Читать книгу Space Physics and Aeronomy, Ionosphere Dynamics and Applications - Группа авторов - Страница 16

The magnetosphere‐ionosphere‐thermosphere (MIT) coupled system is strongly driven by the solar wind. In order to specify the IT response to magnetospheric energy input, the sources of forcing and the physical processes by which the system responds must be understood. The magnetosphere couples to the ionosphere through the magnetic field of the Earth so that magnetospheric electric field, E, and magnetic field, B, couple to ionospheric E and B fields. FACs provide the communication between the magnetosphere and the solar wind to the ionosphere. Solar UV radiation energy maintains ionization, and solar wind coupling with the magnetosphere provides additional energy via fluctuations in the E and B fields.

The energy that enter the IT system from the magnetosphere takes two forms, electromagnetic (EM) in the form of Poynting flux, and kinetic, in the form of precipitating particles. The majority of magnetospheric energy maps to the high‐latitude region of the Earth. Thus, both types of energy occur at high latitudes in the polar cap, auroral zones, and subauroral regions.

The Poynting flux vector, S, is written

(1.1)

where μ₀ is the permeability of free space. The magnetic field, B, includes the Earth’s magnetic field, the steady state contribution from large‐scale currents, and EM wave fields. Only the perturbation wave field, δB, produces energy that dissipates in the ionosphere (Kelley et al., 1991; Richmond, 2010), and only contributions to S from the perturbation magnetic field are considered relevant to energy input to the IT system.

Poynting’s theorem in differential form is written

(1.2)

where energy, W, in the EM wave is

(1.3)

ε₀ is the permittivity of free space, j is the current density, E is the electric field, and j • E is the energy dissipation or conversion rate. It is usually assumed that there is little change in the wave energy with time in comparison with the j.⋅ E term and is usually ignored. This term is commonly referred to as the Joule heating rate, but it includes the energy transferred into bulk kinetic energy (Thayer & Semeter, 2004; Richmond, 2010). The kinetic energy term is usually regarded as less important (Lu et al., 1995) though some would disagree (Thayer & Semeter, 2004).

S, as defined above, is the perturbation Poynting flux, .

Richmond (2010) pointed out that the interpretation of Poynting’s theorem as normally applied for ionospheric energy dissipation requires that the sides of the volume over which the Poynting flux is dissipated be equipotentials. Gary et al. (1994) argued that individual flux tubes can be regarded as the appropriate volume for correct application of Poynting’s theorem.

Figure 1.1, reproduced from Knipp et al. (2004), shows the partition of energy input from 1975 to 2003, based on (1) estimates of power in the solar spectrum from the SOLAR2000 model (Tobiska et al., 2000); (2) estimates of auroral zone particle power based on observations from the Defense Meteorological Satellite Program (DMSP) and NOAA TIROS and Polar‐Orbiting Operational Environmental Satellites (POES); and (3) a combination of the Polar Cap (PC) index (Troshichev et al., 1988) and the Dst index used to parameterize estimates of Joule power. As stated by the authors, the indirect estimates of Joule heat are probably conservative. However Figure 1.1 shows that Joule heat entering the ionosphere is the dominant form of energy input, far exceeding the energy attributed to auroral particle precipitation. The ration of Joule heat to auroral precipitation power ranges from a factor of 2 on average, to greater than an order of magnitude during large magnetic storms.

Figure 1.1 Solar power, particle power, and Joule power from 1975–2003

(reproduced from Fig. 4, Knipp et al., 2004. Reproduced with permission of Springer Nature).

From equation (1.2), Joule heat due to dissipation of electromagnetic energy is

(1.4)

This is equivalent to

(1.5)

where σ is the conductivity (Banks et al., 1981). Radar measurements of the ion convection determine E. Combining E with models of the conductivity, Q, can be calculated. A comparison of the electromagnetic to particle power shows a ratio ranging from 1 (Brekke, 1976; Banks, 1977) to 4 (Ahn et al., 1983b). In these studies, there is a focus on the auroral zone due to the location of radar observatories, which determine not only E but, to a large extent, σ as well (Ahn et al., 1983a).

A more direct measure of incoming EM power is to calculate the Poynting flux entering the ionosphere, which obviates the need to use a model for conductivity. This was done by Huang and Burke (2004) for a superstorm during which four DMSP spacecraft measured large FACs simultaneous with intense electric fields. It should be noted that the calculation of Poynting flux from DMSP usually uses only the two components of horizontal velocity perpendicular to the ram direction. There is uncertainty in the measured ram velocity and this component is omitted. The estimated EM power over a 20‐minute interval measured by DMSP was approximately 500 GW, comparable with the solar power over the dayside hemisphere shown in Figure 1.1. A second point in our discussion is that while all four DMSP spacecraft crossing the current sheet measured magnetometer deflections of greater than 1,200 nT magnetometers, at conjugate locations on the ground magnetometers measured deflections of the order of a few tens of nT. This point will be discussed in more detail below. It is widely accepted that Poynting flux dominates the total energy input in the IT system during periods of magnetic activity. This was confirmed in a detailed study of the energy budget for a magnetic storm on 5–6 August 2011 (Huang et al., 2014). While particle energy may not be dominant in considerations of total energy deposition, it plays a leading role in energy dissipation via conductivity as shown in equation (1.5).

Ionospheric conductivites are described by Maeda (1977) and in the World Data Center A (Kyoto) website (http://wdc.kugi.kyoto‐u.ac.jp/ionocond/exp/icexp.html). A number of models of conductivity and conductance have been derived based on particle precipitation (Fuller‐Rowell & Evans, 1987; Wallis & Budzinski, 1981; Spiro et al., 1982; Roble & Ridley, 1987; Hardy et al., 1987; Robinson et al., 1987; Gjerloev & Hoffman, 2000; McGranaghan et al., 2016, & others). Models have also been developed, which are based on UV observations (e.g., Lummerzheim et al.,1991; Coumans et al., 2004), and which account for proton precipitation (Galand & Richmond, 2001; Coumans et al., 2004). The complexity of the conductivity resulting from particle precipitation has been emphasized by McGranaghan et al. (2016), who point out that three‐dimensional spatial variations in even the average precipitation patterns are normal. When temporal variability is added, the task of capturing conductivity accurately becomes even more challenging.