Research Reports

John Butler has worked with Kieran Hickey and Armin Theissen (PDRAs), research students Enric Pallé Bagó, David García Alvarez, and Anna Maria García Suarez, and meteorology project staff Brenda Morrow, Mark Emerson, and Dierdre McCabe. The main research interests have encompassed the link between the variable Sun and climate, notably identifying a firm link between clouds and cosmic rays; and the flares and angular momentum evolution of late-type stars. During the year John Butler also acted as an external examiner for a PhD submitted to the University of Birmingham, and made a significant contribution to the public understanding of science by means of lectures, talks and media interviews. Towards the end of 2000, he obtained a major Heritage Lottery Fund award to conserve and restore three of the Observatory's historic telescopes and telescope domes.

Clouds and Cosmic Rays

Although evidence for a link between solar activity and climate has been mounting for many years, the physical mechanisms responsible have not been clearly established. The most direct link would be through the known change in the Sun's brightness as a result of the changing solar magnetic field. Such a mechanism can explain about 10-20% of the global warming over the past century. However, a much larger contribution could potentially derive from changes in the albedo and long wavelength radiation of the Earth brought about by changes in cloudiness.

A study of satellite cloud cover factors for the period 1983-1994 has shown that there is a clear correlation between specifically low cloud (1-3km altitude) and the influx of cosmic rays over substantial areas of the Earth. Although the cosmic rays are of Galactic origin, their spectrum and strength are strongly modulated by the solar and interplanetary magnetic fields they encounter en route to Earth. Thus, the connection between the solar magnetic field, cosmic rays and terrestrial clouds is a strong contender as the missing link between solar activity and climate.

We have computed the regression lines of satellite cloud factors and the ionization produced by cosmic rays for a grid of points over the globe and studied the geographical distribution. We find that the correlation is stronger in the eastern hemisphere than the western hemisphere and is stronger in mid-latitudes than equatorial or high-latitude regions. These findings are difficult to explain by the known geographical variation in cosmic-ray flux.

Assuming that the correlations established have been operative over recent centuries, we have computed the net effect of the implied changes in albedo and the Earth's long-wavelength radiation on global temperatures. We find that the predicted changes in cloud factor are capable of explaining much of the current global warming and also the low temperatures observed in the period during the late 17th and early 18th centuries known as `The Little Ice Age'.

A study of ground-based synoptic cloud data has also been made in order to see if the predicted fall in low cloud factors can be seen in historical data. In general, although there are some indications that low cloud is decreasing in the same way that the solar activity - cloud correlation predicts, overall, cloud amounts appear to be increasing. Thus the question of whether or not low cloud has actually changed over past centuries, as predicted, remains unresolved.

Clouds and Sunshine over Ireland

The records of sunshine hours obtained since the late 19th century from four stations (including Armagh) distributed throughout Ireland were analysed. A gradual decrease in sunshine hours has occurred at all four sites since records began. Increasing cloud factors, resulting from enhanced evaporation rates over the Atlantic as sea surface temperatures have risen, is one possible explanation for the decline in sunshine.

A strong negative correlation was confirmed between sunshine factors from ground-based observations and satellite-based cloud factors over Ireland. In addition, it was found that cloud factors over Ireland correlated well with cloud factors over large oceanic areas such as the North Atlantic and mid-high latitudes generally. Thus cloud factors (and similarly sunshine factors) from regions on the boundaries of large oceans which lie in the direction of the prevailing wind could be useful in determining the long-term changes in cloud factors over more extended areas. Knowledge of such long-term variability in the Earth's cloud cover provides important input information for modelling past climate change.

Meteorological Archive

Two grants were obtained during the year, in collaboration with other staff (Martin Murphy, Lawrence Young, Mark Bailey), to establish a Northern Ireland meteorological archive on the world-wide web: one from the Heritage Lottery Fund and the other from the Irish Sailors and Soldiers Land Trust. The object of the project is to make scanned images of the entire Armagh Observatory Meteorological Data Bank available over the internet, in order that the original records can be accessed by educational and research organizations as well as the general public. In addition, calibrated and homogeneous data series will be prepared for maximum and minimum temperatures, rainfall, humidity, sunshine, pressure, cloud factors, and ground temperatures for the Observatory site. This is believed to represent the longest series of homogeneous meteorological data for any site in the UK or Ireland. In addition, temperature data for Dunsink Observatory has been compiled and will be used to bridge a short gap in the Armagh records from May 1825 to December 1832.

Three staff joined the Observatory during the year to work on different aspects of this project, while the growing climate/meteorology team was augmented early in 2001 by the addition of two further staff: Kieran Hickey (PDRA) and student Anna Maria García Suarez. They are to work respectively on calibrating the Armagh climate series and the effect of climate on tree-ring widths, the latter in collaboration with Professor Mike Baillie (QUB).

Late-Type Stars

An escape probability code, originally developed by Stephen Drake, was previously modified and used by D. Jevremovic to give fundamental parameters for flares on Gliese866 based on fits to the observed Balmer decrements. As there were later some doubts as to whether or not the results were realistic, the code has been applied to two other flares, one on the Sun for which direct area information is available, and another on a second dMe star which was independently analysed using another method. In both cases, the escape probability code predicted electron temperatures, electron densities, areas and volumes of the flare plasma close to those found by the other methods. Therefore we are confident that this code is capable of giving reasonable results. It has subsequently been applied to a large flare observed on the star ATMic with the 1.9m telescope at SAAO, and a paper prepared for publication.

John Butler also completed a project on the angular momentum evolution of late-type stars, initiated by the late Brendan Byrne, with PDRA Armin Theissen. Although significant progress was made, particularly on the rich and poorly studied cluster Stock2, the programme could not be completed as originally planned due to difficulties in obtaining the requisite allocations of large telescope time for high-resolution spectroscopy of faint stars.

Restoration of Telescopes and Domes

A grant has been obtained from the HLF for the restoration of three of the Observatory's historic telescopes and domes. This will involve the reconstruction of the original 15-inch Thomas Grubb Equatorial Reflector, the renovation of the Howard Grubb 10-inch Refractor including the provision of a new objective, and the restoration of the Calver/Schmidt telescope to its original Newtonian design. The project will involve the erection of a new dome for the Calver Telescope and the restoration of the Robinson and 1827 Domes. It will be possible for the general public to view the telescopes from a network of wheelchair accessible paths. Site plans have been agreed and planning permission requested.

J.E. Chambers, Research Astronomer

John Chambers has continued to serve as an external supervisor of PhD student Nick Sleep (Open University), together with Mark Bailey and Barrie Jones (Open University), and has worked on extrasolar planets with Monika Kress, Robbins Bell and others of NASA Ames Research Center, California, USA.

The work has focused largely on theoretical studies of extrasolar planets. In particular, he has made a large number of computer simulations to test whether Earth-like planets could form in a wide variety of planetary systems. He has also examined terrestrial-planet formation in known extrasolar systems. The principal conclusion of this work is that stable Earth-like planets can exist orbiting other stars even if these stars have very massive giant planets, provided that the giants lie at least twice as far from the star as the Earth does from the Sun. However, if a star has giant planets moving on highly non-circular orbits, habitable, Earth-like planets are unlikely to exist.

Extending this work with the above collaborators, he has examined whether Earth-like planets in other systems could possess water and organic-rich materials needed to form life. The Earth and other habitable planets probably formed out of very dry material containing little carbon or nitrogen needed to make organics. However, these vital ingredients could have been delivered to a planet when asteroids collided with the planet early in its history. The group has found that this is likely to happen in a range of planetary systems, but by no means all.

Working with Greg Laughlin of NASA Ames, John Chambers has also begun a project to analyse data from observational programmes to detect extrasolar planets. To date, these surveys have found more than 50 low-mass objects orbiting other stars. These objects are most likely to have masses comparable to Jupiter. However, the observational technique used to find them can only place a lower limit on their mass rather than determining their true mass. Laughlin and Chambers are using improved analysis techniques that will make it possible to determine the true masses in many systems for the first time.

John Chambers is also collaborating with Pat Cassen of NASA Ames to make new computer models which will combine theories of planet formation with those of protoplanetary disks. This work is beginning to bridge the gap between these fields and will help improve our understanding of both subjects with important implications for understanding both our own Solar System and planetary systems in general.

J.G. Doyle, Research Astronomer

Gerry Doyle has worked on solar physics with Maria Madjarska (PDRA) and research students Luca Teriaca and Illía Roussev, and in the cool star area with Alex Löbel (a former PDRA), and Darko Jevremovic (PDRA) and David García Alvarez (PhD student).

Solar Plumes

An accurate understanding of the physical conditions in coronal holes is important for constraining models that describe the acceleration of the high speed solar wind. The measurement of line widths can provide information concerning ion temperatures, sub-resolution turbulent motions and velocity fluctuations associated with magnetohydrodynamic (MHD) waves in the corona. Line width variations combined with simultaneous electron density estimates, provides a very powerful diagnostic for the solar corona. Despite the fact that bright plumes are striking features within coronal holes that can extend to more than 30 solar radii from the Sun, the brightest plumes cannot be the source of more than a small percentage of the fast solar wind, because they only account for a tiny fraction of the solid angle subtended by the polar coronal holes. This means that the main source of the fast solar wind lies in the darker inter-plume regions. In order to shed some light on the mechanism responsible for the acceleration of the fast solar wind and to provide constrains for numerical modelling of the coronal holes, we need to know the magnitude and variability of the thermodynamic variables (such as density and temperature), particularly in the inter-plume regions.

Specifically, we have examined spectral time series of the transition region line OV 629Å, observed with the Coronal Diagnostic Spectrometer (CDS) on the SOHO spacecraft in July 1997. Both Fourier and wavelet transforms have been applied independently to the analysis of plume oscillations in order to find the most reliable periods. The wavelet analysis allows us to derive the duration as well as the periods of the oscillations. Our observations indicate the presence of compressional waves with periods in the range 10-25 min. We have also detected a $11 \pm 1$ min periodicity in the network regions of the north polar coronal hole. The waves are produced in short bursts with coherence times of about 30 min. We interpret these oscillations as outwardly propagating slow magneto-acoustic waves, which may contribute significantly to the heating of the lower corona by compressive dissipation and which may also provide enough energy flux for the acceleration of the fast solar wind. The data support the idea that the same driver is responsible for the network and plume oscillations with the network providing the magnetic channel through which the waves propagate upwards from the lower atmosphere to the plumes.

During these observations we also detected by chance a giant macro-spicule at the limb and were able to follow its dynamical structure. Blue and red-shifted emission in the OV line indicates that it is probably a rotating twisted magnetic jet. Emission is also detected in the MgIX 368Å line, at a temperature of 1,000,000K. We also present observations of OVI 1032Å line profiles obtained with the SUMER instrument on SOHO extending from the solar disk to $1.5 R_{\odot}$ above the limb in the north polar coronal hole. Variations of the intensity and line width in the polar plume and inter-plume regions are investigated. We have found an anti-correlation between the intensity and the line width in the plume and inter-plume regions with detailed plume structures being seen out to $1.5 R_{\odot}$ . Possible implications regarding the magnetic topologies of these two regions and related heating mechanisms were discussed. The OVI line width measurements are combined with UVCS/SOHO output to provide an overview of its variations with height extending up to $3.5 R_{\odot}$ . We find a linear increase of the line width from 1 to $1.2 R_{\odot}$ , then a plateau followed by a sharp increase around $1.5 R_{\odot}$ .

Solar UV Explosive Events

Primarily observed in the network lanes at the boundaries of the super-granulation cells, UV explosive events are preferentially found in regions with weak and mixed polarity fluxes that display magnetic neutral lines. However, despite the large amount of observational material on UV explosive events and blinkers, large uncertainties about their basic physical parameters such as electron density and temperature still exist. In this work we explore an electron density diagnostic involving the transition region ion, OV. The derived electron densities in an explosive events are then discussed. In another study, we determine the electron density for a range of solar features using new calculations for the OV line ratio, R=I(761.1Å)/I(760.4Å), in conjunction with observational data obtained with SUMER/SOHO. The densities obtained from this diagnostic are in good agreement with earlier measured values. We conclude from these results that this particular OV ratio is a useful diagnostic for many types of solar features.

Using measured line shifts and electron density enhancements in UV explosive events, an attempt has been made to correlate both observational phenomena, and to associate the observed density enhancements to magnetic reconnection sites. The corresponding local magnetic field strength in these sites is estimated. These values are of the same order as previously measured in photospheric cancelling flux regions.

Solar Spectral Line Broadening

The trend of the line broadening across the disk for optically thin lines has a fundamental importance in constraining which mechanism is operating in heating the corona. Full disk images from SUMER/SOHO taken in HeI, CIV and NeVIII were used to investigate whether there exists a center-to-limb variation in the line width. Both CIV and HeI show such a variation but the higher temperature NeVIII line is relatively constant. For CIV, this corresponds to a $\sim$ 3km s $^{-1}$ difference while HeI is significantly larger particularly at the limb. Whereas this work may suggest that the non-thermal motions are slightly non-isotropic in the transition region and upper chromosphere, with the horizontal unresolved motions exceeding those in the vertical plane, a more probable explanation is that the lines are broadened due to opacity effects. The variation of the CIV 1548Å line width could be explained by increasing the opacity from zero at disk center to $\sim$ 1 at the limb. For HeI the opacity is significantly greater than unity at the limb, implying all mass motions in the chromosphere, transition region and corona are isotropic.

Cool Stars

We modeled the spectral changes of late oxygen-rich Mira variables observed in different pulsation phases. From a combination of variable near-IR spectra and UKIRT spectro-photometry of the $9.7\,\mu$ m silicate dust emission feature in different phases we studied the influence of the changing atmospheric circumstances on the conditions in the circumstellar dust shell. From a detailed modelling of TiO and VO bands in the near-IR spectra, we determined changes of the effective temperature and the effective atmospheric acceleration of the central star. The corresponding model spectral energy distribution is redistributed through the dust shell by means of radiative transfer calculations in order to perform a detailed modelling of shape changes observed in the silicate feature. We showed that the latter were mainly caused by changes in the flux distribution of the incident radiation field with stellar pulsation, whereas intensity changes of the dust emission result from stellar luminosity changes as they are enshrouded by very optically thin dust shells. In the case of the Mira variable, oCet, we computed that the effective temperature increases from $T_{\mbox{\scriptsize {eff}}} = 2,400\,$ K in the minimum phase, to 3,000K $\pm\,$ 100K around the maximum phase. The amplified momentum transfer around maximum light enhances the acceleration of the dust outflow near the dust condensation radius of $\sim$ 6 stellar radii. This produces variations of the terminal dust outflow velocity with phase, by an amount $\Delta v_{\mbox{\scriptsize {infin}}} \simeq 5\,$ km s $^{-1}$ at large distances from the star. The corresponding small changes in flux mean that opacity and gas mass-loss rates, from (2.8-3.2) $\times 10^{-7}\,M_\odot$ yr $^{-1}$ , are sufficient to model the changes in shape observed in the dust emission feature. A comparison with the modelling results for another long-period Mira variable, UOri, has also been completed.

We have also discussed the influence of the non-thermal velocity (micro-turbulence) on the formation of chromospheric lines in the atmospheres of late-type dwarfs. A review of previous work showed a variety of different approaches to the problem leading to different atmospheric structures and consequently different computed line profiles. In that light, we re-examined the formation of the hydrogen Balmer lines and the sodium NaI D lines using twelve different distributions of the micro-turbulent velocity throughout the atmosphere. Our results showed a wide range of possible line shapes. Using the analogy with the solar case and the latest results of the non-thermal component widths as derived from instruments on the SOHO spacecraft we have been able to model H $\alpha$ and the NaI D lines in an active dMe star Gl 616.2.

A Fortran code which computes synthetic light and colour curves of active, spotted stars has been developed. The main feature of this code is that it can simultaneously model the V light curve and the

colour data. It also uses new effective-temperature-colour and Barnes-Evans-like calibrations, temperature and gravity-dependent limb darkening coefficients and different effective surface gravities for the spotted and unspotted photosphere. The code allows for two-component spots, i.e. spots with umbral and penumbral components. Various problematic spot configurations were investigated and we conclude that, in order to be able to differentiate spots with various thermal structures (umbrae, penumbrae, faculae) or polar spots from equatorial bands, the modelling of the infrared colours, especially

and

, is needed.

The active flaring binary CCEri was studied via multi-wavelength observations involving multi-band photometry and ground and space-based spectroscopy. Combining early spectroscopic data with the present implies an orbital period

days. Furthermore, the spectroscopic data suggest spectral types of K7 and M3 for the system. The optical photometry indicated a small spot coverage in late 1989, consistent with data taken a year later that showed CCEri to be entering its brightest-to-date phase. Two flares were detected in the ultraviolet spectral data. These radiated $2.7 \times 10^{24}\,$ J and and $1.6 \times 10^{24}\,$ J in the CIV line alone, each with a total estimated radiative energy budget of $\sim$ $10^{29}\,$ J. For the higher temperature lines, such as CIV, there was no systematic variability with phase. The lower temperature lines show a slight indication of rotational modulation. However, there is a much larger scatter in the individual measurements of the MgII and CIV fluxes than would be expected from measurement errors alone, consistent with an atmosphere showing continual small-scale activity.

C.S. Jeffery, Research Astronomer

Simon Jeffery has worked with Vincent Woolf (PDRA), Regina Aznar Cuadrado (PhD student), Pilar Montañés Rodríguez (PhD student), Lorna Devine (TCD student), and Jacob Samuel (NISTRO scholar). The main area of research involves understanding the late stages of stellar evolution, and the nature and evolution of evolved or burnt-out stars close to the end of their lives as visible objects.

Stellar corpses are the burnt-out remains of stars that have exhausted their reserves of energy and ceased to shine. All that may be seen is a glowing ember as the star gently cools, or the signature of highly energetic material trapped in their enormous gravitational and magnetic fields. Well known examples include black holes, white dwarfs and neutron stars. They are amongst the most exotic objects in the Galaxy, and our mission is to discover how normal and benign stars like the Sun reach such macabre ends. In pursuing it, we also study the origin of elements essential to human life, the physics of matter under extreme conditions, and processes that affect the evolution of entire galaxies.

According to the standard picture, a star like the Sun will eventually swell up to become a red giant as it finishes converting hydrogen to helium. Pausing to convert helium to carbon and oxygen, the expansion then continues until the star sheds its outer layers as a planetary nebula and shrinks to become a white dwarf. This is shown by the solid line in the top right-hand panel of Figure 1.

However, many stars do not fit comfortably into this picture, particularly those that have become so mixed up that even their outer layers have no hydrogen left. These are the helium stars and the hot subdwarfs that together form a major area of the group's research interests.

**Figure 1:** The distribution of stellar luminosity L and temperature T for normal stars divides into several broad bands on the Hertzsprung-Russell diagram (upper left-hand panel). Stars near the top left-hand side of this diagram tend to be of high mass, hot and blue; those near the bottom right-hand side of the diagram tend to be of low mass, cool and red. When nuclear burning occurs at a late stage of evolution, or when a star comes into contact with a close companion, `normal' evolution breaks down. The work of the group focuses on explaining unusual and abnormal stars such as FGSge, V652Her, and RCrB and extreme helium stars.
$\begin{figure} \begin{center} \epsfig{file=csj01.eps,width=14.0cm}\end{center}\par\par\end{figure}$

The problem demands that we study stars that are in transition between hydrogen burning and death. Such phases frequently do not last long -- a few thousand years or less -- and hence such stars are rare. They are also of great interest as examples of stars showing astronomically rapid evolution on time-scales of human concern. For example, observations of one group of luminous extreme helium stars show that they are heating up at rates up to 100 $^\circ$ C per year as they contract to become white dwarfs: a fiery example of stellar global warming!

We must also measure their properties in as much detail as possible. For instance, we would like to know their mass, radius and luminosity at the very least. The chemical composition of their surface layers provides further clues to their history and past evolution, particularly if the material processed by nuclear reactions in their cores has been exposed at the stellar surface. It is also important to know if the star is (or was) one of a pair, a so-called binary star, because such stars can exchange mass with their companion as they evolve.

The team's approach has been to combine high-quality observations with the best possible theoretical models, covering every aspect of stellar structure from the deep interior to the outermost layers of the star's atmosphere. Our theoretical work involves making hypotheses about the origin of given stars. These define boundary conditions for solving the time-dependent equations of stellar structure. Such solutions show long-term evolution in response to changes in chemical composition at different points within the star, and short-term changes (pulsations) in response to instabilities in the energy flow from the star. Our theoretical work also involves the construction of detailed models of the outermost layers of the star and the spectrum of radiation they emit, for detailed comparison with observations.

Major Results

Simon Jeffery and Dr Don Pollacco (QUB) completed an analysis of the radial motion of the surfaces of two pulsating subdwarf B (sdB) stars. Like the Sun, these very evolved stars pulsate non-radially, with a very small amplitude and oscillation periods in the range 2-3 minutes. Unlike the Sun, these stars are very faint, and therefore difficult to observe. Using time-series lasting about 5 hours and comprising approximately 2000 individual observations of each star, several independent modes of oscillation were resolved and the amplitude of the radial motion determined for each case. We also demonstrated that the orbital period of one star, known to be a binary, was between one and three days.

Meanwhile, Regina Aznar Cuadrado and Simon Jeffery completed an analysis of the energy distribution of some 30 sdB stars, using light from the ultraviolet to the infrared. The energy distributions are dominated by the surface temperature of the sdB star and the presence of any cool companion. The fitting software TFIT and BINFIT developed during 1999 was used to derive effective temperatures and relative luminosities for a number of binary sdB stars. The surprising result was that a sizeable fraction of these stars appear to have main-sequence companions, rather than the giant or sub-giant companions previously supposed. This is of significant importance for understanding the evolutionary origin of sdB stars.

An observational programme to measure radius changes in extreme helium stars on both long and short time-scales was also concluded, in collaboration with R.L.C. Starling (a former Armagh summer student, now at Mullard Space Science Laboratory), P.W. Hill (St Andrews) and D. Pollacco (QUB). Approximately 150 spectra obtained with the International Ultraviolet Explorer between 1979 and 1995 were used to measure changes in surface temperature and apparent radius. Over a three-week interval changes due to radial pulsations were measured. These were used to measure directly the masses of radially pulsating helium stars for the first time. The values of $\sim$ $0.9\,M_{\odot}$ are in good agreement with previous estimates and evolution theory.

The second outcome of this work concerned changes over 10-15 years. From quite simple arguments we had predicted heating (or contraction) rates which, depending on the mass of the star, might reach over 100 $^\circ$ C per year. Given the long baseline of our observations, we were able to measure heating rates of between 30 and 120 $^\circ$ C per year for four stars, in excellent agreement with theory. Combining direct observations of pulsation and evolution for the same star (HD168476) places strong constraints on any model for the formation and evolution of extreme helium stars.

TFIT and SFIT

The analysis of stellar spectra is a time-consuming and often subjective task that involves fitting models with many free parameters to observations of varying quality. During 1999, a project to build software for the automatic, efficient and objective analysis of stellar spectra had enabled us to measure properties of both single and binary stars from their observed energy distributions (TFIT and BINFIT). During 2000, this software was extended to the analysis of high-resolution stellar spectra, again for both single and binary stars (SFIT and SFIT_SYNTH).

Other Work

In 1900, FGSagittae (FGSge) was an unknown faint blue star. Since then it has expanded to become a cool giant helium star. With Schönberner (Potsdam), Simon Jeffery has continued work to measure changes in its surface composition during this expansion. This work complements another ongoing project with Pollacco (QUB) to study the chemical evolution of Sakurai's object -- an even more rapidly expanding star that suddenly appeared in 1996. New models have been developed to measure the atmospheric parameters for both stars, which are changing their appearance on astronomically short time-scales, almost literally before our eyes.

Following their successful explanation of the pulsating helium star V652Her as a He+He white-dwarf merger, Hideyuki Saio (Tohoku University) and Simon Jeffery calculated further white-dwarf merger models. These point towards an explanation of extreme helium stars and RCrB stars (a group of stars that fade spectacularly and unexpectedly by factors of 1,000 or more due to the formation of sooty clouds above the surface of the star) as the product of a carbon/oxygen+helium (CO+He) white-dwarf merger.

M.D. Smith, Research Astronomer

Turbulence in Star-Forming Regions

Turbulence is a key concept in modern science. Astrophysical turbulence has fascinating properties and features produced by combinations of extreme physical and dynamical conditions. In particular, turbulence is dominant in molecular clouds, the nurseries for the stars. This implies that an understanding of the origins of stars and stellar systems will require a theory for molecular turbulence. Publication of the analysis of simulations of three-dimensional magnetohydrodynamic self-gravitating supersonic turbulence represents an advance in our understanding of the dynamical processes during star formation.

Unification Scheme

The Unification Scheme is designed to model the evolution of protostars and their environments. The Unification Scheme, drawn out by Michael Smith, was reviewed in the Irish Astronomical Journal and applied in collaboration with others to one of the most well-studied star formation regions in Orion. The synthesis with observations continued with a comparison of data from the Infrared Space Observatory for many very young protostars. Specifically, we were able to model the infrared emission from shock waves produced by jets of fluid from the protostars and demonstrate unequivocally that an evolutionary sequence is consistent across all sets of observations. The results were presented at the meeting entitled ``Star Formation 2000'' at Ringberg Castle in Bavaria.

Star Formation

The formation of massive stars appears to follow different rules from those applying to low-mass stars such as our Sun. Michael Smith began a collaboration during a visit to the Max-Planck-Institut für Radioastronomie in Bonn in August 2000. Speckle interferometry data for one object, S140, allowed us to resolve sub-arcsecond structure and reveal the presence of multiple outflows.

A star forms through the collapse of molecular clumps, termed pre-stellar cores. What triggers a core to collapse? Michael Smith demonstrated that core collapse can be triggered by protostars which form in neighbouring collapsed cores. The protostar ejects bullets and jets which are likely to hit surrounding pre-stellar cores if the population of cores is sufficiently dense. These results were presented at a workshop entitled ``The Modes of Star Formation'' in Heidelberg during September 2000.

The doctoral project of Tigran Khanzadian will be based on observations of star formation regions. Classical optical telescopes reveal dark clouds in our Galaxy, but cannot reveal what is going on inside. Infrared observations, however, provide a glimpse of where stars are forming internally. Moreover, we can catch much more than a glimpse since modern detectors called `wide-field cameras' can observe extremely wide areas. Telescopes of 3-4 metres aperture, focused on a cloud for thousands of seconds, then reveal a detailed pattern of protostars and their activity. Two observing trips to Calar Alto have resulted in an abundance of data. We also obtained time to observe in detail six specific bow-shaped shocks produced by protostars. The idea was to use the UKIRT telescope (Hawaii) with a high-resolution camera (UFTI) to get data which contains information on the physical nature of the shocks (i.e. the images of emission lines from vibrating molecules). Unable to attend the observing run, Chris Davis (UKIRT) performed the observations and has sent the data to Khanzadian.

Several papers on infrared observations and their interpretation were also published during the year. These were studies of the outflows from protostars, including proper motions of clumps over several years which reveal remarkable speeds, the first evidence for the detection of spinning jets, detailed modelling of the ISO data for CepheusA, very high resolution images of Herbig-Haro Object HH1, and a detailed study of the spatial distributions of the excitation of molecules in several outflows.

Origin2000 Supercomputer

A supercomputer was delivered to the Observatory in April 2000, thanks to the JREI programme, funded by PPARC and with SGI as a partner. The arrival of the supercomputer, unique to Armagh, was greeted with much media interest. The `Forge' is an 8-processor Origin2000. It is a supercomputer because the processors can communicate extremely fast, not only with each other but with large banks of data that they need to continuously update. The group immediately began work on three-dimensional turbulence calculations, involving cubes with 352 zones per side, and has now started to simulate molecular jets. The machine will be used to prepare even bigger computational problems which can be run on machines at the UK Astrophysical Fluid Facility (UKAFF), the national fluid simulation facility, which possess 128 processors. Moving images of some of these astrophysical jet calculations are available from the Armagh Observatory web-site (star.arm.ac.uk/~mds/ukaff.html). The UKAFF officially opened in November 2000.

Besides the supercomputer, of value £251,000, there were other grant successes. Dr Alex Rosen, with 3-year PPARC funding, and student Mr Georgi Pavlovski arrived to boost the research on numerical simulations of star forming regions. Furthermore, Michael Smith led a successful bid for a PPARC Theory Visitor Grant which will allow us to bring in computing experts.

Students from schools and universities were able to get the flavour of research in the area of star formation as well as the general nature of Observatory life. Shane Lynch from Trinity College, Dublin, did a 3-month project on Molecular Shocks and Herbig-Haro Objects. Matthew Davis (Friend's School, Lisburn) discovered what could be new Herbig-Haro objects by searching an archive of infrared observations. Michael Smith was a PhD examiner for the thesis of Thomas Stanke (Potsdam). He also updated the research pages for the Observatory's web-site and contributed an article to the annual Irish Scientist.

M.E. Bailey, Director

Mark Bailey worked with David Asher (PDRA), Scott Manley (PhD student) and Sandra Jeffers (MPhil student). Sandra Jeffers used and modified a collision code originally produced by Scott Manley to calculate impact velocities and collision probabilities of comets and various classes of near-Earth asteroid on the Earth and the Moon, and hence determined the frequency distribution of terrestrial and lunar craters according to different assumptions about the population of the different classes of projectile. Sandra Jeffers obtained her MPhil ``Collisional Processes in the Inner Solar System'' (co-supervised by Dr Alan Fitzsimmons, QUB) in July 2000.

Nature of the K/T Projectile

The most famous example, namely the Cretaceous-Tertiary (K/T) boundary, is associated with both an exceptionally large crater (the Chicxulub crater, in the Yucatan peninsula, Mexico), approximately 180km in diameter, and an anomalously high abundance of iridium (Ir) and other platinum group elements. Iridium is not found to any significant degree elsewhere in the Earth's crust, owing to its siderophile nature, but is believed to be a minor constituent of small, primitive bodies such as comets and asteroids. The presence of a large crater together with the anomalous Ir layer at the K/T boundary provide direct evidence associating the mass extinction with a major impact event. The question is, was the impact caused by a comet or an asteroid?

Sandra Jeffers, Scott Manley, David Asher and Mark Bailey, have addressed this question by considering possible constraints on the properties of hypothetical impacting bodies that might produce both the observed amount of iridium and a crater the size of Chicxulub. Assuming that comets and asteroids contain the usual cosmic abundance of iridium, a body at least 6km in diameter is indicated. It is usually assumed that this was an asteroid, but asteroids this large on Earth-crossing orbits are very rare, and probably only run into the Earth on the average once every 300 million years.

The alternative hypothesis is that the projectile was a comet (or cometary fragment), and although many comets are much bigger than typical near-Earth asteroids, on the whole they also collide with the Earth much faster than the $\sim$ 20km s $^{-1}$ associated with asteroids. The difficulty with the cometary hypothesis is that such high-velocity collisions (at mean impact velocities greater than 50km s $^{-1}$ ) completely destroy the incoming body and produce a plume of high-temperature ejecta that escapes the gravitational pull of the Earth. A typical cometary impact, while making a large crater, will tend to leave very little trace of its original make-up on the Earth, in particular it will leave very little iridium.

Our analysis considered both the size and velocity distribution of the different classes of impacting bodies, and estimated the frequency of impacts, both cometary and asteroidal, large and slow enough to deposit sufficient iridium to match observations. The results indicate that an impact by either a Halley-type or a long-period comet could not have caused the K/T extinction, as these bodies arrive far too fast and therefore leave little or no iridium. However, some short-period comets arrive more slowly, and considering their size distribution sufficiently large, slow comets hit the Earth more frequently than similar-sized asteroids. So, by a factor of approximately 3 to 1, it appears that a comet rather than an asteroid probably killed the dinosaurs.

Some caution has to be expressed in the interpretation of this result. First, the geological events surrounding the K/T boundary are exceptionally complex, including climate change (global cooling), volcanism (the outpourings of lava associated with the Indian Decan Traps), and changes or regressions in sea level. As the geologist R.E. Sloan has remarked: ``When you see the late Cretaceous, you figure Murphy was an optimist -- everything went wrong at almost the same time!''.

However, in favour of a more complex cometary picture, instead of the random `stray' asteroid hypothesis, one can also point to evidence for a high abundance of interplanetary dust both below (i.e. before) and above (i.e. after) the K/T boundary layer, suggesting (cf. Bill Napier's work, below) that an enhanced zodiacal cloud may also have been present at that time, perhaps produced by the break-up of a giant comet in a short-period orbit. Moreover, the K/T event sits exactly on one of the approximately periodic 30Myr peaks in the extinction record, and this can be explained as the effect of an increase in the number of short-period comets arising from an increased long-period comet flux from the Oort cloud at that time, owing to the passage of the Sun through the Galactic plane in its orbit around the Galaxy. Our work shows that, on balance, the K/T projectile was more likely to be a comet or cometary fragment than an asteroid, and suggests that a more complete picture of both comets and asteroids is crucial to obtaining a full understanding of the evolution and history of life on Earth.

Near-Earth Objects (NEOs)

It is now recognized that most species of primate are extinct or about to become so, and that in the event of mankind being extinguished there would be little prospect of intelligent life re-appearing on Earth for hundreds of millions of years, if ever. Should intelligent life be a rarity (and the string of special circumstances necessary for mankind to arise on Earth, not least the extinction of the dinosaurs, suggests that it is), such an extinction would make the difference between a Galaxy teeming with human-descended civilizations within a few million years, and one devoid of intelligent life.

These are important issues, and within the past twenty years, near-Earth objects (NEOs) have been identified as the greatest celestial hazard faced by mankind. Every year there is a 1 in 100,000 chance that a one-km diameter object will collide with Earth, causing global catastrophe and over a billion deaths. For the UK alone, that means millions of fatalities. Even the smaller, so-called `Tunguska' objects are believed to hit the Earth every 100 years or so, producing an actuarial risk (again for the UK alone) of several million pounds p.a. These low-frequency, high-consequence events exceed any risk the UK accepts from industries such as nuclear power or dangerous goods transport.

During the summer of 1999 the Government accepted the seriousness of the impact hazard and set up a special commission (the NEO Task Force) to investigate the potential threat posed by comets and asteroids on Earth-approaching orbits. Mark Bailey was invited to review the science of NEOs at one of the first meetings of the Task Force, in January 2000. The Task Force subsequently took expert advice from a wide range of sources (both domestic and abroad), and produced an influential report on the subject (``Report of the Task Force on Potentially Hazardous Near Earth Objects''), published in September 2000.

The Government's response to the NEO Task Force Report, providing comments on each of its 14 key recommendations, was published on 24 February 2001. Political interest in NEOs remains high, and the initial response was followed soon afterwards (7 March 2001) by a debate on the subject in the House of Lords.

What seems perhaps of most significance in these developments is the sea-change of opinion within government circles about the reality of the NEO threat, and the suggestion in the official response that the UK is preparing to play a major role in international efforts to solve the NEO problem. The government has requested a number of organizations to provide background information on the NEO problem (e.g. asking PPARC to provide costings to implement some of the Task Force recommendations) and has promised a further response on the Task Force report later in 2001.

Scientific Administration

Mark Bailey continued to serve as Chair of the Astronomical Science Group of Ireland (ASGI) and as a member of the Royal Irish Academy National Committee of Astronomy and Space Research, the Royal Astronomical Society Education Committee, the Governing Board of the DIAS School of Cosmic Physics, and scientific working groups of IAU Commissions 15 and 20 (both concerning comets and minor planets). He was also invited to advise the Government Task Force on Potentially Hazardous Near-Earth Objects, and others, on various aspects of the NEO risk to civilization, and served as Editor-in-Chief of the journal Earth, Moon, and Planets.

W.M. Napier, Senior Research Fellow

Bill Napier has continued to work on the climatological effects of the accretion of cometary and interplanetary dust on the Earth, and (with Geoffrey Burbidge, University of California, San Diego) on a puzzling anomaly in the redshift distribution of galaxies and distant quasars.

Quasar Redshift Distribution

It has long been known that the redshift distribution of quasars is spiky. For example, distinct peaks have been noted at around z=0.6, 0.96, 1.41 and 1.96. In 1977 it was further claimed that these peaks form a periodic sequence $A+0.089n\log(1+z)$ , where

is a constant. Even more remarkably, it was stated that this periodicity only applied to QSOs which were angularly close to spiral galaxies, say within 40 arcmin. Such a phenomenon is not expected in standard cosmological models, and the claims have generated much controversy. The main problem is that quasars or quasi-stellar objects (QSOs) tend to be discovered preferentially in particular redshift ranges, due to instrumentation and emission-line effects, and the peaks might simply be due to discovery selection effects.

A project to examine this issue was started in 1999, jointly between Bill Napier and Geoffrey Burbidge. New samples of QSOs and rigorous statistical procedures were used throughout.

The first sample comprised 57 redshifts from all known close pairs of QSOs with image separations less than 10 arcsec; the second consisted of 39 QSOs selected through their X-ray emission and proximity to nearby active galaxies; and the third consisted of the 78 QSOs from the 3C and 3CR radio galaxy catalogues. For the first sample, discovery selection effects were controlled by comparing the results with QSOs selected from the general field, while the X-ray and radio QSOs in the second and third samples were not subject to spectroscopic or other discovery selection effects capable of inducing spurious peaks.

Extensive statistical testing, making use of Monte Carlo simulations, was carried out on these datasets, separately and together. Clear evidence was found that the claimed logarithmic periodicity is present in all the samples. The combined datasets confirm the periodicity at a significance level $\sim$ $10^{-5}$ . Further, whereas the periodicity had previously only been seen out to

, i.e. at red-shifts

, the new datasets extended the sequence, periodic in $\log(1+z)$ with peaks separated by 0.089, to higher redshifts

and 4.47. These had been predicted by the formula but never seen before. The long-standing (and controversial) periodicity claim has therefore been confirmed, apparently at a very high confidence level.

Its interpretation remains uncertain. Assuming that the QSO redshifts are primarily cosmological in origin and due to the expansion of the Universe, then the periodicity might be due to large-scale cellular structure, like the system of voids, filaments and walls seen in the distribution of galaxies. In that case the 0.089 coefficient implies a separation between QSOs of around $\sim 500 h^{-1}$ Mpc, where

is the dimensionless Hubble constant $h = H_0/100\,$ km s $^{-1}$ Mpc $^{-1}$ . In a vacuum-dominated Universe, quasi-periodic oscillations in $\log(1+z)$ are predicted to occur, and it may be significant that recent observations of high-redshift supernovae indicate that the energy density of the Universe is indeed dominated by a vacuum term. The constancy of the QSO periodicity to the highest redshifts could then be used to constrain the evolution of the scale factor of the Universe, i.e. to give independent information back to the time of formation of the first stars.

This explanation does not, however, account for the fact that the periodicity is observed only in quasars that happen to lie close to active galaxies or other quasars. Unless an explanation for this latter aspect can be found, it may be that a non-cosmological interpretation of these quasar redshifts is required. Whatever the ultimate explanation, this work (Astron. J. 121, 21-30, 2001) clearly has the potential either to illuminate or confuse the canonical cosmological story.

Gamma-Ray Bursters

A similar exercise was carried out, again jointly with G. Burbidge, to investigate those $\gamma$ -ray bursters whose sources have known redshifts. The sample was quite small (19 out of 500 bursters), and shows no evidence for redshift-periodicity.

Zodiacal Cloud and Climate

The terrestrial planets orbit through a flattish disc of dust in the inner solar system. In favourable conditions this cloud can be seen from the Earth as the zodiacal light, lying along the ecliptic plane. Such dust discs are seen also around many other nearby stars. The evolution of this dust cloud is complex: dust enters the cloud through the disintegration of comets and asteroids, and leaves it by spiralling into the Sun, ejection from the planetary system or destruction by mutual collisions between the particles.

A computer model describing these effects has been developed and applied to examine the temporal behaviour of the cloud. The principal findings are that the zodiacal cloud is highly variable on relatively short time-scales and that the cloud is probably not currently in equilibrium. This work will be a springboard for an investigation into how the Earth's climate responds to prolonged, heavy dustings, which the model predicts should occur on time-scales of tens to hundreds of thousands of years. This may be relevant to the overall assessment of the NEO hazard, and to the ultimate cause of ice-ages.

D.J. Asher, Research Fellow

David Asher continued to work on meteoroid streams, notably the Leonids and June Bootids, extending existing collaborations with Professor Vacheslav Emel'yanenko (South Ural University) and Robert H. McNaught (Australian National University). In addition his work covered a number of other areas of dynamical astronomy, notably the dynamical evolution and origin of near-Earth asteroids (NEAs). In agreement with results of previous authors, he and Mark Bailey found that many observed NEAs have the potential to fall into the Sun on timescales below 1Myr, for a variety of reasons. These include the strong action of Jovian mean-motion resonances, and the interaction (also known to affect orbits such as that of the short-period comet 2P/Encke) of the $\nu_5$ and $\nu_6$ secular resonances.

Perhaps one of the most interesting questions about solar system small bodies is where NEAs came from, and how, if some of them are extinct comets, they can now be entirely within and separate from Jupiter's orbit. David Asher has previously collaborated extensively with Dr Duncan Steel (University of Salford), and together with Mark Bailey extended earlier studies by a former Armagh PDRA, Nathan Harris, and by Steel and Asher, on whether the cometary non-gravitational force could play a significant role in the decoupling of orbits from Jupiter. The new work was able to demonstrate, in a more direct way than before, how the process of decoupling from Jupiter operates, for realistically sized non-gravitational forces. It is hoped in future to extend this work to look at the importance of Sun-grazing states in decoupling.

In addition, David Asher made the first of what will be several visits to the Bisei Spaceguard Center (BSGC) in Japan, to assist in setting up the new project there. When it becomes fully operational, the BSGC will be one of the world's leading observatories for the discovery of NEAs, with a 1 metre telescope and CCD camera covering a 3 degree x 3 degree field of view arriving in 2001; present observations use two smaller telescopes. Asher's role there includes the development of techniques for automated processing of the asteroid data, and will strengthen links between Spaceguard UK and the Japan Spaceguard Association, and between Armagh and international Spaceguard work.

Research during the year has resulted in several research papers, as well as making a large contribution to the public understanding of science by means of popular articles, talks and media interviews.

M. de Groot, Consultant Research Associate

Mart de Groot completed a comprehensive investigation into the evolution of the luminous blue variable star PCygni over the past 400 years, and organized a major international workshop at Armagh to celebrate the 400th anniversary of the discovery of this unusual star. The conference: ``PCygni 2000: Four Hundred Years of Progress'', which involved more than 30 astronomers from more than a dozen countries, was held during 21-23 August 2000. The proceedings of this meeting, the Observatory's millennial conference, are being edited by Drs Mart de Groot and Chris Sterken (Brussels), and will be published during 2001.

The conference took place in the Royal School Armagh. In addition to a welcome reception on the evening of Monday 21 August hosted by Councillor James Clayton, Mayor of Armagh and City District Council, participants visited the Argory, where they enjoyed a tour of that historic building followed by a meal and traditional music.

Mart de Groot, eighth Director of Armagh Observatory (1976-1994), also took the occasion of the conference to announce his formal retirement from astronomy in order to devote more of his time to pastoral duties associated with his church. Mark Bailey replying on behalf of the Observatory management and staff, emphasized that Mart will be very welcome to continue to visit the Observatory, and made a small presentation to mark his contributions to the Observatory during the six years following his retirement from post.