4.2 Visible light observations

Measurements of the plasma properties in the extended corona (i.e., r ≈ 1.5 to 10 R ⊙, where the main solar wind acceleration occurs) require the bright solar disk to be occulted. The coronal emission is many orders of magnitude less bright than the emission from the solar disk, so even the vast majority of “stray light” that diffracts around the occulting edge must be eliminated. Visible-light coronagraphs that combine stray-light rejection with linear polarimetry have the ability to measure the Thomson-scattered polarization brightness (pB) in the corona. The use of pB rather than the total coronal brightness eliminates the contribution from the dust-scattered F-corona, which is believed to be unpolarized up to distances of about 5 R ⊙. Because the coronal plasma is optically thin to the Thomson-scattered photons, pB is proportional to the line-of-sight integral of the electron density ne, multiplied by a known scattering function. Methods for inverting this integral to derive ne as a function of position in various coronal structures have been developed and improved over the years (e.g., van de Hulst, 1950Altschuler and Perry, 1972Munro and Jackson, 1977Guhathakurta and Holzer, 1994Frazin et al., 2007). For coronal holes, the LASCO (Large Angle and Spectrometric Coronagraph) instrument on SOHO has also been used to probe the superradial expansion of open magnetic flux tubes (DeForest et al., 1997Jump To The Next Citation Point2001Jump To The Next Citation Point) and the evolution of transient polar jets (Wang et al., 1998Jump To The Next Citation PointWood et al., 1999). The White Light Coronagraphs on Spartan 201 (Fisher and Guhathakurta, 1995Jump To The Next Citation Point) and on the UVCS (Ultraviolet Coronagraph Spectrometer) instrument aboard SOHO (e.g., Kohl et al., 1995Jump To The Next Citation PointRomoli et al., 2002) have provided electron densities between 1.5 and 5R ⊙ in coronal holes.

Figure 5View Image illustrates a selection of visible-light measurements of the electron density in coronal holes and compares them to similar measurements of streamers and to a semi-empirical model of the chromosphere, transition region, and low corona (Avrett and Loeser, 2008Jump To The Next Citation Point). The blue coronal hole curves were adapted from the results of Fisher and Guhathakurta (1995Jump To The Next Citation Point) (dotted), Cranmer et al. (1999bJump To The Next Citation Point) (solid), Doyle et al. (1999) (dashed), and Guhathakurta et al. (1999a) (dot-dashed). The red curves for equatorial helmet streamers were adapted from the results of Sittler Jr and Guhathakurta (1999) (solid) and Gibson et al. (1999) (dashed).

View Image

Figure 5: Comparison of empirically determined densities in the upper solar atmosphere. Avrett and Loeser (2008Jump To The Next Citation Point) values of electron number density (solid black curve) and total hydrogen number density (dot-dashed black curve) are compared with various visible-light pB electron number densities for coronal holes (blue curves) and streamers (red curves); see text for details.

Note that streamers are denser than coronal holes by about a factor of 10, but the hole measurements themselves can often exhibit variations in the electron density by factors of the order of 2 – 3. Much of this spread is likely to be the result of different lines of sight passing through regions that contain varying numbers of polar plumes (see, e.g., Cranmer et al., 1999bJump To The Next Citation Point2008Jump To The Next Citation Point). Some of the cited pB observations were optimized to avoid bright concentrations of plumes, and others have been purposefully averaged over the full range of coronal hole substructure. It is also possible that absolute calibration uncertainties may still exist between the different instruments used to determine pB and ne, and this could compound the reported range of variation in coronal hole electron densities.

For a steady-state solar wind outflow, the conservation of mass demands that the product of density, flow speed, and cross-sectional area of the flux tube remain constant. Thus, if the magnetic geometry and the electron density are known, mass conservation allows the solar wind outflow speed to be computed. Kohl et al. (2006Jump To The Next Citation Point) used the representative values of ne shown in Figure 5View Image together with a range of estimates for the superradial flux-tube expansion of coronal holes to determine outflow speeds in coronal holes. Figure 41a of Kohl et al. (2006Jump To The Next Citation Point) illustrates the result of this process, which shows a large range of values reflecting the uncertainties in both n e and the flux-tube area factor. Despite these uncertainties, though, the electron densities that became available in the 1990s demonstrated that the fast solar wind accelerates rapidly in coronal holes – probably reaching half of its asymptotic terminal speed (u∞ ≈ 700 –800 kms −1) by heights no larger than 2 –4 R⊙.

The increased sensitivity of the LASCO instrument over earlier coronagraphs revealed a nearly continual release of low-contrast density inhomogeneities, or “blobs,” from the cusps of helmet streamers (Sheeley Jr et al., 1997Jump To The Next Citation PointTappin et al., 1999Wang et al., 2000Chen et al., 2009). These features were seen to accelerate up to speeds of order 300 – 400 km s–1 by the time they reached the outer edge of the LASCO field of view (r ≈ 30 R ⊙); see also Figure 8View Image below. The blobs are typically only about 10% to 15% brighter or dimmer than the surrounding streamer material. Because of this low contrast, these features do not appear to comprise a large fraction of the mass flux of the slow solar wind. However, it is still unclear whether blobs are passive “tracers” that flow with the solar wind speed, or whether they are wavelike fluctuations that propagate relative to the background solar wind reference frame. This diagnostic tool has been much more difficult to apply in coronal holes than it has in the bright streamers, so no firm measurements of the fast wind acceleration yet exist from this technique.

Visible light measurements have also revealed evidence for compressive magnetohydrodynamic (MHD) waves that propagate along open field lines in coronal holes. Intensity oscillations measured by the UVCS and EIT instruments on SOHO were found to have periodicities between about 10 and 30 minutes and are consistent with being upwardly propagating slow-mode magnetosonic waves (DeForest and Gurman, 1998Ofman et al., 19992000). The relative amplitude of the density fluctuations (δn∕n0) for these waves was found to range between about 0.03 and 0.15 (see Cranmer, 2004aJump To The Next Citation Point). This is consistent with measurements of the density fluctuation amplitudes made at larger distances via radio scintillations (Spangler, 2002Jump To The Next Citation Point) and in situ instruments (Tu and Marsch, 1994). There have also been claims that low-frequency oscillations have been measured in H i Lyα emission (Morgan et al., 2004Bemporad et al., 2008Telloni et al., 2009). In these cases, however, it is extremely important to take into account all of the relevant instrumental effects. These measurements still appear to be provisional.

As seen in Section 2 above, coronal holes have long been observed as the sites of thin, ray-like polar plumes. The earliest measurements of polar plume properties were made in broad-band visible light, and these dense strands are often seen to stand out distinctly from the ambient interplume corona. It is not clear, though, to what extent off-limb observations (which integrate over long optically thin lines of sight) ever capture only the “pure” plume or interplume plasmas. Space-based observations from, e.g., Spartan 201 and SOHO improved our ability to measure the physical properties in and between plumes (e.g., Fisher and Guhathakurta, 1995DeForest et al., 1997Cranmer et al., 1999bJump To The Next Citation PointDeForest et al., 2001Jump To The Next Citation Point). Although the brightest plumes are still discernible at the uppermost heights observed by LASCO (i.e., 30 –40 R ⊙), the plume/interplume density contrast becomes too low to measure clearly in interplanetary space (r > 60R ⊙). However, indirect and possibly plume-related signatures in the in situ data have been reported by Thieme et al. (1990), Reisenfeld et al. (1999), and Yamauchi et al. (2002). The disappearance of plumes probably is the result of some combination of transverse pressure balance (i.e., dense plumes expanding to fill more of the available volume; see Del Zanna et al., 1998) and MHD instabilities that can mix the two components (Parhi et al., 1999Andries and Goossens, 2001).

  Go to previous page Go up Go to next page