A general study of energy addition, energy loss, and energy redistribution in the solar wind, for both spherically symmetric and rapidly diverging flow geometries, is presented. It is found that energy addition in the region of subsonic flow increases the solar wind mass flux but either has little effect on (for heat addition) or significantly reduces (for momentum addition) the solar wind flow speed at I AU. In contrast, energy addition in the region of supersonic flow has no effect on the solar wind mass flux but significantly increases the flow speed at I AU. It is also found that both momentum loss in the subsonic region and energy exchange (involving loss in the subsonic region and gain in the supersonic region) can lead to an increase in the asymptotic flow speed. This general study thus places certain constraints on viable mechanisms for driving high-speed solar wind streams and points to a number of specific, selfconsistent studies of such mechanisms that need to be carried out in the future. [Parker, 1958[Parker, , 1963[Parker, , 1964a, Parker [1965, 1969] called attention to the probable need in solar wind models for some type of energy addition above the coronal base, in order to account for observed solar wind characteristics (particularly flow speed) at I AU. However, it was pointed out by Hundhausen [ 1969, 1972] that the inability of theoretical models to produce sufficiently large flow speeds at I AU might not be due to the failure to include energy addition, but rather to the inappropriate distribution of energy among the various modes of energy transport at I AU that was predicted by the models: viz., relative to observations at I AU, models predicted much too large a conductive energy flux and much too small a flow energy flux, but approximately the right total energy flux. Nevertheless, prior to the resolution of the energy distribution problem and often with no heed paid to it a number of authors proceeded to develop solar wind models including some form of energy addition (e.g., Barnes [1969]; Hartle and Barnes [ 1970]; Holzer and Axford [ 1970]; Alazraki and Couturier [ 1971]; Barnes et al. [ 1971]; Belcher [ 1971]; Leer and Axford [1972]; Hollweg [1973, 1978]; Ryan and Axford [1975]; Jacques [1977]; Pneuman [1980]; see also reviews by Hundhausen [1972], Barnes [1974, 1979], Hollweg [1978], and Holzer [1979] and references in these reviews). It appears now that the energy distribution problem has been satisfactorily resolved by a detailed parameter study [Holzer and Leer, 1980], which indicates that conductive solar wind models (without energy source and/or sink terms) describing the fight distribution of energy at 1 AU (viz., essentially all flow energy) cannot produce high-speed solar wind streams and simultaneously satisfy observational requirements on the solar wind mass flux and the coronal base pressure. With the necessity of some type of energy addition and/or loss (above the coronal base) apparently established, there remains a need to address certain questions that have not been...

The outflow of coronal plasma into interplanetary space is a consequence of the coronal heating process. Therefore the formation of the corona and the acceleration of the solar wind should be treated as a single problem. The deposition of energy into the corona through some "mechanical" energy flux is balanced by the various energy sinks available to the corona, and the sum of these processes determines the coronal structure, i.e., its temperature and density. The corona loses energy through heat conduction into the transition region and through the gravitational potential energy and kinetic energy put into the solar wind. We show from a series of models of the chromosphere-transition region-corona-solar wind system that most of the energy deposited in a magnetically open region goes into the solar wind. The transition region pressures and the coronal density and temperature structure may vary considerably with the mode and location of energy deposition, but the solar wind mass flux is relatively insensitive to these variations; it is determined by the amplitude of the energy flux. In these models the transition region pressure decreases in accordance with the increasing coronal density scale height such that the solar wind mass loss is consistent with the energy flux deposited in the corona. On the basis of the present study we can conclude that the exponential increase of solar wind mass flux with coronal temperature, found in most thermally driven solar wind models, is a consequence of fixing the transition region pressure. drostatic corona out to the critical point at some few solar radii where the flow becomes supersonic; it is the properties of this nearly static corona, or more specifically its temperature and density, that define the major properties of the solar wind, namely, the proton speed and the proton flux. It was later realized [Parker, 1965] that classical thermal conduction and enthalpy flux from the inner corona are not sufficient to accelerate the solar wind to the high speeds of Paper number 95JA02300. 0148-0227/95/95JA-02300505.00 over 500 krn/s as first observed by the Mariner 2 spacecraft [cf. Snyder and Neugerbauer, 1964] Thus Parker argued that energy had to be added to the flow beyond the coronal base. Leer and Holzer [1980] showed that this energy must be added beyond the critical point, in the supersonic region of the flow, in order to ensure that the asymptotic flow speed increases. The proton flux, on the other hand, is almost solely determined by the density at the coronal base and by the coronal density scale height, which in the simplest models implies that it is controlled by the "average" coronal temperature. It is relatively straightforward to show that the proton flux at the orbit of Earth r = rE = 1 AU from an isothermal, spherically symmetric electron-proton corona with temperature T can be written (npUp)E -no(2k-•T-T)-3/2( 1 2 exp -:v• + • (1) Here, m v is the proton mass, k is Boltzmanns constant, and vo is the escape speed of the Sun. •is expression shows that the proton...

No abstract

No abstract