Deep Space 1 photometry of the nucleus of Comet 19P/Borrelly

The NASA-JPL Deep Space 1 Mission (DS1) encountered the short-period Jupiter-family Comet 19P/Borrelly on September 22, 2001, about 8 days after perihelion. DS1's payload contained a remote-sensing package called MICAS (Miniature Integrated Camera Spectrometer) that included a 1024 square CCD and a near IR spectrometer with ∼12 nm resolution. Prior to its closest approach of 2171 km, the remote-sensing package on the spacecraft obtained 25 CCD images of the comet and 45 near-IR spectra (L. Soderblom et al., 2002, Science 296, 1087-1091). These images provided the first close-up view of a comet's nucleus sufficiently unobscured to perform quantitative photometric studies. At closest approach, corresponding to a resolution of 47 meters per pixel, the intensity of the coma was less than 1% of that of the nucleus. An unprecedented range of high solar phase angles (52-89 degrees), viewing geometries that are in general attainable only when a comet is active, enabled the first quantitative and disk resolved modeling of surface photometric physical parameters, including the single particle phase function and macroscopic roughness. The disk-integrated geometric albedo of Borrelly's nucleus is 0.029±0.006, comparable to the dark hemisphere of Iapetus, the lowest albedo C-type asteroids, and the uranian rings. The Bond albedo, 0.009±0.002, is lower than that of any Solar System object measured. Such a low value may enhance the heating of the nucleus and sublimation of volatiles, which in turn causes the albedo to decrease even further. A map of normal reflectance of Borrelly shows variations far greater than those seen on asteroids. The two main terrain types, smooth and mottled, exhibit mean normal reflectances of 0.03 and 0.022. The physical photometric parameters of Borrelly's nucleus are typical of other small dark bodies, particularly asteroids, except preliminary modeling results indicate its regolith may be substantially fluffier. The nucleus exhibits significant variations in macroscopic roughness, with the oldest, darkest terrain being slightly smoother. This result suggests the infilling of low-lying areas with dust and particles that have not been able to leave the comet. The surface of the comet is backscattering, but there are significant variations in the single particle phase function. One region exhibits a flat particle phase function between solar phase angles of 50° and 75° (like cometary dust and unlike planetary surfaces), suggesting that its regolith is controlled by native dust rather than by meteoritic bombardment.     

Visible Lightcurve Observations of Comet 19p/Borrelly

The final Deep Space 1 (DS1) mission target, comet 19P/Borrelly, was observedfrom July 28–August 1, 2000 at the CTIO-1.5 m telescope in the R filter. Theobserved lightcurve has a large peak to peak variation between 0.84 mag and1.0 mag. A period of 26.0 ± 1 hr (assuming a double-peaked lightcurve)was found using all five nights. This is in good agreement with the period of25.02 ± 0.5 hr quoted by Lamy et al. (1998) using only 6 points ofHST data and is also consistent with HST data taken around the DS1 encountertime by Weaver et al. (2002).Using the mean magnitude R = 20.8 mag and assuming a 4% albedo, we derivean effective nuclear radius of 2.6 km. The large lightcurve amplitude translates toa long to intermediate axial ratio a/b ≥ 2.2, in excellent agreement with theHST result of a/b ≥ 2.4 (Lamy et al., 1998) and with DS1 images (Soderblom et al., 2002).     

Spin Axis Direction of Comet 19p/Borrelly Based on Observations from 2000 and 2001

We calculate the direction of the rotational angular momentum vector,M, of comet 19P/Borrelly based on rotational lightcurve data from 2000, groundbased imaging of the coma during the Deep Space 1encounter, and the basic near-nucleus coma morphology as revealed by the Deep Space 1 spacecraft. For the most likely direction, we derivea family of solutions (with center at RA = 221°, Dec = -7°) if the direction of M is towards the sunward hemisphere during the Deep Space 1 encounter, whereas if the rotation is of opposite sense, the diametrically opposite family of solutions (with center at RA = 41°, Dec = 7°) would result. We argue that the coma morphology in September 2001 is consistent with the nucleus being a principal axis rotator or one observationally indistinguishable from it. Therefore, for all practical purposes, the direction of the rotational angular momentum vector coincides with the spin axis. We also discuss why the determination of the spin axis direction based on observations from the last apparition is in disagreement with the current result.     

Deep Space 1 at comet 19P/Borrelly: Magnetic field and plasma observations

On September 22, 2001 the Deep Space 1 spacecraft performed a flyby at comet 19P/Borrelly at a solar distance of 1.36 AU leading the Earth by 74^° in longitude. The spacecraft-comet distance at closest approach was 2171 km. The bow shock had a magnetic compression ratio of 2.5 at a distance of 147 100 km from the nucleus. Deep Space 1 first entered the sheath region essentially from the north polar region. Fluctuations from the cometary ion pickup were present throughout the sheath region and even well upstream of the shock, as expected. The magnetic field pileup region had a peak field strength of 83 nT and was shown to be consistent with a pressure equal to the solar wind ram pressure. The peak field location was offset from the time of closest approach. It is uncertain whether this is a spatial or temporal variation. Draping of magnetic fields around the nucleus was sought, but evidence for this was not apparent in the data. A possible explanation is that the interplanetary solar wind was composed of turbulent short-scale fields, and thus the fields were not symmetric about the point of closest approach. During the flyby phase there were in general few intervals of ACE data where there were large scale Parker spiral fields. With the addition of plasma data, the shock properties are investigated. The characteristics of magnetic draping, pileup and fluctuations are explored. These comet 19P/Borrelly results are contrasted with other cometary flyby results.     

The outgassing and composition of Comet 19P/Borrelly from radio observations

Radio spectroscopic observations of Comet 19P/Borrelly were performed during the 1994 apparition and at, and near, the time of the Deep Space 1 flyby in 2001. HCN, CS, CH₃OH, and H₂CO were detected using the 30-m telescope of the Institut de Radioastronomie Millimétrique and the James Clerk Maxwell Telescope, and their production rates relative to water are estimated to be 0.06-0.11, 0.07, 1.7, and 0.4%, respectively. Only upper limits are derived for H₂S and CO. The upper limit for CO/H₂O (<15%) is not very constraining, while the upper limit for the H₂S/H₂O ratio of 0.45% is near the bottom of the range of values measured for other comets. Observations of the OH radical at the Nançay radio telescope provide water production rates a few weeks before the 1994 and 2001 perihelia. Observations of the 1₁₀-1₀₁ water line at 557 GHz with the Odin satellite yield a water production rate of (2.5±0.5)×10²⁸ s⁻¹ on September 22, 2001, at the time of the Deep Space 1 encounter, and (3.3±0.6)×10²⁸ s⁻¹ averaged over the September 22-24, 2001 period. The line shapes are asymmetric and blueshifted by V₀∼−0.18 km s⁻¹ for the best observed HCN lines recorded one week after perihelion. The HCN line shapes, and the similar OH and HCN velocity shifts over the September-November 1994 and August-September 2001 periods, favor anisotropic outgassing towards the Sun. Strong outgassing directed along the primary dust jet seen on visible images is not excluded by the HCN line shapes, but unrealistically high gas expansion velocities are required to explain the line shapes in that case.     

Photometric analysis and disk-resolved thermal modeling of Comet 19P/Borrelly from Deep Space 1 data

Images returned from the Deep Space 1 (DS-1) spacecraft during its encounter with Comet 19P/Borrelly are used to study its disk-integrated and disk-resolved photometry and its thermal properties. A disk-integrated phase function was constructed from a combination of DS-1 images and ground-based observations, giving a geometric albedo of 0.072±0.020 and a phase slope of 0.043 mag deg⁻¹. The shape model of Borrelly [Kirk, R.L., Howington-Kraus, E., Soderblom, L.A., Giese, B., Oberst, J., 2004a. Icarus 167, 54-69] and the ephemerides of DS-1 were used to analyze the disk-resolved photometric data with Hapke's theoretical model. It was found that the surface of Borrelly displays large photometric heterogeneities in its photometric parameters. The single-scattering albedo, w, varies by a factor of 2.5 with an average of 0.057±0.009; the asymmetry factor, g, ranges from almost isotropic (−0.1) to strongly backscattering (−0.7) with an average of −0.43±0.07; the roughness parameter, , is less than 35° for most parts of surface but ranges up to 55° in some areas. Its average is 22°±5°. The observed 1-D temperature profile is modeled well by the standard thermal model (STM) for inactive regions and is found to be consistent with a very low thermal inertia. Water sublimation in the source region of the fan jet is observed to decrease the surface temperature from the STM predictions by 20-40 K. The source areas of two collimated jets could not be determined from either photometric model or thermal model. It is evident that the fan jet activity occurring on Borrelly's surface can be correlated to areas of relatively high albedo, weak backscattering, and high roughness.     

The Deep Space 1 Encounter with Comet 19p/Borrelly

NASA's Deep Space 1 (DS1) spacecraft successfully encountered comet 19P/Borrelly near perihelion and the Miniature Integrated Camera and Spectrometer (MICAS) imaging system onboard DS1 returned the first high-resolution images of a Jupiter-family comet nucleus and surrounding environment. The images span solar phase angles from 88° to 52°, providing stereoscopic coverage of the dust coma and nucleus. Numerous surface features are revealed on the 8-km long nucleus in the highest resolution images(47–58 m pixel). A smooth, broad basin containing brighter regions and mesa-likestructures is present in the central part of the nucleus that seems to be the source ofjet-like dust features seen in the coma. High ridges seen along the jagged terminator lead to rugged terrain on both ends of the nucleus containing dark patches and smaller series of parallel grooves. No evidence of impact craters with diameters larger thanabout 200-m are present, indicating a young and active surface. The nucleus is very dark with albedo variations from 0.007 to 0.035. Short-wavelength, infrared spectra from 1.3 to 2.6 μm revealed a hot, dry surface consistent with less than about10% actively sublimating. Two types of dust features are seen: broad fans and highlycollimated “jets” in the sunward hemisphere that can be traced to the surface. The source region of the main jet feature, which resolved into at least three smaller “jets” near the surface, is consistent with an area around the rotation pole that is constantly illuminated by the sun during the encounter. Within a few nuclear radii, entrained dustis rapidly accelerated and fragmented and geometrical effects caused from extended source regions are present, as evidenced in radial intensity profiles centered on the jet features that show an increase in source strength with increasing cometocentric distance. Asymmetries in the dust from dayside to nightside are pronounced and may show evidence of lateral flow transporting dust to structures observed in the nightside coma. A summary of the initial results of the Deep Space 1 Mission is provided, highlighting the new knowledge that has been gained thus far.     

Imaging Borrelly

The nucleus, coma, and dust jets of short-period Comet 19P/Borrelly were imaged from the Deep Space 1 spacecraft during its close flyby in September 2001. A prominent jet dominated the near-nucleus coma and emanated roughly normal to the long axis of nucleus from a broad central cavity. We show it to have remained fixed in position for more than 34 hr, much longer than the 26-hr rotation period. This confirms earlier suggestions that it is co-aligned with the rotation axis. From a combination of fitting the nucleus light curve from approach images and the nucleus' orientation from stereo images at encounter, we conclude that the sense of rotation is right-handed around the main jet vector. The inferred rotation pole is approximately perpendicular to the long axis of the nucleus, consistent with a simple rotational state. Lacking an existing IAU comet-specific convention but applying a convention provisionally adopted for asteroids, we label this the north pole. This places the sub-solar latitude at ∼60° N at the time of the perihelion with the north pole in constant sunlight and thus receiving maximum average insolation.     

New measures of the orientation of the linear structure in the sunward hemisphere of the coma of Comet 19P/Borrelly: apparitions of 1994, 2001, and 2008

The sky‐projected orientation (position angle) of the axis (line of maximum density or maximum brightness) of the long time‐known, linear structure (LS) in the sunward hemisphere of the coma of Comet 19P/Borrelly is measured on 45 photographs taken by different observers under different projection conditions and covering three consecutive apparitions (1994, 2001, and 2008) for a total time interval of 5174 days. The analysis of the results by a tomographic approach yields an LS axis constantly oriented towards a fixed point in the space, at Right Ascension 214°.4 ± 0°.5 and Declination –7°.0 ± 0°.5 (J2000), corresponding to an obliquity of 103°.5 ± 1° and an orbital longitude of 147°.2 ± 1°, throughout the relevant interval. Such coordinates are close to the ones found by other authors for the spatial orientation of the nucleus spin axis during the apparitions of 1994 and 2001. In the hypothesis of an LS orientation aligned with the nucleus spin axis, the new results confirm the previous ones and show that this orientation remained unchanged during the subsequent 2008 apparition (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

A 3-D global MHD model for the effect of neutral jets during the Deep Space 1 Comet 19P/Borrelly flyby

The Deep Space 1 (DS1) spacecraft passed the sunward side of Comet 19P/Borrelly in 2001. Along its relatively north-south orbit, a set of plasma density and velocity measurement revealed a northward shift of the plasma boundaries and the mass loading peak. Both onboard and ground based telescopes found evidence for asymmetric distribution of the dust and neutrals. In this paper, five mass-loading patterns are studied to present the first study of the effect of non-spherical neutral distribution profiles on the solar wind-cometary plasma interaction environment. Using magnetohydrodynamic simulations, it is found that a combination of Gaussian and cosine neutral jet distribution, with cosine being the major part, can fit the DS1 general plasma measurement well, with a total gas production rate of around 5×¹⁰²⁸ s⁻¹. These model-data comparisons indicate that the general plasma distribution around Comet Borrelly can be explained with its aspherical neutral jet distribution. However, such neutral jets by themselves are insufficient to produce the density offset in the central peak. Kinetic effects, such as finite gyroradius may be required to create the offset plasma peak.     

Thermophysical Modelling of Comet P/Borrelly Effects of Volume Energy Absorption and Volume Sublimation

In this work, we continue revising the theoretical basis ofnumerical models describing the transport of matter andenergy inside a porous dust-ice mixture at low temperature. Amodel of a light-absorbing near-surface layer of a comet nucleus isinvestigated. Gas transport is considered simultaneously with thesolution of the general heat transfer equation. Thequasi-stationary temperature distribution and the H₂O massflux and sublimation rate are computed for a nucleus model ofcomet 19P/Borrelly at the Deep Space 1 (DS1) encounter. Theenergy is deposited in a layer of about 20 particle radii: Thiscorresponds to a solid-state greenhouse effect. The surfacetemperature of the layer-absorbing model as well as the gasproduction rate are significantly smaller than the ones in thesurface-absorbing model. An active fraction of 40–50% would berequired to explain the observed water production rate ofP/Borrelly with our layer-absorption model at the time of the DS1encounter.     

Short-wavelength infrared (1.3-2.6 μm) observations of the nucleus of Comet 19P/Borrelly

During the last two minutes before closest approach of Deep Space 1 to Comet 19P/Borrelly, a long exposure was made with the short-wavelength infrared (SWIR) imaging spectrometer. The observation yielded 46 spectra covering 1.3-2.6 μm; the footprint of each spectrum was ∼160 m × width of the nucleus. Borrelly's highly variegated and extremely dark 8-km-long nucleus exhibits a strong red slope in its short-wavelength infrared reflection spectrum. This slope is equivalent to J-K and H-K colors of ∼0.82 and ∼0.43, respectively. Between 2.3-2.6 μm thermal emission is clearly detectable in most of the spectra. These data show the nucleus surface to be hot and dry; no trace of H₂O ice was detected. The surface temperature ranged continuously across the nucleus from ?300 K near the terminator to a maximum of ∼340 K, the expected sub-solar equilibrium temperature for a slowly rotating body. A single absorption band at ∼2.39 μm is quite evident in all of the spectra and resembles features seen in nitrogen-bearing organic molecules that are reasonable candidates for compositional components of cometary nuclei. However as of yet the source of this band is unknown.     

Radio observations of comet 9P/Tempel 1 with the Australia Telescope facilities during the Deep Impact encounter

We present radio observations of comet 9P/Tempel 1 associated with the Deep Impact spacecraft collision of 2005 July 4. Weak 18-cm OH emission was detected with the Parkes 64-m telescope, in data averaged over July 4–6, at a level of  12 ± 3 mJy km s⁻¹  , corresponding to OH production rate  2.8 × 10²⁸  molecules s⁻¹ (Despois et al. inversion model, or  1.0 × 10²⁸ s⁻¹  for the Schleicher & A'Hearn model). We did not detect the HCN 1–0 line with the Mopra 22-m telescope over the period July 2–6. The 3σ limit of 0.06 K km s⁻¹ for HCN on July 4 after the impact gives the limit to the HCN production rate of  <1.8 × 10²⁵ s⁻¹  . We did not detect the HCN 1–0 line, 6.7 GHz CH₃OH line or 3.4-mm continuum with the Australia Telescope Compact Array (ATCA) on July 4, giving further limits on any small-scale structure due to an outburst. The 3σ limit on HCN emission of 2.5 K km s⁻¹ from the ATCA around impact corresponds to limit < 4 × 10²⁹ HCN molecules released by the impact.     

Radio Line Observations Of Molecular And Isotopic Species In Comet C/1995 O1 (Hale-Bopp)

We present here a review of the radio observations of the remarkable comet Hale-Bopp C/1995 O1 in which most major radio astronomical facilities have been involved. These observations started in August 1995, soon after the discovery of the comet (it was then at ∼7 AU from the sun), and well before its perihelion on April 1st, 1997; they are still going on, hopefully up to end of 1998. Extended cartographies have been obtained using multibeam receivers and on-the-fly techniques. High spatial resolution (a few ″) has been achieved with interferometers. Submillimetric observations are playing an increasing role, and high resolution (R ∼ 10⁶−10⁷) spectroscopy of cometary lines is now performed from decimetric to submillimetric wavelengths. The number of species observed at radio wavelengths now reaches ∼28,when it was ∼14 for comet C/1996 B2 Hyakutake. Most of these species are parent molecules. However, ions have been observed for the first time at radio wavelengths, and their velocities measured. Several isotopic species (involving D,¹³C,³⁴S,¹⁵N) have been sought, allowing isotopic enrichment determinations. The abundances of cometary molecules present many similarities and some differences with the abundances of interstellar molecules in regions where grain mantles are believed to be evaporated to the gas phase (hot cores, bipolar flows). They will be discussed for their implications on the origin of cometary ices and of comets themselves. This revised version was published online in July 2006 with corrections to the Cover Date.     

The Radio Recombination Line Intensities Of Atoms And Ions: Helium,Carbon And Oxygen

The radio recombination line intensities of heavy elements of helium, carbon and oxygen are calculated with accounting for dielectronic recombination. Dielectronic recombination rates are determined accurate to the second order of a perturbation theory and the rates are described as function of principal quantum number for helium-like atom or ion. Balance equations are solved for the departure coefficients from LTE b_n. The collision and spontaneous transition rates are accounted for the balance equations, in which non-equilibrium distribution source is dielectronic recombination. Non-equilibrium amplification coefficients are found as functions of a medium temperature, density and ion charge z = 1–3 for radio recombination lines. Optical depths are calculated for the heavy element low-frequency lines with the numbers 300 > n > 1200. For the chosen electronic temperatures and densities T_e = 0.8× 10⁴–10× 10⁴ K, N_e = 0.05–0.1 cm⁻³ the line optical depth is determined by the values τ_L∼ 0.1× 10⁻⁴–100× 10⁻⁴. Calculated for free-free transition rates, the optical depth is given by using the value τ_ff∼ 10⁻²τ_L.     

Determination of a precise rotation period for the Deep Space 1 target, Comet 19P/Borrelly

The rotation period derived by Mueller and Samarasinha (Mueller, B.E.A., Samarasinha, N.H. [2002]. Earth Moon Planets 90, 463-471) of the Deep Space 1 (DS1) mission target, Comet 19P/Borrelly, using ground-based data from July 28 to August 1, 2000, is improved by two orders of magnitude. This precision is reached in a multistep process.Combining all available ground-based data in 2000 decreases the error by an order of magnitude. Next, assuming that the rotation period did not change between 2000 and 2001, constraints from the HST 2001 data (Weaver, H.A., Stern, S.A., Parker, J.Wm. [2003]. Astron. J. 126, 444-451) yield three possible rotation periods: P = 1.088 ± 0.003 days, P = 1.108 ± 0.002 days, and P = 1.135 ± 0.003 days, which are consistent with our initial derivation of P = 1.08 ± 0.04 days (Mueller, B.E.A., Samarasinha, N.H. [2002]. Earth Moon Planets 90, 463-471).These three periods are further refined and the error bars further improved by another order of magnitude by linking the combined ground-based data from 2000 to the nuclear orientation of Borrelly at the DS1 encounter in 2001 (see Table 2). Due to aliasing, there are seven possible rotation periods around P = 1.088 days, five possible periods around P = 1.108 days, and six possible periods around P = 1.135 days, with precisions of the order of 0.0002 days (≈17 s).     

A McDonald Observatory Study of Comet 19P/Borrelly: Placing the Deep Space 1 Observations into a Broader Context

We present imaging and spectroscopic data on Comet 19P/Borrelly that were obtained around the time of the Deep Space 1 encounter and in subsequent months. In the four months after perihelion, the comet showed a strong primary (sunward) jet that is aligned with the nucleus' spin axis. A weaker secondary jet on the opposite hemisphere appeared to become active around the end of 2001, when the primary jet was shutting down. We investigated the gas and dust distributions in the coma, which exhibited strong asymmetries in the sunward/antisunward direction. A comparison of the CN and C₂ distributions from 2001 and 1994 (during times when the viewing geometry was almost identical) shows that each species is remarkably similar, indicating that the comet's activity is essentially repeatable from one apparition to the next. We also measured the dust reflectivities as a function of wavelength and position in the coma, and though the dust was very red overall, we again found variations with respect to the solar direction. We used the primary jet's appearance on several dates to determine the orientation of the rotation pole to be α=214°, δ=−5°. We compared this result to published images from 1994 to conclude that the nucleus is near a state of simple rotation. However, data from the 1911, 1918, and 1925 apparitions indicate that the pole might have shifted by 5-10° since the comet was discovered. Using our pole position and the published nongravitational acceleration terms, we computed a mass of the nucleus of 3.3×10¹⁶ g and a bulk density of 0.49 g cm⁻³ (with a range of 0.29<ρ<0.83 g cm⁻³). This result is the least model-dependent comet density known to date.     

Radio observations of Comet 9P/Tempel 1 before and after Deep Impact

Comet 9P/Tempel 1 was the target of a multi-wavelength worldwide investigation in 2005. The NASA Deep Impact mission reached the comet on 4.24 July 2005, delivering a 370-kg impactor which hit the comet at 10.3 km s⁻¹. Following this impact, a cloud of gas and dust was excavated from the comet nucleus. The comet was observed in 2005 prior to and after the impact, at 18-cm wavelength with the Nançay radio telescope, in the millimeter range with the IRAM and CSO radio telescopes, and at 557 GHz with the Odin satellite. OH observations at Nançay provided a 4-month monitoring of the outgassing of the comet from March to June, followed by the observation of H₂O with Odin from June to August 2005. The peak of outgassing was found to be around between May and July. Observations conducted with the IRAM 30-m radio telescope in May and July 2005 resulted in detections of HCN, CH₃OH and H₂S with classical abundances relative to water (0.12, 2.7 and 0.5%, respectively). In addition, a variation of the HCN production rate with a period of 1.73±0.10 days was observed in May 2005, consistent with the 1.7-day rotation period of the nucleus. The phase of these variations, as well as those of CN seen in July by Jehin et al. [Jehin, E., Manfroid, J., Hutsemékers, D., Cochran, A.L., Arpigny, C., Jackson, W.M., Rauer, H., Schulz, R., Zucconi, J.-M., 2006. Astrophys. J. 641, L145-L148], is consistent with a rotation period of the nucleus of 1.715 days and a strong variation of the outgassing activity by a factor 3 from minimum to maximum. This also implies that the impact took place on the rising phase of the “natural” outgassing which reached its maximum ≈4 h after the impact. Post-impact observations at IRAM and CSO did not reveal a significant change of the outgassing rates and relative abundances, with the exception of CH₃OH which may have been more abundant by up to one order of magnitude in the ejecta. Most other variations are linked to the intrinsic variability of the comet. The Odin satellite monitored nearly continuously the H₂O line at 557 GHz during the 38 h following the impact on the 4th of July, in addition to weekly monitoring. Once the periodic variations related to the nucleus rotation are removed, a small increase of outgassing related to the impact is present, which corresponds to the release of ≈5000±2000 tons of water. Two other bursts of activity, also observed at other wavelengths, were seen on 23 June and 7 July; they correspond to even larger releases of gas.     

The 1995–2002 Long-Term Monitoring of Comet C/1995 O1 (HALE–BOPP) at Radio Wavelength

The bright comet Hale–Bopp provided the first opportunity to follow the outgassing rates of a number of molecular species over a large range of heliocentric distances. We present the results of our observing campaign at radio wavelengths which began in August 1995 and ended in January 2002. The observations were carried out with the telescopes of Nançay, IRAM, JCMT, CSO and, since September 1997, SEST. The lines of nine molecules (OH, CO, HCN, CH₃OH, H₂CO, H₂S, CS, CH₃CN and HNC) were monitored. CS, H₂S, H₂CO, CH₃CN were detected up to r_h= 3–4 AU from the Sun, while HCN and CH₃OH were detected up to 6 AU. CO, which is the main driver of cometary activity at heliocentric distances larger than 3–4 AU, was last detected in August 2001, at r_h= 14 AU. The gas production rates obtained from this programme contain important information on the nature of cometary ices, their thermal properties and sublimation mechanisms.Line shapes allow to measure gas expansion velocities, which, at large heliocentric distances, might be directly connected to the temperature of the nucleus surface. Inferred expansion velocity of the gas varied as r_h ^-0.4 within 7 AU from the Sun, but remained close to 0.4 km s^-1 further away. The CO spectra obtained at large r_hare strongly blueshifted and indicative of an important day-to-night asymmetry in outgassing and expansion velocity. The kinetic temperature of the coma, estimated from the relative intensities of the CH₃OH and CO lines, increased with decreasing r_h, from about 10 K at 7 AU to 110 K around perihelion.