HI velocity field in the vicinity of the target cloud is simpler to analyze than that of HI disk clouds. Although the analysis has just begun, the filamen- tary HI structures evident at various channels in the 2.5 × 4 deg map (see Fig. 10 for examples) seem promising. These observers will compare their HI data with CO data for this molecular cloud (which covers about 6 square degrees in the mid- dle of the HI map) and see if a relation- ship between the atomic and molecular gas at differing velocities can be deter- mined. The Arecibo telescope is the only instrument in the world that has suffi- cient resolution of extended gas to allow the detailed HI-CO comparison that is needed in order to establish the presence of high-speed atomic flows.

High Velocity Gas in M61: A Tidal Tail? Recent HST observations detected a MgII doublet in absorption along a line of sight through the outskirts of M61 to- wards a radio-quiet QSO (Bowen et al., 1996, ApJ, 472, L77). The MgII absorp- tion lines show a two-component struc- ture, one that is associated with the disk of M61, and a surprising blue-shifted Figure 10: Example 2.5° × 4G° channel maps of the high-latitude molecular cloud, MBM16, for v = -4.2, -1.0 and 2.3 km s-1. They demonstrate the presence of considerable filamentary HI structure. (Courtesy Phil wing which cannot be explained as part Perillat) of M61’s rotation. Such complex absorp- tion line structures are commonly seen in higher redshift systems, but for these were favored as the mechanisms by cloud structure is to identify the possi- systems detailed information on the dis- which a cloud could support itself against ble sources for this energy. tribution and kinematics of the gas is not wholesale collapse. In the last few years, available. Fortunately, low redshift sys- numerical and analytical results have One of the ways of externally inject- tems, such as M61, provide an opportu- shown that whatever the source of these ing energy into a molecular cloud is via nity to investigate the nature of the gas superthermal linewidths, the dissipation high-speed flows from large shock fronts in much larger detail. timescale for waves or turbulence is in the ISM. If this is a ubiquitous fea- Rob Swaters (DTM, Carnegie Inst.), short, of order a dynamical crossing time. ture of non-star forming clouds, then one Wendy Lane (NRL) & Frank Briggs In the light of these computational re- should be able to examine the surround- (Groningen) used follow-up VLA sults, it is now clear that molecular cloud ing medium for the velocity signatures observations to obtain the distribution velocity fields require a continuous en- of the shear flows. Loris Magnani (Geor- and kinematics of the HI. At the position ergy input if they are to be sustained over gia), Ted LaRosa (Kennesaw State) & of the quasar, they detected large cloud lifetimes. Thus, a fundamental Steven Shore (Indiana) chose to search amounts of HI, but only at velocities that problem in understanding molecular for evidence of neutral atomic flows via

March 2001, Number 32 10 NAIC/AO Newsletter correspond to that of the disk. No gas ionized gas that would indicate on-going UGC 4288 was detected at anomalous velocities. star formation; yet it has significant far- 4 However, no HI is seen along the QSO infrared luminosity implying the sight-line at the velocity of the blue- presence of gas and hot dust. Thus, it shifted absorption seen in MgII. The data may be the most interesting example in 2 do provide other clues, which suggest the local Universe of a galaxy’s transition

that the blue-shifted absorption is caused from late-type to early-type after an Flux density (mJy)

by a Magellanic Stream-type interaction or merger. To determine the 0 phenomenon: there is widespread high neutral hydrogen content of this

velocity gas still connected to the disk remarkable galaxy, and thereby provide 29,200 30,200 31,400 of M61, to the north is a low mass (~106 insight into the relation between the HI Velocity (km/s) M
) HI cloud that may be a condensation content and post-starburst phase of within the hypothesized Stream, and the galaxies, Charles Liu (American Figure 11: The HI spectrum of the galaxy, UGC 4288. (Courtesy Karen O’Neil) dwarf companion to M61, NGC 4301, Museum of Nat. Hist.) & Karen O’Neil is clearly elongated towards M61. All (NAIC) recently observed G515 with these features are continuous in velocity, Arecibo. The galaxy was not detected, between 13,000 and 20,000 km s-1 away. and together suggest that a close passage but preliminary results place an upper Perhaps the most impressive find with × 8 of NGC 4301 to M61 has created a limit of MHI < 6 10 M
on the system. this survey is UGC 4288. Previously this Magellanic Stream-type feature. galaxy was believed to be a nearby dwarf However, they leave open the possibility galaxy with HI mass less than ~107 M
. that this HI gas is part of a halo of high Hunting for Massive Galaxies However, Karen & Greg have found it velocity clouds. One of the more extreme ends of the to lie at a distance of 30,220 km s-1, with disk galaxy sequence are the massive a HI velocity width of ~570 km s-1 and Hoping to discriminate between low surface brightness spiral galaxies, a HI mass of 5.2 × 1010 M
(see Fig. these possibilities, this team used the the prototype of which is Malin I. These 11). Arecibo telescope to detect HI associated are characterized by large scale lengths, with the blue-shifted wing seen in MgII, low central surface brightnesses, and In similar vein, Jim Schombert (U of and the tidal feature. A preliminary (typically) high neutral hydrogen Oregon), Karen O’Neil (NAIC), & Jo analysis shows a detection of the blue- masses. In addition to being fascinating Ann Eder (NAIC) are using the Digital shifted component in HI, at a column in their own right, the slow star formation Palomar Sky Survey (DPOSS), com- 17 -2 density of a few times 10 cm . In the rates of these galaxies allow us a unique bined with the 2-Micron All Sky Survey disk of M61 there is evidence for high view into the conditions that likely (2MASS) to identify massive low sur- velocity gas, but they do not see any HI existed during the formation of many of face brightness galaxies that were not in emission between M61 and NGC the massive (HSB) galaxies we see previously identified in other sky sur- 4301, except at the position of the HI today, as well as being important veys, and whose morphologies are sug- cloud already detected in the VLA contributors to the total baryon mass of gestive of Malin 1-class objects. observations. However, some of this gas the local Universe. Preliminary results from this survey have may be at velocities of either M61 or found more than fifteen of these objects, NGC 4301. If that is the case, its In an effort to learn more about these dramatically increasing the number of emission could easily be masked by fascinating objects, two groups have massive, low surface brightness galax- sidelobe contamination due to M61 and been working at Arecibo to identify mas- ies known. NGC 4301. sive LSB spiral galaxies. Karen O’Neil (NAIC) & Greg Bothun (U of Oregon) recently completed a survey of UGC HI in a Damped Lyman-α Absorber HI in the Most Luminous Post- galaxies whose morphologies are indic- Nissim Kanekar (NCRA, India), Tapasi Starburst Galaxy in The Local ative of a high gas content, yet which Ghosh (NAIC) & Jeyaram Chengalur Universe have not been detected previously in (NCRA) have made high-resolution The “E+A” (“Elliptical + Active” = λ21-cm searches out to 10,000 km s-1. Arecibo λ21-cm observations of the post-starburst) galaxy, G515, discovered In all, 50 galaxies were observed, and damped Lyman-α absorber (DLA) to- by Oegerle et al. (1991) is a very HI detected in 27. The detected galax- wards the quasar OI~363. The Arecibo -1 luminous object (MR = -23.1) which has ies range in velocity from 760 km s spectrum yields a spin temperature Ts = been shown to contain a stellar (dwarf systems overlooked in previous 890 ± 160 K, consistent with earlier low- population consistent with a massive, surveys and found now because of Areci- er sensitivity observations of the system. global starburst 1-Gyr old (Liu & Green bo’s increased sensitivity) out through This value of Ts is far higher than spin 1996). Furthermore, it contains no 30,200 km s-1, with the majority lying temperatures measured for the Milky

March 2001, Number 32 11 NAIC/AO Newsletter Way and local spirals, but is similar to the estimated Ts implies that the absorb- of the wide and narrow components as those obtained in the majority of damped er must have a multi-phase medium. the WNM and CNM respectively. absorbers (T >1000 K). For a multi- s They use the measured λ21-cm op- phase medium, the measured spin tem- tical depth and the above estimates of perature is the column density-weighted kinetic temperature to obtain the HI col- Space and Atmospheric Sciences harmonic mean of the temperatures of umn density in the various components. Don Farley individual phases. Hence, Nissim and The total column density in the narrow Jeyaram had conjectured earlier that the components is found to be high T of the z = 0.2212 absorber and s N (CNM)<1.9 ± 0.25 × 1020 cm-2, On-line ISR Database the majority of other DLAs could be ex- HI while that in the wide component is ou can now access Arecibo incoher plained if they were systems like dwarfs N (WNM) > 1.26 ± 0.49 × 1021 cm-2. ent scatter radar data directly from or LSB galaxies which contain larger HI Y Thus, the WNM contains at least 75% the NAIC web site. Just click on Scien- fractions of the warm phase of HI of the total HI in the z = 0.2212 DLA, tific Users, then Space and Atmospheric (WNM), as compared to local spirals. unlike our Galaxy in which the CNM Sciences, then the ISR database link. The high velocity resolution of the and WNM have equitable contributions. Follow the menu directions to get cus- Arecibo spectra (Fig. 12) enables them As Nissim and Jeyaram conjectured ear- tomized plots of Arecibo data. The plots to obtain estimates of the physical con- lier, this accounts for the difference in are generated on line allowing great flex- ditions in the absorbing clouds by fitting the spin temperatures of the z = 0.2212 ibility, although it takes some time to read multiple Gaussians to the absorption pro- system and local spirals, suggesting that and plot the data. Users can adjust the file. The spectra are well fit by a three- the DLA is probably a dwarf or LSB type scales of color images or click on an component model with two narrow and galaxy. Further, this is also in agreement image to get a time series for a particu- one wide components having tempera- with optical studies by Turnshek et al. lar height or height profile for a particu- ± ± tures, Tk1 = 308 24 K, Tk2 = 180 30 (2001). Finally, the total column densi- lar time. Currently not all the data listed ± ± K, and Tk3 = 7600 1250 K, respec- ty in the DLA is found to be NHI~1.45 in the menu is actually available, but we tively. The last of these is in excellent 0.49 × 1021 cm-2, which agrees within are working to get everything into the ± agreement with the expected tempera- the errors with the value of NHI = 7.9 proper format as soon as possible. There tures for the WNM (5000-8000 K). Fur- 1.4 × 1020 cm-2, obtained from the Ly- are also links to descriptions of the World ther, the mere fact that components are man-α profile by Rao & Turnshek Day data and the various analysis pro- seen with much lower temperatures than (1998). This reinforces the identification

[A] [B]

[C] [D]

Figure 12: The Arecibo 0.4-km s-1 resolution λ21-cm absorption spectrum of the z=0.2212 absorber towards OI 363. The panels are as follows; [A] The spectrum (open squares) with the three-component Gaussian fit (see text) shown as a thin solid line; [B] A zoomed-in version of panel [A] to bring out the shallow absorption wing; [C] Residuals after subtracting the two narrow components from the spectrum (open squares). The fit to the wide component (see text) is shown as a thin solid line; [D] Residuals after subtraction of the two narrow components and the wide component from the spectrum. (Courtesy Tapasi Ghosh)

March 2001, Number 32 12 NAIC/AO Newsletter grams. Try this out and let us know about region using both radar feeds. It used feasible. We are now trying to get de- any problems. only about 20% of the transfer capabili- tailed cost estimates for both this feed ty of the entire system, and so there is system and also for the transmitter, ei- room for future applications. The data- ther a new transmitter (the preferred op- Optical Database taking software running on the VME tion) or rebuilding the old ones. A new There have been links to some of the li- crate is a new version of the “MRACF” transmitter would be more reliable and dar data for some time. We are working program, which now samples two re- much easier and less costly to maintain, to make more of these observations ceivers rather than one and sends the data but the price estimates that we have so available. At present, these consist of to the PC for processing rather than us- far are considerably higher than we color range-time plots of Na and K den- ing an array processor in the VME crate. would like. We are very close to having sities. These are the same plots that can an estimate of the (not un-substantial) We have developed a Linux version be viewed with the real-time data moni- cost of rebuilding the old transmitters and of the program that processes the raw tor that runs during experiments. (URL: the time required to do it. http://www.naic.edu/~lidar/ data, and we are now writing the specif- realtime.html) ic processing modules which duplicate the computation that occurred in the ar- 430 MHz Klystron Status ray processor. There will be one of these The rebuilt klystron, after a rather shaky 430 MHz Dual Beam Radar Measure- for each of the current data-taking pro- start, has been performing better and ments grams, allowing a direct comparison better as we use it. It now puts out some- Dual-beam 430 MHz radar was dis- between results taken with the old and what more power than the other cussed at some length in the previous new systems. Once these comparisons klystron—the one that didn’t fail. This Newsletter. At the time of this writing, are complete we will develop improve- performance is very encouraging and we still cannot quite make these mea- ments to the software that will take ad- suggests that we can successfully rebuild surements, but we are tantalizingly close. vantage of the increased computing other klystrons for $60K or so each if We had a failure in the high power col- power and flexibility. they should fail. lector for the Gregorian feed (the device Separating the system into a real- that takes the 430 MHz transmitter pow- time sampling part and a computing part Electron Collision Effects er out of the slotted waveguide). This will make future upgrades easier; we can device had worked before, and we un- switch to faster PCs as they become Turning now to science, some nice derstand what caused the failure, so it is available, for example. However, since progress has been made on the question not a major problem, but it has delayed virtually identical programs exist for of how electron collisions affect the the- other final tests by 2-3 weeks. Very re- Linux PC, Sun OS and Mac OS X, al- ory of incoherent scatter when the radar cently the revised data taking hardware most any computer can be used, includ- beam is nearly perpendicular to the mag- and software needed for the dual beam ing a laptop brought by a visitor. netic field. This work is relevant to mea- system were successfully tested, as de- surements at Jicamarca, not Arecibo, but scribed below. We can also upgrade the sampling Mike Sulzer and Sixto González (both part of the system, for example with dig- NAIC) have developed a theory, via a ital sampling receivers. The idea is to rather complicated numerical simulation, Aeronomy Data-Taking sample and digitize directly at the RF, that appears to explain the effects at Ji- We have successful tested the new raw or more likely some IF, and do all the camarca that have been a puzzle for data transfer capability for the aerono- subsequent filtering and mixing digital- many years. This puzzle was that, after my data-taking system. We can now ly, eliminating the current analog filters fitting the observations to the “standard” sample and process data from the line and complicated cable setups. Existing ISR theory, the analysis often implied feed and the Gregorian simultaneously. technology may or not be adequate for that the Te/Ti ratio was less than unity, The current radar interface does the sam- our needs, but there is little doubt that if which is a physically unreasonable re- pling, and a new card in the VME crate not now, it will be soon. sult. This was true even though the mag- receives the data and transfers it to a dual netic field was taken into account processor 850 MHz Linux PC where it properly, and even if electron collision is stored on disk and processed. Other Ionospheric Interactions effects were included via a simple BGK computers on the local area network can Two separate studies of the HF feed sys- or Fokker-Planck model. Most efforts also access and process the data. The tem discussed in the previous Newslet- at Jicamarca were devoted to looking for recent test ran in exactly the mode re- ter have now been completed and are subtle systematic errors introduced into quired to make velocity, temperature, essentially in agreement. The 2-element the data by electrojet and/or satellite clut- and composition measurements in the F Yagi scheme of feeding the main dish is ter or some other cause. These latter ef-

March 2001, Number 32 13 NAIC/AO Newsletter fects do tend to produce the same sort of uses a single collision frequency param- is largest for the left panel, for which the effect on the deduced Te/Ti, but it has eter for all particle velocities. beam is closest to perpendicular to B, not been possible to explain the prob- and least for the right panel. (We should The simulation must be run many, lems seen well above the F-region peak, mention also that the right panel corre- many times to build up a “library” of the- especially, in terms of systematic errors. sponds to an antenna pattern with large oretical curves for the appropriate range sidelobes and the most serious cross-talk Sulzer & González, in a 1999 paper, of values of electron density, electron and clutter problems.) The corrected pursued the idea that there really was temperature, and angle between the ra- (red) curves do a nice job of eliminating something wrong with the incoherent dar k (line of sight direction) and the values of deduced T /T less than unity scatter plasma theory, or at least the part earth’s magnetic field. Such a library is e i above 350 km or so. We should men- of it that deals with electron Coulomb now available, and it has been incorpo- tion that using a BGK or Fokker-Planck collisions. The BGK model is pretty re- rated into an incoherent scattering least- model, instead of the Sulzer & González alistic for binary collisions between squares-fitting code. simulation, would make corrections in charged and neutral particles, but long Using the new theory, Néstor Apon- the same sense, but not nearly as large range Coulomb collisions are more com- te (NAIC) processed Jicamarca ACF as the corrections shown, leaving de- plicated. The dependence on particle data obtained using the Faraday/ACF duced temperature ratios still significant- velocity is not very well modeled, even mode. A sample of the results is shown ly less than unity, which cannot be by the Fokker-Planck model; at least that in Figure 13, in which the T /T ratio is correct. was the suspicion of Sulzer and e i plotted as a function of altitude for three González. They developed a Monte- So it appears that this new theory re- different antenna pointing positions. In Carlo simulation that accurately calcu- ally does explain the Jicamarca ISR re- this plot, the terms ‘on-axis’, ‘4 degrees’, lates the electron admittance function, sults. This theory may have wider and ‘6 degrees’ have been used loosely taking account of how electrons of a par- implications; there are no doubt other to refer to the different antenna pointing ticular velocity are affected by both elec- plasma phenomena involving electron positions. The on-axis position is the tron-ion and electron-electron collisions Coulomb collisions for which the colli- closest to perpendicular to the earth’s and then summing over all velocities. sion effect has been underestimated. magnetic field (B), less than 2° at 400 The Fokker-Planck model, in contrast, km (an angle that varies slightly with altitude). The 4° and Optical Observations 6° positions point Big Effect Medium Effect Small Effect Airglow observations were included dur- farther away from ing World Day experiments in Decem- 1997/8/27 1998/3/25 1997/8/28 perpendicular to B, 14:03 LT 20:00 LT 14:39 LT ber (a 4-day, Lower Thermospheric with the 4° position 500 Coupling Study) and in a 3-day run in on-axis 4.5 deg 6 deg ° being between 3.6 January. The data collected consisted of ° and 3.2 from B and thermospheric neutral winds from Fab- ° the 6 position being ry-Perot interferometry (FPI) and air- 400 ° between 5.1 and glow intensities, primarily of the atomic 4.5° from B between 1 α

km oxygen O( D) and H emissions. 250 and 550 km. 300 Each panel contains We participated in Bob Kerr’s (Sci-

CC a curve obtained by entific Solutions) and Néstor Aponte’s No CC fitting the radar data (NAIC) experiments in January, in which 200 to the standard mod- four nights of radar observations plus ap- el without electron proximately 10 nights of optical mea- surements were made. A description of 0 1 2 3 0 1 2 3 0 1 2 3 Coulomb collisions Te/Ti Te/Ti Te/Ti (blue line) and a Kerr et al.’s observations is given below. curve obtained from Aponte and co-workers examined the ion the new Sulzer & energy balance of the thermosphere. As with the World Day studies conducted Figure 13: The Te/Ti ratio at Jicamarca for three different antenna posi- González model that tions. In all panels the blue line comes from fits with the standard theory, includes the effects this quarter, to support this work we ob- while the red line was obtained from fits with the theory that includes the of collisions (red served the thermospheric winds by mea- effect of Coulomb collisions. Left panel: on-axis position August 27, 1997 line). suring the O(1D) 630 nm airglow, as well 14:03 LT. Center panel: “4.5 deg.” position March 25, 1998 20:00 LT. as exospheric Hα (656.3 nm) emissions. Right panel: “6 deg.” position August 28, 1997 14:39 LT. (Courtesy Néstor The effect of the The nights were generally clear and the Aponte) collision correction data quality was very good for these

March 2001, Number 32 14 NAIC/AO Newsletter optical viewing con- 0.8 m telescope. Nightly temperature ditions throughout, profiles were estimated from these data and the Hα data will by measuring the slope (scale height) of be a valuable addi- the neutral density profiles. These are tion to a long term shown in Figure 15 for 5 nights and their study of global average between January 17 and 24. Us- change in atmospher- ing our resonance fluorescence Dye and ic hydrogenous com- Alexandrite lidars we also measured the position. Use of the returns from the sodium and potassium nested optical instru- resonance lines, respectively. These me- mentation in concert tallic atoms are distributed between 80 with the topside 430 and 100 km by meteor ablation. Partic- MHz radar measure- ipants in these studies were Rubén Del- ments provides a gado (University of Puerto Rico-Río Figure 14: The Hα column emission measured in the post-midnight pe- unique sample of all Piedras), and Jonathan Friedman, Shikha riod of January 22 is illustrated. These data were taken following a mag- significant species Raizada, and Craig Tepley (all of NAIC). netic disturbance that began 48 hours earlier, and indicate a 10% – 20% compositions and all Overall, about two weeks of nearly con- enhancement of H column abundance above Arecibo as a consequence of physically significant tinuous lidar data were collected during the disturbance. (Courtesy Bob Kerr) dynamics in the up- this January period. per thermosphere, exosphere, and top- studies, as well as for those during the side ionosphere—for the purpose of un- World Day experiments. derstanding the coupling between these atmospheric regions. An ongoing initiative to measure the composition and dynamics of the upper During January 2001, we measured thermosphere, exosphere, and topside the Rayleigh signal backscattered from ionosphere was served during the peri- the neutral atmosphere up to about 90 od January 14 – January 27. Bob Kerr km altitude using our Nd:YAG laser and and Yan Betremieux of Scientific Solu- tions Inc. (SSI), North Chelmsford MA, JANUARY, 2001 were assisted by Jeffrey Hyland of 17-18 18-19 19-20 22-23 23-24 5 day average Tewksbury Memorial High School and 90 SSI for the optical and infrared measure- ments that this program features. The 80 visitors used two Fabry-Perot interfer- ometers to measure the OI 6300 Å emis- 70 sion and the Hα 6563 Å emission, providing winds near the F2 peak and the exospheric H velocity distribution, 60 respectively. Two photometers and the Altitude (Km) Ebert-Fastie Spectrometer were also op- 50 erated. The photometers measured the OI 8446 Å and the Hα nightglow bright- 40 ness, providing diagnostics for (photo- electron model dependent) O 30 composition in the upper thermosphere 150 300 150 200 250 300 and the column abundance of exospher- Temperature (K) Temperature (K) ic H, respectively. An example of Hα (a) (b) photometer data is shown in figure 14, and preliminary analyses indicate that at Figure 15: Nightly and average temperature distribution of the middle atmosphere for January 2001 20% enhancement of H column abun- deduced from Rayleigh lidar. Shown are (a) Temperature profiles compared to the MSIS-90 model (dashed line) for 5 winter nights, and (b) the ‘monthly’ average temperature altitude profile. Deviation below dance occurred following a magnetic dis- about 35 km is due to the fact that the telescope and laser fields-of-view are not completely overlapped in turbance that began on Jan. 20. The this bi-axial system. (Courtesy, Shikha Raizada) period was characterized by excellent

March 2001, Number 32 15 NAIC/AO Newsletter P. Nicholson (Cornell) and others the efforts of Tony Crespo and Víctor Solar System Studies pursued their study of azimuthal asym- Negrón the transmitter is now more reli- Don Campbell metries in the optical and radio wave- able than in the past and Phil Perillat and he combination of the improved length scattering properties of the rings Mike Nolan have made the data acquisi- T sensitivity of the Arecibo 13 cm of Saturn. Delay-Doppler images of the tion software extremely dependable and wavelength planetary radar system and rings were obtained with a range reso- flexible. The portable fast data acquisi- the increased rate of discovery of near lution of approximately 10,000 km, high tion system developed by Jean-Luc Mar- earth asteroids means that the radar sys- enough to separate the echoes from the got with the help of Jeff Hagen has tem is under very heavy demand. The A and B rings. enabled very wide bandwidth data acqui- sition at Arecibo and bistatic observa- fall of last year saw a series of observa- There was major excitment at the tions with the St Croix VLBA antenna tions of Titan, an attempt to detect an- discovery in the fall that two recently and, soon, the 100m Green Bank tele- other satellite of Saturn, Iapetus, discovered small near earth asteroids, scope. A copy of this system has been observations of the Galilean satellites of 2000 DP107 and 2000 UG11, are bina- built for use at the 70m NASA/DSN Jupiter, delay-Doppler imaging of the ry systems. The discoveries involved Goldstone antenna. rings of Saturn, imaging observations of observations with both the Arecibo and numerous near earth and main belt as- Goldstone radar systems. While there teroids and additional imaging observa- have been suggestions from optical ob- Computer Department News tions of the moon at both 13 cm and 70 servations that some near earth asteroids cm. Through early March of this year are binary systems, these are the first Arun Venkataraman there have been observations of a num- confirmed detections. The people in- ber of near earth and main belt asteroids volved in these observations were J-L. Internet 2 getting closer and a series of bistatic runs on Venus Margot (NAIC, now at Caltech), M. ollowing the award to Centennial PR with Arecibo transmitting at 13 cm and Nolan (NAIC), L. Benner, S. Ostro, J. and Qwest Inc of the contract to build several antennas at the NASA/DSN Giorgini (JPL), S. Hudson (WSU) and F the PRISA (Puerto Rico Internet 2 Ser- Goldstone facility receiving the echo. D. Campbell (Cornell/NAIC). Both vices Association) on-island OC-3 net- DP107 and UG11 were very recent dis- The Titan observations by G. Black work with DS-3 uplink to the Abilene I2 coveries and there was no indication (NRAO), D. Campbell (Cornell/NAIC) backbone, work has begun on the fiber from the initial optical observations that and S.Ostro (JPL) confirmed the mea- layout for the Observatory’s “local loop” there was anything very unusual about sured values of the average scattering in the teeth of right-of-way permit diffi- them. This highlights the importance of properties of Titan made in the fall of culties faced by Centennial. Florida In- using the radar to observe as many of 1999. However, the major objective of ternational University’s GigaPOP, these small objects as possible and the the observations was to measure Titan’s managed by FIU’s AMPATH Project, asteroid radar community is very grate- scattering properties as a function of lon- will provide PRISA’s actual link to I2. ful to those users of the telescope who gitude. The longitude of Titan’s sub-earth The Observatory’s central router has are willing to sacrifice some of their point moves approximately 22 degrees been upgraded to support the 155-Mbits/ observing time on short notice to make per (earth) day. Observations were s ATM/SONET interface to Centennial’s these observations possible. scheduled in two groups giving full lon- on-island ATM network. The router will gitude coverage at this 22 degree spac- With the improvements in the rms also channel non-I2 traffic into the Ob- ing. Unfortunately, only about half the surface accuracy of the telescope’s pri- servatory’s existing T1 link to the com- longitude coverage was obtained due to mary reflector, the overall sensitivity of mercial Internet. transmitter problems. Little variation the 13 cm radar system is approaching was seen in the scattering properties but the original goal of the recent telescope the covered region did not include the and radar system upgrading project. The Control Room observing longitudes where a bright feature has resetting of the 1900 tie down cables of While a faster network connection can been observed in near infrared imaging the reflector has increased the 13 cm gain make the remote observing option attrac- by the Hubble Space Telescope and by approximately 20% and made it much tive in the foreseeable future, the new ground-based telescopes using adaptive less dependent on the zenith angle and observer’s workstation in the Control optics systems. On two days there were azimuth of the Dome. The resetting of Room is worth a visit. This is currently suggestions of a quasi-specular compo- the individual panels of the reflector, a Sun Ultra-60 with 1GB RAM, dual dis- nent to the echo but confirmation of this expected by the end of 2001, should give plays (one of them a flat-panel LCD) and will have to await new observations a further 15% increase in the gain. This a gigabit ethernet connection to the net- planned for the fall of this year. increase in gain has been accompanied work servers. And while old-time Ob- by other system improvements. Due to servatory hands can still use the

March 2001, Number 32 16 NAIC/AO Newsletter venerable Analyz data reduction envi- often be seen troubleshooting an auto- ronment (now in its silver jubilee year), Employee of the year 2000 motive electrical problem for a fellow new reduction software is being written Daniel R. Altschuler employee. for the Research Systems Inc. IDL envi- nother year has gone by and we ronment, notably the calibration and Aagain had the difficult task of mapping routines authored by Phil Pe- choosing the Employee of the Year. A rillat and Dr Carl Heiles of UC Berke- committee chaired by Carmen Segarra The Arecibo Observatory, a School Science Project ley and integrated with the User and composed by Víctor Santiago, José Interface. Mike Nolan’s data conversion Chacón, Antonio Nolla (all former re- Jesse Stinebring routines support FITS and CLASS for- cipients of the prize) and Julio Crespo recibo Observatory is in Puerto mat data export; CLASS users can mon- reviewed the nominations by fellow Rico. I decided to focus on how it itor the exported data in realtime. A employees and came up with Víctor Igu- works and runs. I did some research on ina as the winner of this year’s award. what it studies but not much on the his- Data acquisition Víctor, has distinguished himself as tory. My dad is a scientist, and so he goes The VMEbus continues to serve the a tireless worker in the Electronics de- down and works in Arecibo. That is how Observatory well for basic control and partment where his hard work, expertise I decided to study the telescope for the data acquisition, however, it has been and dedication have maintained the 430 Independent Studies Project. I could take apparent for several years that the 32- MHz transmitter so well that its excel- pictures and have interviews in Areci- bit bus would not be sufficient for high- lent reliability is taken for granted by its bo. Arecibo is the biggest (and best) ra- end data acquisition applications; users. Víctor has been with the transmit- dio telescope in the world. moreover, the processing power avail- ter from the start in 1964, first as a tech- I hoped to learn how the Arecibo tele- able on the Observatory VMEbus ma- nician with Domingo Albino, the scope works, some of the measurements chines is exceeded by modern desktop transmitter engineer, and then since 1992 of Arecibo (I got a lot of them), what workstations. As earlier newsletters have alone as engineer and technician. At the Arecibo scientists study, and the history reported, Intel-based PC’s running the same time Víctor has also contributed of the Arecibo Observatory. I learned Linux OS have been used for some re- to the installation of fiber optic and tele- all I wanted to learn. I got more of the cent data acquisition projects, notably the phone lines throughout the Observatory measurements than I expected. At the Wideband Arecibo Pulsar Processor beginning of my project, I thought Areci- (WAPP); these machines feature PCIbus Víctor is always ready to lend a help- ing hand, both on the job and off; he can bo only studied pulsars, stars, and plan- bandwidth and near-1GHz CPUs. Since ets. the custom-built VMEbus components used to drive the Observatory’s radar I didn’t have too system are expensive to duplicate, a hy- much information, brid approach using components from but what I had was both the VMEbus and the WAPP systems very good. I had one was used by Bill Sisk, Jeff Hagen and good book, one web Phil Perillat to build a data acquisition site, and lots of peo- system for dual-beam radar. The “glue” ple let me interview is provided by a custom-built Industry them when I went Pack I/O card that transfers data from a down to Arecibo. VMEbus single-board computer to the When I went to PCIbus computer. Arecibo, I saw all of the different parts of the telescope and Visitors LAN’ed at Observatory how the observatory The planned extension of the Observa- works. It was very tory Local Area Network (LAN) to the difficult at the begin- Visitors’ Quarters should provide wel- ning of the project come relief for guests who like to stay because I had so lit- “connected” at all hours. tle information. But when I went to Arecibo, I got a lot Víctor Iguina receives the award for Employee of the Year, 2000 from more information. Daniel Altschuler. (Photo by Tony Acevedo)

March 2001, Number 32 17 NAIC/AO Newsletter Arecibo is the biggest radio telescope Puerto Rico because there were sink- erator’s job is to make sure that the tele- in the world. The dish is 1,000 feet (305 holes that they could put the dish in so scope is running well and to make sure meters) in diameter. It has a depth of 167 they didn’t have to blast so much. Even that all of the 200 scientists who visit feet (51 meters). The reflector’s surface though they put the dish into a sinkhole, Arecibo each year get their turn on the has 38,778 aluminum panels. Arecibo they still had to use a lot of blasting. The telescope. covers 18 acres (8 hectares) or the same Arecibo Observatory is near the town of I started out by going on the comput- as 26 football fields. Each concrete tow- Arecibo, so if it was placed in Hawaii or er and finding information on Arecibo. er has 9,100 cubic yards of concrete. Two Cuba it wouldn’t be named the Arecibo On December 27, 2000, my Dad flew of the towers are 265 feet (80.772 Observatory. The observatory of Areci- down to Arecibo to do research on pul- meters) tall and the other tower is 365 bo is way up in the mountains of Puerto sars. On December 31, 2000, the rest of feet (106.68 meters) tall. Rico so it is a long and bumpy drive to my family flew down to Arecibo to go get there. The Arecibo Observatory studies lots to Puerto Rico for a vacation, to visit my of different things. It studies the planets, There are many jobs at Arecibo, so Dad, and to do my Independent Studies the ionosphere or atmosphere, pulsars, they need lots of people working there. Project. I asked a lot of people if I could galaxies, quasars, and hydrogen gas in There are lots of jobs the workers have interview them, and they all said yes. I our galaxy. Arecibo has a program that to do like clean the dish of trash, fix interviewed the director of the observa- is called SERENDIP which stands for things on the platform, go under the dish tory, two telescope operators, the direc- to fix cables, and make sure everything tor of scientific staff who is also a radio S earch for is going well in the control room. If you astronomer, and a scientist who uses LI- E xtraterrestrial are under the dish and you hear a big, DAR (a kind of laser) to study the atmo- loud siren noise, that means you have to sphere and ionosphere. I went under the R adio get out of there within 30 minutes be- dish and got some pieces of the reflec- E missions from cause the scientists are going to make tor’s metal panels. I was going to go on microwaves to bounce off of different the platform, but we ran out of time to N earby objects in the solar system. If you are up go on it. I was sad that I couldn’t go on on the platform and you hear another si- the platform, but I was also very relieved D eveloped ren (that’s not as loud as the first), that because the platform is very high up I ntelligent means you have to get out of the way in (even though nobody has died up there 5 seconds because the scientists are mov- and it is very safe). I am very glad I did P opulations. ing the azimuth arm. The telescope op- this project because I have been embar- SERENDIP is a program that uses an antenna that sits at the top of the plat- form, but they haven’t found anything yet! Up 450 feet (137 meters) is the plat- form, and on it is a bow-like structure. It holds the Gregorian dome and the car- riage house. The bow, or the azimuth arm, is 304 feet (93 meters) long. The platform has 26 motors to move the azi- muth arm, the Gregorian dome, and the carriage house to any position within a millimeter precision. The Gregorian dome is six stories high and weighs 75 tons (68 metric tons), but the whole plat- form weighs 900 tons (803.6 metric tons). The Arecibo telescope is in Puerto Rico. The scientists and engineers who planned Arecibo were thinking of put- ting the telescope in Cuba or Hawaii, but they decided to put the observatory in Jesse in front of his exhibit which includes this article plus a model of the Arecibo telescope he made. (Photo Carter McAdams)

March 2001, Number 32 18 NAIC/AO Newsletter rassed that I do not know a lot about as- over lunch. The visiting party consisted Cornell University where he is respon- tronomy and radio telescopes. This of Ron Anderson, Floyd DesChamps, Dr. sible for federal and state relations. project has helped me learn what astron- Robert Palmer, Karen Pearce and omy is and how the biggest and best ra- Stephen Philip Johnson. dio telescope in the world works and Colloquia since the last Newsletter Ron Anderson is a Republican Ap- runs. 16 Feb, 2001 Abel Mendez, UPR, Col- propriations staffer for Chairman James loquium: Arecibo Planetary Habitable Jesse Stinebring is a fourth-grade stu- T. Walsh, chair of the house VA/HUD Zones: The Spatial Distribution of Life dent from Prospect School Oberlin, and Independent Agencies Appropria- in Planetary Bodies Ohio. Jesse has visited the observatory tions subcommittee, the subcommittee twice during the last three years along with jurisdiction over NSF spending. 7 Dec, 2000 Neftali Rivera, UPR, Areci- with his parents Dan and Lynn as part Floyd Deschamps is a Republican Pro- bo, Colloquium: Arecibo Optical Astron- of Dan's regular observing program at fessional staff member with the subcom- omy Observatory the telescope (see contribution to the mittee on Science, Technology and radio astronomy article in this issue). Space of the Senate Committee on Com- 1 Dec, 2000 Hans Moosmüller, Desert Jesse wrote this article for his Indepen- merce, Science and Transportation. The Research Institute, University of Neva- dent Studies project and, at the time of Commerce, Science and Transportation da-Reno, Colloquium: New Optical writing, is currently “defending his the- Committee is one of two authorizing Methods for Monitoring Atmospheric sis” back in Oberlin.—Ed. committees with jurisdiction over NSF Visibility in the Senate — the other two being the 30 Nov, 2000 Mark Chang, UPR, Col- Health, Education, Labor and Pensions loquium: The Magdalena Ridge Obser- Committee. Robert Palmer is the Dem- Visit from congressional staffers vatory - an update ocrat Democratic Staff Director for the Chris Salter House Science Committee. The Science 21 Nov, 2000 Phil Nicholson, Cornell U., n Feb 20, John Harmon and my Committee is the only authorizing com- Colloquium: The discovery of 7 new Oself hosted three “Congressional mittee on the House side with jurisdic- satellites of Uranus and Saturn Staffers”, plus an NSF representative and tion over NSF. Karen H. Pearce is a 6 Nov, 2000 Willem Van Driel, Nançay a Cornell Vice President. The visit in- Legislative Policy Analyst at the NSF’s Observatory Delphine Monnier- cluded a tour of the control room and Office of Legislative and Public Affairs. Ragaigne, Nançay, Colloquium: HI stud- the visitor center, as well as informal dis- Stephen Philip Johnson is Assistant Vice ies of LSB galaxies at Nançay cussions with scientific staff members President for Government Affairs at

Comings and Goings Hector Cruz Retires Jonathan Friedman any of our readers have never met MHéctor Cruz, but his absense may well be noted. Héctor began at the Areci- bo Observatory in 1967 as a temporary employee doing carpentry. He was soon moved to the full time staff in the Main- tenance Department, where he was pro- moted to Trades Foreman in 1971. In 1986 he became Trades Supervisor, where he was in charge of all of those nasty maintenance tasks that nobody wants to deal with. In 1998, Héctor manned the local “fire tower”, stationing himself in the vicinity of the airglow and lidar labs in order to beat back flames from a fire that Group shot taken during the visit by congressional staffers at the visitor center observation deck. From left: raged in the valley below. His dedica- John Harmon, Stephen Johnson, Floyd DesChamps, Karen Pearce, Ron Anderson, Chris Salter, Robert Palmer. tion and good humor will be remembered (Photo by Tony Acevedo)

March 2001, Number 32 19 NAIC/AO Newsletter here, and by those of our visitors who Adios Jean-Luc Margot They have been used successfully at the had the honor of knowing his work at Don Campbell Goldstone NASA/DSN 70m antenna, the the Arecibo Observatory. After two years as a post-doctoral re- NRAO ST Croix VLBA antenna and, of search associate at Arecibo, Jean-Luc course, Arecibo. Margot has moved to sunny California Jean-Luc’s ability, tremendous dedica- taking up the O. K. Earl postdoctoral fel- tion to his work and willingness to help lowship in the Division of Geological anyone who needs his assistance will be and Planetary Sciences at the California sorely missed at Arecibo. We are very Institute of Technology. While at Caltech happy that he will remain connected to Jean-Luc will continue working on such the Observatory and wish him every suc- topics as radar observations of near earth cess in his new position. objects and studying the surface proper- ties of Venus but he has already started using the Keck II optical telescope to search for binary asteroids. He and Mike Brown of Caltech have recently an- Héctor Cruz (Photo by Tony Acevedo) nounced their first discovery of a main belt asteroid with a small companion. Jean-Luc came to Arecibo on comple- Gersom Ortiz Leaves tion of his Ph.D. in the Department of Rey Vélez Astronomy at Cornell in early 1999. Gersom Ortiz worked as an operator at While continuing a number of studies the observatory since 1998. In addition of the polar regions of the moon, the top- to being a diligent operator, Gersom was ic of his Ph.D. thesis, he immediately a popular figure around the control room became heavily involved in the use of and his artistic work was often on dis- the 13 cm radar system for asteroid stud- play with spoof caricatures of various ies. Not content with the traditional de- staff members. Gersom’s presence in the lay-Doppler technique for imaging control room is preserved on film due to asteroids he started investigating the use his “oscar-winning’’ performance in the of interferometry to improve the radar movie: A Day in the Life of the Arecibo images. The use of other telescopes high- Observatory. Gersom left the observa- lighted the need for a wide band porta- tory to pursue a career in engineering; ble direct sampling system so Jean-Luc he is currently studying full time for his set about developing a four channel 20 Masters’ degree. We wish him all the best MHz bandwidth sampling system using and look forward to seeing him from relatively cheap PC or Sun computers. time to time as a visitor to the control room.

New Telescope Operator: Ángel Davíd Rodríguez José Cruz Ángel joined our operations group as a replacement operator for Gersom in Feb- ruary and has been rapidly learning the ropes in the control room. Ángel Davíd was part of the visitor's center staff since its inauguration in 1997. Ángel brings considerable expertise in software devel- opment to the control room and should be of great help to us in a number of soft- ware-related tasks. Ángel Davíd Rodríguez (Photo by Seth Shostak)

March 2001, Number 32 20 NAIC/AO Newsletter

TO:

*address correction requested correction *address

Ithaca, NY 14853-6801 U.S.A. 14853-6801 NY Ithaca,

Cornell University Cornell

504 Space Science Building Science Space 504 NAIC/Arecibo Observatory Newsletter Observatory NAIC/Arecibo

NAIC/AO Newsletter is published three times per year by the National Astronomy and Ionosphere Center. The NAIC is operated by Cornell University under a cooperative agreement with the National Science Foundation.

Duncan Lorimer and Jonathan Friedman, Editors

Address: NAIC/AO Newsletter HC03 Box 53995 Arecibo, PR 00612 Phone: +1-787-878-2612 Fax: +1-787-878-1861 E-mail: [email protected] or [email protected] WWW: http://www.naic.edu