Influence of upstream solar wind on thermospheric flows at Jupiter

The coupling of Jupiter's magnetosphere and ionosphere plays a vital role in creating its auroral emissions. The strength of these emissions is dependent on the difference in speed of the rotational flows within Jupiter's high-latitude thermosphere and the planet's magnetodisc. Using an azimuthally symmetric global circulation model, we have simulated how upstream solar wind conditions affect the energy and direction of atmospheric flows. In order to simulate the effect of a varying dynamic pressure in the upstream solar wind, we calculated three magnetic field profiles representing compressed, averaged and expanded `middle' magnetospheres. These profiles were then used to solve for the angular velocity of plasma in the magnetosphere. This angular velocity determines the strength of currents flowing between the ionosphere and magnetosphere. We examine the influence of variability in this current system upon the global winds and energy inputs within the Jovian thermosphere. We find that the power dissipated by Joule heating and ion drag increases by ~190% and ~185% from our compressed to expanded model respectively. We investigated the effect of exterior boundary conditions on our models and found that by reducing the radial current at the outer edge of the magnetodisc, we also limit the thermosphere's ability to transmit angular momentum to this region.


Abstract
The coupling of Jupiter's magnetosphere and ionosphere plays a vital role in creating its auroral emissions. The strength of these emissions is dependent on the difference in speed of the rotational flows within Jupiter's highlatitude thermosphere and the planet's magnetodisc. Using an azimuthally symmetric global circulation model, we have simulated how upstream solar wind conditions affect the energy and direction of atmospheric flows. In order to simulate the effect of a varying dynamic pressure in the upstream solar wind, we calculated three magnetic field profiles representing compressed, averaged and expanded 'middle' magnetospheres. These profiles were then used to solve for the angular velocity of plasma in the magnetosphere. This angular velocity determines the strength of currents flowing between the ionosphere and magnetosphere. We examine the influence of variability in this current system upon the global winds and energy inputs within the Jovian thermosphere. We find that the power dissipated by Joule heating and ion drag increases by ∼190 % and ∼185 % from our compressed to expanded model respectively. We investigated the effect of exterior boundary conditions on our models and found that by reducing the radial current at the outer edge of the magnetodisc, we also limit the thermosphere's ability to transmit angular momentum to this region.

Introduction
The theoretical background for our study is given in section 2. In sec-  In this section we present a summary of some basic theoretical principles 137 that we use throughout this study. We rely on work that has been conducted and 156 where Σ P is the height-integrated Pedersen conductance, B i is the assumed where ρ i =R i sin θ i (R i is the ionospheric radius), B i =2B J (B J is the equato- where j ||i (θ i ) is the FAC density and the sign corresponds to the north- where σ P and σ H are the local Pedersen and Hall conductivities respectively.

194
Integrating over the height of the thermosphere-ionosphere to get the total 195 equatorward current, we find that Ω T can be defined as where Σ P is the height-integrated Pedersen conductivity where z is altitude. In these expressions Ω T is a weighted average of the ef-198 fective neutral angular velocity ω T throughout the thermosphere-ionosphere, 199 which also contains contributions from meridional winds. area of the ionosphere extracted from planetary rotation, P is given by where τ is the torque per unit area of the ionosphere exerted by the J×B 206 force. The smaller component of this total used to accelerate the magneto-207 spheric plasma is The remainder of this power is dissipated in the upper atmosphere as heat 209 and mechanical work The power P A consists of two components, as shown by Smith et al. (2005).

211
One of these is Joule heating, P J , and the other is ion drag power, P D , which 212 8 is dependent on the sub-corotation of the neutral atmosphere and is then 213 viscously dissipated as heat. These are given by and  given by The equatorial magnetic field in the middle magnetosphere, B ze , and corre-  the flux conservation condition is where F O is the initial profile of the flux function (given by Eq. (19)). Rear-272 ranging to solve for ∆B z where ∆B z <0 for a southward field perturbation, and where i=P or H representing Pedersen or Hall, σ is the conductivity and ρ 312 is the neutral mass density. Adjacent pressure levels enclose constant masses 313 of thermospheric gas (hydrostatic equilibrium assumption). Therefore, the where 325 where Σ P O =0.0275 mho (Nichols and Cowley, 2004) is the background con-326 ductivity due to solar photoionisation, and Σ P j (j ||i ) in mho is an auroral en- whereṀ =1000 kg s −1 is the assumed mass outflow rate from the Io torus 376 and j ||i is the upward FAC density in the ionosphere. Our method for solving these equations is the same as that in SA09 and

Results and Discussion
In this section we present the results obtained from our modelling. We 417 firstly discuss results concerning angular velocities, conductivities and cur-418 rents. Then we proceed to discuss the thermospheric flows and energies.  The variation of height-integrated true Pedersen conductivity Σ P for our 477 three magnetospheric cases is shown in Fig. 3a  Pedersen conductivities (Fig. 3a), FAC densities (Fig. 5) and azimuthally-     whose Ω M profile is poorly constrained (due to paucity of observations).  with Ω J throughout regions IV and most of III (Nichols and Cowley, 2004).

768
The A 45 radial current profile resembles those for expanded cases, due to 769 the magnetosphere sub-corotating to a greater degree (see Fig. 9a). The    corresponding FAC density as a function of latitude is shown in Fig. 10f.

824
The radial current profile is smaller in magnitude than that of B 100 . The  The azimuthal velocity in the high latitude region is shown in Fig. 10g. 842 We expect a slight increase in sub-corotation throughout region III (see 843 Fig. 10a) compared to B 100 . This is evident by comparing Fig. 10g with 844 Fig. 7a where we can see that the region of super-corotation (dark red) has 845 diminished for B 68 . The meridional velocity distribution is shown in Fig. 10h.

846
The high altitude localised accelerated flow in region III is slightly faster than 847 in case B 100 because the pressure gradient and advection terms are 27−40 % 848 larger in this region of B 68 . This would lead to a minimal temperature in-849 crease ∼3 %, most notably in regions II and I (see Fig. 10i).   Table 1), compressed, baseline (average) and expanded.