1:
2: % LaTeX2e (galley):
3: %\documentclass[galley,jgrga]{agu2001}
4: % LaTeX2e (draft):
5:
6: \documentclass[12pt,preprint]{aastex}
7: %\documentclass[preprint2]{aastex}
8:
9: %\titlerunninghead{VELOCITY PROFILES IN SOLAR CORONA}
10:
11: \shorttitle{Velocity Profiles in the Solar Corona}
12: \shortauthors{Qu\'emerais et al.}
13:
14: %\authoraddr{E. Qu\'emerais,
15: %Service d'A\'eronomie, BP3, 91 371 , Verri\`eres le Buisson, France
16: %(eric.quemerais@aerov.jussieu.fr)}
17:
18: %\usepackage{psfig}
19: \usepackage{graphicx}
20:
21: \def\hlya{H Ly-$\alpha$}
22: \def\etal{{\it et al.}}
23: \def\v{{\it Voyager}}
24: \def\vone{{\it Voyager} 1}
25: \def\vtwo{{\it Voyager} 2}
26: \def\deg{$^{\circ}$}
27: \def\lya{Lyman $\alpha$}
28: \def\lyb{Lyman $\beta$}
29: \def\Ha{H$\alpha$}
30: \def\phcs{photon cm$^{-2}$ s$^{-1}$}
31: \def\cc{cm$^{-3}$}
32: \def\quem{Qu\'emerais}
33: \def\rso{$R_{\odot}$}
34:
35: \begin{document}
36:
37:
38: %% ------------------------------------------------------ %%
39: %
40: %% TITLE
41: %
42: %% ------------------------------------------------------ %%
43:
44: \title{Velocity Profiles in the Solar Corona from Multi-Instrument Observations}
45:
46: % Type the title of your manuscript between the
47: % curly brackets in a \title command. Capitalize
48: % only acronyms, first letter of the first word,
49: % first letter of proper nouns, first letter of
50: % the first word after colons, and first letter
51: % of the first word of a subtitle. If the title
52: % exceeds one line, break it so that the first line
53: % is longer than the second line; break the title
54: % before articles, prepositions, and conjunctions.
55: % To break the title, type a double backslash where
56: % you want the break to occur, as shown above.
57:
58:
59: %% ------------------------------------------------------ %%
60: %
61: %% AUTHOR NAMES, AFFILIATIONS, and ALTERNATE AFFILIATIONS
62: %
63: %% ------------------------------------------------------ %%
64:
65: \author{E. Qu\'emerais, R. Lallement, D. Koutroumpa} %, S. Ferron and J.-L. Bertaux}
66: \affil{Service d'A\'eronomie, Verri\`eres le Buisson, France}
67:
68: \and
69:
70: \author{P. Lamy}
71: \affil{Laboratoire d'Astrophysique de Marseille, Marseille, France}
72:
73: %\author{E. Kyr\"ol\"a and W. Schmidt}
74: %\affil{Finnish Meteorological Institute, Helsinki, Finland}
75: %
76: %% ------------------------------------------------------ %%
77: %
78: %% ABSTRACT
79: %
80: %% ------------------------------------------------------ %%
81:
82: \begin{abstract}
83:
84: We present a method to derive outflow velocities in the solar corona using different data sets
85: including solar wind mass flux coming from the SWAN/SOHO instrument, electron density values
86: from LASCO-C2 and interplanetary solar wind velocities derived from ground-based
87: Interplanetary Scintillation Observations (IPS).
88:
89: In a first step, we combine the LASCO electron densities at 6 \rso\ and the IPS velocities, and compare
90: the product to the SWAN mass fluxes. It is found that this product
91: represents the actual mass flux at 6 \rso\ for the fast wind, but not for the slow wind.
92: In regions dominated by the slow wind, the fluxes derived from SWAN are systematically smaller.
93: This is interpreted as a proof that the fast solar wind has reached its terminal velocity
94: at $\sim$ 6 \rso\ and expands with constant velocity beyond this distance. On the contrary, the slow
95: solar wind has reached only half of its terminal value and is thus accelerated further out.
96:
97: In a second step, we combine the LASCO-C2 density profiles and the SWAN flux data to derive velocity
98: profiles in the corona between 2.5 and 6 \rso\ . Such profiles can be used to test
99: models of the acceleration mechanism of the fast solar wind.
100:
101: \end{abstract}
102:
103:
104: \keywords{The Sun --- The Corona}
105:
106:
107: \section{Introduction}
108:
109: There is no consensus on the mechanisms of energy deposition and solar wind acceleration in the
110: low corona, especially in the fast wind, although major results have been obtained with the SOHO
111: instruments. In particular, UV spectroscopy has led to the discovery of high ion temperatures
112: and significantly lower electronic temperatures in coronal holes, as well as strong ion temperature
113: anisotropies (e.g., Kohl et al., 2006; David et al., 1998). A number of models have been developed to
114: reproduce these properties, with damping of high frequency waves and ion cyclotron heating as favoured
115: mechanisms. Because one of the strong difficulties in modeling the solar wind expansion is linked to
116: the heat flux in regions which are not collisional enough for classical transport
117: theory to be valid, kinetic models have also been developed. These models are able to reproduce the
118: acceleration by velocity filtration (e.g., Zouganelis et al., 2005), although they do not reproduce the
119: published temperature anisotropies. Since all different classes of models predict different velocity
120: profiles, accurate measurements of the solar wind flow speed in the acceleration region are required to
121: decide which models are correct.
122:
123: One of the methods to obtain the flow speed profile in the acceleration region relies on the measured
124: coronal density profile as well as the expanding geometry. The velocity profile is simply obtained from
125: the application of the continuity equation, assuming a value for the particle flux coming from the observed region.
126: The flux is inferred from in situ or remote sensing measurements. Munro and Jackson (1977) pioneered
127: this method using Skylab white-light coronographic images of polar coronal holes. Assuming that the solar
128: flux is roughly constant at all latitudes, they used its average ecliptic value. They concluded that there
129: is a substantial energy addition in the low corona, and that the flow accelerates to supersonic speed
130: within 2-3\rso\ .
131:
132: Later on, the analysis of solar backscattered Lyman-alpha radiation led to the derivation of a
133: significantly lower mass flux in the polar regions (Kumar and Broadfoot, 1979; Lallement et al., 1985a).
134: It was shown that the requirement for a non-thermal mechanism is strongly dependent on the density
135: profile and on the out-of-ecliptic mass flux (Lallement et al., 1985b).
136: However, our knowledge of the solar wind conditions outside of the ecliptic plane is very limited as
137: most of the in situ studies of the solar wind are based on data obtained in the ecliptic plane.
138: Because of the small angle (7 degrees) between the ecliptic plane and the solar equatorial
139: plane, this means that most of our knowledge on the solar wind concerns the wind which originates
140: close to the solar equator.
141:
142: %In 1994, during the first latitudinal scan of the Ulysses spacecraft, it clearly appeared that the conditions
143: %in the solar wind were changing dramatically with heliographic latitude. Phillips et al. (1995) showed that
144: %during solar minimum conditions the streamer belt around the solar equator was associated with a slow
145: %solar wind (mean velocity around 400 km/s) while a fast solar wind was found at higher heliographic latitudes.
146: %Very different conditions were reported after the second fast latitude scan made around the solar maximum of
147: %cycle 23 (2001) (Mc Comas et al., 2001).
148:
149: In this article, we present a new approach to the determination of velocity profiles in the solar corona.
150: The steps are the following,
151: \begin{itemize}
152: \item
153: The velocity at large distance from the Sun is obtained from Interplanetary Scintillation data (Kojima et al., 1998).
154: \item
155: Using the SWAN interplanetary Lyman $\alpha$ data, we can derive the ionization rate of hydrogen at 1 AU.
156: After removal of the contributions from photo-ionization and electron impact ionization, we obtain the ionization
157: rate due to charge exchange with the solar wind protons.
158: \item
159: Using the value of the charge exchange cross section, which depends on solar wind velocity, we compute the solar wind
160: flux.
161: \item
162: Assuming radial expansion, we derive the solar wind velocity profile from the values of coronal density and solar wind mass
163: flux. The assumption of radial expansion is not necessary. We could introduce a correction term for non-radial expansion.
164: \item
165: The previous relation is then applied to the density profiles obtained from the LASCO-C2 polarized white light images, between
166: 2.5\rso\ and 6\rso\ .
167: \end{itemize}
168:
169: %In this paper, we will show how the knowledge of the ionization rate of hydrogen is used to derive
170: %the solar wind flux. Combined with the knowledge of the velocity of the solar wind, this yields the
171: %value of the solar wind density at all heliographic latitude. The velocity of the solar wind can
172: %be derived from the Interplanetary Scintillation
173: %of radiowaves (Kojima and Kakinuma, 1990).
174: %
175: %the electron density is derived from LASCO-C2 electron maps assuming a radial expansion.
176:
177: In what follows, we concentrate on a test period from 1996 to 1997 corresponding to a minimum in the solar
178: activity cycle. We briely present the results of Qu\'emerais et al. (2006) for this period. We then show how
179: mean values of density and velocity can be derived. In the last paragraph, we show how this information can
180: be combined with density profiles in the solar corona to give information on velocity profiles.
181:
182: In our conclusion, we discuss the velocity profiles found in the corona and how they relate to fast wind or
183: slow wind conditions. These results are relevant to studies of the mechanism of the solar wind acceleration.
184:
185: \section{Method of Analysis}
186:
187: The Lyman $\alpha$ background is due to the backscatter of solar Lyman $\alpha$ photons by hydrogen atoms
188: present in the interplanetary medium. This UV emission was first identified by instruments on board the OGO-8
189: spacecraft (Bertaux and Blamont, 1971; Thomas and Krassa, 1971). The hydrogen atoms of the local insterstellar
190: cloud can get very close to the Sun because of the motion of the solar system relative to the local cloud.
191: In the vicinity of the sun, the hydrogen atoms are ionized by charge exchange with solar wind protons or
192: photo-ionized by solar EUV radiation or, with a lesser probability, ionized by collision
193: with hot solar wind electrons.
194:
195: In the inner heliosphere the spatial pattern of the full-sky Lyman $\alpha$ background is carved by the
196: ionizing fluxes from the Sun (Jozelyn and Holzer, 1975; Kumar and Broadfoot, 1979).
197: Once a hydrogen atom loses its electron, it cannot backscatter Lyman $\alpha$ photons anymore. This creates a hole
198: in the sky pattern. Areas of the sky where ionization is more effective appear dimmer at Lyman $\alpha$ because
199: there are less hydrogen atoms which can backscatter solar Lyman $\alpha$ photons. The most effective ionization
200: process, roughly 4 times out of 5, is charge exchange with solar wind protons. Even in that case, the hydrogen atom
201: is lost because the hydrogen atom which captured the electron has the velocity of the solar wind proton, i.e.
202: at least a few hundred km/s. The illuminating solar Lyman $\alpha$ line is roughly 1 Angstr\"om wide.
203: In terms of Doppler shift, this corresponds to 150 km/s. This means that a hydrogen atom with a velocity
204: larger than 150 km/s is blind to solar Lyman $\alpha$ photons and this results in a loss of hydrogen atoms able
205: to backscatter the solar atoms exactly as in the case of photo-ionization.
206:
207: Qu\'emerais et al. (2006) published an analysis of the full-sky background maps obtained with SWAN/SOHO
208: between 1996 and 2005. The
209: maps were used to derive the hydrogen ionization rate at 1 AU at all heliographic latitudes. We will use these
210: values to derive the solar mass flux and its latitudinal dependence. The ionization rate of hydrogen is the sum
211: of three terms, the photo-ionization rate, the charge exchange ionization rate, and the electron impact ionization
212: rate. Photo-ionization and charge exchange have a radial dependence like the inverse of the square of the distance
213: to the sun. The photon flux and the solar particle flux have approximately the same radial variation.
214: The electron impact ionization rate also varies with the solar wind electron temperature which is not constant with
215: distance to the sun so its total radial dependence is not like the inverse of the square of the distance to the sun.
216: Using the solar wind electron distribution values obtained by the WIND/waves experiment (Issautier et al., 2004)
217: we have computed the absolute value of the electron impact ionization rate at 1 AU between 1994 and 2001.
218: We have found values between 10$^{-8}$ s$^{-1}$ at solar minimum and a few 10$^{-8}$ s$^{-1}$ at solar maximum.
219: The total ionization rate found by Qu\'emerais et al. (2006) varies between 6 10$^{-7}$ s$^{-1}$ and
220: 1.2 10$^{-6}$ s$^{-1}$, i.e. almost two orders of magnitude larger.
221: Then, because electron impact ionization is small compared to the two other sources of ionization, we will
222: neglect it and assume that the total ionization rate of hydrogen varies like the inverse of the square of the
223: distance to the sun.
224:
225: The total ionization rate $\beta_{tot}$ at 1 AU can be written as
226:
227: \begin{equation} \label{basicbeta}
228: \beta_{tot} = \beta_{exc} + \beta_{phot} + \beta_{eii} = \sigma_{exc}(v_{sw}) \cdot v_{sw} \cdot N _{sw} +
229: \beta_{phot} + \beta_{eii}
230: \end{equation}
231:
232: In the previous equation, we neglect the velocity of the hydrogen atom before the solar wind speed $ v_{sw}$.
233: The charge exchange cross section variation with velocity $\sigma_{exc}(v_{sw})$ is given by
234:
235: \begin{equation} \label{basicbeta2}
236: \sigma_{exc}(v_{sw}) \cdot v_{sw} = \frac{\beta_{tot} - \left( \beta_{phot} + \beta_{eii} \right)}{N _{sw}}
237: \end{equation}
238:
239: Equation \ref{basicbeta2} can be used to derive the solar wind mass flux from the knowledge of $\beta_{exc}$ and
240: the density of the solar wind. In the range 0-1200 km/s the product $\sigma_{exc}(v_{sw}) \cdot v_{sw}$ is a
241: monotonic function of the velocity $v_{sw}$, and can easily be inverted. The charge exchange ionization rate
242: is obtained from the SWAN data analysis once a value for photo-ionization and electron impact ionization has
243: been removed.
244:
245: %The next step in our analysis is to assume that the quantity $\beta_{phot} + \beta_{eii}$ is known and is
246: %equal to a value between $5 \times 10^{-8}$ s$^{-1}$ and $10^{-7}$ s$^{-1}$. Although previous studies
247: %suggest that the hydrogen photo-ionization rate varies with heliographic latitude like the solar illuminating flux
248: %by maybe 10 or 20\% (Pryor et al., 1992, Auch\`ere et al., TBD), we will assume that this quantity is constant
249: %with latitude. The error induced by this assumption is small compared to the value of the total ionization rate.
250:
251: %In that case, the SWAN data can be used to derive the charge exchange ionization rate.
252: The velocity profile $V(R)$ is simply derived using the flux conservation which can be written as follows,
253: where $R$ is the distance from the Sun center (in \rso\ ), and $R_o$ is 1 AU,
254: \begin{equation}
255: N(R) \cdot V(R) \cdot R^2 = N_o \cdot V_o \cdot R_o^2
256: \end{equation}
257:
258: In that case the the velocity profile is given by
259:
260: \begin{equation}
261: V(R) = \frac{\beta_{exc}}{\sigma_{exc}(V_o)} ~ \frac{R_o^2}{R^2} ~ \frac{1}{N(R)}
262: \end{equation}
263:
264: This formalism will be used in the following sections.
265:
266:
267: \section{Data Sets}
268:
269: \subsection{Solar Wind Mass Flux: SWAN data}
270:
271:
272: %Over the last three decades, there have been a number of studies of the interplanetary background. Reviews can
273: %be found in Ajello et al. (1987) or Bertaux (1996). Early studies found that the observed sky background was not
274: %compatible with an isotropic distribution of the ionizing fluxes from the Sun (Broadfoot et al., 197X, Bertaux
275: %et al 198X). It was then suggested that this could be used to study the spatial distribution of the ionizing
276: %fluxes from the sun (Kumar and Broadfoot, 1979; Lallement et al., 1985). The SWAN instrument (Bertaux et al, 1995)
277: %was then selected to fly on the SOHO mission to study this effect. SWAN stands for Solar Wind ANisotropies.
278: The Solar Wind ANisotropy (SWAN) instrument on-board the SOHO spacecraft was developed in a collaboration between
279: Service d'A\'eronomie (Verri\`eres le Buisson, France) and the Finnish Meteorological Institute (Helsinki, Finland)
280: (Bertaux et al., 1995). Its main purpose is to study anisotropies in the solar wind mass flux and their effect on
281: the interplanetary background (Bertaux et al., 1997). Thanks to the exceptional longevity of the SOHO mission, the
282: data obtained by this instrument almost cover a complete 11-year solar activity cycle.
283:
284: The most recent analysis of the SWAN data pertaining to the determination of the ionizing fluxes from the Sun
285: was published by Qu\'emerais et al. (2006). In this study, the authors have determined the absolute value
286: of the ionization rate of hydrogen at 1 AU as a function of heliographic latitude and its variations with
287: heliographic latitude. Their analysis covers the period from 1996 to 2005, i.e. almost a complete solar cycle.
288: The Hydrogen ionization rate is related to the solar wind mass flux as shown in equation (2).
289: After correction for solar EUV photo-ionization and the variation of the charge exchange cross-section with
290: solar wind bulk velocity, the SWAN data give monthly averages of the solar wind mass flux at all latitudes
291: between 1996 and 2005.
292:
293:
294: \subsection{Solar Wind Density: LASCO data}
295:
296:
297: In principle, the solar wind number density at 1 AU can be estimated from in situ measurements performed
298: by numerous spacecrafts proeminently restricted to the ecliptic plane. However results from the Ulysses
299: spacecraft have demonstrated that the density at higher latitudes can be very different from that
300: in the ecliptic (Phillips et al., 1995).
301:
302: We have therefore relied on the LASCO-C2 white light images to obtain an estimate of the solar wind density
303: at all latitudes. The LASCO-C2 polarized radiance images $pB$ are routinely generated from the daily
304: polarization sequences consisting a set of three polarized images taken through the orange filter
305: (540-640 nm) and polarizers oriented at 60${}^o$,0${}^o$ and -60${}^o$.
306: Qu\'emerais and Lamy (2002) have developed a method to invert these images and to produce two-dimensional
307: electron density maps of the solar corona.
308: Compared to our past studies (see also Lamy et al., 1997; LLebaria et al., 1999; Lamy et al., 2002),
309: the maps used for this present studies have benefited from refined corrections which successfully remove
310: the instrumental polarization.
311: The values at 6\rso\ from the center of the Sun have been extracted at all heliographic
312: latitudes. We have then computed a monthly average from the daily maps to obtain values with the same
313: sampling as the ionization rate derived from the SWAN data.
314: We discuss below in which cases the density at 6\rso\ can be simply extrapolated at 1 AU.
315:
316:
317: \subsection{Solar Wind Terminal Velocity: IPS data}
318:
319:
320: As one can see from a quick perusing through the catalog of the omniweb database, various
321: spacecraft have measured the solar wind velocity in the ecliptic plane, but only a few
322: measurements are available at mid and high heliographic latitude.
323: Here we have used the solar wind velocity estimates obtained from Interplanetary
324: Scintillation data (IPS) (Kojima and Kakinuma, 1990). These velocity values are derived from the analysis of
325: scintillations of radio-emissions from compact radio sources (Kojima et al., 1998). Because the distribution of radio
326: sources in the sky is not uniform, the sampling of these measurements is not either. These authors have found that
327: the solar wind speed is roughly constant beyond 0.1 AU from the sun which means that the acceleration occurs before.
328: Here we have averaged the velocity values beyond 0.1 AU from the sun as a function of the heliographic latitude from
329: which the wind originates. Those averages were made on a monthly basis. It should be noted also that there is a 4 month
330: data gap every year due to the weather conditions which prevent the ground-based radio observations.
331:
332:
333: \section{Results}
334:
335: \subsection{Computation of H Ionization Rates}
336:
337: The first step in our analysis is to check whether SWAN results are compatible with the LASCO density
338: values, the IPS velocity estimates and the assumption that the IPS velocity is reached at the outer
339: edge of LASCO-C2. To do that, we compute the charge exchange ionization rate at one AU $\beta_{exc}$
340: where the velocity is derived from the IPS data and the density is equal to the
341: value found from the LASCO images with a radial scaling coefficient. This coefficient expresses the relationship
342: between the density measured by LASCO at 6\rso\ and the density at 1 AU. For a radial expansion, the
343: product $N \cdot V \cdot r^2$ is conserved. If the acceleration is finished, then the velocity is constant and
344: the density varies like $1/r^2$.
345:
346: Figure \ref{quadri96} shows the paramaters and calculations obtained in April-May 1996. The top left panel shows the
347: LASCO-C2 average density at 6\rso\ as a function of heliographic latitude. There is a strong enhancement in the
348: equatorial band. The top-right panel shows the IPS mean velocity as a function of heliographic latitude for
349: the same period. The bottom-left panel shows the ionization rate of hydrogen at 1 AU. The diamonds linked by a dotted
350: line correspond to the SWAN data. The solid line corresponds to the value derived from the scaled LASCO density and the
351: IPS velocity data. The computed value represents only ionization through charge exchange. The bottom-right panel
352: shows the difference of the two curves shown in the bottom-left panel. In this plot, we see two different regimes.
353: In the equatorial band, the computed charge exchange rate is systematically larger than the SWAN total ionization rate.
354: This means that the velocity between 6\rso\ and 1 AU is not constant and that the radially scaled density is
355: overestimated. This also implies that the velocity at 1 AU is larger than the velocity at 6\rso\ which means
356: that in the equatorial band, the solar wind has not completed its acceleration. Outside of the equatorial band,
357: the SWAN ionization rates are systematically larger than the computed charge exchange ionization rate. This is
358: consistent with the fact that in this region the solar wind has reached its terminal velocity and that the density
359: simply scales like $1/r^2$ with solar distance. The difference between the two values can be attributed to the
360: sum of photo-ionization and electron impact ionization.
361:
362: Figure \ref{quadri97} shows a very similar picture. The values shown here are for April-May 1997. Once again, the
363: equatorial belt data suggest that the wind is not completely accelerated while at higher latitudes the ionization
364: rate values agree well.
365:
366: Figure \ref{diffbeta} shows the difference between the SWAN total ionization rate $\beta_{tot}$ and the
367: charge exchange ionization rate $\beta_{exc}$ computed from LASCO and IPS data as a function of the IPS solar
368: wind velocity. All monthly values for different heliographic latitudes have been used. The data are averaged
369: over intervals of 10 degrees in latitude. Two patterns appear clearly. For velocities larger than 500 km/s, the quantity
370: $\beta_{tot} - \beta_{exc}$ is positive. Its mean value is equal $(5.3 \pm 2.3) \times 10^{-8}$ s$^{-1}$.
371: This is consistent with the fact that for the fast wind conditions (here faster than 500 km/s), the acceleration
372: is completed at 6\rso\ and that the solar wind density at 1 AU is correctly derived from the LASCO value
373: scaled by $1/r^2$. In that case, the difference between total ionization rate and charge exchange ionization
374: rate is equal to the sum of hydrogen photo-ionization rate and electron impact ionization rate. The value
375: of $(5.3 \pm 2.3) \times 10^{-8}$ s$^{-1}$ found here is compatible with previously estimated values. For example
376: hydrogen photo-ionization rate estimates for solar minimum conditions
377: range between $5 \times 10^{-8}$ s$^{-1}$ (Rucinski and Fahr, 1991) to $10^{-7}$ s$^{-1}$ (Ogawa et al., 1995).
378: We estimated electron impact ionization rates for hydrogen using the WIND/WAVES data (Issautier et al., 2005) and
379: found values around $10^{-8}$ s$^{-1}$ close to solar minimum.
380:
381: When the IPS velocity value is lower than 500 km/s, we find that the quantity $\beta_{tot} - \beta_{exc}$ is
382: negative. As mentioned above, the flux conservation from $r_i = 6$\rso\ to 1 AU also contains a ratio
383: $V(r_i)/V(1AU)$. This suggests that for slow solar wind conditions, the solar wind has not yet reached its terminal
384: velocity, and that the ratio $V(r_i)/V(1AU)$ is smaller than one, assuming a radial outflow as defined in equation (3).
385:
386: Here we find a clear dichotomy between fast solar wind regions at high latitude and slow solar wind regions near the
387: equatorial region.
388:
389:
390: \subsection{Velocity Profiles in the Corona}
391:
392:
393: Figure \ref{densite} shows the LASCO-C2 electronic density radial profiles in spring 1996 averaged
394: over the same period as the SWAN data. This results in very regular curves where most
395: transient effects like CME's have been smoothed out. The two solid lines show the East limb and West limb density
396: profiles. The dashed line shows the North pole density profile and the dotted line shows the south pole density profile.
397: Within 2.5 \rso\ of the Sun, the data are contaminated by the instrument scatterred light and the uncertainty is large.
398: The profiles can only be used with confidence between 2.5 \rso\ and 6 \rso\ .
399: We see that the pole profiles have densities which are 10 times lower than the equatorial values.
400:
401: Figure \ref{vitesse} displays the velocity profiles derived from the densities in Figure \ref{densite} using equation
402: (4). The flux scaling factor $\frac{\beta_{exc}}{\sigma_{exc}(V_o)}$ is derived from the SWAN data of the same date.
403: The solid line shows the radial dependence of the solar wind as a function of distance for the average of North and
404: South poles. On the right hand we have plotted the corresponding IPS value around 600 km/s. The dotted line shows
405: the individual profiles derived for the north and south poles. They are very similar to the average profile.
406: From these profiles, we see that the solar wind goes from about
407: 200 km/s at 2.5\rso\ up to $\sim$ 600 km/s at 4.5\rso\ . After a distance of 5\rso\ , the increase of solar wind
408: velocity is less pronounced and acceleration is essentially finished at 6\rso\ .
409:
410: The dashed line shows the profile derived using the average of the two equatorial density profiles shown in figure
411: \ref{densite}. Once again the flux scaling ratio is derived from the SWAN equatorial data for the same period.
412: The corresponding IPS velocity value is shown on the right hand part of the plot and has a value close to 400 km/s.
413: We see that for the slow wind, the mean velocity at 6\rso\ is only 200 km/s, i.e. half of the terminal velocity.
414: This result is limited by the assumption of radial expansion which is implicit in equation (4).
415: Although this assumption is quite justified for a polar coronal hole above 5\rso\ or 6\rso\ ,
416: this may not be the case for the equatorial slow wind.
417: In that case a correction term should be applied to take into account the non-radial part of the slow wind expansion.
418: Furthermore, this profile concerns an average of many transient phenomena which may have had very different individual
419: velocity profiles.
420:
421: \section{Comparison with other works}
422:
423: The present work provides time-averaged velocity profiles at all latitudes. Coronal
424: flow velocities have been obtained in specific regions and at specific times in different ways.
425: The outflow velocity has been inferred from analyses of the Doppler dimming of the intensities of the OVI 1032\AA ,
426: 1037 \AA\ , HI L$\alpha$ 1216\AA\ lines, using the UVCS on board SOHO. In particular, Strachan et al. (2000)
427: derived the latitudinal dependence of O5+ outflow velocities between 1.75 \rso\ and 2.75 \rso\ .
428: Gabriel et al. (2005) derive an outflow velocity of 150-200 km/ at 2.5 $R_s$ in polar plumes.
429: Kohl et al. (1998), Cranmer et al. (1999) and Antonucci et al. (2000) show that in a polar hole, the fast wind speed
430: reaches 300-400 km/s near 3\rso\ . On the other hand, Raymond et al. use UVCS observations of a sun-grazing
431: comet at 6.8\rso\ and derive an outflow velocity smaller than 640 km/s in an high speed wind region.
432: They argued that only 50\% to 75\% of the 1 AU proton energy is reached at 6.8\rso\ and, that an additional energy
433: deposition is at work beyond this distance.
434:
435: Above streamers, Strachan et al. (2002) derived a null velocity up to 3.5\rso\ and an abrupt acceleration to 90
436: km/s at 5\rso\ along the streamer axis, while along the streamer legs the velocity has already reached 100 km/s at
437: 2.33\rso\ . Based on expanding features, Sheeley et al. (1997) estimated the acceleration of the slow wind
438: above streamers and found it compatible with a Parker-type isothermal expansion with values
439: ranging from 150 to 300 km/s between 5\rso\ and 25\rso\ .
440:
441: In the open field/fast wind regions, the geometry is extremely simple and the comparison between previous
442: localised measurements and the monthly-averaged SWAN/LASCO/IPS results can be directly performed. Indeed, there
443: is a quite good agreement with SUMER and UVCS results between 2\rso\ and 3\rso\ . The agreement suggests that the
444: fast wind acceleration is rather homogeneous and constant with time. Note that the flow speed we derived is
445: however above the maximum speed at 6.8\rso\ deduced from the Sun-grazing comet.
446:
447: For the slow wind the comparison is not so easy because it originates in and around closed-field regions
448: with a more complicated geometry. It has been shown by Strachan et al. (2002) that the profiles vary strongly
449: from the streamer axis to the legs. Since the streamers vary in angle, latitude, and shape more work is
450: needed to interpret the average profile.
451:
452: \section{Conclusion}
453:
454: In this article,
455: we have shown how we can retrieve velocity profiles in the corona between 2.5\rso\ and 6\rso\ .
456:
457: This analysis has been applied to data from 1996 and 1997. We have used monthly averages because the SWAN inversion technique
458: is based on data covering at least one solar rotation.
459: We have found that during solar minimum, there is a strong difference between the velocity profiles derived for the polar
460: and the equatorial regions of the Sun. The polar regions show a fast solar wind with lower density at 1 AU while the equatorial
461: regions are the source of the slow solar wind with increased density. This features was already well known from analysis
462: of the Interplanetary Scintillation data (Kojima and Kakinuma, 1990) or the Ulysses fast latitudinal scan of 1994
463: (Phillips et al., 1995).
464:
465: Using the LASCO density values derived at 6\rso\ , we have been able to show that the fast wind has a radial expansion with
466: constant velocity beyond 6\rso\ . This confirms the idea that the fast solar wind is accelerated at low altitudes in the corona
467: and that it has already reached its terminal velocity at 6\rso\ . This also means that the density ratio between 6\rso\ and
468: 1 AU is simply scaled by the ratio of the square of the distance to the Sun (equation 3, where $V(R) = V_o$ and R = 6 \rso\ ).
469: On a side note, we can also mention that in this case, a combination of the SWAN flux data with the LASCO density values
470: can be used to derive the velocity value at 1 AU without the need of the information given by the IPS analysis.
471:
472: The slow wind originating from equatorial regions of the Sun shows a different behaviour. First, we have seen that the density
473: ratio between 6\rso\ and 1 AU is not simply equal to the ratio of the square of the distance to the sun. This is shown
474: by computing the flux using the density and the IPS velocity and comparing to the SWAN value. We see that the SWAN value is
475: always smaller which means that the velocity ratio between the outer edge of LASCO-C2 and 1 AU, $V(6R_{\odot})/V_o$,
476: is smaller than one.
477:
478: As shown in Figure \ref{diffbeta}, the threshold between fast and slow winds is $\sim$ 500 km/s. Below that threshold, the
479: wind has not reached its terminal velocity at 6\rso\ while it has reached its terminal velocity at 6\rso\ for values larger
480: than 500 km/s. The velocity profile for the slow wind may be different from the one we have produced if the expansion factor
481: is not purely radial. This may be true at lower altitudes in the corona.
482:
483: Finally, our velocity profiles provide a test of the models of solar wind acceleration
484: to reveal which mechanisms are really effective.
485: In future developments of this work, we will extend this study to all periods of solar activity.
486: The corresponding SWAN data used to derive the solar wind flux at 1 AU have already been analysed by
487: Qu\'emerais et al. (2006).
488: This will allow us to study the variation of the acceleration profile with solar activity.
489:
490: \begin{flushleft}
491:
492: \noindent
493: {\em Acknowledgements}
494:
495: SOHO is a mission of international cooperation between ESA and NASA.
496: SWAN and LASCO-C2 activities in France are funded by CNES with support from CNRS.
497: SWAN activities in Finland are funded by TEKES and the Finnish Meteorological Institute.
498:
499: \end{flushleft}
500:
501: %{\em Please provide any other acknowledgement information.}
502:
503: \begin{thebibliography}{}
504:
505: \bibitem{aje87}
506: Ajello J.M, Stewart A.I., Thomas G.E. \& Graps A. 1987;
507: %Solar cycle study of interplanetary Lyman-alpha variations:
508: %$Pioneer-Venus$ orbiter sky background results;
509: Astrophysical Journal, 317, 964
510:
511: \bibitem{bar93}
512: Baranov V.~B. \& Malama Y.~G., 1993;
513: %Model of the solar wind interaction with the local interstellar
514: %medium - Numerical solution of self-consistent problem;
515: \jgr, 98, 15157
516:
517: %\bibitem{bar98}
518: %Baranov, Vladimir B.; Izmodenov, Vlad V.; Malama, Yury G. (1998);
519: %On the distribution function of H atoms in the problem of the solar wind interaction with
520: %the local interstellar medium;
521: %Journal of Geophysical Research, Volume 103, Issue A5, p. 9575-9586
522:
523: \bibitem{ber71}
524: Bertaux J.L., Blamont J.E., 1971;
525: %Evidence for an extra-terrestrial Lyman-alpha emission, the interstellar wind;
526: Astronomy \& Astrophysics, 11, 200.
527:
528: \bibitem{ber85}
529: Bertaux J. L., Lallement R., Kurt V.G., Mironova E.N.; 1985;
530: %Characteristics of the local interstellar hydrogen determined from
531: %$Prognoz$ 5 and 6 interplanetary Lyman-$\alpha$ line profiles measurements with a hydrogen absorption cell;
532: Astronomy \& Astrophysics, 150, 82.
533:
534: \bibitem{ber95}
535: Bertaux, J. L.; Kyr\"ol\"a, E.;
536: Qu\'emerais, E.; R.; Lallement, R.; Schmidt, W.; et al., 1995;
537: %SWAN: A Study of Solar Wind Anisotropies on SOHO with Lyman Alpha Sky Mapping;
538: Solar Physics, v. 162, p. 403-439.
539:
540: \bibitem{ber96}
541: Bertaux, J. L.; Lallement, R.; Qu\'emerais, E., 1996;
542: %UV Studies and the Solar Wind;
543: Space Science Reviews, v. 78, Issue 1/2, p. 317-328.
544:
545: \bibitem{ber97}
546: Bertaux, J. L.; Qu\'emerais, E.; Lallement, R.; Kyroela, E.; Schmidt, W.; Summanen, T.;
547: Goutail, J. P.; Berthe, M.; Costa, J.; Holzer, T., 1997;
548: %First Results from SWAN Lyman a solar wind mapper on SOHO;
549: Solar Physics, v. 175, Issue 2, p. 737-770.
550:
551: \bibitem{cra}%{1999ApJ...511..481C}
552: Cranmer, S.~R., et al.\ 1999, \apj, 511, 481
553:
554: \bibitem{dav98} %{1998A&A...336L..90D}
555: David, C., Gabriel, A.~H., Bely-Dubau, F., Fludra, A., Lemaire, P., \& Wilhelm, K.\ 1998, \aap,
556: 336, L90.
557:
558: \bibitem{gab} %{2005ApJ...635L.185G}
559: Gabriel, A.~H., Abbo, L., Bely-Dubau, F., Llebaria, A., \& Antonucci, E.\ 2005, \apjl, 635, L185
560:
561: \bibitem{iss05}
562: Issautier, K.; Perche, C.; Hoang, S.; Lacombe, C.; Maksimovic, M.; Bougeret, J.-L.; Salem, C.; 2005;
563: %Solar wind electron density and temperature over solar cycle 23: Thermal noise measurements on Wind;
564: Advances in Space Research, Volume 35, Issue 12, p. 2141-2146.
565:
566: \bibitem{izm01}
567: Izmodenov, Vladislav V.; Gruntman, Mike; Malama, Yury G., 2001;
568: %Interstellar hydrogen atom distribution function in the outer heliosphere;
569: Journal of Geophysical Research, Volume 106, Issue A6, p. 10681-10690.
570:
571: \bibitem{jos75}
572: Joselyn, J. A.; Holzer, T. E., 1975,
573: %The effect of asymmetric solar wind on the Lyman alpha sky background;
574: Journal of Geophysical Research, vol. 80, Mar. 1, 1975, p. 903-907
575:
576: \bibitem{koh98}
577: Kohl, J.~L., et al.\ 1998, \apjl, 501, L127
578:
579: \bibitem{koj90}
580: Kojima,M. and T. Kakinuma, Space Sci. Rev., 53, 173-222, 1990.
581:
582: \bibitem{koj98}%1998JGR...103.1981K
583: Kojima, M.; Tokumaru, M.; Watanabe, H.; Yokobe, A.; Asai, K.; Jackson, B. V.; Hick, P. L.; 1998;
584: Journal of Geophysical Research, vol. 103, p. 1981
585:
586: \bibitem{kohl06} %{2006A&ARv..13...31K}
587: Kohl, J.~L., Noci, G.,
588: Cranmer, S.~R., \& Raymond, J.~C.\ 2006, \aapr, 13, 31
589:
590: \bibitem{kum79}
591: Kumar, S.; Broadfoot, A. L., 1979,
592: %Signatures of solar wind latitudinal structure in interplanetary Lyman-alpha emissions -
593: %Mariner 10 observations;
594: Astrophysical Journal, Part 1, vol. 228, Feb. 15, 1979, p. 302-311.
595:
596: \bibitem{kyr} %{1998JGR...10314523K}
597: Kyr{\"o}l{\"a}, E., Summanen, T., Schmidt, W., M{\"a}kinen, T., Qu\'emerais, E., Bertaux,
598: J.-L., Lallement, R., \& Costa, J.\ 1998, \jgr, 103, 14523.
599:
600: \bibitem{lal85}
601: Lallement, R.; Bertaux, J. L.; Kurt, V. G., 1985;
602: %Solar wind decrease at high heliographic latitudes detected from Prognoz interplanetary Lyman alpha mapping;
603: Journal of Geophysical Research (ISSN 0148-0227), vol. 90, Feb. 1, 1985, p. 1413-1423.
604:
605: \bibitem{lal86} %{1986JGR....91.6751L}
606: Lallement, R., Holzer, T.~E., \& Munro, R.~H.\ 1985, \jgr, 91, 6751
607:
608: \bibitem{lal04}
609: Lallement, R.; Raymond, J. C.; Bertaux, J.-L.; Qu\'emerais, E.; Ko, Y.-K.; Uzzo, M.;
610: McMullin, D.; Rucinski, D. 2004;
611: %Solar cycle dependence of the helium focusing cone from SOHO/UVCS observations. Electron impact rates and associated pickup ions;
612: Astronomy and Astrophysics, v.426, p.867-874.
613:
614: \bibitem{lam97}
615: Lamy, P.L. et al.,
616: %Electronical Densities in Coronal Holes from LASCO-C2 Images,
617: \textit{ESA SP}-\textbf{404}, 491-494, 1997.
618:
619: \bibitem{lam02}
620: Lamy, P.; Llebaria, A.; Quemerais, E.,
621: %Solar cycle variation of the radiance and the global electron density of the solar
622: %corona,
623: Advances in Space Research, Volume 29, Issue 3, p. 373-378, 2002.
624:
625: \bibitem{lle99}
626: Llebaria, A., Lamy, P.L., Koutchmy, S.,
627: %The Global Activity of the Solar Corona,
628: \textit{ESA SP}-\textbf{446}, 441-446, 1999.
629:
630: \bibitem{mun77} %{1977ApJ...213..874M}
631: Munro, R.~H., \& Jackson, B.~V., 1977, \apj, 213, 874
632:
633: \bibitem{pry96}
634: Pryor, W. R.; Barth, C. A.; Hord, C. W.; Stewart, A. I. F.; Simmons, K. E.; Gebben, J. J.; McClintock, W. E.;
635: Lineaweaver, S.; Ajello, J. M.; Tobiska, W. K., 1996;
636: %Latitude variations in interplanetary Lyman-a data from the Galileo EUVS modeled with solar He 1083 nm images;
637: Geophysical Research Letters, Volume 23, Issue 15, p. 1893-1896.
638:
639: \bibitem{que02}%{2002A\& A...393..295Q}
640: Qu\'emerais, E.; Lamy, P., 2002,
641: %Two-dimensional electron density in the solar corona from inversion of white light images -
642: %Application to SOHO/LASCO-C2 observations;
643: Astronomy and Astrophysics, v.393, p.295-304.
644:
645: \bibitem{que06}%{2002A\& A...393..295Q}
646: Qu\'emerais, E.; Lallement, R.; Ferron, S.; Koutroumpa, D.; Bertaux, J.L.; Kyr\"ol\"a, E.; Schmidt, W., 2006,
647: %Interplanetary hydrogen absolute ionization rates: Retrieving the solar wind mass flux
648: %latitude and cycle dependence with SWAN/SOHO maps
649: \jgr, Volume 111, Issue A9, CiteID A09114
650:
651: \bibitem{ray98}
652: Raymond, John C.; Fineschi, S.; Smith, P. L.; Gardner, L.; O'Neal, R.; Ciaravella, A.; Kohl, J. L.;
653: Marsden, B.; Williams, G. V.; Benna, C.; et al.; 1998;
654: %Solar Wind at 6.8 Solar Radii from UVCS Observation of Comet C/1996Y1;
655: \apj, Volume 508, Issue 1, pp. 410-417.
656:
657: \bibitem{ruc91}
658: Rucinski, D.; Fahr, H.J.; 1991;
659: %Nonthermal ions of interstellar origin at different solar wind conditions;
660: Annales Geophysicae (ISSN 0939-4176), vol. 9, Feb. 1991, p. 102-110.
661:
662: \bibitem{stra00} %{2002ApJ...571.1008S}
663: Strachan, L., Panasyuk, A.~V., Dobrzycka, D., Kohl, J.~L., Noci G., Gibson, S.E. \& Biesecker, D.A., 2000,
664: \jgr, vol 105, A2, 2345-2356.
665:
666: \bibitem{stra02} %{2002ApJ...571.1008S}
667: Strachan, L., Suleiman, R., Panasyuk, A.~V., Biesecker, D.~A., \& Kohl, J.~L.\ 2002,
668: \apj, 571, 1008
669:
670: \bibitem{tho71}
671: Thomas G.E.; Krassa R.F.; 1971,
672: % OGO-5 measurements of the Lyman alpha sky
673: % background,
674: Astronomy \& Astrophysics, 11, 218.
675:
676: %\bibitem{tob97}
677: %Tobiska, W. K.; Pryor, W. R.; Ajello, J. M. 1997;
678: %Solar hydrogen Lyman-alpha variation during solar cycles 21 and 22;
679: %Geophysical Research Letters, Volume 24, Issue 9, p.1123.
680: %
681:
682: \bibitem{wit79}
683: Witt, N.; Blum, P. W.; Ajello, J. M., 1979;
684: %Solar wind latitudinal variations deduced from Mariner 10 interplanetary H /1216 A/ observations;
685: Astronomy and Astrophysics, vol. 73, no. 3, Mar. 1979, p. 272-281.
686:
687: \bibitem{woo00}
688: Woods, Thomas N.; Tobiska, W. Kent; Rottman, Gary J.; Worden, John R., 2000;
689: %Improved solar Lyman a irradiance modeling from 1947 through 1999 based on UARS observations;
690: Journal of Geophysical Research, Volume 105, Issue A12, p. 27195-27216.
691:
692: %\bibitem{wit04}
693: %Witte, M. 2004; Kinetic parameters of interstellar neutral helium.
694: % Review of results obtained during one solar cycle with the Ulysses/GAS-instrument;
695: %Astronomy and Astrophysics, v.426, p.835-844.
696:
697: \bibitem{zou05} %{2005ApJ...626L.117Z}
698: Zouganelis, I.,
699: Meyer-Vernet, N., Landi, S., Maksimovic, M., \& Pantellini, F.\ 2005,
700: \apjl, 626, L117
701:
702: \end{thebibliography}
703:
704: \clearpage
705:
706: \begin{figure}
707: \noindent\rotatebox{90}{\includegraphics[height=15.0cm]{f1.eps}}
708: %\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
709: \caption{ Monthly averages of the data used in this analysis. These values were obtained in March and April 1996.
710: For all panels, the x axis shows the heliographic latitude. The top left panel shows the electron density at 6\rso\
711: obtained from the LASCO-C2 polarized white light images. The density is strongly enhanced around the equatorial belt.
712: The top right panel shows the average solar wind velocity as deduced from the interplanetary scintillation data.
713: We clearly see the fast wind at high latitudes and the slow wind around the equator. The bottom left panel shows the
714: ionization rate of hydrogen at 1 AU. The diamonds joined by a dotted line shows the values derived from the SWAN data.
715: The solid line shows the charge exchange ionization found when using the densities (scaled at 1 AU) and the velocities
716: displayed in the top panels. The bottom right shows the difference between the SWAN values and the values obtained by
717: combining the LASCO-C2 and IPS data. Note the excess produced by the SWAN values at high latitudes,
718: i.e., for the fast solar wind.}
719: \label{quadri96}
720: \end{figure}
721:
722: \begin{figure}
723: \noindent\rotatebox{90}{\includegraphics[height=15.0cm]{f2.eps}}
724: %\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
725: \caption{ Same as the previous figure but for data obtained in Spring 1997. Although the numerical values are not identical,
726: we find the same dichotomy between fast wind and slow wind. As shown in the bottom right panel, the flux deduced from the IPS
727: values and and the LASCO-C2 data is too high along the equator. This means that the slow wind has not reached its
728: terminal velocity at 6\rso\ .}
729: \label{quadri97}
730: \end{figure}
731:
732: %\begin{figure}
733: %\noindent\rotatebox{90}{\includegraphics[height=15.0cm]{quadriplot1998.ps}}
734: %%\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
735: %\caption{ caption }
736: %\label{quadri98}
737: %\end{figure}
738:
739: \begin{figure}
740: \noindent\rotatebox{90}{\includegraphics[height=15.0cm]{f3.eps}}
741: %\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
742: \caption{ Difference between the SWAN total ionization rate at 1 AU and the charge exchange rate deduced from LASCO density
743: and IPS velocity data. The x-axis shows the velocity value from the IPS data. In this plot, we have used all available
744: data from 1996 and 1997, averaged monthly. The latitude bins are 10 degrees wide. Here we see the two types of wind.
745: The fast wind ($\geq$ 500 km/s) where the difference is positive and the slow wind where it is negative. The average of
746: the fast wind values give an estimate for the sum of photo-ionization and electron impact ionization rates.}
747: \label{diffbeta}
748: \end{figure}
749:
750: \begin{figure}
751: \noindent\includegraphics[height=12.0cm]{f4.eps}
752: %\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
753: \caption{ Density profiles for Spring 1996 deduced from the LASCO-C2 polarized white light images. The values correspond
754: to one-month averages. The top solid lines give the values for the East and West limbs. The two limbs are very similar
755: which is normal for an average of one month of data. The dashed line shows the north pole density profile and the dotted
756: line shows the south pole profile. There's almost a factor of 10 between polar and equatorial values.}
757: \label{densite}
758: \end{figure}
759:
760: \begin{figure}
761: \noindent\includegraphics[height=12.0cm]{f5.eps}
762: %\psfig{figure=FIGURES/figure2.eps,height=10.5cm,width=15.0cm}
763: \caption{ Velocity profiles for Spring 1996 deduced from the combination of SWAN, LASCO and IPS data. The solid line
764: represents the average of the north and south profiles while the dotted lines show the individual north and south profiles.
765: We see an inflexion around 3.5\rso\ which suggests a change in the acceleration at this altitude. The dashed line
766: shows the profile derived for the equatorial solar wind. This profile was derived by assuming a radial expansion and may
767: have to be corrected for non-radial effects. After 6\rso\ , the lines show the IPS velocity values. For both polar
768: profiles, we see that the profiles reach the terminal value around 6\rso\ , whereas at the equator the velocity at
769: 6\rso\ is only half of its terminal value.}
770: \label{vitesse}
771: \end{figure}
772:
773:
774: \clearpage
775:
776: \end{document}
777:
778: