Jovian Meteoroid Environment Model JMEM: Dust from the Galilean Satellites

Transcription

1 Jovian Meteoroid Environment Model JMEM: Dust from the Galilean Satellites J. Schmidt, X. Liu (U Oulu) M. Sachse, F. Spahn (U Potsdam) R. Soja, R. Srama (U Stuttgart) N. Altobelli, C. Vallat (ESA) (images: NASA) 1

2 Background 2

3 Jupiter s ring system and ring moons (images: NASA) 3

4 Jupiter s ring system and ring moons DUST: micron sized particles, silicates or ice (images: NASA) 3

5 Jupiter s ring system and ring moons DUST: micron sized particles, silicates or ice (images: NASA) (Hamilton&Krüger, 2008) 3

6 Jupiter s ring system and ring moons DUST: micron sized particles, silicates or ice (images: NASA) (Hamilton&Krüger, 2008) 3

7 four large moons Io Ganymede rings Europa Callisto 4

8 5

9 Dust is all over the place: images: NASA) 6

10 Dust is all over the place: * main rings, halo & gossamer ring images: NASA) 6

11 Dust is all over the place: * main rings, halo & gossamer ring * stream particles from Io's volcanoes images: NASA) 6

12 Dust is all over the place: * main rings, halo & gossamer ring * stream particles from Io's volcanoes * dust from the irregular satellites images: NASA) 6

13 Dust is all over the place: * main rings, halo & gossamer ring * stream particles from Io's volcanoes * dust from the irregular satellites * external dust, magnetospherically captured images: NASA) 6

14 Dust is all over the place: * main rings, halo & gossamer ring * stream particles from Io's volcanoes * dust from the irregular satellites * external dust, magnetospherically captured * dust from the Galilean moons images: NASA) 6

15 7

16 7

17 7 hypervelocity impacts of interplanetary dust particles => erode surfaces of atmosphereles bodies

18 (LASP/CU Boulder) LADEE dust detector Luna (Horanyi et al,nature, 2015) 7 hypervelocity impacts of interplanetary dust particles => erode surfaces of atmosphereles bodies

19 Dust from the Galilean moons: -> dust exospheres (Krüger, 1999, Nature) replenish cirumplanetary dust environment Modeling Krivov et al., 2003 Sremcevic et al.,

20 Dust from the Galilean moons: -> dust exospheres (Krüger, 1999, Nature) replenish cirumplanetary dust environment Modeling Galileo DDS data: Krivov et al., 2003 Sremcevic et al., 2003 (Krüger et al, 1999, Nature) (Krüger et al, 2003, Icarus) 8

21 Dust from the Galilean moons: -> dust exospheres (Krüger, 1999, Nature) replenish cirumplanetary dust environment -> relevant for JUICE dust hazard Galileo DDS data: Modeling Krivov et al., 2003 Sremcevic et al., 2003 (Krüger et al, 1999, Nature) (Krüger et al, 2003, Icarus) 8

22 Dust from the Galilean moons: -> ejecta from impacts of interplanetary dust: fill region of Galilean moons -> relevant for JUICE dust hazard NaCl -> water (Postberg et al, 2009, Nature) -> S/C with dust spectrometer: analyse grain composition * Cassini CDA@Enceladus 9

23 Dust from the Galilean moons: -> ejecta from impacts of interplanetary dust: fill region of Galilean moons -> relevant for JUICE dust hazard -> S/C with dust spectrometer: analyse grain composition * Cassini CDA@Enceladus nano-silica -> hydrothermal activity (Hsu et al, 2015, Nature) 10

24 Dust from the Galilean moons: -> ejecta from impacts of interplanetary dust: fill region of Galilean moons -> relevant for JUICE dust hazard -> S/C with dust spectrometer: analyse grain composition * Cassini CDA@Enceladus * Europa Clipper SUDA@Europa 11

25 Dust from the Galilean moons: -> ejecta from impacts -> habitability: of interplanetary dust: non-ice composition fill region of Galilean exchange processes moons -> relevant for JUICE dust hazard -> S/C with dust spectrometer: analyse grain composition * Cassini CDA@Enceladus * Europa Clipper SUDA@Europa 11

26 Galileo Dust Detection Subsystem (DDS) DDS data (H. Krueger) NUMBER DENSITY [km**-3] Io Europa Ganymede RADIAL DISTANCE [RJ] rs/jschmidt/projects/dyne2/galileodata/ddsplot jschmidt Mon Sep 12 11:52: Callisto

27 Galileo Dust Detection Subsystem (DDS) DDS data (H. Krueger) NUMBER DENSITY [km**-3] Io Europa Ganymede RADIAL DISTANCE [RJ] rs/jschmidt/projects/dyne2/galileodata/ddsplot jschmidt Mon Sep 12 11:52: Callisto

28 Galileo Dust Detection Subsystem (DDS) DDS data (H. Krueger) NUMBER DENSITY [km**-3] Io Europa (Hamilton&Krüger, 2008, Nature) Ganymede RADIAL DISTANCE [RJ] rs/jschmidt/projects/dyne2/galileodata/ddsplot jschmidt Mon Sep 12 11:52: Callisto

29 Dust Model for the Galilean Satellites: JMEM 13

30 14 (Liu et al, JGR, 2016)

31 A dust creation, launch to orbit: -> want to test effect of different starting distributions? -> dynamics through Hill-sphere (nearly) generic 14 (Liu et al, JGR, 2016)

32 A dust creation, launch to orbit: B long term orbital evolution: -> CPU expensive. Want to re-use individual trajectories -> statistical weighting with probabilities from A -> want to test effect of different starting distributions? -> dynamics through Hill-sphere (nearly) generic 14 (Liu et al, JGR, 2016)

33 1 st step: ~10 6 particles from surface -> tabulate phase space Hill sphere 15 (Liu et al, JGR, 2016)

34 1 st step: ~10 6 particles from surface -> tabulate phase space Hill sphere 15 (Liu et al, JGR, 2016)

35 1 st step: ~10 6 particles from surface -> tabulate phase space Hill sphere 15 (Liu et al, JGR, 2016)

36 -> for each Galilean moon ~10 CPU hours -> repeat for various starting distributions, if wanted 1 st step: ~10 6 particles from surface -> tabulate phase space Hill sphere 15 (Liu et al, JGR, 2016)

37 -> for each Galilean moon ~10 CPU hours -> repeat for various starting distributions, if wanted 1 st step: ~10 6 particles from surface -> tabulate phase space Hill sphere -> determine abscissas and weights for Gauss integration = weighting of particle trajectories (Liu et al, JGR, 2016) 15

38 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

39 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG planet, inc. higher moments 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

40 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG planetary magnetic field + corotational E field 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

41 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG radiation pressure and Poynting-Robertson drag 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

42 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG plasma drag 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

43 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG gravity of satellites 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

44 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG solar gravity 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) 16 (Liu et al, JGR, 2016)

45 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG solar gravity 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) -> do this for 10 particle sizes from 0.05 micron to 1cm -> integrate until the particle hits a sink: planet, moons, escape from the system -> repeat for each moon: 60,000 CPU hours on large cluster (CSC Espoo) 16 (Liu et al, JGR, 2016)

46 ~r = ~r JUP + ~r EM + ~r RP + ~r DRAG + ~r MOONS + ~r SOLG solar gravity 2 nd step: for given moon do -> long-term integrations (using abscissa values of Gaussian weighting as starting conditions) -> do this for 10 particle sizes from 0.05 micron to 1cm -> integrate until the particle hits a sink: planet, moons, escape from the system -> repeat for each moon: 60,000 CPU hours on large cluster (CSC Espoo) storage: save (osculating) orbital elements ~ once per orbit 16 (Liu et al, JGR, 2016)

47 3 rd step: populate cylindrical grid from elements: Ganymede positions times J Europa 17

48 3 rd step: populate cylindrical grid from elements: Ganymede positions times J increment location in the grid Europa 18

49 3 rd step: populate cylindrical grid from elements: Ganymede positions times J increment location in the grid Europa 19

50 4 th step: co-adding and calibration number density for grains of size r in cell i,j,k of grid 20

51 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr Z d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} number density for grains of size r in cell i,j,k of grid 20

52 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } pos and direction of velocity starting vector 20

53 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid B) grid from 2 nd step d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } pos and direction of velocity starting vector 20

54 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid B) grid from 2 nd step d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } pos and direction of velocity starting vector numerical integration: use Gaussian weights 20

55 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid n total (i, j, k; >r)= X B) grid from 2 nd step d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } moons pos and direction of velocity starting vector numerical integration: use Gaussian weights c moon Z 1 r dr 0 f(r 0 )ñ moon (i, j, k; r 0 ) 20

56 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid n total (i, j, k; >r)= X B) grid from 2 nd step d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } moons pos and direction of velocity starting vector numerical integration: use Gaussian weights c moon Z 1 r dr 0 f(r 0 )ñ moon (i, j, k; r 0 ) 20 plausible power law: e.g. r -3.7

57 4 th step: co-adding and calibration ñ moon (i, j, k; r)dr =dr A) distributions of starting conditions 1 st step Z number density for grains of size r in cell i,j,k of grid n total (i, j, k; >r)= X B) grid from 2 nd step d P moon ( ) N moon( ; i, j, k; r) V cell (i, j, k) where = {,,,, v} {z } moons pos and direction of velocity starting vector numerical integration: use Gaussian weights c moon Z 1 r dr 0 f(r 0 )ñ moon (i, j, k; r 0 ) normalization from Galileo DDS 20 plausible power law: e.g. r -3.7

58 4 th step: co-adding and calibration * do the same for the velocity in a given grid cell ~v total (i, j, k; >r)= X moons c moon Z 1 r dr 0 f(r 0 ) ~v moon (i, j, k; r 0 ) normalization from Galileo DDS 21 plausible power law: e.g. r -3.7

59 4 th step: co-adding and calibration * do the same for the velocity in a given grid cell * to reconstruct the flux on a user specified surface: also store moments for <v 2 > and <v 3 > ~v total (i, j, k; >r)= X moons c moon Z 1 r dr 0 f(r 0 ) ~v moon (i, j, k; r 0 ) normalization from Galileo DDS 21 plausible power law: e.g. r -3.7

60 4 th step: co-adding and calibration * do the same for the velocity in a given grid cell * to reconstruct the flux on a user specified surface: also store moments for <v 2 > and <v 3 > * populating the grid cells takes about 2 CPU hours for all moons, grain sizes X Z 1 ~v total (i, j, k; >r)= c moon moons r dr 0 f(r 0 ) ~v moon (i, j, k; r 0 ) normalization from Galileo DDS 21 plausible power law: e.g. r -3.7

61 Calibration, in practice NUMBER DENSITY [km**-3] Galileo Dust Detection Subsystem (DDS) DDS data (H. Krueger) DDS detected particles >0.3 micron (Krüger et al, 2009) or >0.6 micron RADIAL DISTANCE [RJ] /Users/jschmidt/Projects/Dyne2/GalileoData/DDSPlot jschmidt Mon Sep 12 11:52:

62 Calibration, in practice NUMBER DENSITY [km**-3] Galileo Dust Detection Subsystem (DDS) DDS data (H. Krueger) DDS detected particles >0.3 micron (Krüger et al, 2009) or >0.6 micron use these data to calibrate the model RADIAL DISTANCE [RJ] /Users/jschmidt/Projects/Dyne2/GalileoData/DDSPlot jschmidt Mon Sep 12 11:52:

63 Grain lifetimes Lifetime [years] Grain radius 23 [microns]

64 Grain lifetimes Lifetime [years] small grains ejected by co-rotational E-field Grain radius 23 [microns]

65 Grain lifetimes Lifetime [years] large eccentricities: collisions with moons small grains ejected by co-rotational E-field Grain radius 23 [microns]

66 Grain lifetimes eccentricities small Lifetime [years] large eccentricities: collisions with moons small grains ejected by co-rotational E-field Grain radius 23 [microns]

67 Model: Grains from Io only inertial frame Io Callisto

68 Model: Grains from Io only inertial frame evolution radially outward: ȧ>0 -> plasma drag Io Callisto

69 Model: Grains from Io only inertial frame evolution radially outward: ȧ>0 -> plasma drag ė>0 -> solar radiation Io Callisto

70 Model: Grains from Io only inertial frame evolution radially outward: ȧ>0 -> plasma drag ė>0 -> solar radiation Io inclinations: -> solar radiation + scattering by moons Callisto

71 Model: Grains from Europa only 25 inertial frame

72 Model: Grains from Ganymede only 26 inertial frame

73 Model: Grains from Callisto only 27 inertial frame

74 Calibration of the model with Galileo DDS: 28

75 Calibration of the model with Galileo DDS: grains from Io 28

76 Calibration of the model with Galileo DDS: grains from Io from Europa 28

77 Calibration of the model with Galileo DDS: grains from Io all from Europa 28

78 Calibration of the model with Galileo DDS: grains from Io all from Europa DDS threshold s>0.3 micron (Krüger et al, 2009) 28

79 Calibration of the model with Galileo DDS: grains from Io all from Europa DDS threshold s>0.3 micron (Krüger et al, 2009) -> most are ejecta from Io 28

80 Calibration of the model with Galileo DDS: grains from Io all from Europa 28 DDS threshold s>0.3 micron (Krüger et al, 2009) -> most are ejecta from Io -> little or nothing from Ganymede/ Callisto

81 Calibration of the model with Galileo DDS: 29 DDS threshold s>0.6 micron (Krüger et al, 2009) -> larger contribution by grains from Ganymede

82 All moons: inertial frame

83 Grains > 10 micron 31 inertial frame

84 Vertical cut

85 Only grains from Callisto: -> very low density -> larger vertical spread 33

86 Systematic velocities dust velocity -local kepler velocity 34

87 Systematic velocities dust velocity -local kepler +Ganymede: dust is slower than moons 34

88 Systematic velocities dust velocity -local kepler dust is faster than moon 35

89 In a frame with fixed orientation towards the Sun: 0.6µm grains from Ganymede 36 (Liu et al, JGR, 2016)

90 In a frame with fixed orientation towards the Sun: 0.6µm grains from Ganymede apocenters, pericenters get organized wrto subsolar longitude! 36 (Liu et al, JGR, 2016)

91 In a frame with fixed orientation towards the Sun: 0.6µm grains from Ganymede radiation pressure pumps eccentricity apocenters, pericenters get organized wrto subsolar longitude! 36 (Liu et al, JGR, 2016)

92 `heliotropic dust rings 2.0µm grains from Callisto 37 (Liu et al, JGR, 2016)

93 `heliotropic dust rings 2.0µm grains from Callisto point towards or away from Sun, depending on grain size: $ changes sign 37 (Liu et al, JGR, 2016)

94 Calculation of fluxes: Z J = d 3 vf(~v ) ~v sc ~v H (cos!) we cannot store the full velocity distribution function in each grid point! 38

95 Calculation of fluxes: Z J = d 3 vf(~v ) ~v sc ~v H (cos!) we cannot store the full velocity distribution function in each grid point! and: approximation J = nv does NOT hold in many cases! 38

96 Calculation of fluxes: Z J = d 3 vf(~v ) ~v sc ~v H (cos!) we cannot store the full velocity distribution function in each grid point! and: approximation J = nv does NOT hold in many cases! => save 4 moments of the velocity components and use known algorithm to reconstruct the distribution from the moments 38

97 Calculation of the flux: Example 39

98 Calculation of the flux: Example `true' distribution reconstruction from moments 39

99 Calculation of the flux: Example `true' distribution reconstruction from moments 39

100 e.g. particles >300µm accumulated by JUICE: DDS threshold 40

101 e.g. particles >300µm accumulated by JUICE: DDS threshold VREL~1-4km/s 40

102 e.g. particles >300µm accumulated by JUICE: DDS threshold VREL~1-4km/s 41

103 e.g. particles >300µm accumulated by JUICE: flux: ~ 6 fold for this calibration, 3% chance to have one such impact on 10m 2 surface DDS threshold VREL~1-4km/s 41

104 Fluxes onto the surfaces of the moons: space weathering, chemical alteration

105 Fluxes onto the surfaces of the moons: space weathering, chemical alteration IR/UV color ratio leading trailing Saturn s moon Dione (Schenk et al, 2010) planet

106 Fluxes onto the surfaces of the moons: EUROPA space weathering, chemical alteration

107 Fluxes onto the surfaces of the moons: space weathering, chemical alteration Ganymede 45

108 Fluxes onto the surfaces of the moons: space weathering, chemical alteration Callisto 46

109 Summary * new model/code to simulate dust environment of Jupiter * size-resolved number densities and fluxes 47

110 Summary * new model/code to simulate dust environment of Jupiter * size-resolved number densities and fluxes * applications: -> dust hazard, degradation of SC hardware -> pointing information for instruments on future missions: JUICE (PEP-JoEE, RPWI) Europa Clipper (SUDA) -> characterize external inflow of dust on surfaces, e.g. Io dust on Europa -> space weathering 47

111 Summary * TBD: Sputtering, variable charging processes, different materials (e.g. silicates) 48

112 Summary * TBD: Sputtering, variable charging processes, different materials (e.g. silicates) * future applications of the code to: -> Jovian main and gossamer rings -> dust from the irregular satellites (dark color of Callisto & Ganymede) 48

113 Summary * TBD: Sputtering, variable charging processes, different materials (e.g. silicates) * future applications of the code to: -> Jovian main and gossamer rings -> dust from the irregular satellites (dark color of Callisto & Ganymede) Thank You! 48

114 spare slides 49

115 50 (Liu et al, JGR, 2016)

116 51

117 Northern latitude IO resolution:50x50 levels:20 sources: Western longitude Trailing Leading flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux flux Jupiter