PREDICTION OF FORBIDDEN ULTRAVIOLET AND VISIBLE EMISSIONS IN COMET 67P/CHURYUMOV–GERASIMENKO

The modeled CO(a³Π) rate profiles for different formation processes in the coma of 67P are shown in Figure 1. The number of excited CO(a³Π) molecules produced per unit volume per second is referred to as a volumetric production rate. For equal (5% relative to water) CO₂ and CO volume mixing ratios relative to water in the coma the major production source of CO(a³Π) is electron impact on CO. The electron impact on CO₂ and photodissociation of CO₂ are the next most dominant sources of CO(a³Π). The thermal electron recombination of HCO⁺ and ${\mathrm{CO}}_{2}^{+}$ ions and fluorescence of CO are minor CO(a³Π) production sources. Above 500 km the contributions from photodissociative excitation of CO₂ and electron impact on CO and CO₂ are nearly equal. Since the lifetime of CO(a³Π) is short (∼3 ms, Gilijamse et al. 2007), most of the excited molecules decay to the ground state by spontaneous emission. Hence the radiative decay is the major loss source of this excited state. Other loss processes, such as collisional quenching and ionization by photons and photoelectrons, are smaller compared to radiative decay by several orders of magnitude. The number of species that de-excite to ground state per second by various loss mechanisms is referred to as the loss rate.

Zoom In Zoom Out Reset image size

The cross section for electron impact excitation of various excited states and the calculated suprathermal electron intensity in 67P coma at 10 km radial distance is presented in Figure 2. The suprathermal electron intensity in the energy range between 10 and 15 eV mainly determines the excitation rate of CO(a³Π) through electron impact on CO-bearing species.

Zoom In Zoom Out Reset image size

3.2.1. Atomic Oxygen in ³P, ¹D, ¹S, and ⁵S States

The modeled major production rate profiles of atomic oxygen and atomic carbon in ground states are presented in Figure 3. Below 50 km radial distances, various sources are contributing to the formation of O(³P). The production of atomic oxygen in the ground state is mainly due to strong collisional quenching of O(¹D) with water. The next important source of atomic oxygen is charge exchange of OH⁺ and O⁺ ions with water. Photodissociation of CO₂ and CO are the next important O(³P) production sources. Above 50 km, radiative decay of O(¹D) is the major source of atomic oxygen in the ground state.

Zoom In Zoom Out Reset image size

The calculated major chemical loss rate profile for O(³P) via different destruction mechanisms are presented in Figure 4. The major loss process for the atomic oxygen is due to collisions with OH molecules, which yield atomic hydrogen and molecular oxygen. The atomic oxygen can travel to large distances before getting lost in chemical reactions. Hence, we accounted for transport loss by taking 1 km s⁻¹ as advection velocity.

Zoom In Zoom Out Reset image size

We have accounted for many O(¹S) and O(¹D) formation and destruction processes in the coma as described in Bhardwaj & Raghuram (2012). The modeled major production rate profiles for O(¹S) in comet 67P are shown in Figure 5. The photodissociation of H₂O and CO₂ are equally important sources in producing O(¹S) in the inner coma of comet 67P. Below 100 km, suprathermal electron impact on CO₂ is the next important O(¹S) source. The photodissociation of CO and electron impact on H₂O and CO₂ are minor sources of O(¹S). Electron recombination of H₂O⁺ ion is a minor source of O(¹S) in the inner coma, whereas its contribution is significant at large (>10³ km) radial distances.

Zoom In Zoom Out Reset image size

The calculated O(¹D) production rate profiles for different formation processes are presented in Figure 6. The photodissociation of H₂O is a dominant source of O(¹D) throughout the inner coma. Contribution from other O(¹D) formation processes is minor (<5% to the total). At large radial distances (>10³ km) the contributions from dissociative recombination of H₂O⁺, radiative decay of O(¹S) and photodissociation of OH become significant in the formation of O(¹D).

Zoom In Zoom Out Reset image size

The modeled destruction rate profiles of O(¹S) and O(¹D) are shown in Figure 7. The O(¹S) and O(¹D) atoms are strongly quenched by H₂O up to radial distances of ∼10 and ∼200 km, respectively. Above these radial distances the radiative decay, which leads to forbidden visible emission lines, is the major loss process for the O(¹S) and O(¹D). Collisional quenching of O(¹S) and O(¹D) by other neutrals is smaller compared to H₂O quenching by several orders of magnitude, hence these processes are not shown in the figure.

Zoom In Zoom Out Reset image size

The production of atomic oxygen in the ⁵S state yields a [O i] 1356 Å emission line via immediate decay to ground state (with a lifetime of 185 μs, Johnson 1972). The calculated [O i] 1356 Å emission rates are presented in Figure 8. Electron impact on atomic oxygen is the major production source for [O i] 1356 Å emission followed by electron impact on CO₂ and H₂O.

Zoom In Zoom Out Reset image size

3.2.2. Atomic Carbon in ³P and ¹D States

In case of atomic carbon formation, the photodissociation of CO is the major source of C(³P), as shown in Figure 3. Collisional quenching of C(¹D) is the next important source of atomic carbon in the ground state. All other production processes described in Bhardwaj et al. (1996) contribute little (<5%) to the total. The loss of atomic carbon is mainly due to collisions with OH, which yield atomic hydrogen and CO. The next main loss source is due to collisions with H₃O⁺, which leads to the formation of HCO⁺ and H₂. The model accounts for the transport of atomic carbon in ³P and ¹D states with an advection velocity of 1 km s⁻¹. Transport is the major loss process for atomic oxygen and atomic carbon compared to the total loss due to chemical reactions.

The calculated formation rates of the metastable C(¹D) atom via different production processes are presented in Figure 9. The major formation mechanism for C(¹D) is the photodissociation of CO. At a large radial distance (10³ km), the dissociative recombination of the CO⁺ ion is also an important source of C(¹D). Other formation reactions are smaller compared to photodissociation of CO by more than an order of magnitude. The modeled C(¹D) loss rates presented in Figure 10 show that collisional quenching with water is the dominant loss process up to 300 km radial distance, above which radiative decay takes over. Collisional quenching of CO and CO₂ is a relatively less significant C(¹D) loss process.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The radial emission rate profiles are integrated for each emission line along the line of sight perpendicular to the Sun-comet direction at different radial distances to obtain limb brightness profiles. The model calculated intensity profiles, as a function of projected distance, for the CO Cameron band, atomic oxygen ([O i] 6300+6364, 5577, 2972, and 1356 Å), and atomic carbon emissions ([C i] 1931, 9823, 9850 Å) are shown in Figure 11. Among the calculated emission intensities, the CO Cameron band emission peaks close to the nucleus (<20 km). The calculated intensity profiles of 5577 Å and red-doublet (6300+6364 Å) are flat up to radial distances of 20 km and 200 km, respectively, due to the strong collisional quenching of O(¹S) and O(¹D) with H₂O in the inner coma. The calculated G/R ratio as a function of projected distance is also presented in the same figure on the right Y axis. Since the lifetime of metastable C(¹D) is large (4080 s), the collisional quenching with water makes [C i] 1931, 9850, and 9823 Å emission profiles flat up to 1000 km. The [O i] 1356 Å line is the weakest emission among the calculated emissions (presented in Figure 11 after multiplying a factor of 10).

Zoom In Zoom Out Reset image size

Model calculated intensity profiles, when the comet was at a 3 AU heliocentric distance, are presented in Figure 12 as a function of projected distance. Very close to the nucleus surface, the oxygen red-doublet, green line, CO Cameron, and [O i] 1356 Å emissions are intense. Since the neutral gas production rate is low (5 × 10²⁵ s⁻¹) at 3 AU, the calculated intensity profiles are decreased by two orders of magnitude compared to those at perihelion. Due to the high radiative lifetime (∼110 s), the collisional quenching of O(¹D) is significant for a radial distance up to 20 km, which alters the G/R ratio from 0.9 to 0.1. Inspite of having a high mixing ratio (15%), the role of CO in determining the oxygen visible emission intensities as well as in determining the G/R ratio is insignificant. In the case of the CO Cameron band, most of the emission intensity (>90%) close to the nucleus is mainly via the electron impact excitation of CO. For a radial distance higher than 50 km, the major (∼50%) source for CO(a³Π) is CO₂ via electron impact and photodissociation excitation processes. The long radiative lifetime (∼4080 s) of C(¹D) makes the atomic carbon emission intensity profile flat up to 1000 km.

Zoom In Zoom Out Reset image size

3.4.1. Role of CO₂ and CO Volume Mixing Ratios Relative to Water

By varying the μ_w(CO₂) and μ_w(CO), the contribution of different processes producing CO(a³Π) is calculated at three different projected distances. The calculations are presented in Table 1 by varying μ_w(CO₂) and μ_w(CO) from zero to two, and then to five percent. In all of these cases, the contribution from the photodissociation of CO₂ and electron impact on CO₂ processes are nearly equal. Keeping 1% μ_w(CO₂) in the coma, we varied μ_w(CO) between 1% and 5%. In this case, the contribution of photodissociation of CO₂ and electron impact on CO₂ in the inner coma is <15%, whereas electron impact on CO is about 65%–90%. The contribution from other chemical reactions, such as the electron recombination of HCO⁺ and ${\mathrm{CO}}_{2}^{+}$ ions contribute less than 10% to the formation of CO(a³Π).

Table 1.  The Calculated Contribution of Different Production Processes Producing CO(a³Π) in Comet 67P for Different CO₂ and CO Volume Mixing Ratios Relative to Water

Volume Mixing Ratios Relative to Water (%)		Percentage Contribution at Different Projected Distances (km)
		hν +CO2			eph + CO2			eph + CO			Othersa
CO2	CO	10	102	103	10	102	103	10	102	103	10	102	103
1	1	14.2	14.5	13.9	15.8	15.4	14.7	69.7	68.5	65.4	0.2	1.6	6.0
1	2	9.1	9.3	9.1	9.6	9.5	9.1	81.1	80.1	77.5	0.2	1.1	4.3
1	5	5.0	5.2	5.1	4.8	4.7	4.6	90.1	89.3	87.2	0.1	0.8	3.1
2	1	21.6	21.9	20.6	24.4	23.6	22.1	53.6	52.2	48.8	0.4	2.3	8.5
2	2	15.2	15.5	14.8	16.3	15.9	15.1	68.3	66.9	63.7	0.3	1.7	6.4
2	5	9.0	9.3	9.1	8.7	8.6	8.3	82.1	81.0	78.2	0.2	1.1	4.5
5	1	31.4	31.5	28.9	36.5	34.8	31.7	31.6	30.4	27.7	0.5	3.3	11.7
5	2	25.2	25.5	23.8	28.0	26.9	24.9	46.4	44.9	41.6	0.4	2.7	9.8
5	5	17.3	17.7	16.9	17.4	16.9	15.9	65.0	63.4	59.9	0.3	2.0	7.3
0	1	0.0	0.0	0.0	0.0	0.0	0.0	100.0	99.8	99.0	0.0	0.2	1.0
0	2	0.0	0.0	0.0	0.0	0.0	0.0	100.0	99.8	98.9	0.0	0.2	1.0
0	5	0.0	0.0	0.0	0.0	0.0	0.0	100.0	99.7	98.7	0.0	0.3	1.3
1	0	45.3	44.6	39.5	54.0	50.9	45.0	0.0	0.0	0.0	0.7	4.5	15.5
2	0	45.1	44.5	39.4	54.2	51.1	45.1	0.0	0.0	0.0	0.8	4.5	15.5
5	0	44.5	44.0	39.0	54.7	51.5	45.4	0.0	0.0	0.0	0.8	4.5	15.6

Note.

^aOthers corresponds to the sum of contributions from the dissociative recombination of HCO⁺ and ${\mathrm{CO}}_{2}^{+}$ ions and the resonance fluorescence of CO.

Download table as:  ASCIITypeset image

When μ_w(CO₂) is increased to 2%, the contribution from electron impact on CO is reduced to 50%–80%. About 10%–20% of CO(a³Π) is produced from the photodissociation of CO₂. By increasing μ_w(CO₂) to 5%, the contribution of CO₂, from both photodissociation and electron impact reactions, in producing CO(a³Π) increases to a total of 30%–60%. In the case of equal (5%) μ_w(CO) and μ_w(CO₂), the contribution from CO is around 65% and the rest comes from CO₂-associated reactions in the inner coma.

Modeling results in Table 1 show that in the absence of CO₂ the main source for production of CO(a³Π) is the electron impact of CO, and other processes have negligible contribution to the total. When CO is absent in the coma electron impact on CO₂ and photodissociation of CO₂ processes are producing CO(a³Π) with nearly equal contributions. In this case, at 1000 km projected distances, the contribution of the thermal recombination of HCO⁺ and ${\mathrm{CO}}_{2}^{+}$ producing CO(a³Π) is 15%.

The calculated percentage contribution of different processes in the formation of O(¹S) and O(¹D) are presented in Table 2. The main processes controlling the formation of O(¹S) is photodissociation of H₂O and CO₂. The photodissociation of H₂O is the dominant source of O(¹D) production in the inner coma. Beyond 1000 km radial distances, the photodissociation of OH, the electron recombination of H₂O⁺, and the radiative decay of O(¹S) are also important O(¹D) sources. The calculations presented in Table 2 show that below 100 km radial distance the formation of O(¹D) is mainly (80%–90%) through photodissociation of H₂O. Above these distances this contribution changes to around 50%, while the rest is via photodissociation of OH and dissociative recombination of H₂O⁺ and radiative decay of O(¹S).

Table 2.  The Calculated Contribution of Different Production Processes Producing O(¹S) and O(¹D) in Comet 67P for Different CO₂ and CO Volume Mixing Ratios Relative to Water

Volume Mixing Ratios		Percentage Contribution at Different Projected Distances (km)
Relative to Water (%)		hν + H2O			hν + CO2[hν +OH]			Othersa			G/R Ratio
CO2	CO	10	102	103	10	102	103	10	102	103	10	102	103
1	1	75.1[94.6]b	67.3[83.9]	44.3[54.2]	16.4[0.8]	14.5[5.4]	9.8[24.5]	8.5[4.6]	18.2[10.6]	45.9[21.3]	0.41	0.09	0.05
1	2	74.3[94.5]	66.6[83.9]	43.8[54.2]	16.2[0.8]	14.3[5.4]	9.6[24.5]	9.5[4.7]	19.1[10.7]	46.6[21.3]	0.42	0.09	0.05
1	5	71.7[94.3]	64.2[83.6]	42.1[54.0]	15.6[0.8]	13.8[5.4]	9.3[24.5]	12.6[4.9]	22.0[10.9]	48.7[21.5]	0.44	0.09	0.05
2	1	63.3[93.5]	57.2[82.7]	38.7[53.5]	27.6[0.8]	24.6[5.4]	17.0[24.2]	9.1[5.8]	18.2[12.0]	44.2[22.3]	0.49	0.10	0.05
2	2	62.8[93.4]	56.8[82.6]	38.3[53.5]	27.4[0.8]	24.4[5.4]	16.9[24.2]	9.8[5.8]	18.9[12.0]	44.8[22.3]	0.49	0.10	0.05
2	5	61.2[93.2]	55.2[82.4]	37.1[53.3]	26.6[0.8]	23.7[5.3]	16.3[24.1]	12.2[6.1]	21.1[12.3]	46.6[22.5]	0.51	0.10	0.06
5	1	43.2[90.2]	39.5[79.0]	28.0[51.4]	46.7[0.8]	42.3[5.1]	30.8[23.3]	10.2[9.1]	18.2[15.8]	41.2[25.3]	0.69	0.14	0.07
5	2	43.0[90.1]	39.3[79.0]	27.9[51.4]	46.5[0.8]	42.2[5.1]	30.7[23.3]	10.4[9.1]	18.5[15.9]	41.5[25.3]	0.69	0.14	0.07
5	5	42.5[89.9]	38.7[78.8]	27.3[51.3]	45.8[0.8]	41.6[5.1]	30.1[23.2]	11.7[9.3]	19.7[16.0]	42.6[25.5]	0.71	0.14	0.07
0	1	92.4[95.8]	81.7[85.3]	51.8[54.9]	0.0[0.8]	0.0[5.5]	0.0[24.9]	7.6[3.4]	18.3[9.2]	48.2[20.2]	0.34	0.07	0.04
0	2	90.9[95.7]	80.4[85.2]	51.0[54.9]	0.0[0.8]	0.0[5.5]	0.0[24.9]	9.1[3.5]	19.6[9.3]	49.0[20.2]	0.35	0.07	0.04
0	5	86.8[95.4]	76.8[84.9]	48.6[54.8]	0.0[0.8]	0.0[5.5]	0.0[24.8]	13.2[3.8]	23.2[9.6]	51.4[20.4]	0.36	0.08	0.04
1	0	75.9[94.7]	68.0[84.0]	44.8[54.2]	16.6[0.8]	14.6[5.4]	9.9[24.5]	7.5[4.5]	17.4[10.6]	45.3[21.3]	0.41	0.08	0.05
2	0	63.8[93.5]	57.6[82.7]	39.0[53.5]	27.8[0.8]	24.8[5.4]	17.2[24.2]	8.4[5.7]	17.6[11.9]	43.8[22.4]	0.48	0.10	0.05
5	0	43.2[90.2]	39.5[79.1]	28.1[51.4]	46.7[0.8]	42.4[5.1]	30.9[23.3]	10.1[9.0]	18.1[15.8]	41.0[25.4]	0.69	0.14	0.07

Notes.

^aOthers corresponds to the sum of the contributions from all reactions listed in Tables 1 and 2 of Bhardwaj & Raghuram (2012) for the formation of O(¹S) and O(¹D) except for the photodissociation of H₂O and CO₂. ^bThe values in the square brackets are for O(¹D).

Download table as:  ASCIITypeset image

In the case of O(¹S) production, both photodissociation of CO₂ and H₂O are important formation processes in the inner coma. It is found that the role of CO photodissociation is very small (<5%) in the O(¹S) production. Calculations presented in Table 2 show that for 1% of μ_w(CO₂), below a 100 km projected distance, the contributions in the formation of O(¹S) are 65%–75% from the photodissociation of H₂O, 15% from CO₂ photodissociation, and 20% from other reactions. At a 1000 km projected distance, the photodissociation of H₂O is contributing around 45% and ∼45% contribution mainly from dissociative recombination of H₂O⁺. In this case, the calculated G/R ratio varies between 0.05 and 0.4 for less than 1000 km projected distance. When we increased μ_w(CO₂) to 5%, the contribution from both photodissociation of H₂O and CO₂ is similar (30%–45%) for the projected distances less than 1000 km. In this case, the G/R ratio is found to vary between 0.07 and 0.7.

In the absence of CO₂, the photodissociation of water and dissociative recombination of H₂O⁺ mainly controls the formation of O(¹S) in the cometary coma. In this case, by changing the CO alone between 1% and 5%, it is found that the change in the calculated G/R ratio profile is insignificant. Assuming the absence of CO in the coma, the calculated contributions are not changed from the previous cases whereas the calculated G/R ratio profile is increasing linearly by increasing μ_w(CO₂).

We have calculated the [O i] 1356 Å emission intensity by varying μ_w(CO₂) and μ_w(CO) between 1% and 5%. The calculations show that electron impact on atomic oxygen is an important (50%) excitation process for [O i] 1356 Å emission (see Figure 8). Electron impact on CO₂ is the next important emission source of [O i] 1356 Å line (30% to the total). Below 50 km, the formation of atomic oxygen is through collisional quenching of O(¹D) with water (35%), charge exchange between O⁺, and OH⁺ with water (45%). Above this radial distance, 75% atomic oxygen is produced due to radiative decay of O(¹D). The role of CO₂ and CO in producing atomic oxygen is less than 5%. Hence, by increasing the μ_w(CO₂) in the coma it is found that 30% of this emission line intensity is increased only below 50 km radial distance. The role of CO in producing this emission line is insignificant.

The μ_w(CO) can significantly influence the [C i] 1931 Å emission intensity compared to that of CO₂. By increasing the μ_w(CO) from 1% to 5% it is found that the intensity of this emission line also increases. The [C i] 1931 Å is mainly through the photodissociation of CO (75%) and CO₂ (20%). The role of the resonant scattering of C(¹D) is significant (50%) for radial distances larger than 500 km. Similar effects have also been observed on 9850 and 9823 Å emission lines.

3.4.2. Role of H₂O Gas Production Rate

The ALICE ultraviolet spectrometer on board the Rosetta mission is designed to observe many emission lines from 67P in the wavelength region 750–2050 Å (Stern et al. 2007; Feldman et al. 2011). This range overlaps with part of CO Cameron bands covering 1800–2600 Å, [O i] 1356 Å, and [C i] 1931 Å emission lines (Feldman et al. 2004; Stern et al. 2007). By making limb scan observations, the ALICE spectrograph can be used to derive the spatial distribution of CO and/or CO₂ around the 67P nucleus. ALICE can observe the shortward part of CO Cameron band emission in its longward limit where, unfortunately, the sensitivity is small (Stern et al. 2007). Similarly, the [C i] 1931 Å emission line also falls into the longer end of the ALICE spectral range. It may, however, possible to detect it because of strong resonant fluorescence efficiency of C(¹D) atom. The Rosetta on board OSIRIS is a scientific camera system with 12 discrete filters, which is designed to observe 67P cometary coma over the wavelength rage of 250–1000 nm (Keller et al. 2007). OSIRIS can also map the release of certain daughter species such as OH and [O i], based on observed emission intensities at 3090 Å and at 6300 Å, respectively.

The outcome of this study can be compared with on board ROSINA mass spectrometers in situ measurements of the neutral composition in the coma (Balsiger et al. 2007; Hässig et al. 2015). Space-based observations from Earth, such as from the HST, can also observe these ultraviolet emissions during the Rosetta mission observation period. Several ground-based observatories have been observing comet 67P (http://www.rosetta-campaign.net/planned-observations) in visible and infrared regions to study the spatial distribution of various species. In this context the present modeling work can provide a better understanding of different processes governing the CO Cameron band, atomic oxygen, and atomic carbon emission lines in comet 67P to derive parent neutral composition in the coma.

By making several observations on different comets, Tozzi et al. (1998) demonstrated that there is a strong correlation between 1931 Å emission line intensity and CO column density. The radiative decay of C(¹D) to ground state yields 9823 and 9850 Å emission lines. By observing these emission lines on Hale–Bopp, Oliversen et al. (2002) concluded that they can be used as direct tracers of CO photodissociation in the cometary coma. The model calculations also show that the major production source of C(¹D) in the inner coma is mainly due to the photodissociation of CO and the contribution from other production processes is smaller by an order of magnitude compared to the former (see Figure 9). The model calculated that 1931, 9850, and 9823 Å emission line profiles are flat up to 1000 km projected distances due to the strong collisional quenching of C(¹D) with H₂O. Since these atomic carbon emission lines are mainly controlled by the photodissociation of CO, observed intensity profiles can be used to derive the CO gas production rate in the coma. The calculated emission intensity profile can be useful as a baseline prediction to constrain the CO mixing ratios in the coma for the ALICE observation of carbon emission lines, which can then be compared with the ROSINA observations for the same regions under similar solar illumination.

Measuring atomic oxygen visible emission line intensities is an important diagnostic tool in estimating the water production rate as well as to understand the spatial distribution of H₂O in the cometary coma (Delsemme & Combi 1976, 1979; Fink & Johnson 1984; Schultz et al. 1992; Morgenthaler et al. 2001; Furusho et al. 2006). Decock et al. (2015) analyzed several ESO's VLT observed high-resolution green- and red-doublet emission line spectra on various comets. CO₂ mixing ratios are derived in these comets by comparing the ESO VLT observations with modeled G/R ratio profiles (Decock et al. 2015). The model calculated G/R ratio profile is presented on 67P in Figure 11. By modeling green- and red-double emission intensity profiles on various comets at different heliocentric distances, Raghuram & Bhardwaj (2014) have shown that the G/R ratio value increases linearly by increasing μ_w(CO₂) in the coma, whereas the affect of CO is minor in determining either green- or red-doublet emission intensities. Hence, the observed G/R ratio profile on 67P can be used to constrain CO₂ mixing ratio in the coma.

When a comet is far away from the Sun (3 AU), it is expected to have higher CO and CO₂ volume mixing ratios, which are species associated with low sublimation temperatures (e.g., Mumma & Charnley 2011). The calculations made at 3 AU heliocentric distance (see Figure 12), with mixing ratios of 5% CO₂ and 15% CO, show that atomic oxygen red-doublet emission is the most intense emission in the inner coma. This emission can be observed by Rosetta on board the OSIRIS instrument, which can be subsequently used to derive the water production rate. Unfortunately, there are no filters on OSIRIS to measure [O i] 5577 Å, [C i] 9823 Å, and 9850 Å emission lines (Keller et al. 2007). The predicted oxygen red-doublet intensity along the projected distance (Figures 11 and 12) could be useful in analyzing OSIRIS visible spectra of 67P and, subsequently, deriving H₂O distribution around the nucleus.

The photon cross section for the formation of O(¹S) from H₂O has never been reported in the literature (Huestis & Slanger 2006). In this model, the formation of O(¹S) from photodissociation of H₂O has been accounted for by assuming 0.5% yield for H₂O at Lyα wavelength (Bhardwaj & Raghuram 2012). The spectrometers on board ROSINA can measure the CO₂ number densities during this mission period at different radial distances in the coma. By combining the observed G/R ratio profile with ROSINA CO₂ measurements, it would be possible to constrain the O(¹S) average yield value at solar Lyα. The high-resolution spectroscopic observations, such as the analysis of Decock et al. (2015), can provide information about collisional quenching of O(¹S) and O(¹D) metastable states in the coma of 67P. The observation of both green- and red-doublet emission line widths and G/R ratio profiles along with ROSINA measurements can solve the puzzle that the green line is wider than either of the red-doublet emission lines in comets.

The modeling of production rates of CO(a³Π) has shown that suprathermal electron impact reactions mainly govern the CO Cameron band emission with a contribution of around 75%, whereas CO₂ photodissociation contributes about 25% (see Section 3.1). In the absence of CO, the electron impact on CO₂ is an equally important production source of CO(a³Π) as photodissociation of CO₂ (see Table 1). This suggests that the electron impact excitation mechanism should be considered for the estimation of parent species production rates in the coma. With sufficient CO (≥3%) in the coma, the contribution from electron impact reactions in producing this band emission close to the nucleus (<100 km) is about 80%. The excited state CO(a³Π) is mainly populated in the coma by suprathermal electrons in the energy range between 10 and 15 eV (see Figure 2). Since the major source for the production of CO(a³Π) is electron impact, the observed CO Cameron band emission close to the nucleus would be suitable to track the suprathermal electron intensity (McPhate et al. 1999) rather than CO₂ neutral density.

The [O i] 1356 Å emission line is an excellent tracer for electron impact processes in the coma. Modeling of electron impact excitation processes shows that this emission is mainly due to electron impact excitation of atomic oxygen followed by electron impact dissociative emission of CO₂ and H₂O (see Figure 8). However, the intensity of this emission is weaker by three orders of magnitude compared to CO Cameron band emission. The electron impact excitation cross section for atomic oxygen producing [O i] 1356 Å emission line peaks at 15 eV, whereas for CO₂ and H₂O, it is between 30 and 60 eV. The contribution from atomic oxygen is about 50% and the rest is through CO₂ (30%) and H₂O (20%). Hence, half of the observed emission intensity profile is linked to the suprathermal electron intensity at 15 eV. Recently, ALICE observed several H i, O i, and C i emissions near the cometary nucleus when the comet was at around 3 AU (Feldman et al. 2015). The observation of O i 1356 Å emission line intensity, which is varying between ∼1.5 and ∼3 Rayleigh at 10 km projected distance, (Feldman et al. 2015), is close to our predicted calculation (∼1.5 Rayleigh, see Figure 12). Detailed analysis of this emission line will be presented in future work using the Rosetta measured neutral density distribution around the nucleus.

The Rosetta Plasma Consortium (RPC)/Ion and Electron Sensor (IES) is capable of measuring the electron energy spectra in the energy rage 1 eV/e to 22 keV/e (Burch et al. 2007). Since both CO Cameron band and [O i] 1356 Å emission lines are governed mainly by electron impact excitation reactions, the observed emission intensities may be supportive for the IES measured suprathermal electron intensity at around 15 eV.

Molecular oxygen is the major source for the production of O(¹S) and O(¹D) in the terrestrial atmosphere. The recent discovery of O₂ in 67P's coma by ROSINA/DFMS (Bieler et al. 2015) demands the inclusion of O₂ in the model in order to calculate green- and red-doublet emission intensities. By including 4% molecular oxygen with respect to water production rate, and for the input conditions described in Section 2, the G/R ratio is found to increase by around 20% close to the nucleus (<20 km projected distance). Bieler et al. (2015) observed that the relative abundance of molecular oxygen ranges from 1%–10% with respect to the H₂O production rate. Hence, in order to determine CO₂ abundance based on the G/R ratio, the contribution from molecular oxygen should also be considered. In the case of higher O₂ abundance in comets, the observed G/R ratio can be significantly controlled by photodissociation of O₂ and may lead to the underestimation of the CO₂ mixing ratio.

The estimated diamagnetic cavity on the sunlit side of this comet at perihelion is around 30–40 km (Benna & Mahaffy 2006; Hansen et al. 2007; Koenders et al. 2015). The extent of the diamagnetic cavity depends on the gas production rate and solar wind conditions during the comet perihelion visit. Beyond this cavity, most of the ions are transported toward the tail side due to solar wind interaction. The assessment of solar wind interaction on the emission intensities is beyond the scope of this work, but we would like to discuss the possible sources that can alter the emission intensities. Outside the diamagnetic cavity, the chemical lifetime of neutrals can be significantly altered by charge exchange between solar wind ions and cometary species. Hence, it is expected that the calculated intensities outside of the diamagnetic cavity can be changed based on the solar wind conditions during that time. The electrons outside of the diamagnetic cavity are primarily solar wind electrons or shocked solar wind electrons (Gringauz et al. 1986; Gan & Cravens 1990; Cravens 1991; Reme 1991; Ip 2004). The population of suprathermal electrons outside of the diamagnetic cavity is a complex problem due to admixture of solar wind electrons. However, the radius of collisional zone and diamagnetic cavity are subjected to the gas production rate and solar wind conditions during the comet perihelion passage.

For electron impact driven emissions, such as the CO Cameron band and [O i] 1356 Å, due to strong solar wind interaction, both neutral density and electron population may change outside of the diamagnetic cavity region, thus the observed emission intensities vary significantly. In this region,the solar wind electrons may also contribute to the total emission intensity (Bhardwaj et al. 1990, 1996; Bhardwaj 1999). However, the dissociative recombination CO-bearing ions to the total emission intensity contribute little (<5%), whereas the formation of atomic oxygen contributes significantly (50%), because of charge exchange between O⁺ and OH⁺ with H₂O. We do not expect that the radiative decay and collisional quenching of O(¹D) can be altered significantly due to solar wind interaction.

The evolution of cometary ionosphere around the 67P nucleus has been monitored by the Rosetta Plasma Consortium and Rosetta Orbiter Spectrometer for Ion and Neutral Analysis (ROSINA) instruments. The recent observations of the ROSINA/Double Focussing Mass Spectrometer (DFMS), RPC/Ion and Electron Sensor (IES), and RPC/Ion Composition Analyzer (ICA), when the comet was beyond 2 AU, have shown that, due to the low out-gassing rate, no contact surface is formed and most of the solar wind has directly accessed the 67P's nucleus, though the plasma close to the comet is dominated by cometary water ions (Broiles et al. 2015; Fuselier et al. 2015; Nilsson et al. 2015). Clark et al. (2015) found that the suprathermal electrons are accelerated to several hundreds of eV. The high energetic solar wind charged particles may also be involved in producing the excited atomic and molecular states discussed here.

The formation of O(¹S), O(¹D), and C(¹D) is mainly due to photochemical reactions. The contribution from ions and thermal electron recombination reactions for the inner coma is very small (<5%). Hence, we do not expect the predicted oxygen visible and [C i] 1931 Å line emission intensities to change due to solar wind interaction for the inner coma unless the radial distribution of H₂O is changed.

In this model, we have accounted for main parent oxygen- and carbon-bearing species to compute the emission intensities. However, the contribution from photodissociation and electron impact of other minor species is also possible. In the case of atomic oxygen visible emissions, the dissociation of other oxygen-bearing species, such as HCOOH or H₂CO, are unlikely to be the parent because they cannot decay fast enough to produce O(¹D) and O(¹S) (Festou & Feldman 1981). However, in the case of carbon emissions there could be an involvement from other carbon-bearing species, such as hydrocarbons. Since the major processes governing these emissions are via photochemical reactions, the role of electron temperature for the inner coma is not significant (<5%). The model calculations are done for the gas phase, thus scattering of solar photons by dust grains could be a significant factor in governing these emission intensities.

The recent ROSINA/DFMS observations on 67P show that the cometary coma contains a variety of species with heterogeneous distributions, which vary with time and latitude (e.g., Hässig et al. 2015; Le Roy et al. 2015). The Microwave Instrument on the Rosetta Orbiter (MIRO) mapped around 67P's nucleus when it was at 3.4 AU (Biver et al. 2015; Lee et al. 2015). The water column density in the inner coma (within 3 km from the nucleus) is found to vary even by two orders of magnitude. Luspay-Kuti et al. (2015) further investigated the heterogeneity of 67P's coma by measuring various major (H₂O, CO₂, and CO) and minor (HCN, CH₃OH, CH₄, and C₂H₆) volatile species using ROSINA/DFMS. Our calculated emission intensities may change significantly due to variable neutral densities around comet 67P. Future work will include the use of in situ measured neutral densities from ROSINA sensors to drive model calculated emission intensities. Results could then be compared with ALICE and OSIRIS observations as well as ground-based observations.