EDP Sciences
Free Access
Issue
A&A
Volume 578, June 2015
Article Number L7
Number of page(s) 4
Section Letters
DOI https://doi.org/10.1051/0004-6361/201526132
Published online 08 June 2015

© ESO, 2015

1. Introduction

Solar and Heliospheric Observatory (SOHO) instruments have observed about three thousand near-Sun comets since 1996, the vast majority of which are members of the Kreutz family of sungrazing comets. Most of the comets have been detected by the Large Angle Spectrometric Coronagraph (LASCO; Brueckner et al. 1995), which observes the white-light corona between about 2.3 and 32 R, and by the Solar Wind ANisotropies (SWAN; Bertaux et al. 1995) instrument, a scanning imager that maps the full sky in H i Lyα. The huge envelope of atomic hydrogen around the nucleus, or coma, is mainly produced by the UV photodissociation of cometary water (H2O vapor from ice sublimation of the nucleus or surrounding grains) and its byproduct hydroxyl (OH). Observations of the intensity profile of the resonantly scattered radiation illuminated by solar H i Lyα at 1215.67 Å can provide information on the column density of the atomic hydrogen coma as a function of distance from the cometary nucleus and thus yield an estimate of the water production rate. Since all other constituents in the coma are normally compared to water, it is important to obtain accurate evaluation of production rates.

thumbnail Fig. 1

LASCO C2 image of the solar corona on 1997 May 1 at 17:31 UT showing comet C/1997 H2 approaching the Sun from northwest (inside red circle).

Open with DEXTER

Over the past two decades, several comets have also been spectroscopically detected by the UltraViolet Coronagraph Spectrometer (UVCS; Kohl et al. 1995) instrument onboard SOHO (Raymond et al. 1998, 2002; Uzzo et al. 2001; Povich et al. 2003; Bemporad et al. 2005, 2007; Ciaravella et al. 2010; Giordano et al. 2015). In this Letter, we report the water production rate at perihelion of a non-Kreutz comet observed by UVCS, comet C/1997 H2 (Stezelberger et al. 1997), as determined from the H i Lyα intensity profile obtained when the comet was at a distance of ~0.14 AU from the Sun. The overall contribution of neutral hydrogen atoms created by photodissociation processes from H2O (~20 km s-1) and OH (~8 km s-1) is estimated by fitting a Haser model (Haser 1957) to the observed H i Lyα radiance profile. To obtain a better agreement between the observed radiance profile and the model, however, an additional fraction of low-velocity hydrogen atoms (~4 km s-1) is required. This low-velocity component is attributed to thermalization of fast hydrogen atoms within the collision zone of the comet.

2. Observations

In Fig. 1, we show the visual appearance of comet C/1997 H2 as seen in white light by the LASCO C2 coronagraph. The comet approached the Sun from the northwest quadrant at a position angle (measured counterclockwise from celestial north) of 296.2° projected onto the plane of the sky. Detections were first made with the LASCO C3 coronagraph on 1997 April 29, at 31.43 R from the Sun’s center. According to the orbital data, given in the Minor Planet Electronic Circular (MPEC) 1997-J05, the comet reached a perihelion distance of q = 0.137 AU at tp = 20:00 UT on 1997 May 2. UVCS is a long-slit UV spectrograph consisting of two spectrometric channels (the OVI and LYA channels) for the observation of spectral lines in the UV range with an instantaneous field of view of 42 tangent to the limb of the Sun with nominal spatial resolution of 7′′ per pixel (~5000 km at the Sun). The UVCS telescope mirror and instrument rotation mechanisms can point the spectrograph slits in the radial direction to observe the entire corona at heliocentric distances between about 1.5 and 10 R. The position of the UVCS slit for the observations analyzed in this work was at a projected height of 8.48 R and a roll angle of 110° counterclockwise from north pole. At this slit position, the UVCS observations, composed of a total of 83 exposures of 200 s each, started at 20:11 UT on 1997 May 2 and ended at 01:08 UT on the next day. The comet nucleus entered the slit at the exposure taken between 23:15.4 and 23:18.7 UT. Over the available ranges, only the Lyα, Lyβ and Lyγ lines were detected. The H i Lyα 1215.67 Å line was well observed spectroscopically. Significant emission in the H i Lyβ 1025.72 Å line and weak emission in H i Lyγ 972.02 Å were also observed in the hydrogen inner coma. Figure 2 shows the H i Lyα 1215.67 Å image of comet C/1997 H2, reconstructed from UVCS observations. Details and thorough spectroscopic analysis of the complete set of observations obtained for comet C/1997 H2 by UVCS from 1997 May 1 to May 3 are out of the purpose of this Letter and will be presented elsewhere.

3. Data analysis and modeling

The observed cometary H i Lyα emission may arise from collisional excitation with thermal electrons and/or from resonant scattering of the H i Lyα at 1215.67 Å emitted by the solar chromosphere. To understand its origin, we estimated the observed ratio between the Lyα and Lyβ spectral line intensities. The intensity ratio Lyα/Lyβ is about 400 around the nucleus, roughly doubling up at a distance of 5 × 104 km from the nucleus. Collisional excitation at the high electron temperatures typical of the solar wind would produce a ratio close to 8 (Raymond et al. 1998), while scattering of emission-line photons from the solar disk produces a much higher ratio (because of the very high Lyα/Lyβ ratio of the chromospheric emission) of about 910 (Raymond et al. 1997). Therefore, the observed Lyα in the extended coma is mostly attributable to resonant scattering from neutral hydrogen atoms, while a partial contribution from collisional excitation is possible at a few × 104 km from the nucleus (e.g., Raymond et al. 2002). As suggested by the Referee of this Letter, the Lyα/Lyβ ratio could also indicate the emission of Lyα and Lyβ by electron excitation of H2O molecules. Electrons produced by the ionization of H2O molecules have enough energy to dissociate the H2O molecule and generate H atoms in an excited state, thus producing emission in both Lyα and Lyβ spectral lines with a Lyα/Lyβ ratio that is much lower than for resonance scattering from solar photons, as is actually observed in this observation near the nucleus.

The observed Lyα line emission stems from a superposition of the cometary signal, foreground coronal emission, interplanetary emission and detector dark counts. To identify the cometary signal, the average emission over several exposures taken before the cometary coma entered the UVCS field of view was subtracted. The data were then calibrated using previously generated calibration files that consider such factors as the internal occulter setting and the grating position (Gardner et al. 2002).

thumbnail Fig. 2

Reconstructed image of the H i Lyα light from the comet C/1997 H2 as observed at 8.48 R. The horizontal axis represents the position (expressed in arcsec) along the UVCS slit. Each bin in the vertical direction is given by the product of the single exposure time and the comet velocity in the plane of the sky. Courtesy of S. Giordano.

Open with DEXTER

The hydrogen atoms that scatter H i Lyα radiation are mainly formed by dissociation of molecules outgassed from the cometary nucleus. Hydrogen is produced primarily by photodissociation of water through the following reactions: The cometary outgassing rate of water vapor QH2O can be estimated from the observed Lyα radiance distribution by assuming that the photodissociation of water outside the collision sphere is the only noteworthy process. The production rate, QH2O, is assumed to be constant (stationary model) and isotropic. All particles are assumed to have a radial velocity, and their distribution is spherically symmetric (Haser 1957). The column densities of the two H populations produced by photodissociation of water at a distance r from the nucleus are obtained by integration along the line of sight. If NH1 and NH2 represent the column densities coming, respectively, from the first photodissociation process (Eq. (1)) and from the second photodissociation process (Eq. (2)), they can be expressed (e.g., Mäkinen et al. 2001) as In the above equations, γi and vi (i = H2O,OH,H1,H2) are the inverse scale lengths () and the radial velocities of the respective particle populations, r is the shortest distance between the column and the nucleus and . Typical hydrogen ejection velocities for the two different populations are vH1 = 20 km s-1 and vH2 = 8 km s-1 (e.g., Combi & Smyth 1988; Bertaux et al. 1998). The scale lengths of the hydrogen-producing processes at 1 AU are LH2O(1 AU) = 8.2 × 104 km and LOH(1 AU) = 2.25 × 105 km (Mäkinen et al. 2001), while the scale lengths for the two hydrogen components are given by the product of the respective outgassing velocities and their lifetime τH.

The total intensity, ILyα along the line of sight is given by (5)where (6)are, respectively, the contributions from hydrogen atoms given by Eqs. (1) and (2), Nj (with j = H1,H2) is the column density of the scattering H atoms, and (7)is the g factor, that is, the number of Lyα photons per second fluorescently produced by absorption of chromospheric Lyα photons by each H atom in the line of sight at a distance d = dcomet from the Sun. In Eq. (7), is the frequency-integrated cross section of the transition (f12 = 0.416 being the oscillator strength) that multiplied by the factor yields the wavelength integrated cross section; πF0 is the solar Lyα flux per unit wavelength interval at 1 AU at the Doppler-shifted wavelength corresponding to the heliocentric radial velocity of the comet; pφ is the probability that in a scattering process a photon changes its direction of an angle φ relative to the initial direction. A g factor value g(dcomet) = 0.0953 photons s-1 was determined from daily average values obtained by the SOLar Stellar Irradiance Comparison Experiment (SOLSTICE) onboard NASA’s Upper Atmosphere Research Satellite (UARS), which measures the solar UV radiance. The estimate was corrected for the solar rotation difference between the face of the Sun seen by the instrument and that seen by the comet and for the Doppler shift in the H i Lyα line profile caused by the heliocentric radial velocity of the comet (36 km s-1, as calculated from the given orbital parameters) toward the Sun, on the basis of the solar irradiation profile around the Lyα line as given by Lemaire et al. (1998). We also took the optical thickness for each component (=Njσj) into account with σj representing the scattering cross sections that depend on the velocities of the different populations vHj. Optical depth effects have been included because they can be important near the nucleus where a stronger contribution of the collisional component in the Lyα emission is expected.

Neutral hydrogen atoms released by a comet are lost from the coma through a combination of photoionization by the solar UV radiation, impact ionization by solar wind electrons (Combi et al. 1986) and charge exchange with solar wind protons (Keller 1973; Bertaux et al. 1973). For H atoms released by the comet into the solar wind, their lifetime, τH, has been estimated to be in the range of 1 − 2 × 106 s at 1 AU, varying with the square of the heliocentric distance (e.g., Combi et al. 2005). The actual value, however, depends critically on several factors, such as the time and location, the solar wind density, velocity, electron temperature, the solar UV radiation flux, and the heliographic latitude. Instead of assuming a fixed value for τH, we leave this atomic property as a free parameter, together with the unknown outgassing rate, QH2O, to be solved by compariing the coma model with observations.

thumbnail Fig. 3

H i Lyα radiance profile along the UVCS slit from the exposure containing the nucleus taken near perihelion (red squares and blue diamonds represent data taken north and south of the cometary nucleus). The green dashed curve is a model (Model A) obtained by considering two hydrogen atom populations (vH1 = 20 km s-1, vH2 = 8 km s-1) with QH2O and τH left as free parameters. The green solid curve is a model (Model B) obtained by adding the contribution of a thermalized hydrogen atom population (vH3 = 4 km s-1). See the text for details.

Open with DEXTER

The above model (Model A) was applied and compared to the data obtained with the analysis of the Lyα radiance profile along the UVCS slit from the exposure containing the nucleus. The best-fit to the Lyα radiance profile observed on 1997 May 2 for the crossing at 23:17 UT is shown in Fig. 3 as a dashed line, superposed on the data. The model was obtained through a least-squares approach, yielding a water production rate of QH2O = 7.39 × 1028 molecules s-1 and a hydrogen atom lifetime of τH = 0.85 × 106 s at 1 AU as best-fitting parameters. Although Model A fits the UVCS data quite well (the different symbols in Fig. 3 specify radiance estimate north and south of the cometary nucleus), it underestimates by about 5%−10% the observed radiance between about 1−5 × 104 km from the nucleus. Moreover, the hydrogen atom lifetime is also underestimated, falling below the previously mentioned range of 1−2 × 106 s at 1 AU.

We note that the Haser model assumes a simple radial outflow of the daughter species once they are created, without considering the non-radial motion of the radicals produced after photodissociation or recombinative dissociation of the parent molecules outside the collision zone. A more realistic vectorial model, which assumes isotropic ejection of daughter species following photodissociation of the parent molecule, was introduced by Festou (1981). Both models are frequently used to determine the production rates of the observed daughter species, although neither the Haser nor the vectorial model include any collision effects. As pointed out by the Referee of this Letter, however, the vectorial model is flatter at the center than the Haser model so that its application would probably increase the discrepancy between the model and the UVCS data.

The approximation of two hydrogen atom populations with very narrow velocity distributions is probably reasonable only for small comets, for which most hydrogen atoms are produced outside the collision zone. In fact, Haser-like models obtained with a H-atom distribution consistent with H2O and OH photodissociation yield reasonable results mostly for heliocentric distances >0.8 AU and gas production rates <few × 1028 s-1, where thermalization of suprathermal H atoms can be ignored (Combi et al. 2005). For comets with higher gas production rates, the effect of collisions could redistribute a significant fraction of the high velocity atoms to lower kinetic energy, with significant pile-up at velocities 4 km s-1 (e.g., Combi & Smith 1988). To obtain a better fit to the observed Lyα radiance profile of comet C/1997 H2, a low-velocity component is thus probably required. This low-velocity component could result from thermalization of fast hydrogen atoms within the collision zone.

The effectiveness of the thermalization depends on the number of atoms produced within the region of the coma, where collisions are important, and on temperature. To estimate the expected amount of this low-velocity component, we calculated the fraction of atoms produced within the cometary collision zone (see, e.g., McCoy et al. 1992). The radius of the collision zone, Rc, is obtained by imposing , where σ is the collisional cross section and nH2O( = QH2O/ 4πr2vH2O) the water number density. Substituting this into the above integral and solving for Rc, we estimate Rc ≈ 1200 km for a cross section σ ≈ 2 × 10-15 cm2 and a typical molecular outflow velocity vH2O ≈ 1 km s-1. The scale lengths for photodissociation of water at d = 0.137 AU is thus comparable (~1500 km) in magnitude to the collision distance of water molecules at this distance from the Sun.

Because of thermalization, a considerable fraction of the high-speed (~20 km s-1) atomic H generated by the comet will thus be slowed down to v ≲ 4 km s-1 while the ~8 km s-1 component from the OH radicals would probably survive in large part (see, e.g., Combi & Smith 1988). This new situation can be approximated with a model (Model B) similar to the previous one that does, however, include a third component with speed vH3, left as a free parameter, accounting for the broad thermalized component at low velocities. (We only considered integer values for vH3 varying between 1 to 6 km s-1.)

As in the previous case, this model was applied to least-square fit the radiance of the H i Lyα as observed by UVCS in the exposure where the cometary nucleus entered the slit, leaving the outgassing rate QH2O, the hydrogen lifetime τH at 1 AU, and the speed of the low-velocity component as free parameters. The weighting of these two components has also been left a free parameter in our model since the amount of thermalized high-speed hydrogen atoms is uncertain. As shown in Fig. 3, an outstanding fit to the observed Lyα radiance profile was obtained for Model B with best-fitting parameters QH2O = 6.64 × 1028 molecules s-1 and τH = 0.99 × 106 s, together with the addition of a further low-speed component of 4 km s-1 attributed to thermalization of fast hydrogen atoms within the collision zone. Weights of 0.72 and 0.28 for the highest (~20 km s-1) and lowest (~4 km s-1) speed components, respectively, have been found to better fit the data, showing that a significant fraction of the high speed H atoms have been thermalized.

Finally, we point out that the nucleus of this comet is probably small in view of its relatively modest water production rate and the short distance from the Sun at which it was observed by UVCS. Approximating the cometary nucleus as a sphere, we can provide a rough estimate of its equivalent radius (e.g., Giordano et al. 2015) as , where L = 4.81 × 1011 erg mol-1 is the latent heat of sublimation of ice, A = 0.06 the cometary albedo, Fd the solar flux scaled to the cometary heliocentric distance and NA the Avogadro’s number.

Given the outgassing rate estimated in this work, the above relationship, which only holds in the absence of unobserved fragmentation events, yields req = 160 m.

4. Summary and conclusions

In this work, we analyzed the H i Lyα emission at 1215.67 Å from the extended coma of comet C/1997 H2 that was observed near perihelion at 0.137 AU with the UVCS instrument onboard SOHO. The observed H i Lyα radiance profile of the cometary coma, mainly produced by the water dissociation chain, was compared with a Haser model with hydrogen atom velocities of ~20 km s-1 and ~8 km s-1 (from photodissociation of H2O and OH, respectively). To obtain a better agreement with the UVCS H i Lyα observations in a range between about 1 and 5 × 104 km from the nucleus, a further low-velocity component of 4 km s-1 was, however, required. By taking the contribution from this additional slower component into account, the hydrogen production rate for the crossing near perihelion corresponds to a cometary water production rate of QH2O = 6.64 × 1028 molecules s-1 and a hydrogen atom lifetime at 1 AU of τH ≈ 1.0 × 106 s. The ~4 km s-1 low-velocity component is attributed to thermalization of a substantial fraction (30%) of fast hydrogen atoms within the cometary collision zone.

Acknowledgments

UVCS is the result of a collaborative effort between NASA and ASI, with a Swiss participation. SOHO is a mission of international cooperation between ESA and NASA. The author wishes to express his thanks to the anonymous referee for the very helpful comments and suggestions that significantly improved the paper and to S. Giordano for producing Fig. 2 in this paper and for his valuable comments. This work, made possible by the UVCS/SOHO instrument and Flight Operations team, has been originally suggested and encouraged by E. Antonucci, lead observer for this particular UVCS observation.

References

All Figures

thumbnail Fig. 1

LASCO C2 image of the solar corona on 1997 May 1 at 17:31 UT showing comet C/1997 H2 approaching the Sun from northwest (inside red circle).

Open with DEXTER
In the text
thumbnail Fig. 2

Reconstructed image of the H i Lyα light from the comet C/1997 H2 as observed at 8.48 R. The horizontal axis represents the position (expressed in arcsec) along the UVCS slit. Each bin in the vertical direction is given by the product of the single exposure time and the comet velocity in the plane of the sky. Courtesy of S. Giordano.

Open with DEXTER
In the text
thumbnail Fig. 3

H i Lyα radiance profile along the UVCS slit from the exposure containing the nucleus taken near perihelion (red squares and blue diamonds represent data taken north and south of the cometary nucleus). The green dashed curve is a model (Model A) obtained by considering two hydrogen atom populations (vH1 = 20 km s-1, vH2 = 8 km s-1) with QH2O and τH left as free parameters. The green solid curve is a model (Model B) obtained by adding the contribution of a thermalized hydrogen atom population (vH3 = 4 km s-1). See the text for details.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.