The effect of newly erupting flux on the polar coronal holes

He i 10830 Å images show that early in sunspot cycles 21 and 22, large bipolar magnetic regions strongly affected the boundaries of the nearby polar coronal holes. East of each eruption, the hole boundary immediately contracted poleward, leaving a band of enhanced helium network. West of the eruption, the boundary remained diffuse and gradually expanded equatorward into the leading, like-polarity part of the bipolar magnetic region. Comparisons between these observations and simulations based on a current-free coronal model suggest that:

1. The Sun's polar magnetic fields are confined to relatively small caps of high average field strength, apparently by a poleward meridional flow.
2. The enhanced helium network at high latitude marks the location of relatively strong polar fields that have become linked to the newly erupted bipolar region in that hemisphere.
3. The distortion of the polar-hole boundary is accompanied by a corresponding distortion of the equatorial neutral sheet in the outer corona, in which the amount of warping depends on the magnitude of the erupted flux relative to the strength of the Sun's polar magnetic fields.

     

Magnetohydrodynamic tube waves and high speed solar wind streams

It has been widely conjectured that magnetohydrodynamic (MHD) waves may provide the extra momentum or energy required to explain the high speed solar wind streams that originate in coronal holes. Although the magnetic structuring inherent in this problem has been incorporated into models of the bulk flow, this is not generally true of the associated treatments of wave propagation. In particular, as pointed out by Davila (1985), we might generally expect the magnetic geometry to substantially modify those waves whose wavelength is comparable to the hole width. Using both a geometrical optics and an eigenmode approach, Davila addressed the question of wave propagation in a simple uniform width flux slab model of a coronal hole and concluded

1. the hole may act as a ‘leaky wave guide’, i.e., waves travelling along it may leak into the surrounding corona, but
2. the group velocity of waves with periods in a physically relevant range (around 100 s) is downward, indicating that such waves cannot carry energy into the solar wind and therefore cannot be driving it.

We agree with (i) but argue that (ii) results from a mistaken interpretation of a dispersion relation, and is incorrect. Furthermore, we apply the cylindrical tube leaky wave approach of Cally (1986) to a simple coronal hole model, and find two wavetypes with substantial upward energy fluxes. However, of these, we argue that the so-called ‘trig modes’ (geometry modified fast waves) leak so profusely that they are unable to transport energy over the distance required; the non-axisymmetric ‘thin tube’ modes, though, do not suffer from this disability.     

A survey of coronal holes and their solar wind associations throughout sunspot cycle 20

To gain insight into the relationships between solar activity, the occurrence and variability of coronal holes, and the association of such holes with solar wind features such as high-velocity streams, a study of the period 1963–1974 was made. This period corresponds approximately with sunspot cycle 20. The primary data used for this work consisted of X-ray and XUV solar images obtained from rockets. The investigation revealed that:

1. The polar coronal holes prominent at solar minimum, decreased in area as solar activity increased and were small or absent at maximum phase. This evolution exhibited the same phase difference between the two hemispheres that was observed in other indicators of activity.
2. During maximum, coronal holes occurred poleward of the sunspot belts and in the equatorial region between them. The observed equatorial holes were small and persisted for one or two solar rotations only; some high latitude holes had lifetimes exceeding two solar rotations.
3. During 1963–74 whenever XUV or X-ray images were available, nearly all recurrent solar wind streams of speed ?500 km s^?1 were found associated with coronal holes at less than 40° latitude; however some coronal holes appeared to have no associated wind streams at the Earth.

     

Coronal holes,solar wind streams,and geomagnetic activity during the new sunspot cycle

We have extended our previous study of coronal holes, solar wind streams, and geomagnetic disturbances from the declining phase (1973–1975) of sunspot cycle 20 through sunspot minimum (1976) into the rising phase (1977) of cycle 21. Using daily He I 10830 Å spectroheliograms and photospheric magnetograms, we found the following results:

1. As the magnetic field patterns changed, the solar atmosphere evolved from a structure having a few, large, long-lived, low-latitude coronal holes to one having numerous small, short-lived, high-latitude holes (in addition to the polar holes which persisted throughout this 5-year interval).
2. The high-latitude holes recurred with a synodic rotation period of 28–29 days instead of the 27-day period already known to be characteristic of low-latitude holes.
3. During 1976–1977 many coronal holes were intrinsically ‘weak’ in the sense that their average intensities did not differ greatly from the intensity of their surroundings. Such low-contrast holes were rare during 1973–1975.

An updated Bartels display of the occurrence of holes, wind speed, and geomagnetic activity summarizes the evolution of their characteristics and interrelations as the sunspot cycle has progressed. Long-lived, low-latitude holes have become rare but remain terrestrially effective. The more common high-latitude holes are effective only when the Earth lies at a relatively high heliographic latitude in the same solar hemisphere.     

Keynote address: Outstanding problems in solar physics

Celebrating the diamond jubilee of the Physics Research Laboratory (PRL) in Ahmedabad, India, we look back over the last six decades in solar physics and contemplate on the ten outstanding problems (or research foci) in solar physics:

1. The solar neutrino problem
2. Structure of the solar interior (helioseismology)
3. The solar magnetic field (dynamo, solar cycle, corona)
4. Hydrodynamics of coronal loops
5. MHD oscillations and waves (coronal seismology)
6. The coronal heating problem
7. Self-organized criticality (from nanoflares to giant flares)
8. Magnetic reconnection processes
9. Particle acceleration processes
10. Coronal mass ejections and coronal dimming

The first two problems have been largely solved recently, while the other eight selected problems are still pending a final solution, and thus remain persistent Challenges for Solar Cycle 24, the theme of this jubilee conference.     

Correlation and spectral analysis of daily solar radio flux

Correlation and spectral analysis of solar radio flux density and sunspot number near the maximum of the sunspot cycle has indicated the existence of

1. long period amplitude modulation of the slowly varying component (SVC) of radio emission
2. coronal storage over a period of the order of three solar rotations
3. fast decay (one solar rotation period or less) of gyromagnetic emissions from radio sources
4. shift in location of chromospheric sources compared to those of either the upper corona or the photosphere.

     

Solar-interplanetary modeling: 3-D solar wind solutions in prescribed non-radial magnetic field geometries

A model is presented which describes the 3-dimensional non-radial solar wind expansion between the Sun and the Earth in a specified magnetic field configuration subject to synoptically observed plasma properties at the coronal base. In this paper, the field is taken to be potential in the inner corona based upon the Mt. Wilson magnetograph observations and radial beyond a certain chosen surface. For plasma boundary conditions at the Sun, we use deconvoluted density profiles obtained from synopticK-coronameter brightness observations. The temperature is taken to be 2 × 10⁶ K at the base of closed field lines and 1.6 x 10⁶K at the base of open field lines. For a sample calculation, we employ data taken during the period of the 12 November 1966 eclipse. Although qualitative agreement with observations at 1 AU is obtained, important discrepancies emerge which are not apparent from spherically symmetric models or those models which do not incorporate actual observations in the lower corona. These discrepancies appear to be due to two primary difficulties - the rapid geometric divergence of the open field lines in the inner corona as well as the breakdown in the validity of the Spitzer heat conduction formula even closer to the Sun than predicted by radial flow models. These two effects combine to produce conductively dominated solutions and lower velocities, densities, and field strengths at the Earth than those observed. The traditional difficulty in solar wind theory in that unrealistically small densities must be assumed at the coronal base in order to obtain observed densities at 1 AU is more than compensated for here by the rapid divergence of field lines in the inner corona. For these base conditions, the value ofβ(ratio of gas pressure to magnetic pressure) is shown to be significantly greater than one over most of the lower corona - suggesting that, for the coronal boundary conditions used here, the use of a potential or force-free magnetic field configuration may not be justified. The calculations of this paper point to the directions where future research on solar-interplanetary modelling should receive priority:

1. better models for the coronal magnetic field structure
2. improved understanding of the thermal conductivity relevant for the solar wind plasma.

     

Alfvén waves in the solar atmosphere

We examine the propagation of Alfvén waves in the solar atmosphere. The principal theoretical virtues of this work are: (i) The full wave equation is solved without recourse to the small-wavelength eikonal approximation (ii) The background solar atmosphere is realistic, consisting of an HSRA/VAL representation of the photosphere and chromosphere, a 200 km thick transition region, a model for the upper transition region below a coronal hole (provided by R. Munro), and the Munro-Jackson model of a polar coronal hole. The principal results are:

1. If the wave source is taken to be near the top of the convection zone, where n _H = 5.2 × 10¹⁶ cm^?3, and if B _⊙ = 10.5 G, then the wave Poynting flux exhibits a series of strong resonant peaks at periods downwards from 1.6 hr. The resonant frequencies are in the ratios of the zeroes of J ₀, but depend on B _⊙, and on the density and scale height at the wave source. The longest period peaks may be the most important, because they are nearest to the supergranular periods and to the observed periods near 1 AU, and because they are the broadest in frequency.
2. The Poynting flux in the resonant peaks can be large enough, i.e. P _⊙ ≈ 10⁴–10⁵ erg cm^?2s^?1, to strongly affect the solar wind.
3. ¦δv¦ and ¦δB¦ also display resonant peaks.
4. In the chromosphere and low corona, ¦δv ≈ 7–25 kms^?1 and ¦δB¦ ≈0.3–1.0 G if P _⊙≈10⁴-10⁵ erg cm^?2s^?1.
5. The dependences of ¦δv¦ and ¦δB¦ on height are reduced by finite wavelength effects, except near the wave source where they are enhanced.
6. Near the base, ¦δB¦ ≈ 350–1200 G if P _⊙ ～- 10⁴–10⁵. This means that nonlinear effects may be important, and that some density and vertical velocity fluctuations may be associated with the Alfvén waves.
7. Below the low corona most wave energy is kinetic, except near the base where it becomes mostly magnetic at the resonances.
8. ?₀ < δv ² > v _A or < δB ² > v _A/4π are not good estimators of the energy flux.
9. The Alfvén wave pressure tensor will be important in the transition region only if the magnetic field diverges rapidly. But the Alfvén wave pressure can be important in the coronal hole.

     

Motions and magnetic fields in filaments

Based on the developed method of jointly using data on the magnetic fields and brightness of filaments and coronal holes (CHs) at various heights in the solar atmosphere as well as on the velocities in the photosphere, we have obtained the following results:

The upward motion of matter is typical of filament channels in the form of bright stripes that often surround the filaments when observed in the HeI 1083 nm line.

The filament channels observed simultaneously in Hα and HeI 1083 nm differ in size, emission characteristics, and other parameters. We conclude that by simultaneously investigating the filament channels in two spectral ranges, we can make progress in understanding the physics of their formation and evolution.

Most of the filaments observed in the HeI 1083 nm line consist of dark knots with different velocity distributions in them. A possible interpretation of these knots is offered.

The height of the small-scale magnetic field distribution near the individual dark knots of filaments in the solar atmosphere varies between 3000 and 20000 km.

The zero surface separating the large-scale magnetic field structures in the corona and calculated in the potential approximation changes the inclination to the solar surface with height and is displaced in one or two days.

The observed formation of a filament in a CH was accompanied by a significant magnetic field variation in the CH region at heights from 0 to 30000 km up to the change of the predominant field sign over the entire CH area. We assume that this occurs at the stage of CH disappearance.

     

The Alfvén-wave theory of solar flares

Evidence is discussed showing that a representative solar flare event comprises three or more separate but related phenomena requiring separate mechanisms. In particular it is possible to separate the most energetic effect (the interplanetary blast) from the thermal flare and from the rapid acceleration of particles to high energies. The phenomena are related through the magnetic structure characteristic of a composite flare event, being a bipolar surface field with most of its field lines ‘closed’. Of primary importance are helical twists on all scales, starting with the ‘flux rope’ of the spot pair which was fully twisted before it emerged. Subsequent untwisting by the upward propagation of an Alfvén twist wave provides the main flare energy.

1. The interplanetary blast model is based on subsurface, helically twisted flux ropes which erupt to form spots and then transfer their twists and energy by Alfvén-twist waves into the atmospheric magnetic fields. The blast is triggered by the prior-commencing flash phase or by a coronal wave.
2. The thermal flare is explained in terms of Alfvén waves travelling up numerous ‘flux strands’ (Figure 3) which have frayed away from the two flux ropes. The waves originate in interaction (collisions, bending, twisting, rubbing) between subsurface flux strands; the sudden flash is caused by a collision. The classical twin-ribbon flare results from the collision of a flux rope with a tight bunch of S-shaped flux strands.
3. The impulsive acceleration of electrons (hard X-ray, EUV, Hα and radio bursts) is tentatively attributed to magnetic reconnection between fields in two parallel, helically twisted flux strands in the low corona.
4. Flare (Moreton) waves in the corona have the same origin as the interplanetary blast. Sympathetic flares represent only the start of enhanced activity in a flare event already in the slow phase. Filament activation also occurs during the slow phase as twist Alfvén waves store their energy in the atmosphere.
5. Flare ejecta are caused by Alfvén waves moving up flux strands. Surges are attributed to packets of twist Alfvén waves released into bundles of flux strands; the waves become non-linear and drive plasma upwards. Spray-type prominences result from accumulations of Alfvén wave energy in dome-shaped fields; excessive energy density eventually explodes the field.

     

Inhibition of conductive heat flow by magnetic constriction in the corona and transition region: Dependence on the shape of the constriction

The magnetic field that fills the corona is rooted in a small fraction of the solar surface. The consequent constriction of the field lines inhibits the conduction of heat down from the corona, thereby strongly affecting the energy balance in the corona and transition region. In this paper, we clarify how the shape of the constriction acts together with the amount of constriction to inhibit the heat flow. We analyze the heat flow in model tapered flux tubes in which the plasma properties are constant on cross sections, the plasma is static, and the only energy transfer is by thermal conduction. We find:

1. From the general solution to the model, only two particular solutions are readily applied to the solar atmosphere. One is the steady-state case, appropriate for quiet regions and active regions which are not flaring; the other is the time-dependent case in which no heat enters the hot end, appropriate for conductive cooling of flare loops.
2. In the steady-state case, the inhibition factor Φ/Φ _P (the ratio of the heat flow through a constricted flux tube to the heat flow through the otherwise equivalent unconstricted tube) is simply the ratio of the harmonic mean area of the constricted tube to the area of the unconstricted tube. The inhibition of heat flow thus results from the amount and shape of the constriction.
3. For any given shape and amount of monotonic tube taper, the inhibition factor in the steady-state case is a good estimate of that in the time-dependent case (within a factor of 2).
4. The amount of constriction Γ (the ratio of the flux tube's hot end area to its cold end area) limits the range of possible values of the inhibition factor; the value of the inhibition factor within this range is set by the shape of the constriction. For the steady-state case, Γ ^?1 ≤ Φ/Φ _P ≤ 1. Therefore, the shape of the constriction is as important as the amount of constriction in determining the amount of inhibition.
5. For the linear taper of a cone-shaped tube, the inhibition factor is an especially simple function of the amount of constriction: Φ/Φ _P = Γ ^?1/2. This inhibition-constriction relation for the cone correctly estimates the inhibition to order of magnitude for any tube in which the constriction occurs gradually all along the tube. If the constriction is more concentrated to the hot (cold) end of the tube than in a cone, then the inhibition is greater (less) than in a cone.
6. Because most flux tubes that pass from the corona down through the transition region should not differ greatly from a cone in shape of constriction, the simple inhibition-constriction relation for the cone should correctly estimate the magnitude of the inhibition in quiet regions, active regions, and flare loops. The only observational input needed for this estimate is the amount of magnetic constriction through the transition region.

     

Models of the open solar atmosphere

McWhirter et al. (1975) have presented a standard model for the transition region and inner corona that matches with the Harvard Smithsonian Reference Atmosphere. They assume an open field line configuration and solve numerically the equations of energy and hydrostatic equilibrum. The purpose of the present paper is to generalise their model for the temperature and density as functions of height in several ways and, in particular, to determine the temperature maxium and its location. The effect of varying the following characteristics of the model is determined:

1. Boundary conditions on temperature and density;
2. magnitude of the heating;
3. form of the heating term;
4. divergence of the field lines;
5. presence of subsonic flows, either upward or downward.

If the heating is localised at great altitudes, it tends to produce a narrower and larger temperature maximum at a greater altitude than a uniform heating and even more so than a heating proportional to density. For fixed base conditions, an increase in heating or field line divergence or downflow decreases the coronal temperature and reduces the height of the temperature maximum, while a steady upflow has the opposite effects. A maximum possible upflow was found, beyond which a catastrophe occurs so that no steady hot solution exists.     

A concentric ellipse multiple-arch system in the solar corona

A typical concentric ellipse multiple-arch system was observed in the solar corona during the February 4, 1962 eclipse in New Guinea. The following results have been obtained from analysis of a white-light photograph taken by N. Owaki (see Owaki and Saito, 1967a).

1. The arches are composed of four equidistant components, elliptical in shape, and almost concentric with a prominence at the common center of the ellipses.
2. The prominence and arch system appears to be the lower region of a helmet-shaped streamer.
3. The widths of the arches are observed to increase with height.
4. Analysis was made in the light of three models for the coronal structures that could lead to the observed arches: (a) rod-like concentrations of electrons; (b) tunnel-shaped elliptical shells of electrons; and (c) dome-like ellipsoidal shells of electrons. Electron densities are derived for the models, and the dome-like model is excluded as a possibility for arch systems exhibiting a coronal cavity.
5. The scale height in the arch-streamer region is found to be almost the same as that of the K-corona, suggesting equal temperatures, density distributions, etc. in each region.
6. There is a dark space (a coronal cavity) between the innermost arch and the prominence. The brightness of this cavity is 1/5 that of the adjacent arch. It is 3% brighter than the background corona of the arch-streamer system.
7. A comparison is made between the deficiency of electrons in the coronal cavity and the excess of electrons in the prominence. It is found that the ratio of the excess to the deficiency lies between 0.9 and 40.
8. A comparison between the electron efflux from the ‘leaky magnetic bottle’ possibly formed by rod-shaped coronal arches and the electron influx into those arches from the chromosphere leads us to the conclusion that the rod model is probably valid and that spicules appear to be an adequate supply for the electrons observed in the arches. The tunnel model may be valid, but in that case spicules are probably not the sources of the electrons observed in coronal arches.

     

Relative Velocities and Linewidths in a Coronal Hole and Outside

Relative Doppler velocities and spectral linewidths in a coronal hole and in the quiet Sun region outside have been obtained from Solar and Heliospheric Observatory (SOHO)/Coronal Diagnostic Spectrometer (CDS) observations. Five strong emission lines in the CDS wavelength range (namely, O? iii 599 Å, O?v 630 Å, Ne?vi 562.8 Å, He?ii 304 Å, and Mg?ix 368 Å), whose formation temperatures represent different heights in the solar atmosphere from the lower transition region to the inner corona, have been used in the study. As reported earlier, relative velocities in the coronal hole are generally blueshifted with respect to the quiet Sun, and the magnitude of the blueshifts increases with height. It has been found that the polar coronal hole has larger relative velocities than the equatorial extension in the inner corona. Several localized velocity contours have been found mainly on network brightenings and in the vicinity of the coronal hole boundary. The presence of velocity contours on the network may represent network outflows whereas the latter could be due to localized jets probably arising from magnetic reconnection at the boundary. All spectral lines have larger widths in the coronal hole than in the quiet Sun. In O?v 630 Å an extended low-linewidth region is seen in the coronal hole?–?quiet Sun boundary, which may indicate fresh mass transfer across the boundary. Also polar coronal holes have larger linewidths in comparison with the equatorial extension. Together with larger relative velocities, this suggests that the solar wind emanating from polar hole regions is faster than that from equatorial hole regions.     

Flares and changing magnetic fields

An observational study of maps of the longitudinal component of the photospheric fields in flaring active regions leads to the following conclusions:

1. The broad-wing Hα kernels characteristic of the impulsive phase of flares occur within 10″ of neutral lines encircling features of isolated magnetic polarity (‘satellite sunspots’).
2. Photospheric field changes intimately associated with several importance 1 flares and one importance 2B flare are confined to satellite sunspots, which are small (10″ diam). They often correspond to spot pores in white-light photographs.
3. The field at these features appears to strengthen in the half hour just before the flares. During the flares the growth is reversed, the field drops and then recovers to its previous level.
4. The magnetic flux through flare-associated features changes by about 4 × 10¹⁹ Mx in a day. The features are the same as the ‘Structures Magnétiques Evolutives’ of Martres et al. (1968a).
5. An upper limit of 10²¹ Mx is set for the total flux change through McMath Regions 10381 and 10385 as the result of the 2B flare of 24 October, 1969.
6. Large spots in the regions investigated did not evince flux changes or large proper motions at flare time.
7. The results are taken to imply that the initial instability of a flare occurs at a neutral point, but the magnetic energy lost cannot yet be related to the total energy of the subsequent flare.
8. No unusual velocities are observed in the photosphere at flare time.

     

Observations of moving magnetic features near sunspots

The properties of small (< 2″) moving magnetic features near certain sunspots are studied with several time series of longitudinal magnetograms and Hα filtergrams. We find that the moving magnetic features:

1. Are associated only with decaying sunspots surrounded entirely or in part by a zone without a permanent vertical magnetic field.
2. Appear first at or slightly beyond the outer edge of the parent sunspot regardless of the presence or absence of a penumbra.
3. Move approximately radially outward from sunspots at about 1 km s^?1 until they vanish or reach the network.
4. Appear with both magnetic polarities from sunspots of single polarities but appear with a net flux of the same sign as the parent sunspot.
5. Transport net flux away from the parent sunspots at the same rates as the flux decay of the sunspots.
6. Tend to appear in opposite polarity pairs.
7. Appear to carry a total flux away from sunspots several times larger than the total flux of the sunspots.
8. Produce only a very faint emmission in the core of Hα.

A model to help understand the observations is proposed.     

Rotational Tomography For 3d Reconstruction Of The White-Light And Euv Corona In The Post-Soho Era

Improving our understanding of the mechanisms that energize the solar wind and heat structures in the solar corona requires the development of empirical methods that can determine the three-dimensional (3D) temperature and density distributions with as much spatial and temporal resolution as possible. This paper reviews the solar rotational tomography (SRT) methods that will be used for 3D reconstruction of the solar corona from data obtained by the next generation of space-based missions such as the Solar and Terrestrial Relations Observatory (STEREO), Solar-B and the Solar Dynamics Observatory (SDO). In the next decade, SRT will undergo rapid advancement on several frontiers of 3D image reconstruction:

1. Electron density reconstruction from white-light coronagraph images.
2. Differential emission measure (DEM) reconstruction from EUV images.
3. Dual-spacecraft (STEREO) observing geometry.
4. Fusion of data from multiple spacecraft with differing instrumentation.
5. Time-dependent estimation methods.

Although the principles described apply to many different wavelength regimes, this paper concentrates on white-light and EUV data. Previous work on all of these subjects is reviewed, and major technical issues and future directions are discussed.     

Coronal Holes and Open Magnetic Flux over Cycles 23 and 24

As the observational signature of the footprints of solar magnetic field lines open into the heliosphere, coronal holes provide a critical measure of the structure and evolution of these lines. Using a combination of Solar and Heliospheric Observatory/Extreme ultraviolet Imaging Telescope (SOHO/EIT), Solar Dynamics Observatory/Atmospheric Imaging Assembly (SDO/AIA), and Solar Terrestrial Relations Observatory/Extreme Ultraviolet Imager (STEREO/EUVI A/B) extreme ultraviolet (EUV) observations spanning 1996?–?2015 (nearly two solar cycles), coronal holes are automatically detected and characterized. Coronal hole area distributions show distinct behavior in latitude, defining the domain of polar and low-latitude coronal holes. The northern and southern polar regions show a clear asymmetry, with a lag between hemispheres in the appearance and disappearance of polar coronal holes.     

On the latitude migration of polar faculae in the solar activity cycle period: 1970–1978

Coordinates of polar faculae have been measured and processed using daily photoheliograms of the Kislovodsk Station of the Pulkovo observatory with the final goal of studying their latitude distribution during the solar cycles 20–21. The results obtained are as follows:

1. The first polar faculae emerge immediately after the polarity inversion of the solar magnetic field at the latitudes from 40° to 70° with the average ?-55°.
2. The zone of the emergence of polar faculae migrates poleward during the period between the neighbouring polarity inversions of the solar magnetic field. This migration is about 20° for 8 years, which corresponds to a velocity of 0.5 m s^-1.
3. The maximum number of polar faculae was reached at the activity minimum (1975–1976).
4. The last polar faculae were observed in the second half of 1978 at the latitudes from 70° to 80°.

     

Evidence for a poleward meridional flow on the sun

We define for observational study two subsets of all polar zone filaments, which we call polemost filaments and polar filament bands. The behavior of the mean latitude of both the polemost filaments and the polar filament bands is examined and compared with the evolution of the polar magnetic field over an activity cycle as recently distilled by Howard and LaBonte (1981) from the past 13 years of Mt. Wilson full-disk magnetograms. The magnetic data reveal that the polar magnetic fields are built up and maintained by the episodic arrival of discrete f-polarity regions that originate in active region latitudes and subsequently drift to the poles. After leaving the active-region latitudes, these unipolar f-polarity regions do not spread equatorward even though there is less net flux equatorward; this indicates that the f-polarity regions are carried poleward by a meridional flow, rather than by diffusion. The polar zone filaments are an independent tracer which confirms both the episodic polar field formation and the meridional flow. We find:

1. The mean latitude of the polemost filaments tracks the boundary of the polar field cap and undergoes an equatorward dip during each arrival of additional polar field.
2. Polar filament bands track the boundary latitudes of the unipolar regions, drifting poleward with the regions at about 10 m s^-1.
3. The Mt. Wilson magnetic data, combined with a simple model calculation, show that the filament drift expected from diffusion alone would be slower than observed, and in some cases would be equatorward rather than poleward.
4. The observation that filaments drift poleward along with the magnetic regions shows that fields of both polarities are carried by the meridional flow, as would be expected, rather than only the f-polarity flux which dominates the strength. This leads to the prediction that in the mid-latitudes during intervals between the passage of f-polarity regions, both polarities are present in nearly equal amounts. This prediction is confirmed by the magnetic data.