I. González Hernández

SOLAR SUBSURFACE FLUID DYNAMICS DESCRIPTORS DERIVED FROM GLOBAL OSCILLATION NETWORK GROUP AND MICHELSON DOPPLER IMAGER DATA

We analyze Global Oscillation Network Group (GONG) and Michelson Doppler Imager (MDI) observations obtained during Carrington rotation 1988 (2002 March 30-April 26) with a ring-diagram technique in order to measure the zonal and meridional flow components in the upper solar convection zone. We derive daily flow maps over a range of depths up to 16 Mm on a spatial grid of 75 in latitude and longitude covering ±60° in latitude and central meridian distance and combine them to make synoptic flow maps. We begin exploring the dynamics of the near-surface layers and the interaction between flows and magnetic flux by deriving fluid dynamics descriptors such as divergence and vorticity from these flow maps. Using these descriptors, we derive the vertical velocity component and the kinetic helicity density. For this particular Carrington rotation, we find that the vertical velocity component is anticorrelated with the unsigned magnetic flux. Strong downflows are more likely associated with locations of strong magnetic activity. The vertical vorticity is positive in the northern hemisphere and negative in the southern hemisphere. At locations of magnetic activity, we find an excess vorticity of the same sign as that introduced by differential rotation. The vertical gradient of the zonal flow is mainly negative except within 2 Mm of the surface at latitudes poleward of about 20°. The zonal-flow gradient appears to be related to the unsigned magnetic flux in the sense that locations of strong activity are also locations of large negative gradients. The vertical gradient of the meridional flow changes sign near about 7 Mm, marking a clear distinction between near-surface and deeper layers. GONG and MDI data show very similar results. Differences occur mainly at high latitudes, especially in the northern hemisphere, where MDI data show a counter cell in the meridional flow that is not present in the corresponding GONG data.

Subsurface Meridional Circulation in the Active Belts

Temporal variations of the subsurface meridional flow with the solar cycle have been reported by several authors. The measurements are typically averaged over periods of time during which surface magnetic activity existed in the regions where the velocities are calculated. The present work examines the possible contamination of these measurements due to the extra velocity fields associated with active regions plus the uncertainties in the data obtained where strong magnetic fields are present. We perform a systematic analysis of more than five years of GONG data and compare meridional flows obtained by ring-diagram analysis before and after removing the areas of strong magnetic field. The overall trend of increased amplitude of the meridional flow towards solar minimum remains after removal of large areas associated with surface activity. We also find residual circulation toward the active belts that persists even after the removal of the surface magnetic activity, suggesting the existence of a global pattern or longitudinally-located organized flows.

MERIDIONAL CIRCULATION DURING THE EXTENDED SOLAR MINIMUM: ANOTHER COMPONENT OF THE TORSIONAL OSCILLATION?

We show here a component of the meridional circulation developing at medium-high latitudes (40°-50°) before the new solar cycle starts. Like the torsional oscillation of the zonal flows, this extra circulation seems to precede the onset of magnetic activity at the solar surface and moves slowly toward lower latitudes. However, the behavior of this component differs from that of the torsional oscillation regarding location and convergence toward the equator at the end of the cycle. The observation of this component before the magnetic regions appear at the solar surface has only been possible due to the prolonged solar minimum. The results could settle the discussion as to whether the extra component of the meridional circulation around the activity belts, which has been known for some time, is or is not an effect of material motions around the active regions.

Meridional Flows from Ring Diagram Analysis

In order to search for meridional circulation in the solar envelope, we have applied ring diagram analysis to a set of small regions over the surface. The helioseismic data consist of Solar Oscillation Investigation/Michelson Doppler Imager Dopplergrams taken over a time span of about 50 hr (~3000 images) on 1998 June 20-22. The regions studied cover 115° in latitude centered on the equator and 30° in longitude. We find poleward flows between r/ R=0.97 and the surface. There is no evidence in this depth range for the return path of these meridional flows. The temporal stability of these flows will be discussed after the analysis of a synoptic map obtained using the same technique.

Comparison of Two Fitting Methods for Ring Diagram Analysis of Very High l Solar Oscillations

A new method of fitting tridimensional power spectra of solar oscillations is described and compared with a previous one whose use has been more common. The new method fits the parameters of the Lorentzian profiles in a bidimensional k - ω diagram constructed from an azimuthal average of the tridimensional one. The horizontal velocities are then determined keeping these parameters fixed, greatly reducing the computation time. Both methods are compared for two radial orders (n = 3, 4) of a tridimensional power spectrum obtained for a region of about 15° square around solar disk center. The images used in this work correspond to a 3 day set of 1080 × 1080 pixel intensity images obtained at the Observatorio del Teide on 1994 November 8-10 with the Taiwanese Oscillation Network (TON) instrument. The results of the fitted velocities agree within the estimated errors for the two methods. The reduction of the computing time obtained with the new method makes it convenient for the ring diagram analysis.

Subsurface Flows in and Around Active Regions with Rotating and Non-rotating Sunspots

The temporal variation of the horizontal velocity in sub-surface layers beneath three different types of active region is studied using the technique of ring diagrams. In this study, we select active regions (ARs) 10923, 10930, 10935 from three consecutive Carrington rotations: AR 10930 contains a fast-rotating sunspot in a strong emerging active region while other two have non-rotating sunspots with emerging flux in AR 10923 and decaying flux in AR 10935. The depth range covered is from the surface to about 12 Mm. In order to minimize the influence of systematic effects, the selection of active and quiet regions is made so that these were observed at the same heliographic locations on the solar disk. We find a significant variation in both components of the horizontal velocity in active regions as compared to quiet regions. The magnitude is higher in emerging-flux regions than in the decaying-flux region, in agreement with earlier findings. Further, we clearly see a significant temporal variation in depth profiles of both zonal and meridional flow components in AR 10930, with the variation in the zonal component being more pronounced. We also notice a significant influence of the plasma motion in areas closest to the rotating sunspot in AR 10930, while areas surrounding the non-rotating sunspots in all three cases are least affected by the presence of the active region in their neighborhood.

ACTIVE REGION MORPHOLOGIES SELECTED FROM NEAR-SIDE HELIOSEISMIC DATA

We estimate the morphology of near-side active regions using near-side helioseismology. Active regions from two data sets, Air Force Data Assimilative Photospheric flux Transport synchronic maps and Global Oscillation Network Group near-side helioseismic maps, were matched and their morphologies compared. Our algorithm recognizes 382 helioseismic active regions between 2002 April 25 and 2005 December 31 and matches them to their corresponding magnetic active regions with 100% success. A magnetic active region occupies 30% of the area of its helioseismic signature. Recovered helioseismic tilt angles are in good agreement with magnetic tilt angles. Approximately 20% of helioseismic active regions can be decomposed into leading and trailing polarity. Leading polarity components show no discernible scaling relationship, but trailing magnetic polarity components occupy approximately 25% of the area of the trailing helioseismic component. A nearside phase-magnetic calibration is in close agreement with a previous far-side helioseismic calibration and provides confidence that these morphological relationships can be used with far-side helioseismic data. Including far-side active region morphology in synchronic maps will have implications for coronal magnetic topology predictions and solar wind forecasts.

Ring-diagram parameter comparisons for GONG, MDI and HMI

We examine the differences between ring-diagram mode frequency estimates from samples of Global Oscillation Network Group [GONG], Michelson Doppler Imager [MDI] and Helioseismic and Magnetic Imager [HMI] data, and find that different instruments and analysis pipelines do result in small systematic differences which may not be uniform across the solar disk.

GREEN'S FUNCTIONS FOR FAR-SIDE SEISMIC IMAGES: A POLAR-EXPANSION APPROACH

We have computed seismic images of magnetic activity on the far surface of the Sun by using a seismic-holography technique. As in previous works, the method is based on the comparison of waves going in and out of a particular point in the Sun, but we have computed here Greens functions from a spherical polar expansion of the adiabatic wave equations in the Cowling approximation instead of using the ray-path approximation previously used in the far-side holography. A comparison between the results obtained using the ray theory and the spherical polar expansion is shown. We use the gravito-acoustic wave equation in the local plane-parallel limit in both cases and for the latter we take the asymptotic approximation for the radial dependences of Greens function. As a result, improved images of the far side can be obtained from the polar-expansion approximation, especially when combining Greens functions corresponding to two and three skips. We also show that the phase corrections in Greens functions due to the incorrect modeling of the uppermost layers of the Sun can be estimated from the eigenfrequencies of the normal modes of oscillation.

Comparison of HMI Dopplergrams with GONG and MDI data

We compare sample Dopplergrams from the Helioseismic and Magnetic Imager, the Michelson Doppler Imager and the Global Oscillation Network Group. Each instrument has a distinct static velocity patterm across the disk; once this has been subtracted and the images interpolated to a common grid, the agreement is satisfactory.