Time variability and heterogeneity in the coma of 67P/Churyumov-Gerasimenko

See allHide authors and affiliations

Science  23 Jan 2015:
Vol. 347, Issue 6220, aaa0276
DOI: 10.1126/science.aaa0276


Comets contain the best-preserved material from the beginning of our planetary system. Their nuclei and comae composition reveal clues about physical and chemical conditions during the early solar system when comets formed. ROSINA (Rosetta Orbiter Spectrometer for Ion and Neutral Analysis) onboard the Rosetta spacecraft has measured the coma composition of comet 67P/Churyumov-Gerasimenko with well-sampled time resolution per rotation. Measurements were made over many comet rotation periods and a wide range of latitudes. These measurements show large fluctuations in composition in a heterogeneous coma that has diurnal and possibly seasonal variations in the major outgassing species: water, carbon monoxide, and carbon dioxide. These results indicate a complex coma-nucleus relationship where seasonal variations may be driven by temperature differences just below the comet surface.

Initially, comets were classified depending on the location where they formed in the protoplanetary disc (1, 2). This classification assumed a similar composition of the nucleus within a given formation region. No cometary nucleus composition has been sampled in situ. Rather, it is implicitly assumed that measurements of the outgassing of comets reveal the composition of the volatile components of the nucleus. However, compositional homogeneity of at least one comet was confirmed by studying outgassing from the fragments of the broken-up comet Schwassmann-Wachmann 3 (3). Detailed observations of other cometary comae indicated that there is evidence of heterogeneity. Missions to comet Halley detected release of volatiles in multiple jet-like features that were dominantly seen on the sunlit side of the nucleus (4, 5). The Deep Impact mission detected asymmetries in composition in the coma of Tempel 1 (6). In particular, these remote sensing observations at Tempel 1 indicated an absence of correlation between H2O and CO2 in the coma.

Detailed, close-up cometary images have also showed visible differences between different areas of cometary nuclei. These images suggested that heterogeneity in the coma of a comet may be related to heterogeneity of the nucleus. Observations by EPOXI (Extrasolar Planet Observation and Deep Impact Extended Investigation) at Hartley 2 in 2010 near perihelion indicated that the nucleus is complex, with two different sized lobes separated by a middle waist region that is smoother and lighter in color (7). Outgassing from sunlit surfaces of the nucleus revealed that the waist and one of the lobes were very active. A CO2 source was detected at the small lobe of the comet, whereas the waist was more active in H2O and had a considerably lower CO2 content. Based on these coma observations, it has been tentatively suggested that the heterogeneity in the comet’s nucleus was primordial (7). Seasonal effects could not be ruled out because the observations also showed a complex rotational state for the comet (7). The smaller of the two lobes may have been illuminated differently because of this complex rotation (7). In support of the findings at Hartley 2, there are indications of a heterogeneous nucleus for comet Tuttle and a heterogeneous coma (7, 8).

The Stardust mission to comet P81/Wild 2, on the other hand, showed a large mixing of materials on the scale of grains and therefore a homogenized mix of the refractory material in the comet (9). The results at Hartley 2 and at P81/Wild 2 raise the larger question of whether heterogeneity in the coma is a common feature in comets and whether this reveals an underlying heterogeneity in the composition of the nucleus, which would point to general transport of cometesimals in the early solar system.

In August, the European Space Agency’s mission Rosetta arrived at its target comet 67P/Churyumov-Gerasimenko (67P) after a 10-year journey (10). Rosetta provides an excellent opportunity for long-term study during the comet’s sunward approach to perihelion. The observations presented here are from a 2-month period beginning near the initial encounter at about 3.5 astronomical units from the Sun.

Like Hartley 2, the nucleus of 67P appears complex in shape. The comet 67P consists of two lobes of different sizes, connected by a neck region. The lobes are much larger, more rugged, and darker than the neck region and the overall shape has been compared to a rubber duck (11). The structural similarities of 67P and Hartley 2 suggest the possibility of another heterogeneous comet and, by virtue of the extended observations at 67P, a chance to determine whether heterogeneity in the coma and nucleus are related.

Here, we show compositional variations in H2O, CO, and CO2 at comet 67P observed with ROSINA/DFMS (Rosetta Orbiter Spectrometer for Ion and Neutral Analysis/Double Focusing Mass Spectrometer) (12). ROSINA/DFMS is a mass spectrometer that measures the in situ neutral and plasma coma composition at the position of the spacecraft (see the supplementary materials). During Rosetta’s approach to 67P, ROSINA/DFMS measured the neutral coma composition with a time resolution (>10 measurements per rotation) much finer than the rotation period of the comet of ~12.4 hours (13). In August, the spacecraft scanned the comet at northern summer hemisphere (positive latitudes) from about 10° up to almost 90° (Coordinates: Cheops System). In September, the spacecraft made a similar scan at southern winter hemisphere (negative latitudes) down to about –50°. Two data sets are shown in Figs. 1 and 2 to illustrate the diurnal and latitudinal variations and heterogeneity of the cometary coma.

Fig. 1 H2O, CO, and CO2 measurements for 4 to 8 August 2014.

The upper panel shows the signal on the DFMS detector for H2O, CO, and CO2, and the lower panel shows the latitude and longitude of the nadir view of the spacecraft. At the top is the distance from the spacecraft to the comet. The signal increases with decreasing distance to the comet, and diurnal variations are also visible. CO2 has a different periodicity than H2O, as seen around 4 to 6 August.

Fig. 2 H2O, CO, and CO2 measurements for 15 to 19 September 2014.

Over this 4-day period, the spacecraft remained at a nearly fixed distance from the comet and executed a southern latitude scan from about 0° to –45° latitude. H2O and CO2 have different periodicities, and there are deep minima in the H2O signal. CO follows the CO2 profile with less variation.

During this approach and latitude scan, the H2O, CO, and CO2 signals from the comet increased by more than an order of magnitude, roughly in agreement with a 1/R2 dependence on the coma density, where R is the distance from the comet’s center. Overall, the H2O signal is the strongest; however, there are clearly periods when the CO or CO2 signals rival that of H2O.

Superposed on this general increase in signal are large, diurnal variations for all three neutral species. For H2O, these variations are periodic, initially with half the rotation rate of the comet (~6.2 hours) and then, after 6 August, at the rotation rate (~12.4 hours). This change in periodicity in the signal is interpreted as a latitudinal effect of the sampling position. Peaks occur at ±90° longitude. For the most part, the CO signal follows the H2O signal, but the variations are smaller. CO2 shows a different periodicity. Initially, a CO2 peak is observed in association with an H2O peak, and a second CO2 peak occurs approximately 3 hours later. After August 6, a single CO2 peak is observed; however, this peak is not exactly coincident with the H2O peak. The two CO2 peaks merge, resulting in a shoulder on the main peak and a slight shift of the main CO2 peak relative to that of H2O (~45 min or one measurement point). Statistical uncertainties (Embedded Image) in the signal detected by ROSINA/DFMS are smaller than the dots in Figs. 1 to 3, and contributions to the signal due to spacecraft outgassing (14) are subtracted.

Fig. 3 H2O, CO, and CO2 profiles for 18 September 2014.

The Sun is shining on the comet from the top middle of the pictures. The snapshots of the spacecraft view of the comet show that H2O peaks are observed when the neck region is in view. The separate CO2 peak and the deep minimum in H2O occur when the spacecraft views the larger of the two lobes and the neck region is blocked. [Shape model credit: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA].

The diurnal variations at half the rotation rate of the comet that are seen in August are also observed at southern latitudes in the September time frame (Fig. 2). The H2O peaks in Fig. 3 are nearly equal, and there is a deep minimum between the two peaks. As in the first data set, CO follows H2O. However, there is much less variation in CO than in H2O, resulting in times when the CO signal is greater than that for H2O. The best example of the differences between H2O and CO2 are seen just after 18 September (Figs. 2 and 3). The nearly equal H2O peaks and the deep minimum in the H2O signal are evident, as is the clear offset between the second CO2 and H2O peaks.

We have combined the signal and the spacecraft perspective over the 18 to 19 September 2014 window to illustrate which side of the comet is in view when the peaks occur (Fig. 3). The peaks in H2O signal are observed when the neck of the comet is in view of the spacecraft. The deep minimum in H2O signal is observed when the spacecraft views the southern hemisphere of the larger of the two lobes. This large lobe blocks a direct view of the neck of the comet. The separate, second CO2 enhancement is observed when the spacecraft views the underside of the body of the larger of the two lobes of the comet. The CO signal in the second rotation of the comet follows the CO2 profile, and CO and CO2 have very similar intensities.

We see from this data (Figs. 1 to 3) that the coma composition of 67P is highly heterogeneous. H2O, CO, and CO2 variations are strongly tied to the rotation period of the comet and to the observing latitude. At large negative latitudes, the H2O signal varies by at least two orders of magnitude (Fig. 3). Also, the H2O minima are not as deep when the spacecraft is at mid and high positive latitudes because there is a view of the neck region over the edge of the larger lobe (see Fig. 1 and the observations on 15 September in Fig. 2).

The separate CO2 peak also occurs when the spacecraft views the bottom of the larger of the two lobes of the comet (see Fig. 3 at 5 hours). CO follows H2O at positive latitudes and follows both H2O and CO2 at negative latitudes.

The separate CO2 peak, the large variations in the H2O signal, and the weaker variations in CO result in large changes in the relative concentration of H2O, CO, and CO2 in the heterogeneous coma of 67P (see the supplementary materials). For example, the CO/H2O number density ratio is 0.13 ± 0.07, and the CO2/H2O ratio is 0.08 ± 0.05 in the last H2O peak on 7 August at 18 hours in Fig. 1 (measured high in the northern summer hemisphere). However, The CO/H2O ratio changes from 0.56 ± 0.15 to 4 ± 1 and back to 0.38 ± 0.15 within the second cometary rotation (Fig. 3), between 12 and 24 hours on 18 September, measured low in the southern winter hemisphere. Similarly, the CO2/H2O ratio changes from 0.67 ± 0.15 to 8 ± 2 and back to 0.39 ± 0.15 over the same rotation. These are large changes within a short amount of time, which indicate a strongly heterogeneous and time-variable coma.

The similarities in the structure of the nuclei and the heterogeneous comae of 67P and Hartley 2 are striking. The behavior in terms of the H2O dominant outgassing at the neck versus CO2 outgassing at one of the lobes described here was also found for Hartley 2 (7).

The compositional differences in the Hartley 2 coma were interpreted as evidence for a heterogeneous cometary nucleus (7). However, seasonal effects could not be ruled out. With observations over a wide range of latitudes at 67P, we can distinguish between compositional differences and seasonal effects; to do so, we have mapped the CO2/H2O density ratio from 17 August through 22 September onto the shape model (Fig. 4).

Fig. 4 The nadir point for each pair of CO2/H2O measurements over the time period from 17 August through 22 September was mapped to the model surface.

The mapping is shown for the bottom side of the larger of the two lobes of the comet, and cometary latitudes run approximately vertically in this. The layering is due to spacecraft rastering above the comet nucleus. A high ratio is measured for the lower part that is poorly sunlit in southern hemisphere winter. [Shape model credit: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA].

Although a direct mapping of the signal observed in the coma onto the comet surface is oversimplified, a generalized interpretation reveals features of the outgassing of the comet. Seasonal effects on the CO2/H2O ratio are clearly evident (Fig. 4). On the upper half of the comet, the CO2/H2O ratio is less than 1, indicating a higher sublimation of H2O from positive latitude regions that receive more illumination during northern hemisphere summer on the comet. A broad region of high CO2/H2O ratio occurs at negative latitudes in the winter hemisphere, likely the result of deep minima in the H2O signal (such as the one shown in Fig. 3 on 18 September at 4 hours). This winter hemisphere of the comet is poorly illuminated by the Sun. With limited illumination, this region of the comet nucleus may be considerably colder than other regions, including the neck and smaller lobe. The temperature at and below the surface of the nucleus may be sufficient to sublimate CO and CO2 but not sufficient to sublimate water. The weak, periodic illumination of this region may be sufficient to drive CO and CO2 sublimation, producing the separate CO and CO2 peak (Fig. 3 at 18 hours). However, the compositional asymmetry in the two H2O peaks cannot be explained in a similar way and might be the strongest indication for heterogeneity in the comet nucleus. The strong heterogeneity in the coma of comet 67P is likely driven by seasonal effects on the comet nucleus. However, the smaller variation of CO and CO2 compared with H2O might indicate that CO and CO2 ices sublimate from a greater depth, whereas H2O ice sublimates closer to the surface and experiences more direct temperature differences due to sunlight. Furthermore, that lack of overall correlation between H2O, CO, and CO2 implies that the outgassing from the nucleus is not correlated or that CO and CO2 are not strictly embedded in H2O. For Tempel 1, material was found in layers and supports the above idea (15).

In summary, the coma composition has been measured over many rotational periods of the comet and a wide range of latitudes with high time resolution and compositional detail. Concentrations of the three molecules change over the rotational period of the comet and indicate a strongly heterogeneous coma. For the most part, H2O dominates, but CO and CO2 can at times dominate in the coma. These observations also indicate that there are substantial diurnal and latitudinal variations in the coma. Peaks in the H2O signal are observed, along with deep minima at high negative latitudes when the neck region of the nucleus is blocked from view of the spacecraft. A separate peak in CO2 signal occurs when the winter hemisphere of the larger lobe of the comet faces the spacecraft. The diurnal and latitudinal variations suggest that compositional differences in the coma may be seasonal and may indicate different subsurface temperatures in the nucleus.

Further observations may distinguish seasonal effect from nucleus heterogeneity. As the comet approaches the Sun, the overall temperature of the nucleus will increase, and as the seasons change, there may be considerable changes in the H2O, CO, and CO2 outgassing, with the current high CO2/H2O ratio region shown in Fig. 4. In addition, differences in the sublimation of species similar in sublimation temperatures could demonstrate the extend of heterogeneity in the nucleus independent of seasonal changes.

Supplementary Materials

Materials and Methods

Supplementary Text


References and Notes

  1. Acknowledgments: The authors thank the following institutions and agencies, which supported this work: Work at the University of Bern was funded by the State of Bern, the Swiss National Science Foundation, and the European Space Agency (ESA) PRODEX Program. Work at the Max-Planck Institut für Sonnensystemforschung (MPS) was funded by the Max-Planck-Gesellschaft (MPG) and Bundesministerium für Wirtschaft und Energie (BMWI) under contract 50QP1302. Work at Southwest Research institute (SwRI) was supported by subcontract 1496541 from the Jet Propulsion Laboratory (JPL) and under NASA prime contract NNX148F71G. Work at BIRA-IASB was supported by the Belgian Science Policy Office via PRODEX/ROSINA PEA 90020. Work at Imperial College London has been partially funded by the Science and Technology Facilities Council (STFC). This work has been carried out thanks to the support of the A*MIDEX project (ANR-11-IDEX-0001-02) funded by the Investissements d’Avenir French Government program, managed by the French National Research Agency (ANR). This work was supported by CNES grants at IRAP, LATMOS, LPC2E, Laboratoire d’Astrophysique de Marseille (LAM), and CRPG and by the European Research Council (grant 267255). A.B.-N. thanks the Ministry of Science and the Israel Space agency. Work at the University of Michigan was funded by NASA under contract JPL-1266313. Work by J.H.W. at SwRI was supported by NASA JPL subcontract NAS703001TONMO710889. ROSINA would not give such outstanding results without the work of the many engineers, technicians, and scientists involved in the mission, in the Rosetta spacecraft, and in the ROSINA instrument over the past 20 years whose contributions are gratefully acknowledged. Rosetta is an ESA mission with contributions from its member states and NASA. We thank the OSIRIS team for giving permission to use the shape model. We acknowledge herewith the work of the ESA Rosetta team. All ROSINA data are available on request until it is released to the PSA archive of ESA and to the PDS archive of NASA.

Stay Connected to Science

Navigate This Article