Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 19940022903: Studies for Io's extended atmosphere and neutral clouds and their impact on the local satellite atmosphere and on the planetary magnetosphere"

See other formats


Studies for Io's Extended Atmosphere and Neutral Clouds 
and Their Impact on the Local Satellite Atmosphere 
and on the Planetary Magnetosphere 


William H. Smyth 


Atmospheric and Environmental Research, Inc. 
840 Memorial Drive 
Cambridge, Massachusetts 02139-3794 


December 9, 1993 


Final Report for the Period 
December 10, 1992 to December 9, 1993 

Prepared for 
NASA Headquarters 




1 R rpofl Ho 


vrrnmrnt A((fUi(m Ho 


TECHNICAL REPORT STANDARD TITLE PACE 


e c i p • r r» » % Co»oloq No. 


4 T »rTe o*>(j Swbt.ll • Studies rcr Id's Extended 
Atmosphere and Neutral Clouds and Their 
Impact on the Local Satellite Atmosphere and I Perform. n 9 
on the Planetar y Magnetosphere * 


7 Author ( «) 

William H. Smyth 

9. Pfflofmmg Onjontiotton Nome onj Addreil 

Atmospheric and Environmental Research, Inc. 
840 Memorial Drive 
Cambridge, MA 02139-3794 


12. Sponsoring Agency Home ond Address 

NASA Headquarters 
Headquarters Division 
Washington, DC 20546 


IS. Supplementory Notes j 


8 . Per lo» m 

ng Orgomiotion Report No 

10 Work Uo 

it No 

11 Control 

or Gron? No. 

NASW- 

-4771 

13. T ype o 1 

Report and Period Cohered 

Final Report 

12/10/92 - 12/9/93 

1 4. Sponsor* 

ng Agency Code 


16. Abfttroct 

The research performed in this project is divided in two main investigations: (1) the 
synthesis and analysis of a collection of independent observations for Io's sodium 
corona, its sodium extended atmosphere, and the sodium cloud, and (2) the analysis of a 
(System III longitude correlated) space-time "bite-out" near western elongation in the 
1981 sodium cloud images from the JPL Table Mountain Sodium Cloud Data Set. For 
the first investigation, modeling analysis of the collective observed spatial profiles has 
shown that they are reproduced by adopting at lo's exobase a modified sputtering flux 
speed distribution function which is peaked near 0.5 km/s and has a small high-speed 
(15-20 km/s) nonisotropic component. The nonisotropic high-speed component is 
consistent with earlier modeling of the trailing directional feature. For the second 
investigation, modeling analysis of the "bite-out" observed near western elongation (but 
not eastern elongation) has shown that it is reproduced in model calculation by adopting 
a plasma torus description for the sodium lifetime that is inherently asymmetric in 
System III longitudes of the active sector and that also has an east-west asymmetry. 
The east-west and System 111 longitude asymmetries were determined from independent 
observations for the plasma torus in 1981. The presence of the "bite-out feature only 
near western elongation may be understood in terms of the relative value for sodium of 


17. Key Wordft (Selected by Avthor(ft)) 

18. Distribution Stotement 

satellite corona and extended 


atmospheres, plasma-neutral 


interactions in magnetospheres 

1 — — — m ” 


Il9. Security Clot « if. (of fb»* report) 1 20. Security Clofttif. (of this page) |?1- No. of Poge* 


Unclassified 


Unclassified 


* For *»lr by (hr Or«(in^SmKf for F rdrr* 1 Sctrniific mnd Trrhmc*l Informanon. Sf>r t ng f i r I d , Virginia 221^1. 











TABLE OF CONTENTS 


Page 

Standard Title Page i 

Table of Contents i> 

I. INTRODUCTION 1 

II. IO'S SODIUM CORONA AND SPATIALLY EXTENDED CLOUD 2 

2.1 Observational Data for Sodium 2 

2.2 Analysis of the Sodium Observations 4 

III. SYSTEM III ASYMMETRY IN IO'S SODIUM CLOUD 4 

3.1 Observational Data for Sodium 4 

3.2 Observational Data for the Io Plasma Torus 6 

3.3 Analysis of the Sodium Observations 7 

REFERENCES 

APPENDICES 

A. Io's Sodium Corona and Spatially Extended Cloud: A 

Consistent Flux Speed Distribution 

B. Correlating System III Longitudinal Asymmetries in the 

Jovian Magnetosphere and the Io Sodium Cloud 


ii 




I. 


INTRODUCTION 


The overall objective of this research has been to undertake certain studies for the 
sodium corona, its spatial extension, and the more distance sodium cloud by analysis of 
their ground-based observations. The major objectives of these studies have been to 
improve significantly our understanding of the nature of the escape of sodium from Io, 
the nature of the space-time variabilities of the extended sodium atmosphere beyond the 
immediate corona, and also the nature of Io's sodium cloud and its interactions with the 
plasma torus. The interactions of Io’s extended atmosphere/neutral clouds and the plasma 
torus are coupled in interesting ways that have yet to be fully understood, but which are 
recognized as being fundamentally important in both the atmosphere and magnetospheric 
disciplines. Because atomic sodium is brighter in its emission lines than the much more 
abundant atomic oxygen and atomic sulfur by a factor of several orders of magnitudes 
and has therefore been observed almost exclusively in the past two decades, studies for 
sodium provide the primary avenue for probing the behavior of neutrals in the Io-Jupiter 
system. 


In this report, studies for sodium are divided into two different investigations 
which are discussed in Section II and Section III. In Section II, the major objectives are 
(1) the synthesis of a number of different and independent sodium observations covering 
a radial range from Io's nominal exobase of ~1.4 satellite radii to east-west distances from 
Io of -100 satellite radii and (2) the determination from this collective data set of a 
consistent flux speed distribution for sodium at the satellite exobase. This investigation is 
discussed in detail in a recently submitted paper entitled "Io’s Corona and Spatial 
Extended Cloud: A Consistent Flux Speed Distribution" which is included in Appendix 
A. In Section III, the major objective is the analysis and explanation of a System III 
longitude correlated "bite-out" in the south portion of the sodium cloud when Io is near 
western elongation. This "bite-out" was previously documented in the JPL Table 
Mountain Io Sodium Cloud Data Set (Goldberg et al. 1984). This investigation is 
discussed in detail in a preliminary version of a paper entitled "Correlating System III 
Asymmetries in the Jovian Magnetosphere and the Sodium Cloud” which is included in 
Appendix B. Because of the two papers in the Appendices, the text of the annual report 
will be limited to an overview summary of these two investigations. 


1 




II. IO'S SODIUM CORONA AND SPATIALLY EXTENDED CLOUD 


2.1 Observational Data for Sodium 

A data set, composed of different ground-based observations for Io’s sodium 
corona and spatially extended sodium cloud and covering the spatial range from Io's 
nominal exobase of 1.4 satellite radii to east-west distances from Io of ±100 satellite 
radii, is assembled and found to be internally consistent. The data set is composed of 
three parts: (1) the novel 1985 eclipse measurements of Schneider et al. (1991) acquired 
from —1.4 to —10 satellite radii from Io, (2) the 1985 east-west emission data of Schneider 
et al. (1991) acquired from ~4 to ~40 satellite radii from Io, and (3) sodium cloud image 
data acquired from ~ 10 to -100 satellite radii from Io by a number of different observers 
in the 1976 to 1983 time frame. 

The interleaved novel ground-based observations of Io's sodium corona and the 
near extended cloud were obtained in 1985 by Schneider (1988; Schneider et al. 1987, 
1991) when the Galilean satellites of Jupiter underwent mutual eclipse. These 
observations provide for the first time valuable information about the density gradient of 
atomic sodium within the Lagrange sphere of Io (i.e., a radius of 5.85 satellite radii, Rio) 
and also provide spatial information beyond this boundary that extends into the nearer 
portion (i.e., 6-40 Ri 0 from Io's center) of the sodium cloud. Two different types of 
observations were acquired: eclipse observations and emission observations. Eclipse 
observations measured the absorption feature at the D-line wavelengths as seen from 
Earth in the spectra of the reflected sunlight from the disk of another Galilean satellite 
(either Europa or Ganymede for measurements of interest here) that was produced by 
sodium in Io's atmosphere as Io eclipsed the sun from the viewpoint of the other Galilean 
satellite. Since the equivalent width of each absorption profile can be directly converted 
to a column abundance of sodium along the optical path, successive measurements during 
one solar eclipse of either Europa or Ganymede by Io produce a spatial profile of the 
column density near Io. The column density distributions determined from the different 
eclipse measurements are essentially symmetric about Io within the Lagrange sphere 
radius (5.85 Rio, i.e., the gravitational grasp of the satellite), providing a one-dimensional 
(time independent) radial profile as noted earlier by Schneider et al. (1991). Emission 
observations measured the solar resonance scattered D-line intensity emitted by sodium 


2 




atoms in the near cloud environment of Io. In additional to spectral information, which 
will be only briefly considered here, these observations provide a one-dimensional spatial 
profile of the D-line emission brightnesses along a slit that is oriented east-west (i.e., 
perpendicular to the spin axis of Jupiter) and that contains (or very nearly contains) Ios 
disk. The east-west sodium brightness profiles for the emission data are distinctly 
different in front of lo and behind Io and are time-dependent. Eclipse data yield the most 
accurate information very near Io (i.e., 1.4 to 6 Rj 0 from Io's center), while emission data 
yield the most accurate information at larger distances (i.e., 4 to 40 Rio) as discussed in 
detail by Schneider et al. (1991). 

In order to extend the spatial range coverage of the 1985 emission data, spatially- 
extended east-west D 2 emission profiles for sodium from ~10 to -100 satellite radii from 
Io were extracted from sodium cloud image observations. This was accomplished by 
using a number of the sodium cloud images in D 2 emission lines which were acquired in 
the 1975-1984 time interval (Murcray 1978; Murcray and Goody 1978; Matson et al. 
1978; Goldberg et al. 1980, 1984; Morgan 1984, private communication; Goldberg and 
Smyth 1993) where Io was centered behind an occulting mask typically 10 to 12 arc sec 
across (i.e., covering a radial distance from the center of Io of about 10 Ri 0 ) in order to 
block the bright disk-reflected sunlight from the satellite. Brightnesses in the immediate 
vicinity of the mask were usually spatially distorted in the resulting image but were 
consistently measured to be a few kiloRayleigh (kR), similar to the brightnesses of the 
1985 emission data at 10 Ri 0 - 

A comparison of the space-time behavior of the 1985 emission data and the 
sodium cloud image data was undertaken. They were found to be consistent. The east- 
west sodium brightness profiles as determined from the emission data are distinctly 
different in front of Io and behind Io and exhibit the time-dependent behaviors of the 
forward cloud and the trailing north-south oscillating directional features, which are 
correlated with the Io geocentric phase angle, the Io System III longitude, and the east- 
west asymmetry of the plasma torus. In this regard and of particular interest, an observed 
brightness enhancement of the trailing cloud portion of the east-west emission data 
profile, which is also correlated with an observed enhancement in the full width half 
maximum of the measured spectral line shape, is identified as occurring when the 
(higher-speed) trailing directional feature crossed the east-west position of the observing 
slit (i.e., the null position between north and south). 


3 




2.2 Analysis of the Sodium Observations 


The spatial sodium profiles acquired in the 1985 eclipse and emission data are 
analyzed together with the more spatially-extended east-west profiles extracted from 
earlier sodium cloud image observations in order to investigate the nature of the flux 
speed distribution of sodium atoms at Io exobase. The combined data set is analyzed 
using the sodium cloud model of Smyth and Combi (1988a,b). At the satellite exobase, 
the speed dispersion of either an isotropic Maxwell-Boltzmann flux speed distribution or 
an isotropic classical sputtering flux speed distribution (which has a broader dispersion) 
is shown to be inadequate in fitting the combined data set. In the absence of the trailing 
cloud enhancement at the null condition, an isotropic modified sputtering flux speed 
distribution provides an excellent fit to the combined data set for a sodium source of 1.7 x 
lO^G atoms sec -1 (see Figures 1 lb and 12b, in Appendix A). To fit also the enhanced 
trailing cloud profile at the null condition, an additional enhanced high-speed (-15-20 km 
sec -1 ) sodium population is required which is nonisotropically ejected from the satellite 
exobase so as to preferentially populate the trailing cloud rather than the forward cloud. 
Such a nonisotropic high-speed population of sodium is consistent with the earlier 
modeling analysis of the directional features by Pilcher et al. (1984). The two- 
dimensional distribution of the sodium about Io determined by the model for the isotropic 
modified sputtering flux speed distribution is presented (see Figure 15, in Appendix A). 
A complete sodium source rate speed distribution function at the satellite exobase, which 
also includes the even higher-speed (-20-100 km sec' 1 ) charge-exchange source (Smyth 
and Combi 1991) at Io that populates the sodium zenocorona (or magneto-nebula) far 
from Jupiter, is also presented (see Figure 16, in Appendix A). 


III. SYSTEM III ASYMMETRY IN IO'S SODIUM CLOUD 

3.1 Observational Data for Sodium 

The "bite-out" of the sodium cloud near western elongation was discovered in the 
1981 Region B/C image data reported by Goldberg et al. (1984), which are part of the 
larger JPL Table Mountain lo Sodium Cloud Data Set. The time evolution of this "bite- 
out" was excellently documented in a sequence of these images acquired on May 13, 
1981 which were included in a 16 mm movie showing the changing D 2 cloud image on 
the sky plane as Io moved about Jupiter (Goldberg et al. 1982; Goldberg 1983). The 
entire 1981 Region B/C data set consists of 263 images of the sodium cloud which were 


4 




recorded simultaneously in both the D 2 (5890A) and Dj (5896A) wavelength emission 
lines over 14 nights with an image integration time of approximately 10 minutes. Three 
consecutive images were usually added to improve signal to noise. These co-added 
images produced a time-averaged picture of the sodium cloud over an Io geocentric 
phase angle of about 4° and an lo System III magnetic longitude angle of about 14° and, 
therefore, approximate reasonably well instantaneous snapshots of the Io sodium cloud. 
The range of the geocentric phase angles and System III magnetic longitude angles of Io 
covered by the entire 1981 Region B/C data set is summarized in Appendix B graphically 
in Figure 1 and numerically in Table 2. Most of the 263 images comprising this coverage 
have undergone some type of partial data processing so that the time evolution of the 
cloud can be followed qualitatively. The dots and squares in Figure 1 show the midpoint 
conditions for 34 images that have undergone complete data reduction and calibration. 
Observational conditions and Image ID Numbers for these 34 images are summarized in 
Appendix B in Table 3. 

The time evolution of the "bite-out" in the south portion of the sodium cloud near 
western elongation is illustrated in Appendix B in Figure 2 for a sequence of four images 
on May 13, 1981 and in Figure 3 for a sequence of four images with nearly the same Io 
geocentric phase angle but different Io System III longitudes. In the top image of Figure 
2 for which Io has a System III magnetic longitude angles of about 168°, the "bite-out" 
was not present. For System III magnetic longitude angles of about 180°, the "bite-out" 
however, began to appear in the form of a deficiency of sodium south of Io and to the 
outside of its orbit that slices across the image at an angle of positive slope, as is 
illustrated in the second image of Figure 2. The "bite-out" became more pronounced as 
the System III longitudinal angle increases. The same general pattern is repeated in other 
1981 west cloud data. Examination of a number of other sodium cloud images obtained 
in the 1977 to 1984 time frame indicates that the "bite-out" is a long-term and stable 
feature of the sodium cloud near western elongation. 

Examination of sodium cloud images for Io east of Jupiter, when the satellite is in 
the same System III longitude angle range for which the "bite-out" occurs in the west 
cloud, however, does not reveal a corresponding bite-out" feature. In fact, little or no 
change is observed in the east cloud, as illustrated in Figure 3 of Appendix B. Because 
the excitation mechanism for sodium emission in the D-lines is solar resonance 
scattering, the deficiency of sodium brightness in the "bite-out" represents a deficiency in 
the sodium abundance. This suggests that the "bite-out" may be caused by an east-west 


5 




asymmetry in the plasma torus together with a longitudinally asymmetry in the "so- 
called" active sector (-200-300 degrees System 111 longitude) of the plasma torus (Hill, 
Dessler and Goertz 1983) in which the electron-impact ionization lifetime of sodium 
would then have to be enhanced. Such an enhancement would require an increase in 
some combination of the electron density and temperature in the active sector and also in 
the plasma torus west of Jupiter. Any increase in electron density would increase the 
emission brightness in the plasma torus of the S + ions (6716 A and 6731 A) and the S'*"*’ 
ions (9531 A), since their electron impact excitation mechanism is primarily sensitive to 
the electron density for the electron temperatures in the plasma torus near Io. The 
sodium cloud images in the JPL data set have indeed recently been shown by Goldberg 
and Smyth (1993) to be brighter east of Jupiter and hence exhibit an east-west asymmetry 
which is anticorrelated with the well-documented east-west asymmetry of the plasma 
torus emission brightness. The presence of the "bite-out feature then suggests that the 
brightness of the sodium emission lines and ion emission lines should also be naturally 
anticorrelated in System III longitude. 


3.2 Observational Data for The Io Plasma Torus 

A review of observations for ion emissions from the plasma torus indicates that a 
physically corresponding System III brightness enhancement in these plasma torus S + 
emission lines, which is anticorrelated with the "bite-out sodium brightness, has indeed 
been observed in the active sector from ground-based telescopes since 1976. A 
comparison of these plasma observations for the System III asymmetry indicates that 
generally two separate longitudinally asymmetric brightness components appear to exist 
in the torus. The first component is centered at -170 to 200 degrees, with an angular 
width of - 90 degrees or more, and is a permanent feature of the torus. When compared 
to the normal brightness of the plasma torus at the diametrically opposite magnetic 
longitude angle, the first component is typically 2-5 times brighter (and at extremes, 10 
times brighter) with significant brightness changes occurring on a time scale of months to 
years. The second component is centered at -280-290 degrees, has a comparable angular 
width to the first component, and is sometimes present and sometimes absence in the 
torus. The brightness of the second component may be comparable to or less than 
brightness of the first component. 


6 




For our immediate purpose of understanding the "bite-out” in the sodium image 
data, the presence of both a longitudinal and east-west brightness asymmetry in the 
plasma torus is fortunately reasonably well documented in the winter to spring of 1981 by 
the extension amount of ground-based observations acquired by Morgan (1983) and his 
analysis and modeling of these data (Morgan 1985 a,b). The System III longitudinal 
variation of the SII (6716A, 6731 A) emissions was time dependent in the measurements 
of Morgan (1983; 1985 a,b). It changed from an ordered single peak ~ 2.5 times brighter 
in a broad region centered near 180° System III longitude in run I (Feb 14-17) and run 2 
(Mar 20-23) to a double-peaked structure with peaks located near 180° and 290° in run 3 
(April 21 - 24) and run 4 (May 2-4). In Figure 4 of Appendix B, the longitudinal 
asymmetry for run 4 is shown and reveals that near 180° and 300° system III longitude 
the two intensity peaks west of Jupiter were brighter and vary by about a factor of 2 while 
the intensity peaks east of Jupiter were much dimmer and had a smaller amplitude 
variation. This double-peaked structure west of Jupiter was also independently verified 
in the April 1981 SII image data of Pilcher et al. (1985). 

3.3 Analysis of the Sodium Observations 

To study the time-variable "bite-out" signatures of the sodium cloud, it is necessary 
to have a model that contains, to a reasonable level of accuracy, a description of the atom 
dynamics, the atom sink processes, the atom excitation mechanism, and the atom source 
velocity distribution at the satellite exobase. An updated version of the sodium cloud 
model of Smyth and Combi (1988a,b) is adopted here for this purpose. The sodium cloud 
model of Smyth and Combi (1988b) includes the dominant electron impact ionization sink 
for sodium which is based upon a three-dimensional space-time dependent description of 
the plasma torus for a tilted and offset magnetic dipole field in the presence of an east-west 
electric field. The east-west electric field provides an east-west asymmetry in the plasma 
properties which is required to explain the anticorrelated east-west brightness asymmetry of 
the sodium and ion emissions. The sodium cloud model has been improved (see Appendix 
B) to include an inherently asymmetric plasma torus description based upon the 1981 
plasma torus data of Morgan (1985 a,b). The sodium source at Io’s exobase has recently 
been shown by Smyth and Combi (1993) to be well represented by a modified sputtering 
flux speed distribution function which has its maximum value at ~0.5 km sec' 1 (i.e., a sub- 
escape speed from lo's exobase). For atoms in the sodium cloud of which all have already 
escaped from Io, previous modeling efforts (Smyth and Combi 1988b) have shown, 
however, that the basic nature of the forward cloud and also the escaped sodium within 


7 




about one half Jupiter radii north and south of Io can be reasonably well reproduced by 
assuming a monoenergetic 2.6 km sec'* source at a nominal Io's exobase of 2600 km, 
where the escape speed is ~2.0 km sec 1 . This source is quite adequate for our present 
purposes of investigating the "bite-out” feature but will understandably (see Pilcher et al. 
1984) not reproduce the trailing-cloud directional feature due to its absences of higher- 
velocity components (i.e., -10-20 km sec" 1 ) or the brightness gradient in the more distance 
portions of the forward cloud (Smyth and Combi 1993). 

To analyze the "bite-out" in the sodium cloud image data, model calculations 
were performed for select sodium cloud images both east and west of Jupiter. For 
sodium cloud images west of Jupiter, the "bite-out" feature was not produced when an 
inherently symmetric description of the plasma torus was adopted. The "bite-out" feature 
was, however, produced with its proper System III longitude correlation when the 
inherently asymmetric description of the plasma torus based upon the 1981 plasma 
measurements of Morgan (1985 a,b) was adopted. For sodium cloud images east of 
Jupiter, the "bite-out" feature was not produced for either the inherently symmetric or 
inherently asymmetric description of the plasma torus. The "bite-out" feature occurs for 
images west of Jupiter because the sodium lifetime within the enhanced System III 
longitude region is sufficiently reduced so as to become smaller than or comparable to the 
transport time for sodium atoms to pass through this region. The "bite-out" feature does 
not occur for images east of Jupiter because the sodium lifetime within the enhanced 
System III longitude region is somewhat larger than the transport time for sodium atoms 
to pass through this region. East of Jupiter, the sodium lifetime near Io is larger because 
of the outward radial displacement (by the east-west electric field) of the sharply 
decreasing inward radial gradient of both the plasma torus properties and the radial 
structure of the System III longitudinal enhancement. 


8 




REFERENCES 


Goldberg, B. A., Y. Mekler, R. W. Carlson, T. V. Johnson, and D. L. Matson (1980). lo’s 
Sodium Emission Cloud and the Voyager 1 Encounter. Icarus 44, 305-317. 

Goldberg, B. A., R. W. Carlson, G. W. Gameau, T. V.Johnson, S. K. LaVoie, J. J. Lorre and 
D. L. Matson (1982). Dynamics of the Io Sodium Cloud. BAAS 14, 762. 

Goldberg, B. A. (1983). Dynamics of the Io Sodium Cloud, 16 mm movie produced for display 
at the National Air and Space Museum, Smithsonian Institute, Washington, D. C. 

Goldberg, B. A., Garneau, G. W. and LaVoie, S.K. (1984) Io's Sodium Cloud. Science, 226, 
512-516. 

Goldberg, B. A. and W. H. Smyth (1993). The JPL Table Mountain Io Sodium Cloud Data Set, 
paper in preparation. 

Hill, T. W., A. J. Dessler and C. K. Goertz (1983). Magnetospheric Models, In Physics of the 
Jovian Magnetosphere, (A. J. Dessler, Ed.), pp 106-156, Cambridge Press, N. Y. 

Matson, D. L. , B. A. Goldberg, T. V. Johnson, and R. W. Carlson (1978). Images of Io’s 
Sodium Cloud. Science 199, 531-533. 

Morgan, J. S. (1983). Low Resolution Spectroscopy of the Io Torus, Ph.D. thesis, Dept of 
Astronomy, Univ. of Hawaii. 

Morgan, J.S. (1985a). Temporal and Spatial Variations in the Io Torus, Icarus 62, 389- 
414. 

Morgan, J.S. (1985b). Models of the Io Torus, Icarus 63, 243-265. 

Murcray, F. J. (1978). Observations of Io’s sodium cloud. Ph.D. Thesis, Dept, of Physics, 
Harvard University, Cambridge, Massachusetts. 

Murcray, F. J., and R. Goody (1978). Pictures of the Sodium Cloud. Ap. J. 226, 327-335. 


9 




Pilcher, C. B., Smyth, W. H., Combi, M. R., and Fertel, J. H. (1984) Io's Sodium Directional 

Features: Evidence for a Magnetospheric-Wind-Driven Gas Escape Mechanism. A JjL JL - 
287, 427-444. 


Pilcher, C. B., J. H. Fertel and J. S. Morgan (1985). [SII] Images of the Torus. Ap. J. 291, 377- 
393. 

Schneider, N. M. (1988) Sodium in Io's extended atmosphere. Ph.D. Thesis, Department of 
Planetary Science, The University of Arizona. 

Schneider, N. M., Hunten, D. M., Wells, W. K., and Trafton, L. M. (1987) 

Eclipse measurements of Io's sodium atmosphere. Science, 238, 55-58. 

Schneider, N.M., Hunten, D.M., Wells, W.K., Schultz, A.B., and Fink, U. (1991) The 
Structure of Io’s Corona, Ap J., 368, 298-315. 

Smyth, W. H., and Combi, M. R. (1988a) A general model for Io’s neutral gas cloud. 

I. Mathematical description. Ap. J. Supp., 66, 397-41 1. 

Smyth, W.H., and Combi, M.R. (1988b) A General Model for Io’s Neutral Gas Cloud. 

II. Application to the Sodium Cloud. Ap. J. 328, 888-918. 

Smyth, W.H., and Combi, M.R. (1991) The Sodium Zenocorona, J. Geophys. Res. 96, 22711- 
22727. 

Smyth, W.H. and M. R. Combi (1993). Io's Sodium Corona and Spatially Extended Cloud: a 
Consistent Flux Speed Distribution. Icarus, submitted. 


10 




Appendix A 


Io’s Sodium Corona and Spatially Extended Cloud 
A Consistent Flux Speed Distribution 




Io's Sodium Corona and Spatially Extended Cloud : 


A Consistent Flux Speed Distribution 


William H. Smyth^ 


Michael R. Combi^ 


Submitted to Icarus 
October 29, 1993 


1. Atmospheric and Environmental Research, Inc., Cambridge, MA 02139 


2. Space Physics Research Laboratory, University of Michigan, Ann Arbor, Michigan 48109 


Number of manuscript pages : 46 
Number of Figures : 16 
Number of Tables : 5 


Key words : lo, satellite exospheres, gas tori 



Running heading : lo's Sodium Corona and Spatially Extended Cloud 


For editorial correspondence and proofs please contact: 
William H. Smyth 

Atmospheric and Environmental Research, Inc. 
840 Memorial Drive 
Cambridge, MA 02139 

Telephone number : (617) 547-6207 
Fax number : (617) 661-6479 
Internet : smyth@aer.com 


2 


ABSTRACT 


A data set, composed of different ground-based observations (or lo s sodium 
corona and spatially extended sodium cloud and covering the spatial range from los 
nominal exobase of 1.4 satellite radii to east-west distances from lo of ±100 satellite radii, 
is analyzed collectively to investigate the velocity distribution of sodium at the exobase. 
The data set is composed of the novel 1985 eclipse measurements of Schneider el al. 
(1991) acquired from 1.4 to -10 satellite radii from !o, the 1985 east-west emission data of 
Schneider et al. ( 1991) acquired from ~4 to ~40 satellite radii from lo, and sodium cloud 
image data acquired from ~10 to -100 satellite radii from lo by a number of different 
observers in the 1976 to 1983 time frame. The column density distributions determined 
from the different eclipse measurements are essentially symmetric about lo within the 
Lagrange sphere radius (5.85 satellite radii, i.e., the gravitational grasp of the satellite), 
providing an one-dimensional radial profile as noted earlier by Schneider cr al. ( 199 1 ). At 
distances beyond the Lagrange sphere radius, however, the apparent change in the 
observed column density slope is shown in model calculations to occur in the forward 
cloud, but not trailing cloud portion, of the column density profile. 1 he east-west sodium 


brightness profiles as determined from the emission data are distinctly dillerent in front of 
lo and behind lo and are shown to exhibit the time-dependent behaviors of the forward 
cloud and the trailing north-south oscillating directional features, which am correlated with 
the lo geocentric phase angle, the lo System III longitude, and the east-west asymmetry of 
the plasma torus. In this regard and of particular interest, an observed brightness 
enhancement of the trailing cloud portion of the east-west emission data profile, which is 
also correlated with an observed enhancement in the full width half maximum of the 
measured spectral line shape, is identified as occurring when the (higher-speed) trading 
directional feature crossed the east west position of the observing sht (i.c . the null position 



between north and south). The combined data set is analyzed using the sodium cloud 
model of Smyth and Combi (1988a,b). At the satellite exobase, the speed dispersion of 
either an isotropic Maxwell-Boltzmann flux speed distribution or an isotropic classical 
sputtering flux speed distribution (which has a broader dispersion) is shown to be 
inadequate in fitting the combined data set. In the absence of the trailing cloud 
enhancement at the null condition, an isotropic modified sputtering flux speed distribution 
provides an excellent fit to the combined data set for a sodium source of 1.7 x 10 26 atoms 
sec*. To fit also the enhanced trailing cloud profile at the null condition, an additional 
enhanced high-speed (-15-20 km sec' 1 ) sodium population is required which is 
nonisotropically ejected from the satellite exobase so as to preferentially populate the 
trailing cloud rather than the forward cloud. Such a nonisotropic high-speed population of 
sodium is consistent with the earlier modeling analysis of the directional features by Pilcher 
et al. (1984). A complete sodium source rate speed distribution function at the satellite 
exobase, which also includes the even higher-speed (-20-100 km sec *) charge-exchange 
source (Smyth and Combi 1991) at Io that populates the sodium zenocorona (or magneto- 
nebula) far from Jupiter, is presented. 


4 


I. INTRODUCTION 


Novel ground-based observations of lo's sodium corona and near extended cloud 
were obtained in 1985 by Schneider (1988; Schneider et al. 1987, 1991) when the Galilean 
satellites of Jupiter underwent mutual eclipse. These observations provide for the first time 
valuable information about the density gradient of atomic sodium within the Lagrange 
sphere of Io ( i.e., a radius of 5.85 satellite radii, Rio ) and also provide spatial information 
beyond this boundary that extends into the nearer portion (i.e., 6-40 Ri 0 from lo's center) 
of the sodium cloud. Two different types of observations were acquired: eclipse 
observations and emission observations. Eclipse observations measured the absorption 
feature at the D-line wavelengths as seen from Earth in the spectra of the reflected sunlight 
from the disk of another Galilean satellite (either Europa or Ganymede for measurements of 
interest here) that was produced by sodium in lo's atmosphere as Io eclipsed the sun from 
the viewpoint of the other Galilean satellite. Since the equivalent width of each absorption 
profile can be directly converted to a column abundance of sodium along the optical path, 
successive measurements during one solar eclipse of either Europa or Ganymede by Io 
produce a spatial profile of the column density near Io. Emission observations measured 
the solar resonance scattered D-line intensity emitted by sodium atoms in the near cloud 
environment of Io. In additional to spectral information, which will be only briefly 
considered here, these observations provide a one-dimensional spatial profile of the D-line 
emission brightnesses along a slit that is oriented east-west (i.e., perpendicular to the spin 
axis of Jupiter) and that contains (or very nearly contains) lo's disk. Eclipse data yield the 
most accurate information very near Io (i.e., 1.4 to 6 Ri 0 from lo's center), while emission 


5 



data yield the most accurate information at larger distances (i.e., 4 to 40 R[ 0 ) as discussed 
in detail by Schneider et al. (1991). 

In this paper, the spatial sodium profiles acquired in the 1985 eclipse and emission 
data are analyzed together with more spatially-extended east-west profiles extracted from 
earlier sodium cloud image observations in order to investigate the nature of the flux speed 
distribution of sodium atoms at Io exobase. The observational information is presented, 
compared, and evaluated in section 2. Modeling of the spatial profiles of the sodium data is 
undertaken in section 3. Discussion and conclusions are presented in section 4. 

2. OBSERVATIONS 

Of particular interest in this paper are the five higher quality eclipse measurements 
and the nine higher quality emission observations obtained at the Catalina Observatory 1 .5 
meter telescope using the LPL echelle spectrograph that were presented by Schneider et 
a/.(1991). The dates, times, orbital angular parameters of Io, spectral ID numbers, and the 
numbering of these observations adopted in this paper are summarized in Table 1. Four of 
the five eclipse observations were acquired when Io was east of Jupiter (i.e., an Io 
geocentric phase angle within 90 ± 90 degrees), and only one was acquired when Io was 
west of Jupiter (i.e. , within 270 ± 90 degrees). Seven of the emission observations were 
obtained when Io was east of Jupiter, and only two were obtained when lo was west of 
Jupiter. In addition, select sodium cloud image observations for Io near its orbital 
elongation points, which were acquired in 1976 by Murcray (1978; Murcray and Goody 
1978), in 1981 by Goldberg et al. (1984; Goldberg and Smyth 1993), and in 1983 by 
Morgan (1984, private communication), are used to extract east-west D 2 brightness 
profiles that overlap the spatial range of the emission data and extend it to ±100 Rio- 


6 


2.1 Eclipse Observations 


The column density profiles for all five of the eclipse observations are presented 
collectively in Figure 1 and follow directly from the information given by Schneider et al. 
(1991) in their Table 3. Only one lower-bound data point from eclipse 4 at a distance from 
the center of Io of 1. 17 Ri 0 is excluded since it is well within the nominally exobase radius 
of 1.4. Rio The spatial profile for eclipse 2 obtained on September when Io was very near 
eastern elongation (i.e., 90 degrees Io phase angle) is highlighted in Figure 1 and seen to 
be similar to the other four spatial profiles. A comparison of the two sides of each eclipse 
measurement shows no detectable difference in the slopes of their separate profiles within 
the Lagrange sphere radius of Io. This was noted earlier by Schneider et al. (1991) and 
thus effectively reduces the eclipse data to a radially symmetric profile. The dashed line 
shows the power law fit to the data points contained within the Lagrange radius and has the 

form 


N (1.4 <r <5.85) = 2.55xl0' 2 r 248 

where N is the column density in units of atoms cm'^ and r is the distance from the center 
of Io in units of the Rio (i.e., 1815 km). This is the same power law obtained earlier by 
Schneider et al. and undercuts the data points outside the Lagrange radius where the eclipse 
observations are deemed to be less reliable. Although the eclipse data beyond the Lagrange 
sphere contain more vertical scatter, they also appear to have possibly a reduced slope 
Such a change in slope might be caused by the domination of the planetary gravitational 
field beyond the Lagrange radius of Io and will be discussed further in the modeling 
analysis section. 


7 



2.2 Emission Observations 


The D 2 brightness profiles for the nine emission observations are shown 
collectively in Figure 2. This information, which was previously published only in a 
graphical format (Schneider et al. 1991), is summarized numerically in Table 2 as provided 
by Schneider (1990, private communication). The spatial profile for Emission 4, which 
was obtained on September 14 when lo was nearest eastern elongation, is highlighted, and 
its east and west profiles are identified separately. Excluding all data points inside of 4 Ri 0 
where the seeing and instrumental effects artificially flatten the profile, the power law fit to 
the remaining emission brightness data in Figure 2 is given by 

I D (r > 4) =101 r' 145 

where I n is the D 2 brightness in kiloRayleighs (kR) and r is the distance from the center 
of Io in Rj 0 units. This D 2 brightness of ~ 100 kR as r approaches Io's surface is 
consistent (Brown and Yung 1976) with the maximum sodium column density of ~1 x 
1 0 12 atoms cm' 2 deduced from the eclipse data in Figure 1. 

For more insight into the emission data, it is instructive to make a direct comparison 
of the relative behavior of the nine emission observations. Since the spatial coverage of the 
D 2 brightness is very nonuniform, this comparison can be readily accomplished by simply 
inspecting the columns in Table 2. The nonuniform spatial coverage in Table 2 occurs 
because of different distance intervals adopted to obtain an average brightness value (given 
different signal to noise ratios) for each data point in the profile and because of signal drop- 
out associated with constraints imposed on positioning the slit profile on the CCD detector 


8 


during interleaved eclipse and emission measurements. Using the emission data for Io east 
of Jupiter (i.e., seven out of the nine emission observations) and for Io west of Jupiter 
(i.e., the remaining two emission observations), the relative brightness of the forward and 
trailing sodium cloud near Io may be monitored as a function of the Io geocentric phase 
angle by comparing in each observation the two brightness profiles east and west of Io. In 
this comparison, it will be seen that systematic changes in the D 2 brightness are generally 
consistent with the Io phase angle dependence observed in Io sodium cloud image data over 
the last two decades. 

When Io is east of Jupiter and has a phase angles less than about 65 degrees, a 
brighter forward sodium cloud is observed to the east of Io and a dimmer trailing cloud is 
observed to the west of Io (Goldberg et al. 1984). This is the pattern exhibited by the 
emission 1 observation on August 27 which has an average Io phase angle of 61.4 degrees. 

When Io is east of Jupiter and has phase angle between about 65 and 85 to 90 
degrees, east and west profiles are observed (Goldberg etal. 1984) to be very- similar since 
the brighter forward sodium cloud is now swinging through and approximately aligned 
along the observer's line of sight. This is the pattern exhibited by the emission 2 
observation on August 27 which has an average Io phase angle of 72.2 degrees. For the 
emission 4 observation acquired 18 days later on September 14 for an average Io phase 
angle of 87.7 degrees, the reverse pattern, however, is present with the trailing cloud 
being brighter than the forward cloud. Inspection of the line profile shapes for each data 
point of the emission 4 observation on September 14 indicates that its full width half 
maximum (FWHM) is highly asymmetric east and west of Io with a much larger FWHM 
present at larger distances from Io in the east profile (i.e., in the trailing cloud). The 
emission 3 observation acquired a day earlier for Io west of Jupiter (and essentially a mirror 


9 



geometric observation to that on September 14) and the emission 6 observation acquired for 
Io west of Jupiter a day later on September 15, also both exhibit a similar highly 
asymmetric values for the FWHM which are larger in the trailing cloud. This reversal of 
the normal pattern by the emission 3, the emission 4, and the emission 6 observations 
therefore all occur when additional high speed sodium was present in the trailing portion of 
the cloud outside of Io’s orbit The Doppler signature of this additional high speed sodium 
was, however, not seen in the trailing portion of the cloud in the emission 5 observation 
which was also acquired on September 14 only about one and one-half hours after the 
emission 4 observation (i.e., an increase in the lo phase angle and hence the cloud’s sky- 
plane projection angle of only about 13 degrees). 

When Io is east of Jupiter and has a phase angle greater than about 85 to 90 
degrees, the forward sodium cloud is observed in image data to be brighter to the west of 
Io and the trailing cloud to be dimmer and to the east of Io (Goldberg et al. 1984). This is 
the pattern exhibited by the emission 7, emission 8, and emission 9 observations which 
have average Io phase angle of about 117, 122, and 143 degrees, respectively. A 
comparison of the forward and trailing cloud profiles for the emission 5 observation, which 
has an average Io phase angle of about 101 degrees, is not possible because there is only 

one data point west of lo. 

When Io is west of Jupiter and has a phase angle greater than about 235 degrees, 
the forward sodium cloud is generally observed to be brighter and to the east of Io. and the 
trailing cloud is generally observed to be dimmer and to the west of Io (Goldberg a al. 
1984). Examination of the profiles for the two out of the nine emission observations for Io 
west of Jupiter (i.e., emission 3 and emission 6 with lo phase angles of about 277 and 295 
degrees, respectively), shows, however, the reverse pattern. As noted above, these two 


observations, together with the eastern emission 4 observation, all exhibit this reverse 
pattern and also have a highly asymmetric spatial profiles east and west of Io for their 
FWHM values, with much larger FWHM values present at larger distances from Io in the 
trailing cloud . These asymmetric spatial profiles of the FWHM suggest additional high 
speed sodium in the cloud outside of Io's orbit may be responsible for this reverse pattern. 

A final comparison in Table 2 can be made for the nearly mirror geometric 
measurements for the emission 7 observation with an Io phase angle of about 1 17 degrees 
and the emission 6 observation with an Io phase angle of about 295 degrees. From Table 
2, the brightness of the forward cloud at a distance of 10-16 Rj 0 can be seen to be larger for 
Io east of Jupiter. This is consistent with the well known east-west intensity asymmetry 
first discovered near Io by Bergstralh et a/.(1975, 1977) and identified at larger distances 
from Io by Goldberg and Smyth (1993). A similar comparison between the approximately 
mirror geometric measurements for the emission 4 observation with an Io phase angle of 
about 88 degrees and the emission 3 observation with an Io phase angle of about 277 
degrees has not been considered since the absolute calibration of the emission 3 observation 
is in question (Schneider 1990, private communication). 

A closer graphical examination of each of the nine profiles in Figure 2 shows that 
the vertical spread of the whole data set, particularly for r >10, is a result of the 
superposition of the separate emission observations, where for each observation the slit 
profiles east and west of Io have two well defined but generally different slopes and 
different absolute brightnesses. This can readily be shown by a power-law fit analysis 
( I D (r > 4) = A | of the separate east and west profiles for each emission observation 

for Io east of Jupiter, where the exponent, p, and amplitude, A , are summarized in Table 
3. These fits omit data points inside four satellite radii of Io's center, where they are 



artificially flattened, and thus depict the decay in the brightness profile primarily outside of 
the Lagrange radius. In Table 3, the emission observations are arranged in increasing lo 
phase angle, and the spatially-projected location of the forward cloud (F), the symmetric 
turning point (S) of the cloud, and the trailing cloud (T) are identified for the east and west 
profiles. The power law slopes of all the forward clouds are similar and have an exponent 
value of ~ 1 6. The power law slopes of the trailing clouds after the symmetric turning 
point, excluding the September 14 emission observation at a phase angle of 100.6 degrees, 
are also similar and have an exponent value of ~ 2.0. Tire power law slopes of the clouds 
at the symmetric turning point, excluding the September 14 emission observation of the 
trailing cloud at a phase angle of 87.7 degrees, are similar and have an exponent value of ~ 
1.8. The two power law slopes of the trailing clouds on September 14 do not follow the 
pattern and have smaller exponent values of ~ 1.23 and 1.27. Not including the two 
emission observations on August 27 for lo east of Jupiter, which are at or before the 
symmetric turning point and have the reverse east-west projection orientation for their 
forward and trailing clouds, the remaining five profiles for lo east of Jupiter are presented 
graphically in Figure 3 along with their power law fits. 

The brightness profiles for the forward cloud (i.e. west of lo) in Figure 3 are very 
tightly confined. The brightness profiles for the trailing cloud (i.e., east of lo) in Figure 3, 
however, are significantly different and drop off more rapidly with increasing lo phase 
angle. The brightest (i.e., least steep) of these profiles is for the emission 4 observation 
(87.7 degrees lo phase angle) that exhibited a much larger FWHM value in the trailing 
cloud for the data points at these larger distances from lo. The next brightest profile is for 
the emission 5 observation (100.6 degrees lo phase angle) acquired -1.5 hours later on the 
same day which does not show the asymmetry in the FWHM values. The remaining three 
rather tightly confined trailing cloud profiles in Figure 3 are for the emission 7. emission 8, 


and emission 9 observations that have a very similar slope of P~2 as noted earlier. This 
behavior of the trailing clouds suggests that both the presents of higher speed sodium 
atoms as well as rapidly changing projection effects of the cloud on the sky plane as the lo 
phase angle increases are likely involved in its explanation. This behavior will be examined 
further and explained in the next section. In contrast, the well confined behavior of the 
forward sodium cloud and the absence of high speed sodium atom signatures suggest that 
the forward cloud is, as expected, primarily associated with the escape of low speed 
sodium from Io. A comparison of the brightness profiles in Figure 3 with other 
observations of the sodium cloud is undertaken below. 

2.3 Sodium Cloud Image Observations 

A large number of sodium cloud images in the Dj and D 2 emission lines were 
acquired in the 1975-1984 time interval (Murcray 1978; Murcray and Goody 1978; Matson 
etal. 1978; Goldberg et al. 1980, 1984; Morgan 1984, private communication; Goldberg 
and Smyth 1993) where Io was centered behind an occulting mask typically 10 to 12 arc 
sec across (i.e., covering a radial distance from the center of Io of about 10 Ri 0 ) in order to 
block the bright disk-reflected sunlight from the satellite. Brightnesses in the immediate 
vicinity of the mask were usually spatially distorted in the resulting image but were 
consistently measured to be a few kiloRayleigh (kR). These brightness are similar to those 
of the emission data at 10 R[ 0 in Figure 2. 

A simple review of these sodium cloud images shows that the forward sodium 
cloud is generally more spatially extended and brighter than the trailing cloud. An 
exception to this occurs, however, when the high-speed trailing directional-feature, which 
oscillates north and south behind the satellite with its inclination correlated (Pilcher et al. 



1984) with the System III longitude angle of lo, passes through the east-west line drawn 
through lo (i.e., the null location of the directional feature). The observed correlations of 
the north inclination, the south inclination, and the null locations of the directional feature 
with the System III longitude of lo are shown in Figure 4 as determined by Goldberg and 
Smyth (1993) and are consistent with those obtained earlier by Pilcher et al. (1984). The 
directional feature changes from a south to north inclination for a null System III longitude 
angle of lo near 165 degrees and change from north to south for a rather poorly defined 
null System III longitude angle of lo somewhere between about 320 and 25 degrees. This 
changing north-south orientation of the trailing directional feature is illustrated in Figure 5 
by three sodium cloud images, where an east-west oriented scale centered on lo of length 
+100 Rio is shown for reference. In Figure 5, a north directional feature is present in 
image A (247 degrees lo System III longitude), a south directional feature is present in 
image B (104 degrees lo System III longitude), and a slightly north but near-null 
directional feature is present in image C (178 degrees lo System III longitude). The spatial 
extension and brightening of the trailing cloud profile along the east-west oriented (dashed) 
line, when the directional feature is near the null location, are readily apparent. 

The forward cloud has been observed in images to distances of 100 satellite radii 
and more ahead of lo, where the cloud brightness levels are then only a few hundred 
Rayleighs. Analysis of the D 2 images of Murcray (1978) by Smyth and McElroy (1978 
see their Fig. 4) indicated that when lo was near eastern elongation the one kR brightness 
level in the forward cloud occurred about 60 R| 0 ahead of lo. Examination of a number of 
additional images indicates that the observed brightnesses of the forward cloud at this 
distance appear many times to be lower, although generally it has been difficult to be 
precise because the cloud is usually not measured to brightnesses less than about 0.2 to 0.5 
kR. Using fourteen images of the sodium cloud for lo near its orbital elongation points that 



are summarized in Table 4, a range (or an bounding envelope) of values for the D 2 east- 
west brightness profile of the forward and trailing clouds has been determined and is 
shown in Figure 3 by different shaded area. 

For the forward cloud profile in Figure 3, the shaded area determined from the 
sodium cloud images occurs for radial distances greater than about 20 R| 0 and can be seen 
to be brighter and to have a less steep slope than the emission data profiles. The lower 
boundary of the forward sodium cloud image profile area intersects the emission data just 
inside of 10 Rio, which is near the Lagrange radius of lo and is where the slope of the 
eclipse data in Figure 1 appears to become less steep. For Io near the elongation point, 
only two emission profiles occur and are too short to overlap the sodium cloud image data, 
with the one for an Io phase angle of 87.7 degrees extending only to a radial distance of 
-10 Rio and the other for an Io phase angle of 100.6 degrees containing only one point at a 
radial distance of -15 Ri 0 - For Io somewhat beyond eastern elongation, the remaining 
three emission data profiles (i.e., for Io phase angles of 1 17.2, 121.6, and 143.1 degrees) 
extend radially to -30 R[ 0 and fall below the sodium cloud image profile area, which is 
more representative of conditions near elongation. For Io near elongation, the question of 
consistency of the forward cloud brightness profiles from the emission data and the sodium 
cloud image data will be addressed in the modeling analysis section. 

For the trailing cloud profile in Figure 3, two different shaded areas have been 
extracted from the sodium cloud image information in Table 4 to quantify us D 2 east-west 
brightness profile . The extracted profile areas represent the two basic orientations for the 
directional feature : (1) when the directional feature is either north or south (lower area) and 
(2) when the directional feature is at the null or near null location (upper area). These two 
areas establish a range (or an bounding envelope) for the values of the D 2 east-west 


1 5 



brightness profile of the trailing sodium cloud for lo near its orbital elongation point. As 
expected, the shaded area for the directional feature at the null location is both brighter and 
less steep than the shaded area for the directional feature with either a north or south 
inclination. As noted in Figure 5, the D 2 east-west brightness profile of the trailing sodium 
cloud is more closely confined to lo than the forward cloud when the trailing directional 
feature is inclined either north or south and has a less steep brightness gradient when the 
directional feature is at the null location. 

The lo System III longitude correlated inclination of the directional feature is 
thought to be produced by high speed (-15-20 km sec' 1 ) sodium atoms escaping from lo in 
a nonisotropic manner. From the analysis of Pilcher etal. (1984), the nonisotropic escape 
flux is diminished at the poles relative to the equator and is also diminished in the forward 
and backward directions of the satellite's motion in comparison to the perpendicular 
direction of motion to lo’s orbit in the satellite orbit plane . For these high speed sodium 
atoms, the lo System III longitude correlation is produced because atom trajectories, upon 
reaching the spatial region behind lo, are synchronized in such a way with the time- 
dependent sodium electron impact ionization sink of the oscillating plasma torus, so as to 
either go under it, go through it, or go over it, thereby producing an alternating north-south 
reduction in the sodium density. 


In comparing the trailing emission profiles in Figure 3 with the trailing sodium 
cloud image data, it is important to identify the orientation of the directional feature in the 
emission data as classified by Figure 4 and also to note if there were enhancements in the 
measured Doppler signature of the trailing D 2 east-west brightness profiles. This 

information is summarized in Table 5. For the three emission profiles where an enhanced 
Doppler spectral profile (i.e., FWHM) was measured, the directional feature was at a null 


or near null localion, and also the trailing D 2 east-west brightness spatial profile was cither 
similar or dominant (i.e., brighter and less steep) when compared to the forward cloud 
spatial profile. Two of these emission profiles occurred when lo was west ofjup.ter and 
an: not included in Figure 3. The third profile occurred for lo at a phase angle of 87.7 
degrees. This is the brightest and least steep of all the profiles in Figure 3 and has a power 
law fit that is along the top boundary of the upper area for the sodium cloud image data 
which corresponds to the null condition. The presence of this brightest trailing profile is 
thus explained as a case where the directional feature was at a null condition. In Figure 3. 
the trailing emission profile for an lo phase angle of 100.6 degrees, acquired only about 
1.5 hours later on the same night and having a south inclined directional feature, has a 
power law fit that is essentially along the top boundary of the lower area for sodium cloud 
image data which corresponds to a north or south inclined directional features. The 
remaining three emission profiles (i.e., for lo phase angles of 117.2, 121.6, and 143.1 
degrees) are for lo somewhat beyond eastern elongation and lie near or just below the 
lower boundary of the lower area for sodium cloud images. The trailing emission profile 
data are therefore quite consistent with the sodium trailing cloud image data 

3. ANALYSIS of the observations 

Modeling analysis of the one-dimensional sodium distribution described in the 
previous section by the eclipse data, the emission data, and the sodium cloud image data 
will now be undertaken. One-dimensional profiles are calculated using the numerical 
sod, urn cloud model of Smyth and Comb, (1988 a.b). when: the electron tmpac, ion, rat. on 
sink for sodium is determined for a 7 degree tilled corotating plasma torus with an offset- 
dipole planetary magnetic field in the presence of a nominal (i.e.. -2.8 mV tn> in lo's 
frame) east-west electric field. Preltminary modeling reported earlier by Smyth and Comb, 



(1987) for the first radial profile published by Schneider et al. (1987) for the sodium 
column density determined by two eclipse measurements showed that it could be fit very 
well from ~4 to 10 Ri 0 by adopting a escaping, monoenergetic (2.6 km sec' 1 ), and 
isotropic sodium flux from the satellite exobase (assumed 1.4 Ri 0 , where the escape speed 
is ~2 km sec -1 ), which was determined independently from modeling typical forward cloud 
image data in the D 2 emission brightness range of —1-3 kR. A similar conclusion is 
reached from a comparison in Figure 6 of this monoenergetic model run with the complete 
set of eclipse data from Figure 1. In the radial interval from -2 to 15 Rj 0 , the effect of the 
sodium lifetime on the calculated profile for this monoenergetic source is illustrated in 
Figure 7, where it is seen to alter primarily the absolute magnitude rather than the slope of 
the column density profile. To fit in addition the eclipse data in Figure 6 between the 
nominal exobase (1.4 R to ) and -4 Ri 0 , a properly chosen dispersion of initial velocities for 
the source flux distribution with non-escaping components is required. For the larger 
spatial distances beyond the Lagrange radius, covered by the east-west profiles for the 
emission data and the sodium cloud image data, the dispersion of initial velocities in the 
source for speed components greater than 2.6 km sec 1 may also need to be included. 
Indeed, the eclipse measurements for the corona near lo, the emission measurements that 
extend into the near sodium cloud, and the sodium cloud image derived profiles that reach 
to distances of ±100 Ri 0 , provide a set of spatially overlapping observations that will be 
used to study and constrain the initial velocity dispersion of the sodium source atoms at the 
exobase. 

To investigate the nature of the initial velocity dispersion of the sodium source, two 
different source flux speed distributions discussed by Smyth and Combi (1988b; see their 
Appendix D) are considered: (1) a Maxwcll-Boltzmann flux distribution and (2) a modified- 



sputtering flux distribution. The Maxwell-Boltzmann flux distribution 0(v;l) is based on 
the Maxwell-Boltzmann velocity distribution and is defined as follows: 


tfv.THAtf 4 -) 2 — (— 

Re v t v t 


where 


V T = ( 


2kT , 

m 


n 


is the most probable speed of the velocity distribution for an atom of mass m. The 
Maxwell-Boltzmann flux distribution is proportional to the local velocity integrated flux <p 0 
referenced here to the satellite radius R s not the exobase radius R E and depends upon one 
parameter, the exobase temperature T (or alternatively v T ), which determines both the 
most probable speed v m 


and the speed dispersion of the flux distribution. The modified-sputtering flux distribution 
0(v;a,v M ,v b ) is proportional to the local velocity integrated flux <t> 0 and depends upon 

three parameters: an exponent a, a velocity parameter v b , and the velocity parameter v M . 


R< 


<^(v;a,v M ,v b ) = 0 o (— M 


1 


R (; v b D(a. v M /v b ) v 


(— ) 3 ( 


v 2 + v 2 


)° 


l-( 




I 9 


where D(a,v M /v b ) is a normalization constant (see Smyth and Combi 1988b). The 
exponent a primarily determines the dispersion of the distribution, which has a greater 
high-speed population as a decreases. The exponent a has a value of 3 for a classical 
sputtering distribution (i.e., representing a complete collisional cascade process) and a 
value of 7/3 for a Thomas-Fermi modified-sputtering flux distribution (i.e., representing 
the limit of a single elastic collisional ejection process), where the latter distribution is based 
upon a Thomas-Fermi differential scattering cross section. The velocity parameter v b is 
related nonlinearly to the most probable speed v m of the flux speed distribution and 
primarily determines v m (see Smyth and Combi 1988b, Appendix D). The velocity 
parameter v M primarily determines the maximum speed for the flux distribution and 
depends upon the maximum relative speed (and masses) of the plasma torus ion and 
sodium atom. The Maxwell-Boltzmann flux distribution and the modified-sputtering flux 
distribution, normalized to unit area, are each shown in Figure 8 for two different values of 
their parameters. The four flux distributions in Figure 8 will be utilized in the subsequent 

modeling analysis. 

In calculating the column density and the D 2 emission brightness in the numerical 
sodium cloud model, a smaller two-dimensional sky-plane grid centered on Io (±15 R| 0 ) is 
used to cover a spatial scale near the satellite more appropriate to the eclipse data while a 
much larger two-dimensional sky-plane grid centered on lo is used to cover a larger spatial 
scale more appropriate for the emission data and the sodium cloud image data. A one- 
dimensional profile for the eclipse data is obtained from the smaller two-dimensional sky- 
plane grid by extracting an average radial profile. This average radial profile will be called 
the calculated eclipse profile and will be denoted by the filled circles in Figures 9-14. A 
one-dimensional east-west D 2 brightness profile (and also a corresponding column density 


20 



profile) for the emission data and the sodium cloud image data is obtained from the larger 
two-dimensional sky-plane grid by selecting only the east-west grid elements that occur in 
the grid row containing lo. In Figures 9-14, the calculated east- west brightness and 
column density profiles are denoted by filled triangles for the forward cloud profile and by 
filled squares for the trailing cloud profile. To construct an eclipse or east-west profile, 
monoenergetic model calculations are performed for 18 different nonuniformly-spaced 
speeds ranging from 0.4 km sec'* to 10 km sec*. Profiles for speeds beyond 10 km sec* 
are determined by a inverse speed extrapolation of the model results. The individual 
profiles for the different speeds are appropriately weighted for a given source flux speed 
distribution and then added to obtain the final spatial profile. Model calculations are 
performed for an Io geocentric phase angle of 92.9 degrees and an Io System III longitude 
angle of 48.6 degrees. These satellite conditions are similar to those for the emission 4 and 
eclipse 2 observations of Table 1, which are the observation closest to the eastern 
elongation point. This choice is also appropriate for all the eclipse data within the Lagrange 
sphere, which has no discernible dependence on these two Io related angles, and for the Io 
sodium cloud image data which have east-west profile areas in Figure 3 that are 
representative of the satellite near its orbital elongation points. 

Model calculations for a Maxwell-Boltzmann flux distribution are presented in 
Figure 9 and Figure 10. In Figure 9a and Figure 9b, the calculated and measured sodium 
column density for the eclipse observations are compared for two different exobase 
temperatures. In Figure 10a and Figure 10b, the calculated D 2 emission brightness profiles 
for the forward and trailing clouds arc compared to the emission 4 measurement profiles 
and to the sodium cloud image data profile areas for the same two exobase temperatures 
adopted, respectively, in Figures 9a and 9b. 



For a Maxwell -Boltzmann flux distribution with a most probable speed of v m - 
1.3 km sec 1 (i.e., an exobase temperature of -1600 K, see Figure 8) and a total flux <p 0 
(referenced to Io's surface area) of 3.0 x 10 8 atoms cm 2 sec 1 (i.e., a total source of -1.2 
x 10 26 atoms sec 1 ), the model calculated column density and D 2 emission brightness 
profiles are shown in Figure 9a and Figure 10a, respectively. In Figure 9a for radial 
distances from the exobase to the Lagrange sphere, the model column density for the 
calculated eclipse profile (filled circles) provides an excellent fit to the eclipse observations 
(open circles) and also compares very favorably with the east-west column density profiles 
calculated for the forward (filled triangle) and trailing cloud (filled squares). In Figure 9a 
beyond the Lagrange sphere, however, all three of these calculated profiles fall below the 
eclipse observations. At and beyond about 8 Ri 0 , the calculated east-west forward and 
trailing profiles rise above the calculated eclipse profile (because the column density is no 
longer spherically symmetric about Io), with the forward cloud profile having the largest 
column density and showing a distinct change in its slope compared to the trailing cloud 
profile. The corresponding model profiles for the D 2 emission brightness are given in 
Figure 10a. For both the forward and trailing profiles, the calculated eclipse and calculated 
east-west profiles are in good agreement with each other inside the Lagrange radius, with a 
maximum brightness of about 200 kR near the exobase. The calculated east-west profile 
threads the three emission 4 data points for the forward cloud, but falls well below the 
emission 4 data points in the trailing cloud. For both the forward and trailing clouds at 
larger radial distances, the calculated east-west profiles fall well below the areas for both 
the forward and trailing cloud images. This behavior indicates that there is a deficiency in 
the high-speed population of this source flux speed distribution at the exobase. 


22 


Model calculations were therefore performed for a Maxwell-Boltzmann flux 
distribution with a higher most probable speed of v m = 2.0 km sec 1 (t.e., a exobase 
temperature of -3700 K, see Figure 8) and with a total flux <p 0 of 1.8 x 10 8 atoms cm 2 
sec' 1 (i.e., a total source of -0.75 x 10 26 atoms sec' 1 ) and are shown in Figure 9b and 
Figure 10b. For the D 2 emission brightness profiles in Figure 10b, the calculated east-west 
profile now threads the center of the forward cloud image area for a radial distance up to 
about 70 R [ 0 and the lower trailing cloud image area for a radial distance of about 25 R [ 0 
before it falls off too steeply. This improved fit at larger radial distances, however, reduces 
the D 2 emission brightness at the exobase to about 80 kR in Figure 10b and causes the 
calculated eclipse profile in Figure 9b to fall below the measured eclipse profile for radial 
distances inside of about 3 Ri 0 - The Maxwell-Boltzmann flux distribution therefore cannot 
fit both the corona profile near Io and the sodium cloud east-west profiles at large distances 
from the satellite. A flux distribution that has a broader dispersion with enhanced 
populations for both the low-speed and high-speed atoms is required. The modified- 
sputtering flux distribution, which has a broader dispersion, is therefore considered in the 

remainder of the paper. 

Model calculations for a classical sputtering flux distribution (i.e., a=.d and a 
modified-sputtering flux distribution (i.e., a=7/3, having even a larger high-speed 
population) are presented in Figure 1 la and Figure 1 lb for the eclipse observations and in 
Figure 12a and Figure 12b for the emission 4 and sodium cloud east-west D 2 emission 
brightness profiles. For the two flux distributions, the most probable speed is, 
respectively, 1.0 km sec' 1 and 0.5 km sec 1 , and the total flux <p Q is, respectively. 3.2 x 

10 8 atoms cm' 2 sec' 1 (i.e., a total source of -1.3 x 10 26 atoms sec 1 ) and 4.2 x 10 8 atoms 
cm' 2 sec' 1 (i.e., a total source of -1.7 x 10 26 atoms sec 1 ) . In Figure 11. both sputtering 
flux distributions provide a reasonably good fit to the observed column density profile from 


23 



the exobase to radial distances of ~8 Rj 0 , just beyond the Lagrange radius. A careful 
comparison of Figure 1 la and Figure 1 lb shows that the calculated sodium column density 
for the classical sputtering flux distribution is slightly larger within the Lagrange radius and 
is somewhat smaller beyond the Lagrange radius (as is to be expected), with the largest 
differences occurring in the trailing cloud profile beyond about 10 Rio- 

For these two sputtering flux distributions, the same general behavior is seen in 
comparing the two calculated D 2 emission brightness profiles in Figure 12a and Figure 
12b. For the classical sputtering flux distribution in Figure 12a, the calculated D 2 emission 
brightness profiles for the forward profile is slightly above the measured data point inside 
the Lagrange radius, matches the two measured data points beyond the Lagrange radius, 
and then threads the forward cloud image area nicely between about 20 Ri 0 and 80 Rj 0 
before it falls too rapidly and drops below this area. An improved fit for the forward 
profile is provided by the modified sputtering distribution (a=7/3) in Figure 12b. The 
calculated D 2 emission brightness profile for the forward profile in Figure 12b matches the 
measured data point inside the Lagrange radius, matches the two measured data points 
beyond the Lagrange radius, and then nicely threads the forward cloud image area all the 
way to 100 Rio- For the trailing profile and the classical sputtering flux distribution in 
Figure 12a, the calculated D 2 emission brightness profile matches the measured data point 
inside the Lagrange radius, is slightly below the two measured data point outside the 
Lagrange radius, and then threads the lower of the two trailing cloud image areas nicely 
between about 15 Ri 0 and 35 Rj 0 before it falls too rapidly and drops below this area. An 
improved fit is also provided for the trailing profile by the modified sputtering distribution 
in Figure 12b. The calculated D 2 emission brightness profile for the trailing profile in 
Figure 12b matches the measured data point inside the Lagrange radius, is slightly below 


the two measured data point outside the Lagrange radius, and then threads the lower of the 
two trailing cloud image areas nicely all the way to 100 Rio- The lower of the two trailing 
cloud image areas corresponds to the east-west sodium cloud profile when the directional 
feature is oriented either north or south (i.e., above or below the east-west line or null 
location). For this non-null condition, it is particularly noteworthy that the spatially- 
isotropic ejection of sodium from the exobase with a modified sputtering flux distribution 
with a=7/3 provides an excellent fit to the combined eclipse, emission, and forward/trailing 
sodium cloud image profile data from 1.4 Rj 0 to 100 Ri 0 . 

The measured trailing data points in Figure 12b were, however, acquired at the null 
condition and are consistent with the upper of the two trailing cloud image areas also 
acquired at the null condition. In order to fit the trailing profile for the null condition, it is 
then clear that a flux distribution is required with an even more enhanced high-speed 
population than the modified sputtering flux distribution with a=7/3. Since the modified 
sputtering flux distribution for a=7/3 corresponds to the limit of a single collision cascade 
process described by a Thomas-Fermi cross section (see Smyth and Combi 1988b), 
reducing the value of a to a smaller value becomes somewhat physically questionable but 
may be done for practical purposes of illustrating the impact of an enhanced high-speed 
population in the model calculation. 


For a modified sputtering flux distribution with a=2, a most probable speed of 0.4 
km sec' 1 , and an isotropic exobase source rate of 1.9 x 10^6 atoms sec' 1 (i.e., a total flux 
<p 0 of 4.7 x 10 8 atoms cm' 2 sec' 1 ), the model-data comparison is shown in Figure 13 for 

the eclipse column density and in Figure 14 for the emission and sodium cloud east- west 
D 2 emission brightness profiles. In Figure 13, the sputtering flux distribution provides a 
reasonably good fit to the observed column density data points with only a small departure 



very near the exobase and produces a column density profile beyond 10 R[ 0 that is 
significantly enhanced compared to the a=7/3 case in Figure lib. In Figure 14, the 
calculated D 2 emission brightness profile for the forward profile is significantly above the 
measured data point both inside and outside the Lagrange radius and is above or in the very 
top of the forward cloud image area all the way to 100 Rio- The additional enhanced high- 
speed population of the a=2 modified sputtering flux distribution is too large and therefore 
not consistent with the observed forward profile. In contrast for the trailing profile in 
Figure 14, the calculated D 2 emission brightness profile matches the measured data point 
inside and outside of the Lagrange radius very well and then threads the upper of the two 
trailing cloud image areas nicely all the way to ~90 Rio- This demonstrates that the trailing 
cloud can be fitted with an enhanced high-speed population of sodium atoms in the flux 
distribution. It also immediately demonstrates that the flux distribution at the exobase must 
be nonisotropic with some of the enhanced high-speed population weighted more toward 
vector directions that will preferentially populate the trailing cloud rather than the forward 
cloud. This nonisotropic requirement for a flux distribution for speeds of ~20 km sec * is 
consistent with the conclusion reached by the modeling analysis of observations for the 
north-south oscillating directional features by Pilcher et al. (1984). In this analysis, their 
-20 km sec' 1 sodium was constrained to be initially directed at near right angles to Ios 
orbital motion and hence was angularly deficient in the forward direction of motion of the 
satellite and (more importantly) in the trailing apex direction, which preferentially populates 


the forward cloud. 


4. DISCUSSION AND CONCLUSIONS 


The composite spatial information for sodium obtained by combining the eclipse 
observations (radial distances from Io of 1.4 R Io to ~10 R lo ), the emission observations 
(east-west distances of ±4 Ri 0 to ±40 R (o ) and the sodium cloud observations (east-west 
distances of ± 10 Rj 0 to ± 100 Rj 0 ) has been analyzed to extract a basic description for the 
flux speed distribution at the satellite's exobase. An isotropic modified-sputtering flux 
speed distribution with a=7/3, a most probable speed of 0.5 km sec 1 , and a source 
strength of 1.7 x 10 26 atoms sec 1 provided a very good fit to these composite observations 
when the directional feature is either north or south and hence not contributing to east-west 
profile of the trailing cloud. It is remarkable that these observations, acquired by a number 
of ground-based programs over very different spatial scales and at different times during 
the 1975-1985 decade, are so self consistent. Near Io, the two-dimensional sodium 
column density produced by this modified sputtering distribution as calculated by the 
sodium cloud model in the profile analysis above is shown in Figure 15 and can be seen at 
larger distances from Io to become nonspherical and more confined near the satellite plane. 
This flattening near the satellite plane is the merging of the near Io corona into the sodium 
cloud and is caused naturally by orbital dynamics beyond the satellite Lagrange sphere 
where the gravity of Jupiter is dominant. This merging and its spatial extension into the 
forward sodium cloud is reproduced very well by the isotropic modified sputtering 
distribution. The forward cloud portion of the east-west emission data profiles has a rather 
tightly confined slope that, in the absence of the trailing cloud enhancement at the null 
condition, is less steep and is brighter than the trailing cloud profiles. In order, however, 
to reproduce the extended east-west profile in the trailing sodium cloud when the directional 
feature is in the satellite plane (i.e., the null condition), additional nomsotropic high-speed 



sodium is required and is generally consistent with the earlier modeling of the directional 
feature (Pilcher et at. 1984). 

The sodium atoms ejected from Io's exobase as described above by the modified 
sputtering flux distribution have speeds primarily in the range from 0 to a few 10 s km 
sec' 1 . This neutral flux distribution represents the spatially integrated effect of the 
incomplete collisional cascade process that occurs from the collisional interactions of 
heavy-ions in the corotating plasma torus with neutrals in Io's atmosphere. This flux speed 
distribution can be alternatively described as a source rate speed distribution by multiplying 
it by the satellite surface area. In addition to these ion-neutral elastic collisional encounters, 
resonance charge exchange between plasma torus sodium ions and neutral sodium in Io s 
atmosphere (i.e., Na + Na + -» Na+ + Na) is also responsible for producing sodium 
atoms with higher speeds relative to Io. These speeds are centered about the corotational 
ion speed (-60 km sec 1 ) relative to Io’s motion and have a dispersion reaching from 
several 10's km sec' 1 to -100 km sec 1 . This higher-speed source of sodium atoms 
together with some lower speed (-15-20 km sec 1 ) sodium atoms associated with the 
nonisotropic source for the directional feature have been shown in modeling studies (Smyth 
and Combi 1991; Flynn et at. 1992) to accurately reproduce the image of sodium in the 
(D 1 +D 2 ) emission line brightness measured on a radial scale of ± 400-500 planetary radii 
about Jupiter by Mendillo etal. (1990). This large scale distribution of sodium has been 
called the magneto-nebula or sodium zenocorona. Modeling studies for this image 
indicated that the higher-speed source was -2 x IO 26 atoms sec’ 1 while the lower speed 
source was -1 x 10 26 atoms sec 1 . More recent observations and analysis for the sodium 
zenocorona images (Flynn et al. 1993) have shown that the source strength for the higher- 
speed sodium source is time variable with values usually in the range -2-4 x IO 26 atoms 
sec 1 . In addition, a recent analysis by Wilson (1993, private communication) of images 


28 



for the narrow forward jet of sodium observed by Schneider et al. (1991) has shown that 
this spatially distributed source is time variable and typically contributes a few times 10 25 
atoms sec" 1 to the higher speed sodium source for the zenocorona. A typical total source 
rate speed distribution for sodium at Io's exobase can therefore be constructed by 
combining the modified sputtering source rate distribution determined in this paper with the 
charge exchange source rate distribution for the zenocorona (see Smyth and Combi 1991). 
This total source rate speed distribution function is shown in Figure 16, where the lower 
(solid line) and upper (dashed line) curves correspond, respectively, to the sodium 
zenocorona higher-speed source rate of 2.2 x 10 26 atoms sec' 1 as determined by Smyth and 
Combi (1991) from January 1990 image data and 4 x 10 2 ^ atoms sec' 1 as determined by 
Flynn et al. (1993) from February 1991 image data. 

The total source rate speed distribution functions at lo's exobase expected for other 
atomic species, such as K, O, and S, can be constructed in a similar fashion to sodium by 
adopting the estimated source rates for the modified sputtering distribution and the charge 
exchange distribution given by Smyth and Combi (1991). The main differences in the total 
source rate speed distribution functions will arise from the relative importance of the source 
strengths for the modified sputtering and charge exchange processes. The relative 
importance of the higher speed source to the lower speed source would be much smaller 
than sodium for K, because of the smaller number of K + ions in the plasma torus, and 
would be significantly larger than sodium for O and S, because of the much larger number 
of O and S ions in the plasma torus. 

Future studies for the sodium flux speed distribution at lo's exobase are anticipated 
using a much larger emission data set for sodium (-100 profiles) acquired in 19S7 and 
recently reduced by Schneider (1993, private communication) to a form suitable for 


29 



analysis. Because of the much larger data base, it will be possible in these studies to 
analyze the combined spatial and spectral information in order to refine the nonisotropic 
nature of the flux distribution and also to search for possible east-west and System III 
modulations in the flux speed distribution. Once this information is determined for 
sodium, the implications for the more abundant species in Io's atmosphere will be 
particularly important in other related studies for the many faceted and complex phenomena 
in the lo-Jupiter system. 


30 


ACKNOWLEDGMENTS 


We are grateful to N. M. Schneider for helpful discussions and for providing the 
numerical data for the 1985 emission observations. This research was supported by the 
Planetary Atmospheres Program of the National Aeronautical and Space Administration 
under grant NAGW-3585 to the University of Michigan and under contracts NASW-4416 
and NASW-4471 to Atmospheric and Environmental Research, Inc. 



REFERENCES 


Bergstralh, J. T„ D. L. Matson, and T. V. Johnson 1975. Sodium D-Line Emission from Io: 
Synoptic Observations from Table Mountain Observatory. Astrophys. J. Lett. 

195, L- 131 -L 1 35. 


Bergstralh, J. T., J. W. Young, D. L. Matson, and T. V. Johnson 1977. Sodium D-Ltne 
Emission from Io: A Second Year of Synoptic Observation from Table Mountain 
Observatory. Astrophys. J. Lett. 211, L51-L55. 

Brown, R. A., and Y. L. Yung 1976. Io, Its Atmosphere and Optical Emissions. In Jupiter 

Studies of the Interior, Atmosphere, Magnetosphere, and Satellites (T. Gehrels, Ed.), 
pp. 1 102-1 145. Univ. of Arizona Press, Tucson. 


Flynn, B., M. Mendillo, and J. Bumgardner 1992. Observations and Modeling of the 
Jovian Remote Sodium Emission. Icarus 99, 1 15-130. 


Flynn, B„ M. Mendillo, and J. Bumgardner 1993. The Jovian Sodium Nebula: Two Yeao> 
of Ground-Based Observations, preprint. 


Goldberg, B. A., Yu. Mekler, R. W. Carlson, T. V. Johnson, and D. L. Matson 1980. lo's 
Sodium Emission Cloud and the Voyager 1 Encounter. Icarus 44, 305-317. 


Goldberg, B. A., G. W. Garneau, and S. K. LaVoie 1984. Io's Sodium Cloud. Science 



Goldberg. B. A. and Smyth. W. H. 1993. The JPL Table Mountain Io Sodium Cloud Data 
Set. Paper in preparation. 


Matson. D. L„ B. A. Goldberg, T. V. Johnson, and R. W. Carlson 1978. Images of los 
Sodium Cloud. Science 199, 531-533. 

Mendillo, M„ J. Baumgardner, B. Flynn and W. J. Hughes 1990. The Extended Sodium 
Nebula of Jupiter. Nature, 348, 312-314. 

Murcray, F. J. 1978. Observations of lo's Sodium Cloud, Ph. D. Thesis, Dept, of 
Physics, Harvard University. 

Murcray, F. J. and R. M. Goody 1978. Pictures of the Io Sodium Cloud. Ap. J., 226, 

327- 335. 

Pilcher, C. B.. W. H. Smyth, M. R. Combi, and J. H. Fertel 1984. Io's Sodium Directional 
Features: Evidence for a Magnetosphenc-Wind-Driven Gas Escape Mechanism. Ap. J 
287, 427-444. 


Schneider. N. M. 1988. Sodium in lo's Extended Atmosphere. Ph.D. Thesis, Department 
of Planetary Sciences, Univ. of Arizona. 

Schneider. N. M., D. M. Hunten, W. K. Wells, and L. M. Trafton 1987. Eclipse 
Measurements of lo's Sodium Atmosphere. Science, 238. 55-58. 



Schneider, N. M„ D. M. Hunten, W. K. Wells, A. B. Schultz, and U. Fink 1991. The 
Structure of Io's Corona. Ap. J 368, 298-315. 

Smyth, W. H. and M. R. Combi 1987. Nature of Io’s Atmosphere and its Interaction with 
the Planetary Magnetosphere. BAAS, 19, 855. 

Smyth, W. H. and M. R. Combi 1988a. A General Model for Io's Neutral Gas Cloud. 

I. Mathematical Description. Ap. J. Supp., 66, 397-41 1. 

Smyth, W. H. and M. R. Combi 1988b. A General Model for Io's Neutral Gas Clouds. 

II. Application to the Sodium Cloud. Ap. J., 328, 888-918. 

Smyth, W. H. and M. R. Combi 1991. The Sodium Zenocorona. JGR, 96, 22711-22727. 

Smyth, W. H. and B. A. Goldberg 1993. The Io Sodium Cloud: Space-Time Signatures of 
East- West and System III Longitudinal Asymmetries in the Jovian Magnetosphere. 
Paper presented at the Io: An International Conference, San Juan Capistrano, 
California, June 22-25. 

Smyth, W. H„ and M. B. McElroy 1978. Io's Sodium Cloud: Comparison of Models and 
Two-Dimensional Images. Astrophys. J. 226, 336-346. 


34 


1985 Io Eclipse and Emission Measurements 


a 

o 


C s 
° 3 

>> 49 
^ -O 

O 8 

cx 

13 

W 


CN 


CN 


vO 


r- oo 


Os 


wo 


g 

g 

o 

<u 

cx 

00 


-a 

3 


00 O 1 
c J 


5 s - 

T3 

is 


OO 

vO 

CN 

CN 

CO 

(N 


CO VO 

r- 

oo 

OS 

CO 

O 

r— < 

WO 


CO CO 

wo 

txQ 

OO 

o 

JZ 

JC 



~a 

Tt Tt 
-C -C 


LTi 

wo 

WO 

WO 

WO 

wo 


wo wo 

wo 

oo 

oo 

OO 

OO 

OO 

OO 


OO OO 

OO 


49 

0X3 

c 


CO 

OO 

Tt 

CO 

co 


Tt 

p 

Tt 

<N 

CO 

CN 


CO 

' OX) 

o 

CN 

wo 

CO 

CO 

<N 

CN 

CO 


CO 

d 

CO 

WO 

CN 


vd 

<N 

1 C3 

s 

-hi 

44 

-H 

-hi 

44 

+1 

-hi 

-hi 

-hi 

44 

+1 

-hi 

4-1 

-hi 

DC 


Os 

OS 

1 

r- 

Tt 

wo 

vO 



wo 

vC 

Os 

CO 

Os 



os 

CN 

wo 

Tt 


d 

CO 

CO 

d 

CN 

d 

— ■ 

d 

oi 



<N 

CO 

MD 

Os 

CO 

wo 


wo 

Tt 

» — > 

vO 

OO 

o 

wo 







CO 


- — ' 

CN 

CN 

CO 

CO 


-gJi 


r- O — « 


r- <04 Tt 

r- wo 

SO 

b so 


d — — 

d 

d Tt d 

d — ’ 

— • 

c c 

49 ^ 




«— * 


49 

« < 

oo OO 
C 09 

-H 44 44 

44 

44 44 -hi 

4-1 -hi 

-hi 

8 % 

rt x) 

CX w 

Tt CO CN 

vO 

wo so 

r-~ Tt 

wo 


d c3 CN 

so vo 

vd 

r- 

O co Q 
OO Os o 

Tt gs 
Os O 

8 

o CX 


<N 

— 

CN CO 



Tt o O 
^ (N m 
r- r- oo 

o o o 


s 

o 


wo vO vO 
Tt CN — ■ 
CN CO tT 

o o o 


Xg 

CO wo 
O O 


i — Tt o r- 

0 O 04 O 

+1 -hi -hi -hi 

04 vO CO ■ — • 

1 ■ »“ -1 Os cO 
— CN CN Tt 


O — vO Tt 
co O wo cO 
04 CO CO VO 
O O O O 




CO 

Tt 

wo 

. — . 

CO 





« — « 

CN 

CN 



t. 

u 


u 

u. 

49 

r- 

CN 

J8 



95 

£ 

s 

a 

CO 

E 

E 

E 

E 

£ 

3 

49 

o 

49 

49 

49 


OO 

cx 

a. 

a. 

a. 

a. 


3 

49 

49 

49 

49 

49 


< 

oo 

OO 

00 

oo 

oo 


UD 

CD 

CD 

tT> 

00 

CO 

CD 

CD 

UO 

co 

CvJ 

CD 

CD 

LO 

CO 

CO* 

CO 

CD 

LO 

CO 

LO 

00 

CD 

LO 

CO 


00 

*o> 

to 

00 

CD* 

r- 

CD 

to 

00 


a; 

LO 

Q. 

T3 

uj 


. Eclipse 2 : 85h1 03, 85h1 04, 85h105, 85h1 06, 85hl 07, 85h108, 85hl09, 8Sh1 10, 8Sh112 

. Eclipse 3 : 8Sh153, 85h1 54, 85hl 55. 85h 1 57, 85hl 59, 85h162, 85h163 

I. Eclipse 4 : 85h287, 85h288, 85h289, 85h290, 85h291, 8Sh292, 85h293, 85h294, 85h295 

'• Eclipse 5 : 85h441, 85h442, 85h443, 85h444, 85h445, 85h446, 85h447, 85h448, 85h449, 85h450 



Radial Emission 1 Emission 2 Emission 4 Emission 5 Emission 7 Emission 8 Emission 9 Emission 3* Emission 6 

Distance 61.4° 72.2° 87.7° 100.6° 117.2° 121.6° 143.1° 276.6° 294.7° 

From lo 27 August 27 August 14 September 14 September 23 September 23 September 23 September 13 September 15 September 
(satellite radii) (H5gl88) (85gl96) (85hl02) (85hl 13) (85h433) (85h436) (85h457) (85h032) (85hl52) 



r~~ 

04 

Nt 

vq 

oo 

vq 

O 

r- 

Nt 

oc 

oc CJ 

O' o- 
co 04 



r- 

nO 

vri 

DC 

CO 

co 

04 04 

« ■ 


<N Ol 

04 

0 1 

04 







O ON 

ON ON 

ON 

vn 

o~ 

04 

to 04 

CO ON 

*— * 

CO 

oo 

ON 

Tt — 

OO ON 

ON 

ON 

NO 





calibration uncertain 








Table 3 


Power Law Fit to the D 2 Emission 


Profiles for Io East of Jupitert 



to (Jeocentric 

Io System III 


Phase Angle 

Longitude 

Date 

(deg) 

(deg) 

-august 27 

61.4 

29.9 


72.2 

65.1 

eptember 14 

87.7 

31.4 


100.6 

73.6 

September 23 

117.2 

267.6 


121.6 

281.9 


143.1 

352.9 


t Profile points inside of 4 R Io are excluded 
F = forward cloud; S = symmetric turning point; 


Exponent 

Amplitude (kR) 

East 

West 

East 

West 

Profile 

Profile 

Profile 

Profile 

1.67 (F) 

1.57 (T) 

191 

124 

1.85 (S) 

1.80 (S) 

169 

142 

1.23 (S/T) 

1.80 (S/F) 

89 

188 

1.27 (T) 

— 

66 

— 

2.05 (T) 

1.57 (F) 

342 

135 

1.96 (T) 

1.54 (F) 

283 

138 

2.16 (T) 

1.64 (F) 

374 

165 


= trailing cloud. 



OMGfNAL PAGE tS 
Of POOR QUALITY 


1985 Io Emission Measurements 



'll) 






»— 






cd 





Ti ™ 

X <19 

U D U 

? 

u 

OX) 

OX) 

00 


c a- 
cd a ^ 

o C 

<2 o 
"5b c 

c 

‘cd 

ailin 

no 

c 

cd 

o o c 
c c c 

CQ.Sf 

W U OT 

c 

b 

b 

b 


cd 






1 b 






1 




forward 

forward 

forward 

Dominan 

Spatial 

Profile 

similar 

symmetri 

similar 

trailing 

trailing 

■3 § 

C <19 -b 
o *-* 

^ rd 

U cd C 

lull 

outh 

-C 

C 

o 

c 

null 

outh 

null 

XZ X _ 

t: C -r 
° ° c 

C9 19 CJ 

to 

•— « 

CO 


c c 

.b Uh c 


D 




Q O 


c 




, , 






o 

— -T) 

CO Tf 

CO 

Ol 

o 

CO 04 Oi 

o Systerr 
Longitui 
Range 
(deg) 

oi oi 

oi 

oi — 

Oi 

oi *— 1 04 

44 +1 

-H 

-hi +1 

+1 

-hi +1 4-1 

CTn — ; 


xf vO 

o 

'OO'O) 

CTn io 


— ' Oi 

CO 

l — — 04 

• — * 1 

04 vC 

o 

O O' 

vo 

vO oc to 




CO 

04 04 Oi 

b <U - . 

O- — 

o- 

O' 

r- 

0-0 0 

b OX) 

c C OJ ^-s 

O — •* 

O 

o o 

O 

o o o 

>{ c o 

4H 

+1 

+1 +1 

+1 

41 41 4-1 

^ u cd xi 

oo a 

or 

XT 04 

VO 

O' 

O' 

04 lC — 

’ oi 

x> 

^ 55 

oo O 

tT 

O' ^ Oi 

x: 

lO o- 

O' 

O' 

• — < 04 "CT 

O Cl, 


04 

* * 

04 




Oi 

rr 

to 

Oi 



< 

•— 1 

* — 1 

04 


o- 

l-. 

u. 

u 

u 

cj 

V 

<U 

<D 

19 

OI 

X) 

X 

X) 

X) 

cd 

Q 

to 

E 

E 

E 

E 

3 

19 

o 

V 

19 


04) 

Qu 

a. 

Cl 

O. 



19 

o 

19 

09 


< 

oo 

oo 

CO 

OO 

c 






o 






cc 

oo 

— 04 

c-n 

rr VO 

vC 

O' oc O' 


E 

UJ 



FIGURE CAPTIONS 


FIG. 1. 1985 Eclipse Data. The sodium column density is shown as a function of the 
radial distance from the center of Io for the five eclipse measurements acquired by 
Schneider et al. (1991). The position of the nominal exobase and the average Lagrange 

radius are indicated. 

FIG 2. 1985 Emission Data. The sodium D 2 emission brightness in units of 

kiloRayleighs is shown as a function of the east-west distance from the center of Io along 
the observing slit for the nine emission observations summarized in Table 3 and acquired 
by Schneider et al. (1991). 

FIG. 3. Selected East and West Brightness Profiles for the 1985 Emission Data. The 
spatial profiles both east and west of Io for the sodium D 2 emission brightness in units of 
kiloRayleighs are shown as a function of the distance along the observing slit from the 
center of Io for five emission observations identified by their satellite geocentric phase 
angle. These five profiles occur when Io is east of Jupiter and just past the satellite phase 
angle where the forward cloud has its symmetric turning point so that the trailing cloud 
profiles are all to the east of Io and the forward cloud profiles are all to the west of Io. A 
power law fit to each profile is also shown. At larger distances from Io, an envelope for 
the east-west D 2 emission profile acquired from sodium image data is shown by the shaded 
area. For the trailing profile, the shaded area is divided into two parts, where the lower 
area corresponds to sodium cloud data where the directional feature is oriented either north 
or south and where the upper area corresponds to the directional feature oriented along the 
east-west direction (i.e., the null condition). 


40 


FIG. 4. System III Correlation for the North-South Orientation of the Sodium Cloud 
Directional Feature. The north and south orientation of the directional feature as 
determined by Goldberg and Smyth (1993) from a set of sodium D-line emission image 
observations is shown as a function of the System III longitude angle of lo. Note that 
image observations for Io east and west of Jupiter are identified, respectively, by open and 

filled squares. 

FIG. 5. Io Sodium Cloud Images. Three calibrated D2 emission images of the Io sodium 
cloud from the JPL Table Mountain Data Set are shown to proper scale with Jupiter and Ios 
orbit as viewed from earth in 1981 (Smyth and Goldberg 1993). The Io System III 
longitude and corresponding orientation of the trailing directional feature in image A are 247 
degrees and north, in image B are 104 degrees and south, and in image C are 178 degrees 
and only very slightly north. An east-west spatial scale of ±100 planetary radii about Io is 
shown for reference. Contour levels for the D2 brightness, from outside to inside, are 0.2, 

0.5, 1, 2, 4, 6, 8, and 10 kR. 

FIG. 6. Monoenergetic Sodium Cloud Modeling of the 1985 Eclipse Data. The eclipse 
data points of Schneider et at. ( 1991) are shown by the open circles. The dots are 
calculated by the Io sodium cloud model of Smyth and Combi (1988b) for their case C 
description of the plasma torus with an east-west electric field of 2.8 m V m' 1 in the lo frame 
and for an isotropic and radial ejection of 2.6 km sec' 1 sodium atoms from an assumed 
exobase of 2600 km radius for a source strength of 6.2 x 10 25 atoms sec' 1 . The calculation 
was performed for an Io geocentric phase angle of 92.9 degrees and an Io System III 
longitude angle of 48.6 degrees. The sodium source used here is typical of that required for 
this monoenergetic calculation to fit the D2 emission brightness of the sodium cloud images 


at larger distances from Io. 



FIG. 7. Effect of Lifetime on the Sodium Column Abundance. Two model calculations 
for the column abundance are compared with the eclipse data points (shown by the open 
circles) of Schneider et al. (1991). The plasma torus description adopted in the Io sodium 
cloud model is described in Fig. caption 6. The eclipse data points are fit outside and 
somewhat inside the Lagrange radius by the model for the minimum lifetime cas e (filled 
circles), which occurs for an Io phase angle of 90 degrees (eastern elongation), an Io 
System III magnetic longitude angle of 179 degrees, and a sodium source rate of 0.96 x 
1 q 26 atoms sec' 1 . The same source rate is assumed for the maximum lifetime CSS S (filled 
squares), which occurs for an Io phase of 270 degrees (western elongation) and an Io 
System III magnetic longitude angle of 299 degrees. 

FIG. 8. Flux Speed Distribution Functions for Sodium at Io’s Exobase. Maxwell- 
Boltzmann flux speed distribution for sodium are shown for a most probable speed, v m , of 
1.3 km sec' 1 (short dashed line) and 2.0 km sec’ 1 (longer dashed line). Modified 
sputtering flux speed distributions are also shown for a =3 and a most probable speed of 
1.0 km sec 1 (dotted line) and for a =7/3 and a most probable speed of 0.5 km sec' 1 (solid 
line). All of the flux speed distributions are normalized to unit area. 

FIG. 9. Model Calculations for the Io Eclipse Data Using a Maxwell-Boltzmann Flux Speed 
Distribution. The atomic sodium column density profile near Io determined from the 1985 
eclipse data by Schneider et al. (1991) is shown by the open circles. The model calculated 
column density profiles are shown by solid dots for the (cylindrically-averaged) corona, by 
solid triangles for the forward cloud along the east-west slit direction, and by solid squares 
for the trailing cloud along the east-west slit direction. These column density profiles were 
calculated using the Io sodium cloud model of Smyth and Combi (1988b) for their case C 



description of the plasma torus (see Fig. caption 6) and for an Io geocentric phase angle of 
92.9 degrees and an Io System III longitude angle of 48.6 degrees, which arc similar to the 
emission 4 observation conditions in Table 1. Sodium was ejected uniformly from an 
assumed exobase of 2600 km radius with a velocity dispersion for a Maxwell-Boltzmann 
flux distribution, where in (a) v m = 1.3 km see l and ^ = 3.0 x 10« atom cm'2 sec'l, and 
in (b) v m = 2.0 km sec* 1 , and <p 0 = 1.8 x 10 8 atom cur 2 sec' 1 . 

FIG. 10. Model Calculations for the East- West D2 Brightness Profiles Using a Maxwell- 
Boltzmann Flux Speed Distribution. The east-west D2 brightness profile near Io in both the 
trailing and forward cloud directions as determined by the emission 4 data of Schneider et al. 
(1991) are shown by the open circles. The east- west profile envelopes in both the trailing 
and forward cloud directions as determined from the sodium cloud image data are shown by 
the shaded areas (see Fig. caption 3). The descriptions of the symbols for the calculated 
profiles, the sodium cloud model and plasma torus, and the Maxwell-Boltzmann flux 
distribution in (a) and (b) are the same as in Fig. caption 9. 

FIG. 11. Model Calculations for the Eclipse Data Using a Modified Sputtering Flux Speed 
Distribution. The atomic sodium column density profile near Io determined from the 1985 
eclipse data by Schneider et al. (1991) is shown by the open circles. The model calculated 
column density profiles are shown by solid dots for the (cylindrically-averaged) corona, by 
solid triangles for the forward cloud along the east- west direction, and by solid squares for 
the trailing cloud along the east-west direction. These column density profiles were calculated 
using the lo sodium cloud model of Smyth and Combi (1988b) for their case C description of 
the plasma torus (see Fig. caption 6) and for an Io geocentric phase angle of 92.9 degrees 
and an Io System III longitude angle of 48.6 degrees, which arc similar to the emission 4 
observation conditions in Table l. Sodium was ejected uniformly from an assumed exobase 



of 2600 km radius with a velocity dispersion for a modified sputtering flux distribution, 
where in (a) a - 3, v m = 1.0 km sec' 1 , and <p 0 = 3.2 x 10 8 atom cm' 2 sec’ 1 , and in (b) a - 

7/3, v m = 0.5 km sec' 1 , and <p 0 = 4.2 x 10 8 atom cm' 2 sec" 1 . 

FIG. 12. Model Calculations for the East-West D2 Brightness Profiles Using a Modified 
Sputtering Flux Speed Distribution. The east-west D2 brightness profile near Io in both the 
trailing and forward cloud directions determined by the emission 4 data of Schneider et al. 
(1991) are shown by the open circles. The east-west profile envelopes determined from the 
sodium cloud image data are shown by the shaded areas (see Fig. caption 3). The 
descriptions of the symbols for the calculated profiles, the sodium cloud model and plasma 
torus, and the modified sputtering flux distribution used in (a) and (b) are the same as in Fig. 
caption 1 1 . 

FIG. 13. Model Calculations for the Eclipse Data Using a Modified Sputtering Flux 
Speed Distribution. The atomic sodium column density profile near Io determined from 
the 1985 eclipse data by Schneider et al. (1991) is shown by the open circles. The model 
calculated column density profiles are shown by solid dots for the (cylindrically-averaged) 
corona, by solid triangles for the forward cloud along the east-west direction, and by solid 
squares for the trailing cloud along the east-west direction. These column density profiles 
were calculated using the Io sodium cloud model of Smyth and Combi (1988b) for their 
case C description of the plasma torus (see Fig. caption 6) and for an Io geocentric phase 
angle of 92.9 degrees and an Io System III longitude angle of 48.6 degrees, which are 
similar to the emission 4 observation conditions in Table 1. Sodium was ejected uniformly 
from an assumed exobase of 2600 km radius with a velocity dispersion for a modified 
sputtering flux distribution, where a = 2, v m = 0.4 km sec'*, and <f> 0 = 4.7 x 10 8 atom 

cm' 2 sec" *. 


4 4 


FIG. 14. Model Calculations for the East- West D2 Brightness Profiles Using a Modified 
Sputtering Flux Speed Distribution. The east- west D2 brightness profile near Io in both 
the trailing and forward cloud directions determined by the emission 4 data of Schneider 
ct al. (1991) are shown by the open circles. The east-west profile envelopes determined 
from the sodium cloud image data arc shown by the shaded areas (see Fig. caption 3). The 
descriptions of the symbols for the calculated profiles, the sodium cloud model and plasma 
torus, and the modified sputtering flux distribution are the same as in Fig. caption 1 3. 

FIG. 15. Two-Dimensional Nature of the Sodium Column Density in Ios Corona. 
Contours for the two-dimensional column density in Io's corona are shown in the sky- 
plane of the earth as determined from the sodium cloud model calculation for the modified 
sputtering flux speed distribution described in Fig. 11(b). The vertical and horizontal 
directions are the projected directions that are, respectively, perpendicular and parallel to 
the semi-major axis of the Io's orbital ellipse on the sky plane. The scale is in kilometers, 
and the small tick marks are separated by 1000 km. Io's location and size are shown to 
scale by the black circle. The sodium column density contours in units of 10 1 1 atoms cm 2 
are, from inside to outside, 7, 5, 3, 2, 1, 0.7, 0.5, 0.3, and 0.2. 

FIG. 16. Total Source Rate Speed Distribution Function for Sodium at Io's Exobase. The 
total source rate speed distribution function at Io's exobase, in units of 10 26 atoms sec’ 1 
(km/sec) 1 , is composed of three separate source rate speed distributions as discussed in the 
text and is shown for a smaller (solid line) and larger (dashed line) source strength for the 
higher-speed zenocorona source centered about 57 km sec" 1 . The decomposition of the 
solid curve into its three separate source rate speed distributions is shown in the cutout and 
is determined by combining (1) the isotropic modified sputtering source rate distribution 



(dotted line in the cutout) fora = 7/3, v m = 0.5 km sec 1 and a source strength of 1.75 x 
10 26 atom sec 1 , (2) the nonisotropic lower-speed source rate distribution (short dashed 
line in the cutout) for the sodium zenocorona centered about 20 km sec' 1 , with a source 
strength of 1.1 x 10 26 atoms sec 1 as determined by Smyth and Combi (1991), and (3) the 
nonisotropic higher-speed source rate distribution (longer dashed line in the cutout) for the 
sodium zenocorona centered about 57 km sec 1 , with a source strength of 2.2 x 10 26 atoms 
sec- 1 as determined by Smyth and Combi (1991). The source rate speed distribution 
(dashed line) with the larger source strength for the higher-speed zenocorona source is 
determined in the same fashion with the exception that the higher-speed zenocorona source 
is increased to 4.0 x 10 26 atoms sec 1 , so as to exhibit the typical time-variable source 
range of 2-4 x 10 26 atoms sec 1 reported by Flynn et al. (1993). 


4 6 


COLUMN DENSITY (cm-2) 


1985 Eclipse Data 



RADIAL DISTANCE FROM 10 (10 RADII) 


r i i <’ l 



D2 INTENSITY (kR) 


1 00 


1 0 


1 


0.1 


1985 Emission Data 




LU 

00 


O 

ss 


x 


H- 

+ 


I 


I 


^j- Emission 1,2 t 3,5.6.7,8.9 - 

+ East Profile Sept 1 4 (Emission 4) ~ 

O West Profile Sept 1 4 (Emission 4) _ 



co 

z> 

o 


LU 

o 



g 

o 

5 


1 ^ 1 1 1 I 




1 


♦ 



•H- 


1 0 


' 1 r 

1 00 


DISTANCE FROM 10 (10 RADII) 


Ki«jur<? 0 



1985 EMISSIOM DATA 


0)0 0 ) 0 ) 0 ) 
<D CD <D CD CD 

D D D D t) 



Figure 





a East Cloud 
■ West Cloud 


Figure A 







Maxwell-Boltzmann Flux Distribution 





O 

CO 



— 

psl 

o 

o 

o 

o 

o 

v 

oO 

O 

x 

O 

X 


X 

o 

X 

o 

X 

Q 

A 

O 

o 

O 


( z _uid suioic) A1ISN3Q NWmOD WfliaOS 


RADIAL DISTANCE FROM 10 (R Io ) 

Figure 9a 




rm i H i '"'ji' 

/■ X 

/ y 


i 

► ■ 
► ■ 

► ■ 

► ■ 


► v 

o°:r 




SniOVU 33NVU9VH 


3SV0OX3 


pvl 

— 

o 

O 

o 

o 

r • A 

o 

y 

o 

X 

X 

o 

X 

o 

O 

o 



— 



I 


RADIAL DISTANCE FROM 10 (R Io ) 





RADIAL DISTANCE FROM 10 (R to ) 

Figure 10a 


7' T "T "T 7 


T -r - r i p irr Tii r |'i y \ 




111 1 1.1. J _!.... 1 l*‘l 11 : 1 



(>!>)) SS'IN.I.HWISI <'<*1 


RADIAL DISTANCE FROM 10 (R to ) 

Figure 10b 




Classical Sputtering Flux Distribution 

Most Probable Speed = 1.0 km/s 
Velocity Dispersion : a = 3.0 
Isotropic Exobase Source Rate = 1.3 x 10^6 atoms/s 



( £_uid SUIOIG) AXISN3Q NWHIOD WflldOS 


RADIAL DISTANCE FROM 10 (R l0 ) 




ECLIPSE DATA 

Modified Sputtering Flux Distribution 

Most Probable Speed = 0.5 km/s 
Velocity Dispersion : a = 7 /3 
Isotropic Exobase Source Rate = 1.7 x 10^6 atoms/s 



o 

o 

o 


o 

d 


o 


( z a ud siuoic) A1ISN3Q NlAimOD WfliaOS 


RADIAL DISTANCE FROM 10 (Rio) 

Figure lib 


Classical Sputtering Flux Distribution 

Most Probable Speed = 1.0 km/s 
Velocity Dispersion : a - 3.0 
Isotropic Exobase Source Rate = 1.3 x 10^6 atoms/s 



RADIAL DISTANCE FROM 10 (Ri 0 ) 

Figure 12a 





— 00 
•tH 

< .3 

2) ti. 

_q bO 

a 

-j S 

-i—i 

ij_ 3 

a 

00 

_ T3 
a> 


•a 

- O 

2 



( z aud siuojg) A1ISN3G NWmOD WfllCIOS 


RADIAL DISTANCE FROM 10 (R, 0 ) 



I! X 


"S -2 2 

H. 52 5 

2 S D 



smavH 30Nvy9vn 


3SVGOX3 


3SW0OX 3 


smovy 30NvyDV3 


iLi-l-L-l-l— I 1- 


jm.i .i-i— *- 


liuj.i_i- i i iiu-i-i-iVi — i 


on) ssmmoiwi <’a 


RADIAL DISTANCE FROM 10 (Rio) 

Figure 14 




0000 0 10000 20000 






siq aiey aojnos 


Figure 16 


Appendix B 


Correlating System III Longitudinal Asymmetries in 
the Jovian Magnetosphere and the Io Sodium Cloud 




Correlating System III Longitudinal Asymmetries in the Jovian 
Magnetosphere and the Io Sodium Cloud 


William H. Smyth 1 


Bruce A. Goldberg^ 


Preliminary Version 


1 . Atmospheric and Environmental Research, Inc., Cambridge, MA 02 1 39 

2. Jet Propulsion Laboratory, Pasadena, California 91103 

Number of manuscript pages : 36 
Number of Figures : 9 
Number of Tables : 5 

Key words : Io, satellite exosphere, gas tori, Jupiter's magnetosphere 


1 



Running heading : Correlating System III Longitudinal Asymmetries 


For editorial correspondence and proofs please contact: 
William H. Smyth 

Atmospheric and Environmental Research, Inc. 
840 Memorial Drive 
Cambridge, MA 02139 

Telephone number : (617) 547-6207 
Fax number : (617) 661-6479 
Internet : smyth@aer.com 


2 


ABSTRACT 


An investigations is undertaken to study an observed space-time "bite-out" feature 
that occurs in the south portion of the Io sodium cloud when Io is near western elongation 
and is in the System III longitude angle interval between ~ 200* - 300* (i.e., in the so called 
"active sector"). This "bite-out" feature was discovered in the 263 images acquired in 1981 
for the sodium cloud (Region B/C) by Goldberg et al. (1984). These sodium cloud images 
are a subset of the larger JPL Table Mountain Io Sodium Cloud Data Set obtained over the 
1974 to 1981 time frame. Examination of the 1981 sodium cloud images shows, 
surprisingly, that the "bite-out" signature does not occur (with the exception of only an 
occasional hint of it) when Io is near eastern elongation and is in the active sector. At the 
time of the sodium observations, an inherent System III asymmetry in the plasma torus was 
documented independently through ground-based observations of plasma torus ion 
emission lines by Morgan (1985 a,b). This inherent System III asymmetry has been used 
to improve the description of the plasma torus in the sodium cloud model of Smyth and 
Combi (1988b). The improved sodium cloud model has been used to study the "bite-out" 
signature. The modeling studies show that the time evolution of this "bite-out" signature 
can be reproduced and that it reflects the space-time dependent electron-impact ionizauon 
sink for sodium produced by the combined east-west and System III asymmetries in the 
Jovian magnetosphere. In particular, the "bite-out" feature for Io near western elongation 
is correlated with an inherent System III asymmetry in the Io plasma torus in the presence 
of an east-west electric field. The presence of an east-west electric field in the planetary 
magnetosphere has been inferred from a number of different plasma torus ion and neutral 
emission line brightnesses and was incorporated earlier in the sodium cloud model (Smyth 
and Combi 1987; 1988b) to explain the east-west intensity asymmetry first discovered in 
the slit-averaged brightness data for the sodium emissions very near Io (a subset of the 
larger JPL Table Mountain Io Sodium Cloud Data Set) reported by Bergstralh et al. (1974, 
1975). This east-west intensity asymmetry is also exhibited in the 1981 Region B/C image 
data set at larger distances from Io that are beyond the observational occulting mask 
centered on the satellite. The sodium cloud, because of its short electron impact ionization 
lifetime in the plasma torus, therefore provides an additional avenue to monitor and study 
these east-west and System III structures of the magnetosphere in past as well as in future 
observational data. 


3 



I. INTRODUCTION 


Studies of observational data for the time-dependent morphology of the D-line 
emissions of lo's sodium cloud on the sky plane have provided an excellent remote sensing 
tool for probing the nature of the local and escaping atmospheres of the satellite and their 
manifold interactions with the planetary magnetosphere. In the two decades since the 
startling 1972 discovery by Brown (1974) of D-line emissions from Io, a large number of 
increasingly higher quality spectral and spatial observations of the sodium cloud have been 
obtained from ground-based facilities. The JPL Table Mountain Io Sodium Cloud Data 
Set, the most massive such data set acquired to date, was briefly summarized earlier by 
Goldberg et al. (1984) and more extensively in the companion paper (Goldberg and Smyth 
1992, hereafter Paper I). This data set provides a valuable information base for three 
reasons. First, it documents various aspects of the nature and temporal variability of the 
sodium emission over a seven year time span from 1974 to 1981. Second, it provides a 
framework for the 1979 Jupiter-encounter data set acquired by the Voyager 1 spacecraft not 
only because of the time period covered, but also because images of the sodium cloud were 
acquired on and near the closest approach date of March 5, 1979. Third, it establishes a 
unique comparison data set for the more recent and higher spatial/time resolution images 
that have begun to be acquired over the last few years and are anticipated to be obtained 
with improved instrumentation in the next several years. 

Observations described in Paper 1 may be divided into two major and spatially 
complementary regions: Region A and Region B/C. As used here. Region A includes 
spatial emission from lo's disk and from the satellite's gravitationally bound atmosphere 
(i.e., radius ~ 6 satellite radii or 0. 1 5 Jupiter radii) and has a maximum brightness near the 
satellite of -100 kiloRayleighs (kR). Region B/C includes sodium D-line emission that 


4 


extends several Jupiter radii from Io, that is characterized primarily by a forward cloud and 
that has a brightness which varies from many 10's of kR nearer the satellite to ~ 0.2 kR at 
the nominally measured outer cloud boundary. Observations for Region A were acquired 
by single slits in 1974-1979 and 1981, by multislits in 1976-1978, and two-dimensional 
images in 1981. Two-dimensional images for Region B/C were obtained in 1976-1979 
and in 1981. As summarized in Table 1, the complete data has led to the identification 
and/or subsequent study of a number of the spatial features or characteristics of the sodium 
emission near Io, near Io's orbit, and far from Io's orbit. 

Two entries in Table 1, the east-west asymmetry of the forward cloud and the 
System III asymmetry of the sodium cloud near western elongation, are the only features 
for which there are to date no modeling analysis studies. The east-west intensity 
asymmetry of the sodium cloud is documented in the JPL data set spatially beyond the 
gravitational confines of the near Io data of Bergstralh et al. (1975, 1977), which also 
exhibited an east-west asymmetry, and is almost certainty caused by the same mechanism, 
namely an east-west asymmetry of the plasma torus sink for sodium. The east-west 
intensity asymmetry of the sodium cloud will not be addressed here. The System III 
asymmetry observed in the shape of the sodium cloud when Io is near western elongation 
is also very well documented in the JPL data set. This asymmetry, a bite-out or zone of 
deficient sodium in the cloud which occurs south of the satellite and to the outside of its 
orbit when the Io System III longitude angle is between ~ 200° - 300°, was first described 
by Goldberg et al. (1984). The purpose of this paper is to analyze the observational data 
for this System III "bite-out" and to provide an explanation for this phenomenon. It will be 
shown that the "bite-out" is a long-term feature of the sodium cloud near western but not 
eastern elongation and that it may be explained by an inherent System III asymmetry in the 
Io plasma torus in the presence of an east-west electric field. 


5 


The remaining paper is organized into four sections. In section 2, observational 
data for the sodium cloud and plasma torus are discussed. In section 3, the sodium cloud 
model used in the analysis of the "bite-out of the sodium cloud is described. An inherent 
System III asymmetry based upon independent 1981 plasma torus observations is 
incorporated in the model to describe the spacetime dependent lifetime for sodium. In 
section 4, select images are analyzed using the model. Discussion of the analyses and 
conclusions drawn in the paper are given in section 5. 

2. OBSERVATIONAL DATA 

2.1 Io Sodium Cloud 

The "bite-out" of the sodium cloud near western elongation was discovered in the 
1981 Region B/C data by Goldberg et al. (1984). The time evolution of this "bite-out" was 
excellently documented in a sequence of these images acquired on May 13, 1981 which 
were included in a 16 mm movie showing the changing D 2 cloud image on the sky plane as 
Io moved about Jupiter (Goldberg et al. 1982; Goldberg 1983). The entire 1981 Region 
B/C data set consists of 263 images of the sodium cloud which were recorded 
simultaneously in both the D 2 (5890A) and D( (5896A) wavelength emission lines over 14 
nights with an image integration time of approximately 10 minutes. Three consecutive 
images were usually added to improve signal to noise. These co-added images produced a 
time— averaged picture of the sodium cloud over an Io geocentric phase angle of about 4 
and an Io System III magnetic longitude angle of about 14° and, therefore, approximate 
reasonably well instantaneous snapshots of the Io sodium cloud. The range of the 
geocentric phase angles and System III magnetic longitude angles of Io covered by the 


6 



entire 1981 Region B/C data set is summarized graphically in Figure 1 and numerically in 
Table 2. Most of the 263 images comprising this coverage have undergone some type of 
partial data processing so that the time evolution of the cloud can be followed qualitatively. 
The dots and squares in Figure 1 show the midpoint conditions for 34 images that have 
undergone complete data reduction and calibration. Observational conditions and Image ID 
Numbers for these 34 images are summarized in Table 3. A more detailed discussion of 
the 1981 Region B/C data set may be found in Paper I. 


The time evolution of the "bite-out" in the south portion of the sodium cloud near 
western elongation is illustrated in Figure 2 for a sequence of four of the completely 
processed images of Table 3 that were acquired on May 13, 1981. In the top image of 
Figure 2, for which Io has a System III magnetic longitude angles of about 168°, the bite- 
out” was not present. For System III magnetic longitude angles of about 180°, the "bite- 
out" however, began to appear in the form of a deficiency of sodium south of Io and to the 
outside of its orbit that slices across the image at an angle of positive slope, as is illustrated 
in the second image of Figure 2. The "bite-out” became more pronounced as the System 
HI longitudinal angle increases as can be seen in the third and fourth images of Figure 2. 


Examination of a number of other sodium cloud images indicates that the "bite-out" 
is a long-term feature of the sodium cloud near western elongation. The same "bite-out" 
pattern is also present in a sequence of four images from Table 3 covering the time period 
in 1981 from April 29 to June 14, which have very nearly the same geocentric phase angle 
but which have an increasing lo System III longitude angle range similar to Figure 2. 
Examination of the partially and completely processed images acquired on June 14. which 
overlap 31 days later (see Figure 1 and Table 2) the observational conditions for some of 
the images acquired in the May 13 image sequence of Figure 2. also shows that this same 


7 



behavior is exhibited. In addition, examination of partially and completely processed 
images near western elongation acquired on May 6, 1981 (see Figure 1) again reveals the 
same "bite-out" behavior of the cloud south of lo. The south "bite-out" is also consistent 
with two 1977 images and one 1979 (Voyager 1 encounter) image acquired near western 
elongation which were included in a paper by Goldberg et al. (1980) describing the 1976- 
1979 Region B/C data set even though the integration times for these earlier images were ~ 
1 - 2 hours (see Paper I). The south "bite-out" of the west cloud is also present in the very 
high quality, but unpublished sodium cloud images, acquired on 13 June 1983 and 16 
August 1984 by J. S. Morgan (1984, private communication). The south "bite -out" of the 
west cloud would therefore appear to be a long-term and stable feature of the sodium cloud. 

It is natural to ask if there is a sodium deficiency south of Io when the satellite is 
near eastern elongation and has the same System III longitude angle range for which the 
"bite-out" occurs in the west cloud. In Figure 1, east cloud images obtained on April 28, 
May 5 and May 12 (i.e., at seven day intervals and one day earlier than the west cloud 
images on April 29, May 6 and May 13) and on June 6 (i.e., two weeks before the west 
cloud images on June 14) are well suited to address this question. This comparison, which 
was undertaken in Paper 1, is summarized here. 

Select and fully processed east images of the sodium cloud from Table 3 acquired 
on April 28, May 5 and May 12 are shown in Figure 3 for an Io System III longitude angle 
in the range from 221.3 to 293.9 degrees. For April 28, the three fully processed images 
in Table 3 have a sodium cloud that is brighter north of Io (see first two images in Figure 
3). The image at an Io System III angle of 221.3 degrees is much more asymmetric north 
to south than the latter two similar images (SIP 415/43-45 and SIP 415/49-46) at a larger Io 
System III longitude angles near 300 degrees. It is, however, difficult to determine if the 


8 


first image at 221 .3 degrees has a "bite-out" feature, although it is clear that the latter image 
at 293.9 degrees does not have this feature. For May 5, the images are fairly symmetric 
north and south of lo for an lo System III angle of 204.2 degrees (i.e., image SIP 418/13- 
13 in Figure 3), are slightly enhanced north of the satellite for a longitude angle of 259.2° 
(i.e., image SIP 418/24-26 in Figure 3) , and arc symmetric north and south of the satellite 
for a longitude angle of 299.6° (image SIP 418/31-33, not shown). Examination of a 
number of additional images on May 5 that have not been fully reduced reinforces this 
behavior and indicates that the cloud is symmetric for lo System III longitude angles of 
210°, that it becomes brighter north of lo for larger angles, and that the asymmetry then 
decreases and is absent by an angle of 294°. This spatial morphology of the south deficient 
zone of sodium in the east cloud is, however, characterized by a steeper gradient in the 
brightness for the whole sodium cloud south of lo rather than a sharp and inclined 
boundaiy for the deficient zone as is present in the south "bite-out" of the west cloud. The 
June 6 image (SPI 424/10-12, not shown), which has almost identical lo phase and System 
III angles as the May 5 image (SIP 418/31-33), exhibits only a slight north brightness 
enhancement compared to no enhancement in the May 5 image. In contrast, the May 12 
images in Figure 3 do not show the south zone of deficient sodium that is indicated in the 
May 5 images and perhaps suggested by the April 28 images. Examination of all processed 
images for May 12 and additional images that have not been fully reduced shows that the 
cloud brightness is essentially symmetric north and south of the satellite for lo System III 
longitude angles ranging from ~ 210° to 302° and does not show an inclined boundary for 
the deficient zone as is present in the south "bite-out' of the west cloud one day later. 

The different north-south behavior of the east cloud images on May 5 and May 12 
for the same range of lo System III longitude angles (~ 180° - 300°) is surprising in the 
light of the consistency and repeatability of the south "bite-out signature that occur in the 


9 



west cloud on May 6 and May 13 over this same range of lo System 111 longitude angles. 
The apparent long-term stability of the "bite-out" in the west cloud south of Io and its 
complete absence or at best very slight presence in the east cloud, suggesting a temporal or 
more erratic behavior exists in the east, raise some very interesting questions about the 
properties of the plasma torus which are discussed below. 

2.2 Io Plasma Torus 

The observed south "bite-out" in the west sodium cloud occurs when Io is in a 
System III longitude range of ~ 180 to 290 degrees so that the centrifugal symmetry plane 
and hence the most dense portions of the plasma torus are south of the satellite. Because 
the excitation mechanism for sodium emission in the D-lines is solar resonance scattering, 
the deficiency of sodium brightness in the "bite-out" represents a deficiency in the sodium 
abundance. This suggests that the "bite-out" is caused by a longitudinally asymmetry in the 
"so-called” active sector of the plasma torus (Hill, Dessler and Goertz 1983) in which the 
electron-impact ionization lifetime of sodium would then have to be enhanced. Such an 
enhancement would require an increase in some combination of the electron density and 
temperature in the active sector. Any increase in electron density would increase the 
emission brightness in the plasma torus of the S’ 1- ions (6716 A and 6731 A) and the S ++ 
ions (9531 A) since their electron impact excitation mechanism is primarily sensitive to the 
electron density for the electron temperatures in the plasma torus near Io. The brightness 
of the sodium emission lines and ion emission lines would then be naturally anticorrelated. 

A physically corresponding System III brightness enhancement in these plasma 
torus S + emission lines, which is anticorrelated with the "bite-out" sodium brightness, has 
indeed been observed in the active sector from ground-based telescopes since 1976. This 


enhancement has also been observed on later dates in the S ++ ion emission line. Pertinent 
S + spectral data for this brightness asymmetry have been acquired from ground-based 
observations in 1976 (Trafton 1980; Trauger, Miinch and Roesler 1980), in 1977 and 1978 
(Trafton 1980; Pilcher and Morgan 1980), in 1979 (Trafton 1980; Morgan and Pilcher 
1982) and in 1981 (Morgan 1983, 1985 a,b). Image data for S + and/or S ++ emissions 
exhibiting this System III brightness asymmetry were also obtained from ground-based 
observations in 1980 (Roesler et al. 1982), in 1981 (Trauger 1984; Pilcher and Morgan 
1985; Oliversen, Scherb and Roesler 1991), in 1982 (Trauger 1984), in 1983 (Trauger 
1984; Pilcher and Morgan 1985), in 1984 (J. S. Morgan 1984, private communication; J. 
T. Trauger 1985, private communication), in 1985 (J. T, Trauger 1985, private 
communication), in the 1988-89 apparition (F. Scherb 1990, private communication; J. T. 
Trauger 1991, private communication), in the 1989-90 and 1990-91 apparitions (J. T. 
Trauger 1991, private communications; N. Thomas 1991, private communication), and in 
1992 (Rauer et al. 1993). Most of the image data acquired in the last five to ten years have, 
however, not been analyzed or published to date. There are indeed a number of interesting 
and outstanding questions about the time variability of the radial and longitudinal structure 
of the plasma torus for which these observations are well suited and, consequently, as we 
shall see, for which these observations have corresponding time varying counterparts in the 
sodium "bite -out". 

A comparison of these plasma observations for the System III asymmetry indicates 
that generally two separate longitudinally asymmetric brightness components appear to 
exist in the torus. The first component is centered at ~ 170 to 200 degrees, with an angular 
width of ~ 90 degrees or more, and is a permanent feature of the torus. When compaxed to 
the normal brightness of the plasma torus at the diametrically opposite magnetic longitude 
angle, the first component is typically 2-5 times brighter (and at extremes. 10 times 



brighter) with significant brightness changes occurring on a time scale of months to years. 
The second component is centered at -280-290 degrees, has a comparable angular width to 
the first component, and is sometimes present and sometimes absence in the torus. The 
brightness of the second component may be comparable to or less than brightness of the 
first component. 

The stability of the "bite-out" in the west sodium cloud and the complete absence or 
at best very slight and erratic behavior of deficient sodium south of Io in the east cloud 
suggest that east-west differences in the properties of the plasma torus, in addition to 
longitudinal asymmetries, also play an important role in shaping the Region B/C data 
summarized in Figure 1. The sense of this behavior is that a significant enhancement of the 
sodium lifetime in the active sector is always present near Io when the satellite is west of 
Jupiter (i.e., near western elongation at 270° Io geocentric phase angle) and never or at best 
only slightly present near Io when the satellite is east of Jupiter (i.e., near eastern 
elongation at 90° Io geocentric phase angle). As discussed in Paper I, the fully reduced 
sodium cloud images of Table 3 do, in fact, exhibit an east-west intensity asymmetry for 
the sodium brightness in a circular annulus centered on Io with an inner radius of 19,000 
km and an outer radius of 21,000 km. The average D 2 brightness in the annulus for the 
sodium cloud images east of Jupiter was - 40% larger than the brightness for the sodium 
cloud images west of Jupiter. This behavior is consistent with the existence of a small east- 
west electric field in the magnetosphere (Barbosa and Kivelson 1983; Ip and Goertz 1983; 
Goertz and Ip 1984) which for a constant electric field simply displaces the plasma torus 
toward the east and produces a slightly hotter and more dense plasma at Io's orbit west of 
Jupiter. 



The presence of such an east-west electric field was shown earlier to provide a 
consistent explanation (Smyth and Combi 1987) for the anticorrelation of the larger sodium 
D-line emission intensities east of Jupiter as measured even closer to Io in the discovery 
data of Bergstralh et al. (1975, 1977) and the larger intensities west of Jupiter of both the 
plasma torus S+ optical emission line (6731 A) measured from a ground-based telescope 
(Morgan 1985 a,b) and the S ++ extreme ultraviolet (EUV) emission line measured from the 
Voyager 1 spacecraft (Sandel and Broadfoot 1982). In the past several years, the 
documentation of this east-west intensity asymmetry has been expanded to include a 
number of additional emission lines and species as summarized in Table 4. Note in Table 4 
that with the exception of neutral sodium, which is excited by solar resonance scattering, 
the west to east brightness ratio for all other emission lines is greater than unity as expected 
for electron impact excitation and the presence of an east-west electric field. More recent 
analysis (Sandel and Dessler 1988; Dessler and Sandel 1992) of the EUV plasma toms data 
acquired by the Voyager 1 and 2 spacecrafts suggests that the east-west electric field is, on 
the average, larger west of Jupiter than east of Jupiter and may also be dependent in 
magnitude on the System [II longitude angle. 

For our immediate purpose of understanding the "bite-out" in the Region B/C image 
data of Figure 1, the longitudinal structure and presence of an east-west electric field in the 
plasma torus is fortunately reasonably well documented in the winter to spring of 1981 by 
ground-based observations summarized in Table 5. Of these plasma toms observations, 
the extension amount of data acquired by Morgan (1983) and his analysis and modeling of 
these data (Morgan 1985 a,b) provide the most relevant information base for the bulk of the 
sodium observations in Figure 1 which occur in mid to late spring of 1981. An east-west 
asymmetry in the radial emission profiles of the optical SII (6716A, 67 3 1 A) and Oil 
(3726A) lines was documented in the observations of Morgan with higher intensities 



measured on the west side of Jupiter. This east-west asymmetry in both the shape and 
intensity of the SII (6731 A) spectral data was independently verified (see Morgan 1985b) 
by 1981 image data of the SII (6731 A) emission acquired by Oliversen (1983; Oliversen et 
al. 1991). An adequate explanation of the east- west asymmetry of the ion data in the 
modeling analysis of Morgan (1985b) was obtained by a convective motion caused by an 
east-west electric field. The magnitude of the east-west electric field required to give a 
reasonable fit to the optical ion data was similar to that required to explain the east-west 
Voyager EUV (685A) east-west intensity asymmetry and the anticorrelated east-west 
intensity asymmetry of the D-line emissions from the sodium cloud (see Smyth and Combi 
1987). The System III longitudinal variation of the SII (6716A, 6731 A) emissions was 
time dependent in the measurements of Morgan (1983; 1985 a,b). It changed from an 
ordered single peak ~ 2.5 times brighter in a broad region centered near 180° System III 
longitude in run 1 (Feb 14-17) and run 2 (Mar 20-23) to a double-peaked structure with 
peaks located near 180° and 290° in run 3 (April 21 - 24) and run 4 (May 2-4). In Figure 4, 
the longitudinal asymmetry for run 4 is shown and reveals that near 180° and 300° system 
III longitude the two intensity peaks west of Jupiter were brighter and vary by about a 
factor of 2 while the intensity peaks east of Jupiter were much dimmer and had a smaller 
amplitude variation. This double-peaked structure west of Jupiter was also independently 
verified in the April 1981 SII image data of Pilcher et al. (1985). In the modeling analysis 
of Morgan (1985b), intrinsic longitudinal intensity changes of about a factor of two were 
required to match the observed System III longitudinal asymmetries. The observations in 
the spring of 1981 of the System III longitudinal and east-west asymmetries in both the 
plasma torus and sodium cloud data suggest that these signatures arc directly related. This 
relationship will be quantified and included in the sodium cloud model discussed in the next 


section. 



3. MODEL DESCRIPTION 


To study the time-variable signatures of the sodium cloud discussed in section 2, it is 
necessary to have a model that contains, to a reasonable level of accuracy, a description of the 
atom dynamics, the atom sink processes, the atom excitation mechanism, and the atom 
source velocity distribution at the satellite exobase. An improved version of the sodium 
cloud model of Smyth and Combi (1988b) using the framework of the more general model 
description in Smyth and Combi (1988a) is adopted here for this purpose. The orbital 
dynamics of sodium atoms in the cloud model are determined by the gravitational fields of 
both Io and Jupiter and by the acceleration of solar radiation pressure produced as atoms 
undergo resonance scattering in the Dj and D 2 emission lines. All three factors were 
included in the earlier 1988 version of the sodium cloud model. The sink for sodium atoms 
in the circumplanetary magnetosphere near Io's orbit is determined by their spacetime 
dependent interactions with the plasma torus, which are dominated by electron impact 
ionization. In the earlier model, the sodium lifetime description was for electron impact 
ionization and was based upon a tilted and offset magnetic dipole field in the presence of a 
constant east-west electric field (~2.8 mV nr 1 in Io’s frame) for a heavy ion magnetospheric 
plasma with Voyager 1 encounter conditions. This plasma torus description, although 
spatially dependent upon three variables (i.e., radial L distance, magnetic latitude, and east- 
west angular location about Jupiter), was otherwise inherendy longitudinally symmetric. The 
model of Smyth and Combi (1988b) has been improved here to include an inherently 
asymmetric plasma torus description. The description of the inherently asymmetric plasma 
torus is discussed below. The atom excitation mechanism for sodium, which is solar 
resonance scattering and depends upon the instantaneous atom-sun radial-velocity component 
because of the Di and D 2 Fraunhoffer features in the solar spectrum, was implemented in the 
earlier model. The sodium source at Io's exobase has recently been shown by Smyth and 


1 5 

<2 2 - 



Combi (1993) to be well represented by a modified sputtering flux speed distribution 
function which has its maximum value at -0.5 km sec' 1 (i.e., a sub-escape speed from Io's 
exobase). For atoms in the sodium cloud, of which all have already escaped from Io, 
previous modeling efforts (Smyth and Combi 1988b) have shown, however, that the basic 
nature of the forward cloud and also the escaped sodium within about one half Jupiter radii 
north and south of Io can be reasonably well reproduced by assuming a monoenergetic 2.6 
km sec'* source at a nominal Io's exobase of 2600 km, where the escape speed is -2.0 km 
sec *. This source is quite adequate for our present purposes of investigating the "bite-out" 
feature but will understandably (see Pilcher et al. 1984) not reproduce the trailing-cloud 
directional feature due to its absences of higher-velocity components ( i.e., -10-20 km sec *) 
or the brightness gradient in the more distance portions of the forward cloud (Smyth and 

Combi 1993). 

As discussed in section 2, an examination of the 1981 plasma torus data (Morgan 
1985 a,b; Pilcher and Morgan 1985; Oliversen et al. 1991) indicated S + and S ++ emission 
enhancements in the torus near Io's orbit in the System III longitudinal region of 180-300 
degrees. The brightness enhancements appear to be attributed primarily to an increase in the 
electron density of the so-called S + "ribbon feature (Trauger 1984) or field-aligned feature 
(Pilcher, Fertel and Morgan; 1985; Pilcher and Morgan 1985; Morgan and Pilcher 1982) that 
is located inside of Io's L-shell and also to a S ++ plasma extension of this feature beyond 
Io's orbit. In 1981 during the time of the sodium cloud measurements, the longitudinal 

asymmetry in the torus had two components, the first one centered at a System III longitude 
angle A ( ' J -180-200 degrees and the second one centered at A^-280-290 degrees. The 

electron impact ionization lifetime of sodium will thus have an inherent longitudinal 

dependence. 



The explicit functional dependence of the sodium lifetime, T, on the System III 
longitude angle, X m , and the radial L distance in the plasma torus coordinate frame of 

Smyth and Combi (1988b, see Appendix C) is given as follows: 


where 


r 


(a 

2 " 


l + 


_Jfl HL 


R(L)- 





J 



for L - L 0 < 0 
for L - L 0 > 0 


( 1 ) 


( 2 ) 


and was suggested from the modeling of plasma torus spectra and image data by J. S. 
Morgan (1987, private communication). Here T 0 = v~' is the lifetime description for sodium 
determined by case C of Smyth and Combi (1988b). The lifetime T 0 includes no inherent 
longitudinal asymmetry in the plasma torus. For the sodium lifetime T , the parameters OC- 

determine the relative longitudinal strength of the two asymmetric plasma torus components 
while the parameters <3 ) determine their angular widths. The radial dependence R(L) of the 

asymmetry is centered at L = L 0 and has an inner scale length o ^ that differs from its outer 
scale length During the 1981 JPL observations of the sodium cloud, best estimated 

values for these parameters are given as follows: 


A (l) = 180* 


A (2) = 280* L 0 = 5.65 

tu 


I 7 



( 3 ) 


<*1=1 «2 =1 <*«n«r= 0 - ,2 5 

cr, = 45° cr 2 = 45' <^ = 1.0 

where L, L 0 , <t Uuiu t and a Mla are understood to be measured in units of Jupiter radii. 

Using the parameters values given by (3), the electron density is enhanced in the 
active longitudinal region by a factor of two (as in the best plasma torus model of Morgan 
1985b) over the Voyager 1 value it has at the diametrically opposite longitudinal angle in the 
plasma torus. At Io’s location, the sodium lifetime as a function of the satellite geocentric 
phase angle and System III longitude angle is correspondingly reduced over much of this 
angular domain by the same factor as is illustrated in Figures 5a and 5b, respectively, by 
comparison of the sodium lifetimes T 0 and r in the plasma torus. The plasma torus 

coordinate frame and its radial coordinate L are defined so as to render the plasma torus 
description in a preferable frame where, for a given east-west angular location about Jupiter 
(e.g. eastern elongation), the two dimensional properties of the plasma torus (and hence also 
the sodium lifetime) are independent of the System III longitudinal angle, in the absence of an 
actual inherent System III asymmetry in the torus. This is illustrated in Figure 6, where the 
sodium lifetime x 0 at eastern and western elongations are different because of east-west 

asymmetries in the plasma torus but are independent of the System III longitude angle. In 
contrast for r, the sodium lifetimes at eastern and western elongations, although also 
different because of east-west asymmetries in the torus, are now different for each value of 
the System III longitude angle. 


I 8 


4. ANALYSIS OF OBSERVATIONS 


To understand the importance of the plasma torus sink in the formation and time 
evolution of the "bite-out" feature, model calculations are first undertaken for the sequence of 
four D 2 emission images in Figure 2 observed on May 13 1981. Model calculations are 
shown in Figure 7 based upon the inherently asymmetric plasma torus sink illustrated in 
Figure 5b and in Figure 8 based upon the inherently symmetric plasma torus sink illustrated 
in Figures 5a and 6. The model calculations in Figure 8 provide a base-line case from which 
to identify in Figure 7 the effects of the inherently asymmetric plasma torus on the evolution 
of the "bite-out" feature. In Figures 7 and 8, the source is that for an isotropic and uniform 
monoenergetic (2.6 km sec' 1 ) flux of sodium atoms ejected radially from Io's exobase, 
where the atom flux is 2 x 10 8 cm 2 sec’ 1 referenced to Io’s surface area in all cases. 
Comparison of Figures 7 and 2 indicates that both the basic character and the Io System III 
longitude evolution of the south "bite-out" feature are reasonably well simulate by the model. 
Indeed, this absence of the "bite-out" and its time evolution in Figure 8 shows that the "bite- 
out" is produced solely by the dependence of the sodium sink on the inherently System III 
asymmetry of the plasma toms. The excess length of the sodium cloud in model calculation 
compared to the observational data occurs because of the assumed constant and isotropic flux 
of sodium ejected from Io's exobase. A more realistic treatment of the source would provide 
a better match to the obsen'ational data but would not alter the conclusions determined for the 
time evolution of the "bite -out" feature. 

To assess the impact of the east-west and inherently System III longitude 
asymmetries on the sodium cloud for Io near eastern elongation and the possible formation of 
a "bite-out" feature, two model calculations for the sodium cloud at an Io geocentric phase 
angle of 90.5* and an Io System III longitude angle of 221.3* are presented in Figure 9. In 



the model calculations, the inherently asymmetric plasma torus is included in the sodium 
sink. For the monoenergetic 2.6 km sec* 1 sodium source, both an isotropic (i.e., spherically 
symmetric on Io's exobase) and a more restrictive band source area (i.e., deficient in sodium 
ejected from the leading and trailing apex areas of Io's exobase) are assumed at a nominal 
exobase radius of 2600 km, where the escape speed is ~2.0 km sec 1 . The same flux is 
adopted as in Figure 7. The Io geocentric phase and Io System III longitude angles are those 
appropriate for the first April 28 image in Figure 3, and differ only appreciably for the latter 
angle from the second May 5 image in Figure 3 and the May 12 image at an Io geocentric 
phase angle of 91.2* and an Io System III longitude angle of 301.8*, not shown in Figure 3. 
The model images show no evidence of the "bite-out" feature in spite of the fact that Io was 
in the active sector of the plasma torus where the "bite-out" occurred when Io was near 
western elongation. Comparison of the two model images in Figure 9 illustrates how the 
band source reduces both the length and brightness of the forward cloud substantially from 
the symmetric source, but additional adjustments are required in the assumed constant source 
strength to match the brightness contours more exactly. 


5. DISCUSSION AND CONCLUSIONS 

The model images in Figure 9 show an east cloud that exhibits little or no asymmetric 
distribution of sodium north or south of Io when the satellite is in the active sector of the 
plasma torus, in contrast to the south "bite-out" feature in the model images in the west cloud 
in Figure 7. Additional model images (not shown) for May 5 east image (SIP 418/24-26) 
and the May 12 east image (SIP 420/30-32) that cover the Io System III longitude angle 
range of the asymmetric sink substantiate the results of Figure 9. The reason there is no 
deficient sodium signature south of Io in the east cloud model is because the east-west electric 


20 


field has displaced Io radially inward from the location of the asymmetric sink and thereby 
has literally removed the sodium from the spatial region of enhanced ionization. This is 
readily apparent by studying Figure 6 where the sodium lifetime in the plasma torus at eastern 
and western elongations is shown in relationship to lo's motion (dark oval) during one Io 
System III longitude period. The Io System III longitude angle on the oval in the centrifugal 
equator plane at the smaller radial displacement is 1 10* and increases as the satellite moves 
upward along the oval and reaches 200* at the top of the oval. The longitudinally asymmetric 
plasma torus sink for sodium is centered on the centrifugal plane at a radial distance of 5.65 
planetary radii near the middle of the oval at eastern elongation (at the location of the black 
dot) and somewhat left of the oval at western elongation (at the location of the black dot). 
The small inner radial scale height G^ of the inherent System III enhancement causes the 

enhancement to decay so rapidly that it is unable to reach appreciably lo’s location in the 
1 10*- 290* sector in the east, while the outer radial scale height G^ of the inherent System 

III enhancement allows the enhancement to be felt at almost full force at los location in the 
1 10*- 290* sector in the west. 

The apparent time variability or erratic behavior of the east cloud and the stability of 
the west cloud regarding the zone of deficient sodium south of Io are therefore likely to be 
related to a difference (or time variability) in the strength of the east-west electric field east 
and west of Jupiter. It is clear from Figure 6 that the west cloud is in a rather stable 
relationship with the location of the inherently System III plasma torus enhancement for small 
modulations of the electric field strength west of Jupiter. This is not. however, true for the 
east cloud because of the very small scale height <J mna and the rather close radial proximity 

of lo’s orbit to the System III enhancement in the plasma torus. Small changes in the east- 
west electric field east and west of Jupiter can therefore modulate the radial inward 
displacement of lo relative to the radial location of longitudinally asymmetric plasma torus 



sink and hence modulate the east cloud sodium population south of the satellite accordingly. 
More recent studies indeed suggest that the electric field may be time variable (Sandel 1985; 
Sandel and Dessler 1988; Dessler and Sandel 1992) and therefore provide a reasonably 
consistent explanation for the absence or erratic nature of the "bite-out" in the east cloud and 
the stable presence of the "bite-out" in the west cloud. 

Additional studies of the 1981 sodium cloud observations (Region B/C) need to be 
undertaken by extending and refining the analysis initiated here. The complete set of 1981 
calibrated images should be carefully modeled in order to determine the sodium source 
strength and its time variability and dependence on the Io geocentric phase and System III 
longitude. The analysis of the 1981 sodium cloud images (Region B/C) may also be readily 
applied to the 1976-1979 sodium cloud images (Region B/C). This would allow us to extend 
the time line and test the stability or time variability of various morphological features of the 
cloud and their corresponding physical properties and structures in the plasma torus. The 
1976- 1979 Region B/C data set includes two images on the Voyager 1 encounter day as well 
as a number of images around this encounter time. Studies of these images would be most 
valuable in linking the whole sodium cloud data set to the Voyager 1 encounter conditions 
determined by the spacecraft and providing a more solid base of understanding for the 
upcoming tours of the circumplanetary environment of Jupiter by the Galileo spacecraft. 


22 



ACKNOWLEDGMENTS 


We are grateful to J. S. Morgan for many helpful discussions and for sharing with 
us his Io sodium cloud and plasma torus image data prior to publication. We also wish to 
thank M. L. Marconi for providing computational support for the model calculations. This 
research was supported by the Planetary Atmospheres Program of the National 
Aeronautical and Space Administration under contracts NASW-3949, NASW-4416, and 
NASW-447 1 to Atmospheric and Environmental Research, Inc. 


23 



REFERENCES 

B allester, G. E. 1989. Ultraviolet observations of the atmosphere of lo and the plasma torus, 
Ph.D. thesis, The John Hopkins University. 

Barbosa, D. D. and M. G. Kivelson 1983. Dawn-dusk electric field asymmetry of the Io plasma 
torus, Geophys. Res. Lett. 10 , 210-213. 

Bergstralh, J. T., D. L. Matson, and T. V. Johnson 1975. Sodium D-line emission from Io: 

synoptic observations from table mountain observatory. Ap. J. Lett. 195 , L-131-L135. 

Bergstralh, J.T., J. W. Young, D. L. Matson, and T. V. Johnson 1977. Sodium D-line 
emission from Io: A second year of synoptic observation from table mountain 
observatory. Ap. J. Lett. 211, L51-L55. 

Brown, R. A. 1974. Optical line emission from Io. In Exploration of the Planetary System: 
Proceedings IAU Symp. No. 65 (A. Woszczyk, and C. Iwaniszewska, Eds.), pp. 527- 
531. D Reidel Publ. Co., Dordrecht. 

Brown, R. A. and D. E. Shemansky 1982. On the nature of S II emission from Jupiter's hot 
plasma torus. Ap. J . 263, 433-442. 

Carlson, R. W., D. L. Matson, and T. V. Johnson 1975. Electron impact ionization of Io's 
sodium emission cloud. GRL 2, 469-472. 

Dessler, A.J. and B.R. Sandel 1992. System III variations in apparent distance of the Io plasma 
torus from Jupiter. Geophys. Res. Lett. 19, 2099-2103. 

Flynn, B., M. Mendillo, and J. Baumgardner 1992. Observations and modeling of the jovian 
remote neutral sodium emission. Icarus 99, 1 15-130. 

Goertz, C.K. and W.-H. Ip 1984. A dawn-to-dusk electric field in the Jovian magnetosphere. 
Planet. Space. Sci. 32, 179-185. 

Goldberg, B. A., R. W. Carlson, D. L. Matson, and T. V.Johnson 1978. A new asymmetry 
in Io’s sodium cloud. BAAS 10, 579. 


Goldberg, B. A., Y. Mekler, R. W. Carlson, T. V. Johnson, and D. L. Matson 1980. Io’s 
sodium emission cloud and the Voyager 1 encounter. Icarus 44, 305-317. 

Goldberg, B. A., R. W. Carlson, G. W. Garneau, T. V. Johnson, S. K. LaVoie, J. J. Lorre and 
D. L. Matson 1982. Dynamics of the lo sodium cloud. BAAS 14, 762. 

Goldberg, B. A. 1983. Dynamics of the lo sodium cloud, 16 mm movie produced for display at 
the National Air and Space Museum, Smithsonian Institute, Washington, D. C. 

Goldberg, B. A., G. W. Garneau, and S. K. LaVoie 1984. Io’s sodium cloud. Science 226, 
512-516. 

Goldberg, B. A. and W. H. Smyth 1993. The JPL table mountain lo sodium cloud data set, 
companion paper. 

Hill, T. W., A. J. Dessler and C. K. Goertz 1983. Magnetospheric models. In Physics of the 
Jovian Magnetosphere , (A. J. Dessler, Ed.), pp 106-156, Cambridge Press, N. Y. 

Ip, W.-H. and C. K. Goertz 1983. An interpretation of the dawn-dusk asymmetry of UV 
emission from the lo plasma torus. Nature 302, 232-233. 

Matson, D. L. , B. A. Goldberg, T. V. Johnson, and R. W. Carlson 1978. Images of Io’s 
sodium cloud. Science 199, 531-533. 

McGrath, M.A. and R. E. Johnson 1987. Magnetospheric plasma sputtering of Io's 
atmosphere. Icarus 69, 519-531. 

McGrath, M. A., H. W. Moos, D. F. Strobel and G. E. Ballester 1990. The lo plasma torus 
dawn-dusk brightness asymmetry. Paper presented at the Magnetospheres of the Outer 
Planets Fred Scarf Memorial Symposium, Annapolis, Maryland, August 20-24. 

Mendillo, M., J. Baumgardner, B. Flynn, and W. J. Hughes 1990. The extended sodium 
nebula of Jupiter. Nature 348, 312-314. 

Morgan, J. S. 1983. Low resolution spectroscopy of the lo torus, Ph.D. thesis. Dept of 
Astronomy, Univ. of Hawaii. 

Morgan, J.S. 1985a. Temporal and spatial variations in the lo torus. Icarus 62, 389-414. 


2 5 



Morgan, J.S. 1985b. Models of the lo torus, Icarus 63 , 243-265. 

Morgan, J. S. and C. B. Pilcher 1982. Plasma characteristics of the lo torus. Ap. J. 253 , 406- 
421. 

Murcray, F. J. 1978. Observations of Io’s sodium cloud. Ph.D. Thesis, Dept, of Physics, 
Harvard University, Cambridge, Massachusetts. 

Murcray, F. J., and R. Goody 1978. Pictures of the sodium cloud. Ap. J. 226, 327-335. 
Oliversen, R. J. 1983. The lo plasma torus: its structure and sulfur emission spectra, Ph. D. 

thesis. Dept of Physics, Univ. of Wisconsin-Madison. 

Oliversen, R. J., F. Scherb and F. L. Roesler 1991. The sulfur torus in 1981. Icarus 93 , 53-62. 
Pilcher, C. B. 1980. Transient sodium ejection from lo. BAAS 12, 675. 

Pilcher, C. B. and W. V. Schempp 1979. Jovian sodium emission from region C2. Icarus 38, 

1 - 11 . 

Pilcher, C. B. and J. S. Morgan 1980. The distribution of fSII] emission about Jupiter. Ap. J. 
238, 375-380. 

Pilcher, C. B. and J. S. Morgan 1985. Magnetic longitude variations in the lo torus. Adv. Space 
Res. 5, 337-345. 

Pilcher, C.B., W. H. Smyth, M. R. Combi, and J. H. Fertel 1984. Io's sodium directional 
features: Evidence for a magnetospheric-wind-driven gas escape mechanism. Ap. J. 

287, 427-444. 

Pilcher, C. B., J. H. Fertel and J. S. Morgan 1985. ( S II J images of the torus. Ap. J . 291, 377- 
393. 

Rauer, H., T. Bonev, K. Jockers, and N. Thomas 1993. Low resolution spectra of the lo plasma 
torus two days after ULYSSES encounter. Preprint. 


26 


Roesler, F. L., R. J. Oliversen, F. Scherb, J. Lattis, T. B. Williams, D. G. York, E. G. Jenkins, 
J. L. Lowrance, P. Zucchino and D. Long 1982. Fabry-Perot/CCD observations of [SIN] 
and [SII] emissions from the Jupiter plasma torus. Ap. J. 259 900-907. 

Sandel, B. R. and A. L. Broadfoot 1982. Io's hot torus- a synoptic view from Voyager, J. 
Geophys. Res. 87, 212-218. 

Sandel, B.R. and A.J. Dessler 1988. Dual periodicity of the Jovian magnetosphere. J. Geophys. 
Res. 93, 5487-5504. 

Scherb, F. and W. H. Smyth 1993. Variability of [OI] 6300 A emission near lo. J. Geophys. 

Res. 98, 18729-18736. 

Schneider, N. M. 1988. Sodium in Io's extended atmosphere. Ph.D. Thesis, Department of 
Planetary Sciences, University of Arizona. 

Schneider, N. M., D. M. Hunten, W. K. Wells, A. B. Schultz and U. Fink 1991a. The structure 
of Io's corona, Ap.J. 368, 298-315. 

Schneider, N. M., J. T. Trauger, J. K. Wilson, D. I. Brown, R. W. Evans, and D. E. 

Shemansky 1991b. Molecular origin of Io's fast sodium. Science 253, 1394-1397. 
Sieveka, E.M. and R. E. Johnson 1985. Nonisotropic coronal atmosphere on Io. 

J. Geophys. Res. 90, 5327-5331. 

Smyth, W.H. 1979. Io’s sodium cloud: Explanation of the east-west asymmetries. Ap. J. 234, 

1 148-1153. 

Smyth, W. H. 1983. Io’s sodium cloud: Explanation of the east-west asymmetries. II. Ap. J. 

264, 708-725. 

Smyth, W.H., and M. R. Combi 1987. Correlating east-west asymmetries in the jovian 
magnetosphere and the lo sodium cloud. Geophys. Res. Lett. 14, 973-976. 

Smyth, W.H. and M.R. Combi 1988a. A general model for Io's neutral gas cloud. 

I. Mathematical description. Ap. J. Supp. 66, 397-41 1. 

Smyth, W.H., and M. R. Combi 1988b. A general model for Io's neutral gas cloud. 


27 



II. Application to the sodium cloud. Ap. J. 328, 888-918. 

Smyth, W.H. and M. R. Combi 1991. The sodium zenocorona. JGR 96, 2271 1-22727. 

Smyth, W.H. and M. R. Combi 1993. Io's sodium corona and spatially extended cloud: a 
consistent flux speed distribution. Icarus, submitted. 

Smyth, W. H. and B. A. Goldberg 1993. Correlating system III longitudinal asymmetries in 
the jovian magnetosphere and the Io sodium cloud. Paper in preparation. 

Smyth, W. H., and M. B. McElroy 1978. Io’s sodium cloud: Comparison of models and two- 
dimensional images. ApJ. 226, 336-346. 

Trafton, L. 1980. The Jovian SII torus: its longitudinal asymmetry. Icarus 42, 1 1 1-124. 

Trafton, L., T. Parkinson, and W. Macy, Jr. 1974. The spatial extent of sodium emission 
around Io. Ap. J. 190, L85-L89. 

Trafton, L., and W. Macy, Jr. 1978. On the distribution of sodium in the vicinity of Io. Icarus 
33, 322-335. 

Trauger, J. T. 1984. The Jovian nebula: A post-Voyager prespective. Science 226, 337-341. 

Trauger, J. T., G. Munch and F. L. Roesler 1980. A study of the Jovian [SII) nebula at high 
spectral resolution. Ap.J. 236, 1035-1042. 

Wehinger, P. A., S. Wyckoff, and A. Frohlich 1976. Mapping of the sodium emission 
associated with Io and Jupiter. Icarus 27, 425-428. 

Wilson, J. K. and N. M. Schneider 1993. Io's fast sodium: Implications for molecular and atomic 
escape, preprint. 


28 


43 

u 


c 

o 

cd 

c 

D 

X 

XJ 


M 

C 

C/2 

s 

o 

H 

cd 

E 

m 

cd 

Oh 


D 

O 

CO 

u 

cd 

CX 

6 

cd 

E 

C/3 

Cd 

CU 


J 

3 ci 

9 E 

g 53 

1) 

H 3 *5 

05 o 


DO 

.£ 00 
5b c 
c c 

cd > 

x <u 

U> 

« O s» 

,-p'E ° 

S ° 


E 

£ 


o 

& 


E 

o 

*> *> 

O cu 


c 


CO 


a 

in O 

S ’§ 

£••6 

cd 

cd o5 

E 5 a 

c /2 cd 

i2 o 

CL 00 


« 

£ 


cd 

- E 
E <q 


■g’ 


DO 

CO c 

"8 1 
03 X 

cx u 

CO 

x ^ p 
&o-o g 

£ 53 H 


* ss 


53 a 

UJ 5 
"O E -1 
c 

Cd c 0/ 

afi’fi 

►h cd u 

E E | 

13 ~ E 
*-* v2 tr 

C /2 di ^ 
>^> 2 
CO > < 


■g 


=3 

X) 

■fi 


43 

■ss J 

43 * C/3 

ClJS' C 
oo -a c 

cd ■— t 

-c -a ^ 

OO cd X 

a o 

Xcoco 


•a'S 

c a 

°-s § 

oQo 

43 'C/3 

CX^ c 
CO —3 c 
cd *3 

x -a £5 

DO cd X 

x cx o 

3 -. co oo 


E 

43 


cx 

a 

•s 

s 


E 

u 

E 


C/2 

43 

3 

CO 

Di 


43 

■g 


VO 

oo 

ON 


_ C 
CO ^ 

S g 
^Jc 

— O 

• ** *—J 

ON _ 


On 


cd 

x: -* 

X 43 

> 

£.2<§ 

co oo ^ 


r- 

oo 

On ^ 
— ' CO 
W ON 
a On 

o — « 

C/3 N — " 

C .- 
X X 
O £ 

|2 o 

cd ~o 

*£ S 

cd 


‘a 

o 


x 

E 

oo 


oo 

r- 

ON 


£ <*> 

- ^ Jd 


43 

C 

o 


X) 

oo 

oo 

On 


X 

E 

o 


>> 

o 

^ c> ^ _ 

( 1 _oo^ 
«3<3' , 0 

o*£ -S-S 
^ 2 >, >, >, 
& & 6 6 E 

O ^ CO OO GO 


43 

> 

o 

•8 

53 

43 

E 

cd 


X 

oo 

oo 

On 


r- 

oo 

ON 


CO 

OO ~Q 

2 £ 
.. o 

ON (J 

r- _ 
2 c 

W cd 

-£ -S 
>, >> 

E E 

co oo 


OO 

On 


cd 

43 

v- 

43 

X 

a 


43 

CX 

cd 

cu 

m 

IE 

H 


co 


43 

*43 

C 

X 

o 

co 

*o 

c 

cd 

c 

O 


On 

On 


is 

C O 
C On 


O 

U 

T3 

s 


C 

E^ 

CO Xu 


o 

43 


OO 

ON 


ON 

rsi 


C/3 

O 

•a 


o 

c: 

3 


3 

d 

OO 


3 

3 

— Cd 

H 


% 


a 


y3 

C 

tJ 

U» 

I 

X 

U 

*o 

u 

l 

X 

5 


r- 

r- 

ON 


VO 

r- 

ON 


/— v cd 

Tf 4-» 

r- 43 

2 £ 
v — • Id 

C ^ 

£ DO 
O *- 
43 

CQ CQ 


ON 

ON 


13 

43 

u> 

43 

*2 

’5 

C 

X 

CJ 

co 



43 

> 

o 

X 

cd 

C/2 

cd 

43 

E 


u 

’fi 

u 

E 


fi 

X °°^ 

•S r- 

>»On 

EC 

CO ; 

X3 cd 
C w 
Cd 43 

DO DO 
*-, u. 
43 43 
X X 

32 2 
o o 
OO 


CO 

On 


CO 

On 

On 


oc 

ON 


cd 

r: do do 
4J u L 
u- 43 43 
43 X X 

x *o *o 
32 c o 
£00 


DO DO 
u. u. 
43 43 
X X 

2 2 
o o 
OO 


43 

X 


oc 

ON 

>o 

ou *— > 
On >> 

J-3 


■ — 1 

£ 

C E 

CU 

OO 


cS 

X °° 

. #. 

oo 


“O 

cd "O 

O 

ON 

, . 

C 

^ C 

oo 


43 

C3 

43 cd 

Ov 



ON 

On 


43 
2 
43 43 
C C 
O XZ X 
2 u u 
Cu CO co 


, 43 
"O 


43 


ON 

r- 

5? 2 

^2 g-o 

2^E§ 

' — ^ u 43 
* Cd X 

• 

43 X X) 

^ 5 c 

43 cd cd 
DO c 
£ O 43 

X cc - c - 


cd 

s 

o 


"O 

c 

43 cd 43 



6* 

E 

*o 

a 

o 

TD 

3 

o 

U 

•rl 

oo 

< 

C/3 

A 3 


m 

t 

2 

U- 

O 



cd 

c 

o 

u 

o 

8 


e a 

>i p 

U 

*o 

^ 2 
> w> 

NJL/ 

£ 

1 

1^1 

E 

43 

IX 

g' 53 

D «j 





< E 
&!<§ 

{3 

l 

o 

C ^ 
o < 

^ s 

C cd 

cn 

c3 

T3 

« < 
<C oo 

2 

c 

o 

X 

u c 

E -a 
E 3 

c 

o 

X 

cd 

DO 


«_ 

O 

cd 

Z3 

cd 

c 

e 

o?-1 
^ « 

StScg- 

XU 

43 

E 

c 

E 

° £ 
o 0h 

CJ 43 
C X 

■§1 

a 

o 

U 

J2 o 

s a 

E -c 
2 cu 

2 

s 

DO 

C 

-u 

- E 

gl 

C 

j2 

UJ 

E 

43 

13 

i 

X 

43 

C 

6 

43 

c 

o 

O 

o 



v 

a. co 

l 

i2 


1 

H 

^ oo 
co ±: 

•— * 
C/2 

43 

o 

uu 

DO 

cd 

s 


<N 


CO 




OHiQtNAL PAG€ f 6 
Of WO* QUALITY 



Table 2 

1981 Region B/C Images: Observing Chronology 


O 




r- On m 


i M >n oo oo ^ O 


^m *7 . — - (\j ^n| co ro * — ' <o co *~m co *n 

co (N co(N co co cm 


•— 1 t CM 
CO CO 


io^^^ONO\^r^a\a\NOoo--;0\-; 

NOONoo^^rir^^r^cor^NOvS^cor- 

cooon’— • cm <— ihHoa^ONVo^o^'-'Os 

*— i cM CO *— 1 CO CM *— < CO <N CM CM ^ CM 


(ciO(Nto«no«or;h ^ 

CM ^^^•n^^(N(NCO^O‘0'0(N 


loor-'o^o 
» -rf O uo in cm 


S-S S' 
$?3;§| 

s; w) 

^ c 
o 
J 


coor^«ncMvncM^^invOcovOOp^t 
iowr6oodr^(>c>^Vvdr^'o^o6oo 
r-'OON^o^O'oo^ONr'fS'-' r^o 
CM cO CM CO CN »—< <s| f— < «—i «— i COCO CM <N 


(NvOco^ONM-o^^avcsr^ojoor^ 

«n o\ JQ ^ »n Q cm 2 f 00 rf — rj *n <n 

Noooob;oot^^^'OvTj-^«ovoC>oo 

04 ^ CM CM CM CM CM CM CM 


cooocooo-^ooor-t^ino^oor-oooo 

O^cococM^^^p^cNcococMTrco 

r^in^vb^cocococococococoincoco 


o a 
O O O 


»o«nvO'Oooa'^‘nvO(Nco^»o»o^^ 
CM CM CM — ^ ^ 


t System III ( 1965) 


Table 3 

1981 Region B/C Images: Fully Processed by MIPL 


Start Conditions End Conditions Mid-Point Conditions 


Date 

of 

Time 

Io 

Phase 

Angle 

Magnetic 
Longitude 
of lot 

Tune 

Io 

Phase 

Angle 

Magnetic 
Longitude 
of lot 

Io 

Phase 

Angle 

Magnetic 
Longitude 
of lot 

Image ID 
Number 

-Observations 

am 

(deg) 

(deg) 

am 

(deg) 

(deg) 

(deg) 

(deg) 

(Tape/frames) 

i April 

5:00 

291.1 

303.2 

5:36 

296.3 

319.8 

293.7 

311.5 

SIP 410/13-15 

8 April 

4:49 

88.0 

213.1 

5:24 

92.9 

229.4 

90.5 

221.3 

SIP 415/27-29 


7:28 

110.3 

286.9 

7:58 

114.5 

300.8 

112.4 

293.9 

♦SIP 415/43-45 


7:39 

111.9 

292.0 

8:07 

115.8 

305.0 

113.9 

298.5 

♦SIP 415/44-46 

9 April 

3:48 

283.5 

131.3 

4:19 

287.9 

145.6 

285.7 

138.5 

SIP 416/4-6 

4:22 

288.3 

147.0 

4:53 

292.7 

161.4 

290.5 

154.2 

SIP 416/7-9 

4 May 

3:25 

217.6 

216.2 

3:55 

221.8 

230.0 

219.7 

223.1 

SIP 417/8-10 

4:20 

225.4 

241.6 

4:53 

230.1 

256.8 

227.8 

249.2 

SIP 417/13-15 


4:56 

230.5 

258.2 

5:33 

235.8 

275.3 

233.2 

266.8 

SIP 417/16-18 


5:49 

238.1 

282.7 

6:28 

243.6 

300.7 

240.9 

291.7 

SIP 417/20-22 


7:19 

250.9 

324.3 

7:52 

255.6 

339.5 

253.3 

331.9 

SIP 417/27-29 


9:06 

266.1 

13.7 

9:50 

272.4 

34.1 

269.3 

23.9 

SIP 417/36-39 

"5 May 

3:59 

65.9 

178.9 

4:30 

70.3 

193.3 

68.1 

186.1 

SIP 418/10-12 

4:32 

70.5 

194.2 

5:15 

76.6 

214.2 

73.6 

204.2 

SIP 418/13-16 


6:34 

87.7 

250.9 

7:10 

92.7 

267.6 

90.2 

259.3 

SIP 418/24-26 

___ 

8:03 

100.2 

292.2 

8:35 

104.7 

307.0 

102.5 

299.6 

SIP 418/31-33 

6 May 

5:36 

283.6 

170.4 

6:10 

288.5 

186.1 

286.1 

178.3 

SIP 419/19-21 

6:11 

288.6 

186.6 

6:43 

293.2 

201.4 

290.9 

194.0 

SIP 419/22-24 

2 May 

3:28 

46.4 

153.5 

3:55 

50.2 

166.0 

48.3 

159.8 

SIP 420/6-8 


4:10 

52.3 

173.0 

4:40 

56.5 

186.9 

54.4 

180.0 

SIP 420/10-12 


5:05 

60.0 

198.5 

6:28 

71.6 

237.0 

65.8 

217.8 

SIP 420/15-20 


6:30 

71.9 

238.0 

7:09 

77.4 

256.1 

74.7 

247.1 

SIP 420/21-23 

— 

8:28 

88.5 

292.7 

9:07 

93.9 

310.8 

91.2 

301.8 

SIP 420/30-32 

1 3 May 

3:28 

250.1 

100.3 

4:03 

255.1 

116.4 

252.6 

108.4 

SIP 421/21-23 

4:43 

260.8 

134.9 

5:04 

263.7 

144.6 

262.3 

139.8 

SIP 421/27-28 

— 

5:42 

269.1 

162.2 

6:07 

272.7 

173.8 

270.9 

168.0 

SIP 421/32-33 


6:09 

273.0 

174.7 

6:47 

278.4 

192.3 

275.7 

183.5 

SIP 421/34-36 


6:49 

278.7 

193.2 

7:23 

283.5 

208.9 

281.1 

201.1 

SIP 421/37-39 


7:25 

283.8 

209.8 

7:58 

288.5 

225.1 

286.2 

217.5 

SIP 421/40-42 


8:00 

288.7 

226.0 

8:35 

293.7 

242.2 

291.2 

234.1 

SIP 421/43-45 


8:25 

292.3 

237.6 

9:00 

297.2 

253.8 

294.8 

245.7 

SIP 421/45-47 

June 

4:19 

100.2 

292.4 

4:53 

105.0 

308.1 

102.6 

300.3 

SIP 424/10-12 

T4 June 

3:47 

284.0 

212.6 

4:28 

289.8 

231.5 

286.9 

222.1 

SIP 425/6-9 

4:17 

288.2 

226.5 

5:05 

295.0 

248.7 

291.6 

237.6 

SIP 425/9-12 


t System III (1965) 

* images processed are redundant 


31 



Table 4 

Observed East- West Brightness Asymmetries in the Jupiter System 


Observational 


Species 

Mode 

Na 

Ground based 

S ++ 

Voyager 1 

S + 

Ground based 

o + 

Ground based 

o 

IUE 

s 

IUE 

S+ 

IUE 

S++ 

IUE 

o 

Ground based 


Emission Line 
(A) 

West/East 
Brightness 
Ratio 

5889, 5895 

~0.8 

685 

- 1.3 

4069 

-2 

6731 

- 1.8 

3726 

-2.2 

1356 

- 1.4 

1479 

- 1.7 

1814 

- 1.5 

1900, 1914 

- 1.3 

1256 

- 1.8 

1729 

- 1.3 

6300 

- 1.5 


Reference 

Bergstralh et al. (1975, 1977) 
Sandel and Broadfoot (1982) 
Morgan (1985a,b) 

Morgan (1985a,b) 

Ballester (1989) 

Ballester (1989) 

McGrath et al. (1990) 
McGrath et al. (1990) 

Scherb and Smyth (1993) 


3 2 


Table 5 

1981 Observations of the Io Plasma Torus 


“a 


% V 


C/3 

VO 


V V V N V N ^ V V "> 


•< 
K vo 

c/3 rl 

vo 


% 'V N % N N 




o< 

a ^ 

c/3 o 

Tf 


\ N N N 


o< 

a <* 

VO 

C/>o 




o< 

a ov 

O r- 

m 




~3 


v v v v 


<s 

cu 

C3 

Q 

Oj 

c 


U 

VI 

JO 

a 


r- 

<N 


VO 

<N 


- m 
fS ^ 

— . CM 
-CM 

VO CM CO 
— • .CM 

Tf CM CM 

— .CM 


r 

<N 


CM 

vn 


o 

CM 

CO 

• o 


vn o 

CM CM 

co 

CM 

CM 

c9 

p 

x: 

n 

« cJ 

o — 

nO 

-O 

■c 

o 

L 

-O 

£ 

1 ^ 

Sa 'B. 


c3 e3 

o 

uu 

< 

,o 

u. 

2 

< s 

s < 

o 

Uu 

2 2 


CM 

oo 

ON 


9 


& 


J* 


E 

o 

-C 

00 

*0 

s 

c 

* 

o 

fc— 

GQ 


.O 

cJ 

VO 

OO 

ON 


00 

Im 

o 

2 


vn 

oo 

ON 


o 

-O 

u 


& 


ro 

on 



FIGURE CAPTIONS 


FIG. 1. Observing Parameters for the 1981 Io Sodium Cloud Image Data for Table 
Mountain Observatory. The solid lines show the complete angular coverage for the data as 
listed in Table 1. The dots and squares show the mid-point conditions for the 34 completely 
reduced images listed in Table 2. 

FIG. 2. 1981 May 13 Region B/C Image West Cloud Sequence. Four D 2 emission images 
of the sodium cloud in Table 3 as measured on the sky plane are shown in contour plot 
format and in proper spatial relationship to Jupiter, lo's location, and lo's circular orbit 
which appears as a thin ellipse. Contour brightness levels in kiloRayleighs, from outside to 
inside, are spaced as follows: 0.2, 0.5, 1.0, 2.0, 5.0, 10.0, 20.0. 

FIG. 3. 1981 Region B/C Images East of Jupiter. Six D 2 emission images of the sodium 
cloud in Table 3 as measured on the sky plane east of Jupiter in 1981 are shown in contour 
plot format and in proper spatial relation to Jupiter, lo's location, and Io circular orbit. 
Contour brightness levels in kiloRayleighs, from outside to inside, are spaced as follows, 
0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0 and 20.0. 

FIG. 4. System III Variability of the Io Plasma Torus East and West of Jupiter. The 
intensity of the Io plasma torus in the 6731 A emission line of S + measured near 5.5 Jupiter 
radii from the planet's center by Morgan (1985b) in Run 4 (May 2-4, 1981) is shown as a 
function of the System III longitude angle. Lines connect data points taken on the same 
night. Open symbols were taken to the east of Jupiter and filled symbols to the west of 
Jupiter. 


34 


FIG. 5. Lifetime of Sodium at lo's Location. The sodium lifetime at lo's location in units of 
hours is shown as a function of the lo System III magnetic longitude angle and the Io 
geocentric phase angle for a plasma torus description that is in (a) inherently symmetric and 
in (b) inherently asymmetric in the System III longitude angle. 

FIG. 6. Lifetime of Sodium in the Plasma Torus at Eastern and Western Elongation. The 
sodium lifetime , in units of hours, is shown in the coordinate frame of the plasma torus for a 
plasma torus description that is inherently symmetric in System III longitude. The location of 
Io in the coordinate frame of plasma torus for these two elongation pictures varies 
periodically as a function of the System III magnetic longitude of the satellite and is shown 
by the dark oval. The radial location where the peak enhancement of the inherently System 
III longitude asymmetry occurs is indicated by the black dot. 

FIG. 7. West Sodium Cloud Model Calculations for a Longitudinally Inherently Asymmetric 
Plasma Torus Sink. Model calculations for the D 2 emission brightness of the four images in 
Figure 2 are shown with the same adopted contour levels. An inherendy asymmetric plasma 
torus description in System 111 longitude and a monoenergetic (2.6 km sec 1 ) uniform and 
isotropic sodium source ejected from an exobase radius of 2600 km are assumed. I he 
sodium flux, referenced to lo's surface area, is 2 x 10^ atoms cm ^ sec . 

FIG. 8. West Sodium Cloud Model Calculations for a Longitudinally Inherently Symmetric 
Plasma Torus Sink. Model calculations for the D 2 emission brightness of the four images in 
Figure 2 are shown with the same adopted contour levels but for an inherently symmetric 
plasma torus description in System III longitude. The sodium source is the same as in Figure 


3 5 


7. 



FIG. 9. East Sodium Cloud Model Calculations for a Longitudinally Inherently Asymmetric 
Plasma Torus Sink. Two model calculations for the D 2 emission brightness of the first April 
28 image of Figure 3 are shown on the sky plane with the same adopted contour levels of 
Figure 3. The plasma torus sink for sodium is the same as in Figure 7. For both the 
isotropic and band sources, a monoenergetic (2.6 km sec - *) ejection of sodium was assumed 
with the same exobase radius and uniform flux as adopted in Figure 7. 


36 



SYSTEM HI MAGNETIC LONGITUDE 0 



Figure 1 




BRIGHTNESS OF THE SODIUM CLOUD 

( JPL IMAGE DATA 1 - MAY 13, 1981) 
(CONTOUR LEVELS'- 0.2, 0.5, 1,2,5, 10, kR) 


a 


LU 

Q 

Z> 


LlI 
CO o 


>- 

CO 


;z 
o 

O - 1 


OJ 

Q 





(H) M|suaiu| 


Plane of Sky Longitude 




Lifetime of Sodium in the Plasma Torus at lo’s Location 

(Plasma Torus : Asymmetry in System III Longitude) 



(6ap) oi JO 339NV 3SVHd 9iaiN39039 


90 180 270 360 450 540 630 

SYSTEM HI MAGNETIC LONGITUDE OF 10 (deg) 




ELECTRON IMPACT LIFETIME OF SODIUM IN THE PLASMA TORUS 

(HOURS) 



Figure 




2 brightness of the sodium clou 

(MODEL) 

SINK : LONGITUDINALLY ASYMMETRIC 
SOURCE : ISOTROPIC, MONOENERGETIC 



2 BRIGHTNESS OF THE 10 SODIUM CLOUD 

(MODEL) 

SINK : LONGITUDINALLY SYMMETRIC 
SOURCE : ISOTROPIC, MONOENERGETIC 



245.7 



d 2 brightness of the sodium cloud 

(MODEL) 

SINK '• LONGITUDINALLY ASYMMETRIC 


H 

2 

UJ 

o 

3 

UJ 


\— 

* — 

to 

o 

> 

2 

to 

O 

O 

►H 

_J 


rO 

CM 

CM 


rO 

CM 

CM 




o 

UJ 

<r 

o 

f- 

cr 

UJ 

3 

2 

O 

2 

to 

>- 


to 


O 

2 

< 

CD 


Figure