K. G. Ivanov
Institute of Terrestrial Magnetism, Ionosphere, and Radio Wave Propagation, Troitsk, Moscow Region, Russia
The solar wind at the Earth's orbit is known to be a continuous sequence of large-scale solar plasma streams moving away from the Sun. Identification of sources of these streams is an important problem of solar-terrestrial physics. Recently, prerequisites for substantial progress in solving this problem have emerged. Indeed, for one thing, almost all possible solar sources of the interplanetary plasma (coronal streamer and coronal holes, flare- and filament-active regions of the lower corona) are now known in principle; second, continuous longtime Earth orbit plasma measurement data have been accumulated, on the one hand, and solar observation data, on the other hand; third, efficient techniques have been developed for identification of interplanetary disturbances at the Earth's orbit determined by coronal holes [Krieger et al., 1973; Neupert and Pizzo, 1974; Nolte et al., 1976], the coronal (heliospheric) streamer [Gosling et al., 1981], suddenly disappearing filaments and flares [Gosling et al., 1977; Ivanov, 1995; Ivanov and Harshiladze, 1994].
However, the use of these prerequisites for solving the problem of "earmarking" the interplanetary plasma streams to solar sources encounters, in our view, two obstacles.
First, one has to get over the traditional notion that, as a rule, an Earth orbit interplanetary plasma stream is determined by a single dominating solar source, be it a coronal hole, a disappearing filament, or a flare.
Stability of this notion in a significant way is due to a definition of the complex streams introduced by NASA specialists [Behannon et al., 1991; Burlaga, 1975; Burlaga et al., 1987]. According to this definition, the complex streams of the magnetospheric plasma are formed under the interaction of two or more speedy streams from different sources or the same nonstationary source. This definition is limited because it does not take into account the complex streams formed under the interaction of two slow streams both between themselves and with the speedy streams. The latter category of complex streams was recently referred to as separate simple and irregular streams [Burlaga, 1975]. Moreover, the above definition was specified: complex streams at the Earth's orbit seldom occur (in 25% of the total number of observed streams [Burlaga, 1975, Figure 1]). Later on the complex streams defined in the above way were considered to be related to relatively rare events of strong magnetic storms [Burlaga et al., 1987].
However, if one widens the notion of complex streams, including the streams formed under interaction of separate slow streams between themselves and also of slow and speedy streams, one may assume that the solar wind in the Earth's orbit consists of a sequence of complex streams, and inside each of them there is a strong interaction of two, three, or more streams from solar sources of different origin. This definition includes in a natural way the complex streams formed in known pair interactions of some speedy streams from coronal holes, flare- and filament-active regions with the slow stream in the coronal streamer belt [Crooker and Cliver, 1994; Ermolaev, 1995; Gonzales et al., 1994; Wei, 1991].
One of the goals of this paper is to introduce a wider definition of complex streams, generalizing modern ideas on the real and assumed interactions in the interplanetary medium between both slow and speedy streams from solar sources of different nature, and to develop correspondingly more detailed classification of the streams.
Second, this "earmarking" is very difficult to realize since it requires handling of all the information available on all presumably important solar sources. Actually, as a rule, various solar flares and disappearing filaments are observed in the Sun in the days preceding the appearance of this or that stream of interplanetary plasma at the Earth's orbit. The principal difficulty is to choose from this set of events the geoeffective ones, that is, the events that are responsible for the particular stream. Up to now there was no commonly accepted methods of identification of such events (we will call them the "earmarking" methods). Moreover, up to now, as far as we know, no methods based on quantitative criteria of the choice of geoeffective flares and filaments were suggested. However, recently the methods based on quantitative measures of geoeffectivity (hierarchy of the radii of flare geoeffectivity and the index of disappearing filament geoeffectivity) have been suggested. These methods were preliminarily tested in the earmarking of 51 shock waves of 1978 to their solar sources [Ivanov, 1995]. That is why the second goal of this paper is earmarking to the Sun of the streams observed in 1978 and partly in 1979, using the above indicated new methods and the necessary information on flares, filaments, coronal holes, and the heliosphere current sheet.
The third goal of this paper is to apply the above suggested classification of complex streams to the results of this earmarking to reveal the streams of various classes and to obtain variations of concentration n, velocity V, and temperature T of protons typical for each class.
As was noted above, this classification has a specific goal, namely, based on the known types of solar sources of interplanetary plasma and admitting the possibility of interaction of these streams on their way from the Sun, it is sought to specify all the classes, expected in the Earth's orbit, of complex streams generated by a variety of combinations of the solar sources. There is direct and indirect evidence of the existence of at least four solar sources of interplanetary plasma: The first source is identified as a belt of coronal rays [Vsekhsvyatskiy, 1943] associated with the heliospheric current sheet (HCS) [Korzhov, 1978], whose quasi-stationary streams are detected in the Earth's orbit from the heliospheric streamer properties [Gosling et al., 1981] and the sectoral structure of the IMF [Hoeksema et al., 1983; Korzhov, 1978]. We will conventionally term this source and the corresponding stream an HCS source and HCS stream. The second source is identified as coronal holes (CH) [Bell and Glazer, 1957; Krieger et al., 1973; Munro and Withbroe, 1972; Neupert and Pizzo, 1974; Nolte et al., 1976], regions of decreased emission in the extreme ultraviolet and soft X ray band responsible for quasi-stationary rapid CH streams. The third source is identified as active regions (AR), flocculi fields of the current cycle, responsible for sporadic (explosion-type) eruptions of plasma associated with solar flares (sf) [Allen, 1944; Chapman and Bartels, 1940; Ivanov, 1995; Mustel, 1964; Newton, 1944] and suddenly disappearing filaments (SDF) [Ivanov and Harshiladze, 1994] of these very active regions. Arrival of an sf stream at the Earth's orbit can be detected [Ivanov, 1982] from the appearance of fast magnetic clouds and shock waves, and the SDF streams are characterized by a dense, cold, and slow plasma [Ivanov and Harshiladze, 1994]. The fourth source is identified as filaments outside AR, whose sudden disappearances are the source of sporadic SDF plasma streams [Bednarova-Novakova and Halenka, 1974; Gosling et al., 1977; Ivanov and Harshiladze, 1994; Joselyn and McIntosh, 1981; Newton, 1936; Wright and McNamara, 1987] with the same thermodynamic characteristics as SDF from AR. Thus we have HCS, CH, sf, and SDF streams of interplanetary plasma. For simplicity, we assume that there is no substantial difference between SDF streams from AR and outside AR.
Note also the following relationships between the characteristic speeds of the streams: vsf > vCH > vSDF > vHCS. Then, besides the four types of isolated streams, there are also a possible six types of double (determined by interaction of sources of two types) streams: SDF-HCS, CH-HCS, CH-SDF, sf-HCS, sf-SDF, and sf-CH. There are four types of triple streams: CH-SDF-HCS, sf-SDF-HCS, sf-CH-HCS, and sf-CH-SDF. There is one type of a complex stream sf-CH-SDF-HCS, whose formation involves participation of the sources of four types.
In all, 15 types of plasma streams are expected in the Earth's orbit.
The complex streams formed by series of flares [Dryer et al., 1978; Ivanov, 1982], recurrent streams [Burlaga et al., 1990], and filament-filament interactions are referred in this classification to corresponding stream classes formed by sf, CH, and SDF solar sources.
This classification is applicable to the 1978 and partial 1979 data for isolating streams of different types.
To identify the solar sources, we used HCS data [Hoeksema and Scherrer, 1985], CH data [Sanchez-Ibarra and Barrasa-Paredes, 1992], solar flare data (Solar Geophysical Data, 1978), and SDF data [Nauka, 1978; Wright, 1991]. Near-Earth plasma streams were determined from the data on density (n), velocity (v), and temperature (T) of protons [Couzens and King, 1986]. The principal criterion for stream revelation was the presence of one dominating maximum in nmax variations contrary to determination of a stream by vmax [Burlaga, 1975].
The earmarking methods to sf and SDF sources have been developed recently and described in detail by Ivanov . As far as we know, they were the first formal earmarking methods of the near-Earth disturbances to sf and SDF based on quantitative criteria of geoeffectivity of these sources. They were checked under the earmarking of 51 shock waves in 1978 to their solar sources [Ivanov, 1995]. The earmarking results were used to classify the streams according to the classes shown in the preceding section.
Typical variations of n, V, and T for streams of each class (except the sf-CH-SDF class, represented by one event) were obtained by the superposition epoch method, in which the hour either with nmax (SDF-, CH-SDF, sf-SDF, sf-CH-SDF-HCS streams) or with the crossing of HCS (with respect to the rest of the streams) was chosen as a reference point.
The results of the identification of streams of different types are listed in Table 1. Overall, 88 streams were identified, earmarked to solar sources, and classified, 158 solar sources participating in the formation of these streams. Most frequently observed were double streams, 47 events (54%); then triple streams, 21 events (24%); isolated (filament only), 13 events (14%); and complex, 7 events (8%). Filaments were the most frequent participants in the formation of streams (81 streams), followed by HCS (55 streams), sf (32 streams), and CH (30 streams). None of the isolated streamer, flare, or hole streams was found, nor were the double (hole-streamer or flare-streamer).
Out of the four expected types of isolated streams, only one type was detected. They are filament streams. Three other types (flare, hole, and streamer) were not found in 1978 at the phase of increasing solar activity. Thus one can assume that such streams either should not be expected at all at the Earth's orbit, or they appear (nobody knows how frequently) at other phases of solar activity. Evidently, the current concept of solar-terrestrial relations, largely based on ideas of isolated streams, should be revised. The fact that, out of all the isolated types possible, only SDF streams were found is further evidence of their important role at this phase of the cycle.
Characteristics of an isolated SDF stream in the Earth's orbit were for the first time ascertained (Figure 1a). Nearly bimodal (with nmax 20 cm-3 ) distribution of concentration versus time of passage of the stream (about 20 hours) is seen in the head portion of the stream. In the wake of this afflux, the concentration decreases while the speed and temperature experience smooth changes during about 1-1.5 days, with vmax/v0 110%, Tmax/T0 = 200% of the prestream values of v0 400 km s-1 and T0 = 4 104 K. Maxima of v and T come in 15-20 hours after the moment of nmax.
It follows from the foregoing that an SDF stream consists of head and tail parts, and it can be assumed that these parts are substantially a filament eruption and posteruptive effluence, respectively.
Note that Gosling et al.  were the first to discover and describe the so-called noncompressive cold density enhancements (NCDE). These authors also suggested that NCDEs are signatures of filament streams in the Earth's orbit, a fact that we recently were able to corroborate by comparing SDF in the Sun with NCDE at the Earth [Ivanov and Harshiladze, 1994]. In their study, Gosling et al.  did not yet, quite naturally, consider the contribution from different solar sources to the streams with NCDEs; therefore the average characteristics of NCDEs [Gosling et al., 1977] differ somewhat from an isolated SDF stream. Indeed, in the case of an SDF stream (Figure 1a), there is a quite definitive two-phase dynamics of n, v, and T variations, and the median undisturbed level v0 = 400 km s-1 is certainly higher than that in NCDEs.
As for the nature of posteruptive effluents, the following explanation is possible. It is well known that under SDF eruption long-living (about a day) coronal holes appear [Harvey and Sheeley, 1979; Rust, 1983; Solodyna et al., 1977]. It is suggested that these transient coronal holes may be a source of an increase of the solar wind velocity after a filament eruption [Kozuka et al., 1994; Rust, 1983]. Evidently, the feature of an SDF stream, which we call a posteruptive effluent, is (detected for the first time) a transient increase of the solar wind velocity due to the effluent from a coronal hole, as was suggested by Rust .
To conclude this section, let us recall that in isolating the median SDF stream, we, for simplicity, did not distinguish SDFs of active regions from SDFs outside these regions.
The filament-streamer (SDF-HCS) (Figure 2a), flare-filament (sf-SDF) (Figure 2b), and hole-filament (CH-SDF) (Figure 2c) streams, as well as the purely filament (SDF) stream, consist of a head plasma afflux and a tail speed and temperature rise. These affluxes have somewhat different forms of variations of n but a maximum value nmax const = 21-25 cm-3 for all streams. Moreover, in all streams (except CH-SDF), vmax = const = 400 km s-1, Tmax = 105 K. Double streams differ from isolated SDF streams in their more abrupt increase in v and their having a pronounced maximum of temperature Tmax in decay of n. This property is most clearly displayed by the CH-SDF stream (Figure 2c). It can be assumed that hot plasma with Tmax is in the region of interaction of appropriate rapid and slow fluxes (solid sections in Figures 2a-2c). Of course, the fact itself of the existence of a hot region at the boundary of rapid and slow streams, as well as the interpretation of this fact in terms of the theory of interaction, is not entirely new. Burlaga  introduced a special term "stream interface." What is new is specification of the notion "slow solar wind." In our case, this is a stream from another solar source: in two events (Figures 2b and 2c) from SDF and in one event (Figure 2a) from HCS.
Note (see Table 1) that SDF-HCS streams (27 events) constitute the most numerous class of streams observed in 1978. These streams, however, need further, refined classification taking into account the mutual position of SDF and HCS in the Sun. Figure 2a shows one of the modifications of an SDF-HCS stream generated upon the disappearance of an SDF situated east of the vertical HCS. Four events were averaged by superposing the moments of crossing HCS. The stream is seen to be almost completely observed upon crossing HCS (east of it).
The sf-CH stream (Figure 2d) looks like a typical blast wave. It is generally interpreted as an isolated sf stream propagating in a quiet solar wind [Hundhausen et al., 1970]. In our classification, the role of a quiet wind is played by the stream from the coronal hole. This stream is seen to feature extremal n, v, and T.
These streams feature a more complex form (Figures 3, Figure 1b) than the streams from two sources. This difference is more clearly seen in flare-filament-streamer (sf-SDF-HCS) and flare-hole-filament (sf-CH-SDF) streams: several abrupt steps and extrema can be noted determined by shock waves (S), interaction boundaries (I), and the filament front (F). The sf-SDF-HCS streams are represented by a modification, in which the flares (sources) were situated east of HCS. Averaging was performed over five events with superposition of the moments of crossing HCS. A shock wave and interaction boundary can be seen situated in front of, and behind, HCS.
Two other streams with HCS, CH-SDF-HCS and sf-CH-HCS, are largely similar to CH-SDF and sf-CH, respectively. In averaging, nmax and bow shock were superposed for CH-SDF-HCS and sf-CH-HCS streams, respectively. The sf-CH-SDF type is represented by only one stream (Figure 3c) with adequately identifiable boundaries F, S, and I. Appropriate sources in the Sun (Figure 4) illustrate, by their arrangement, the contribution of each of them to this complex stream in the Earth's orbit.
The sf-CH-SDF-HCS stream (Figure 1b) is similar to sf-CH and sf-CH-HCS streams: there is an abrupt front and rapid decay of n and T. The abrupt change in n, however, is extraordinarily great, whereas the abrupt changes in v and T are less than those in sf-CH and sf-CH-HCS streams. Besides, the profile of v is similar to that in an sf-CH-SDF stream. In this type of stream, strong fluctuations in n and T are observed in the decay of these quantities.
In the classification proposed here, however, there is room for improvement. First, it is not clear to what extent the simplification of excluding the difference of SDFs in ARs of the current and preceding cycles is valid. Second, there still remains the open question of two more types of possible solar sources: the quasi-stationary plasma streams from ARs [Mustel, 1964, 1980] and the relatively stable filaments [Bednarova and Halenka, 1974; Kiepeheuer, 1947; Waldmeier, 1946]. Therefore the number of types of solar sources can increase from four to seven. Finally, it must be kept in mind that for a classification to be complete, it should take into account not only the types of sources, but also the quantitative and qualitative differences within each type (as is the case, for example, in the works by Nauka , Sanchez-Ibarra and Barrasa-Paredes , Solar Geophysical Data (1978), and Wright and McNamara ), time series of sporadic eruptions [Ivanov, 1982], and the mutual position of the sources.
Nevertheless, this classification, even as it is, has proved to be useful. It allows one to step aside from the current arguments that one solar source of a given type generates in the Earth's orbit an interplanetary plasma stream corresponding only to this source and propagating through a quiet solar wind. As we have seen for the example of data for 1 year, the overwhelming majority of streams in the Earth's orbit were formed as a result of interaction of two, three, or four solar sources. As for the "background" solar wind, the question as to its nature have no unambiguous answer yet. To understand the hydrodinamical sense and solar sources of the plasma between plasma streams, one has to analyze the results of earmarking to solar sources of a rather durable and continuous sequence of the interplanetary plasma streams observed in the Earth's orbit.
In this respect, further revision of previous works concerning the earmarking of interplanetary disturbances in the Earth's orbit to solar sources will, in our view, be needed, as well as new methods of earmarking, interpretation and modeling.
It is worth attracting attention to the prominent role of the disappearing filaments in formation of the solar wind streams in 1978, indeed: (1) only disappearing filaments bore in the interplanetary medium the streams free of the influence of other sources (sf, CH, or HCS); (2) only disappearing filaments are completely represented in all combinations of the sources responsible for the complex streams identified above; (3) filaments participated in the formation of almost all streams of 1978.
The above results and also our previous studies of the disappearing filaments [Belov and Ivanov, 1997; Ivanov, 1996; Ivanov and Harshiladze, 1994] demonstrate that the number of such filaments is much higher than those presented in the well-known Wright  catalogue. Actually, in this catalog, only events satisfying particular criteria are chosen, and many filaments with small dimensions and low intensity are not presented, even though they are geoeffective because of their position relative to the Earth and HCS. In this respect, the estimate of the input of SDF sources into the mass flux of the solar wind is needed. According to Gosling et al.  the NCDE mass flux is comparable to that in the high-speed solar wind streams, but low frequency of their appearance (about 3 times per month) limits by about 10% their input in the total mass flux. However, the input may increase significantly, if all the disappearing filaments are taken into account [Ivanov, 1996; Ivanov and Harshiladze, 1994], and under some conditions in the Sun, become predominant [Ivanov and Harshiladze, 1997].
On the basis of (1) the assumption that every individual interplanetary plasma stream in the Earth's orbit can be formed as a result of the interaction of various combinations of four types of solar sources (flares, coronal holes, disappearing filaments, and coronal rays), (2) the 1978 solar and interplanetary experimental data, and (3) the current methods of identification of the solar source effects in the Earth's orbit, we have obtained the following results:
1. In 1978, 85% of all streams observed in the Earth's orbit were generated by two or more solar sources.
2. Most active are the filament (SDF) sources that took part in the formation of 95% of the streams. Then follow streamer (HCS) sources (75%), flare (sf) sources (45%), and CH sources (40%).
3. SDF sources are responsible for 100% of isolated streams, while no isolated sf, HCS, and CH streams are identified.
4. The role of a slow solar wind, depending on the combination of sources, was played by HCS streams (75%), SDF streams (60%), and CH streams (15%); the role of speedy streams was played by sf streams (45%), CH streams (35%), and SDF streams (75%).
5. Isolated SDF streams last for about 2 days and consist of a head afflux (the eruption itself) and a subsequent small increase in v and T (posteruptive stream).
6. Streams generated by two sources (except the sf-CH stream) differ from SDF streams in a more or less abrupt increase in T to a maximum in decay of n. This is a region of warming-up (interaction) of the two streams.
7. The sf-CH and sf-CH-HCS streams have the form of a blast wave.
8. The form of streams generated by three sources is more complex, which is displayed by the occurrence of two maxima in n, v, and T, identified with filament eruptions, interaction boundaries, and shock waves.
Bednarova, B., and F. Halenka, Results of investigations of solar and geomagnetic activity obtained at the Geophysical Institute, CSAS, Geofys. Sbornic, 22 (424), 277, 1974.
Behannon, K. W., L. F. Burlaga, and F. Hewish, Structure and evolution of compound streams at < 1 AU, J. Geophys. Res., 96, 21,213, 1991.
Bell, B., and H. Glazer, Geomagnetism and the emission-line corona 1950-1953, Smithson. Contrib. Astrophys., 2 (5), 51, 1957.
Belov, A. V., and K. G. Ivanov, Forbush-effects in 1978: Role of suddenly disappearing filaments, Geomagn. Aeron., 37 (3), 32, 1997.
Burlaga, L. F., Interplanetary stream interfaces, J. Geophys. Res., 79 (25), 3747, 1974.
Burlaga, L. F., Interplanetary streams and their interaction with the Earth, Space Sci. Rev., 17, 327, 1975.
Burlaga, L. F., K. W. Behannon, and L. W. Klein, Compound streams, magnetic clouds, and major geomagnetic storms, J. Geophys. Res., 92, 5725, 1987.
Burlaga, L. F., W. H. Mish, and Y. C. Whang, Coalescence of recurrent streams of different sizes and amplitudes, J. Geophys. Res., 95, 4247, 1990.
Chapman, S., and F. Bartels, Geomagnetism, vol. 1, 542 pp., Oxford Univ. Press, New York, 1940.
Couzens, D. A., and J. H. King, Interplanetary Medium Data Book 1977-1985, 800 pp., World Data Center A, Rockets and Satellites, Greenbelt, Md., 1986.
Crooker, N. U., and E. W. Cliver, Postmodern view of M regions, J. Geophys. Res., 99, 23,383, 1994.
Dryer, M., et al., Dynamics MHD modeling of the solar wind disturbances during the August 1972 event, J. Geophys. Res., 83, 532, 1978.
Ermolaev, Yu. I., Velocities and temperatures of protons and alpha particles in various types of solar wind streams, Kosm. Issled., 33, 381, 1995.
Gonzales, W. D., et al., What is a geomagnetic storm?, J. Geophys. Res., 99, 5771, 1994.
Gosling, J. T., et al., Noncompressive density enhancements in the solar wind, J. Geophys. Res., 82 (32), 5005, 1977.
Gosling, J. T., et al., Coronal streamers in the solar wind at 1 AU, J. Geophys. Res., 86 (A7), 5438, 1981.
Harvey, J. W., and N. R. Sheeley, Coronal holes and solar magnetic fields, Space Sci. Rev., 23, 139, 1979.
Hoeksema, F. T., and P. H. Scherrer, The solar magnetic field 1976 through 1982, Rep. UAG 94, 220 pp., World Data Center A, Washington, D. C., 1985.
Hoeksema, J. T., F. M. Wilcox, and P. H. Scherrer, The structure of the heliospheric current sheet: 1978-1982, J. Geophys. Res., 88, 9910, 1983.
Hundhausen, A. J., S. J. Bame, and M. D. Montgomery, Large-scale characteristics of flare-associated solar wind disturbances, J. Geophys. Res., 75 (25), 4631, 1970.
Ivanov, K. G., A study of some interplanetary shock wave tendencies, Space Sci. Rev., 23, 49, 1982.
Ivanov, K. G., Solar sources of interplanetary shock waves, 1978, Geomagn. Aeron., 35 (3), 24, 1995.
Ivanov, K. G., Index of filament increases of plasma density at 1 AU, Geomagn. Aeron., 36 (1), 150, 1996.
Ivanov, K. G., and A. F. Harshiladze, The effect of disappearing solar filaments on the density of near-Earth interplanetary medium and geomagnetic activity, Geomagn. Aeron., 34 (6), 28, 1994.
Ivanov, K. G., and A. F. Harshiladze, Filament-generated solar wind (abstract), in International Symposium on STCP, Paros, Greece, 1997.
Joselyn, F. A., and P. S. McIntosh, Disappearing solar filaments: A useful predictor of geomagnetic activity, J. Geophys. Res., 86 (A6), 4555, 1981.
Kiepeheuer, K. O., A slow corpuscular radiation from the Sun, Astrophys. J., 105, 408, 1947.
Korzhov, I. P., Three-dimensional structure of the interplanetary magnetic field, Astron. Zh., 55 (1), 96, 1978.
Kozuka, Y., et al., The dynamical characteristics of a disappearing filament associated interplanetary disturbance observed in early May, 1992, in Proceedings of the Second SOLTIP Symposium, edited by T. Watanabe, p. 73, Ibaraki Univ., Japan, 1994.
Krieger, A. S., A. F. Timothy, and E. C. Roelof, A coronal hole and its identification as the source of a high velocity solar wind stream, Sol. Phys., 29 (2), 505, 1973.
Munro, R. H., and G. L. Withbroe, Properties of a coronal hole from EUV observations, Rep. TR-31, 20 pp., Harvard Coll. Observ., Cambridge, Mass., 1972.
Mustel, E. R., Quasi-stationary emission of gases from the Sun, Space Sci. Rev., 3, 139, 1964.
Mustel, E. R., Quasi-stationary emission of gases from active regions of the Sun and coronal holes, Astron. Zh., 57, 225, 1980.
Nauka, Solar Data, nos. 1-12, 980 pp., St. Petersburg, 1978.
Neupert, A. S., and V. Pizzo, Solar coronal holes as sources of recurrent geomagnetic disturbances, J. Geophys. Res., 79 (25), 3701, 1974.
Newton, H. W., Eruptive dark flocculi and their possible relation to magnetic disturbances, Observatory, 59 (1), 51, 1936.
Newton, H. W., Solar flares and magnetic storms, Mon. Not. R. Astron. Soc., 104 (1), 4, 1944.
Nolte, J. T., et al., Coronal holes as sources of solar wind, Sol. Phys., 46, 303, 1976.
Rust, D. M., Coronal disturbances and their terrestrial effect, Space Sci. Rev., 34, 21, 1983.
Sanchez-Ibarra, A., and M. Barrasa-Paredes, Catalogue of coronal holes 1970-1991, Rep. UAG 102, 70 pp., World Data Center A, Boulder, Colo., 1992.
Solodyna, C. V., A. S. Krieger, and J. T. Nolte, Observations of the birth of a small coronal hole, Sol. Phys., 54, 123, 1977.
Vsekhsvyatskiy, S. K., Solar forecast of geomagnetic activity, Astron. Zh., 20 (5-6), 34, 1943.
Waldmeier, M., An attempt at an identification of the M region, J. Geophys. Res., 51, 538, 1946.
Wei, F. S., Influence of the heliospheric current sheet on the propagation of flare-associated interplanetary shock waves, in Proceedings of the First SOLTIP Symposium, edited by S. Fischer and M. Vandas, p. 294, Astron. Inst. Czech. Acad. Sci., Prague, 1992.
Wright, C. S., Catalogue of solar filament disappearances 1964-1980, Rep. UAG 100, 62 pp., World Data Center A, Washington, D. C., 1991.
Wright, C. S., and L. F. McNamara, The relationship between disappearing solar filaments, coronal mass ejections, and geomagnetic activity, Sol. Phys., 108, 383, 1987.