Overview

One of the three U.S.-provided instruments on the Solar and Heliospheric Observatory (SOHO) is the Michelson Doppler Imager (MDI) which is the key element of the Solar Oscillations Investigation (SOI), a collaboration between Stanford University and the Lockheed Martin Solar and Astrophysics Laboratory. The Michelson Doppler Imager for Triana (MDI-T) will address the same science goals as MDI but will benefit from experience gained in the MDI program. In order to provide MDI-T on a rapid schedule and at low cost, MDI-T is a nearly identical copy of MDI. MDI is described in detail by Scherrer et al. (1995).

MDI-T will be one of a set of coordinated solar instruments on Triana. The set, called T-SIP, or Triana Solar Instrument Package, is a subset of the instruments that flew on SOHO. The T-SIP instruments provide the core of the SOHO science and will be operated in cooperative observing programs to best leverage the joint capabilities. The T-SIP instruments will measure important inputs to the solar-terrestrial system.

The primary science goal of MDI-T is to determine the properties of the solar interior by exploiting a number of techniques of helioseismology. Helioseismology uses the relationship between acoustic wave propagation speed and the thermodynamic of the solar interior to determine the solar internal structure and dynamics. Acoustic waves with periods near five minutes propagating in all directions permeate the solar interior. These waves are refracted upwards toward the solar surface by the downward temperature gradient and they are reflected downward at the surface by the sharp gradient in density. The acoustic waves are observable as small periodic Doppler shifts in photospheric absorption lines. The observations of trapped mode frequencies and wave advection are the input data for comparison with physical models of the solar interior and for numerical inversions that provide information about the thermal structure and motions in the interior.

In addition to observing the oscillations with high spatial resolution over the entire visible hemisphere, MDI and MDI-T observe photospheric magnetic fields. MDI measures only the line-of-sight component of the field. By the simple addition of two additional half-waveplates in a 6-position polarization analyzer wheel (as in TRACE), MDT-T provides the capability of observing the horizontal component of the field as well. This is simple to implement, and a very useful additional tool to help characterize the eruption and interaction of magnetic fields in active regions.

MDI and MDI-T both have two imaging magnifications, selected by the exposure control shutter. In MDI, the location on the solar disk of the higher-magnification region, about 1/9 of the solar disk area, was fixed prior to launch. MDI observations demonstrated the great utility of the higher resolution observations for many science objectives. MDI-T will provide a simple mechanism to reposition the high-resolution field in the east-west direction on the Sun. This will allow 6-8 day duration observations of convection flows beneath and around sunspots, as compared to the 2-day limit with MDI.

Other minor differences between MDI and MDI-T are primarily to meet the differences in spacecraft interface requirements.

The SOI-MDI program required the dedicated efforts of an international science team as well as the Stanford and Lockheed instrument teams. The task of deducing stellar interior properties from observations of surface motions is not trivial and will continue to require significant effort. To enhance the science return from SOI-MDI, all MDI data have been provided to all members of the scientific community via easy-to-use Web data request forms. The MDI-T program will continue that policy. Additionally, the MDI-T team is dedicated to bringing the excitement of touring the interior of our Sun to the public, and to help incorporate solar and space science in general into the educational system.

MDI obtained an exceedingly useful series of observations that have provided new insights into the solar interior and into the interactions of magnetic fields and convective processes. MDI-T will continue those kinds of observations with some minor but important improvements; it will be well suited to observe the solar interior during the activity maximum in 2001 and beyond.

Historical Perspective

In the 1960's it was discovered that the entire surface of the Sun is constantly in motion and that local areas of the photosphere oscillate with a period of about five minutes. Initially the motions were thought to be local responses to convective plumes below the surface and the five-minute period just the natural buoyancy frequency of the atmosphere. In the early 1970s theoreticians suggested that the oscillations were the atmosphere's response to global modes in the interior. Within a few years these predictions were verified and the discipline of helioseismology was born. By precisely measuring the frequencies of the modes it is possible to determine temperature, density, equation of state, elemental abundances, convective mixing, rotation, and mass flows as a function of radius, longitude, and latitude, throughout much of the solar interior. It is thus possible to verify models of stellar interiors and stellar evolution by direct measurement. A secure understanding of the solar interior is essential to establishing whether the observed solar neutrino flux deficit is due to an incomplete understanding of conditions in the solar interior or of the fundamental physics of elementary particles. The measurements of flow patterns and rotation in the solar convection zone are key to the development of models of stellar convection and the generation of magnetic fields by dynamo action.

MDI on SOHO

The Michelson Doppler Imager (MDI) on SOHO (Scherrer et al., 1995) was designed to exploit the opportunities created by helioseismology which could not be accomplished by ground-based observations. MDI made measurements of the photospheric velocity and intensity each minute. These and the magnetic field were computed on board from filtergrams observed at 5 wavelengths near the 6768 Å Ni I spectral line. A pair of tunable Michelson interferometers following fixed Lyot and broadband filters made the wavelength selection, resulting in a 100 mÅ passband. The instrument shutter selected between 4" and 1.2" resolution on a 1024x1024 CCD detector.

Data for two complete, compressed images could be downlinked each minute during high telemetry bandwidth intervals; alternatively sub-rasters of nearly any combination of full-disk and high-resolution observations in velocity, intensity, line depth, or magnetic field could be transmitted.

The unique and important aspect of the MDI experiment is the capture of long, uninterrupted sequences of images of uniform quality, high precision, and good spatial resolution. These characteristics, coupled with flexible observing modes and the provision of various complementary synoptic observations, make it a very powerful tool for exploring the Sun's interior and the sources of solar variability. The table below lists the basic observables of the MDI and MDI-T instruments.

Observables for MDI and MDI-Triana

MDI was successful during the SOHO prime mission phase. Continued access to MDI observations was one of the key goals of the SOHO Solar Maximum Extended Mission which was interrupted on 24 June 1998. During the initial SOHO mission MDI contributed to several new discoveries:

• New constraints on interior rotation and sound-speed profiles (Schou et al., 1998; Birch & Kosovichev, 1998; Dziembowski et al, 1998; Kosovichev et al., 1997, 1998; Toutain et al. 1998).
• The polar jet stream (Schou et al., 1998).
• Meridional flow with depth (Giles et al., 1997; Schou & Bogart, 1998; González Hernández, 1998).
• Zonal flows with depth (Kosovichev and Schou, 1997; Schou et al., 1998).
• Characterization of supergranulation (Kosovichev & Duvall, 1997).
• Detection of seismic response to flares (Kosovichev & Zharkova, 1998).
• Precise measurement of the seismic radius (Schou et al., 1997).
• Relativistic corrections to the equation of state (Elliott & Kosovichev, 1998).
• First convective flow maps below surface (Duvall et al., 1997).
• Thermal structure beneath sunspots (Kosovichev, Duvall & Scherrer, 1997).
• Detection of Giant Cells (Beck et al.,1998).
• Magnetic carpet (Hagenaar et al., 1998).
• Association of magnetic field emergence and coronal brightenings (Tarbell et al., 1997).
• Strong clumping of magnetic flux in coronal holes (DeForest et al., 1998).

MDI-T, MDI on Triana

The purpose of MDI-T is to complete the studies begun during the SOHO mission and to extend those studies through solar maximum into the declining phase of the present cycle. In particular, MDI-T will study:

• Global Properties of the Solar Interior
  • Radial Stratification of the Solar Interior-Determine the spherically symmetric components of the mean radial structure of the Sun in pressure, density, and sound speed. The prime goal is to focus on the structure of the energy-generating core, the region of convective overshoot, and the sub-surface turbulent boundary layer. These are regions where significant deviations from the standard solar model have been detected from the MDI data. These studies are important for inferring the Sun's chemical composition and understanding the problems of the deficit of solar neutrinos and the Li and Be abundances. These studies also allow investigation of the thermodynamical properties of matter under the extreme conditions found in the Sun and of the mechanisms of energy transport in stellar interiors.
  • Internal Rotation-Determine the rotation rate as a function of radius and latitude. This knowledge is essential for understanding solar and stellar evolution. The interaction between rotation and convection is critical to understanding solar activity, the generation of magnetic fields, and the nature of the solar dynamo. The properties of the solar tachocline (the transition region between the radiative and convective zones) and its variations during solar maximum will be determined. Longer time series will allow us to accurately measure the rotation rate of the solar core.

    eps version. Figure MDI_1 is the spherically symmetric solar interior rotation determined from the MDI medium-l data. The figure shows a shear layer in the region that separates the differentially rotating convection zone and the more rigidly rotating core. The sharp rotation gradient just beneath the surface and the high latitude jet are also shown. MDI-T will continue this series of observations and allow determination of variations between solar minimum and maximum activity levels.
    

  • Solar Background -Determine the global characteristics of small scale solar convective phenomena, and identify the sources and mechanism of excitation of solar oscillations. The determination of properties of the correlated background discovered by MDI (Nigam et al., 1998) is important for understanding the physics of solar oscillations.
  • Detection of g Modes-Continue the search for g modes begun with MDI, GOLF, and VIRGO on SOHO. There were hints of g-mode oscillations detected with GOLF and possibly with MDI and VIRGO during the analysis of the 2-year dataset available from SOHO. MDI-T data will help to either verify these modes or to continue the search by adding a longer baseline and reducing further the noise in the low-frequency band.

• Local Properties of the Solar Interior
  • Interior and Surface Flows-Meridional flow, now detected 70,000 km beneath the surface, is likely an important component of the solar cycle. Measurements will be pushed to greater depths and the indication from surface Doppler measurements that the flow varies over the solar cycle will be checked. The zonal flows, or torsional oscillations, propagate in latitude over the solar cycle. Their connection to the solar cycle is unclear, but it is important to determine the depth dependence of the flow as well as its time variation. The high-latitude jet discovered with MDI data will be studied further, to look for time or location variations. Searches will be made for other, more difficult to detect, jet streams.
  • Supergranulation-Smaller scale convective elements and magnetic field elements are advected to supergranulation boundaries. MDI-T will determine the characteristics of this key flow pattern at and beneath the surface. The tracking of the high-resolution field will enable better observations of the supergranules, whose one-day lifetime is uncomfortably close to the time for regions to traverse the fixed field of MDI's high-resolution field.
  • Large-Scale Surface Flows-Evidence is beginning to mount for large-scale convection cells detectable in the Doppler residuals (Beck et al., 1998) and by tracking of supergranules, as well as in seismically inferred near-surface flows (Schou & Bogart, 1998). These features span up to 50 degrees in longitude or more, and may play a significant role in the organization and distribution of active regions. Analysis of MDI-T data will allow us to continue to explore these surface flow patterns and the associated thermal structures during a different part of the solar cycle.
  • Active Region Tomography- Some of our greatest uncertainties about solar activity are the nature of the subsurface structure of magnetic fields. Do sunspots have a single monolithic flux tube below the surface, or is the "spaghetti" model valid? Active region tomography using the several variants of acoustic imaging has been used with MDI data for the first views of this region. MDI-T will allow detailed study of the internal thermal and convective structure and evolution of active regions.

    ps version. Figure MDI_2 shows a frame from one of the three sequences of an erupting active region observed by SOHO/MDI with enough coverage to perform time-distance 3-D tomographic analysis of the sub-surface sound speed and flows. The figure shows the photosphere in white light rendered partly transparent with vertical and horizontal cuts of sound speed inversions. The three sunspots emerged during the observing sequence. The red coloring shows regions of increased wave speed and the blue shows cooler regions just beneath the sunspots. The warmer/faster region extends from 3000 km to about 8000 km beneath the photosphere with a horizontal extent comparable to the entire active region. The observation was made in July 1996.
    
    

  • Flare-Induced Seismic Events-MDI data provided the first observations of an acoustic response to flares. The detection of scattering of the flare-induced seismic waves on sunspots will allow the determination of the sub-surface structure of active regions. Important goals are understanding the mechanism of the seismic response, momentum and energy transport into the Sun's interior, and the relation between the seismic and atmospheric (Moreton) waves. Detection of seismic waves generated by a flare and converging at the flare antipode is an exciting prospect of flare seismology, with promises of application to the prediction of solar activity.

    ps version.
    Figure MDI_3 shows the response of the photosphere to the flare of 9 July 1996. The figure shows a map of line-of-sight velocity in the region around the flare. Darker colors are motion into the sun and lighter colors are upwards motion. The flare caused an impulse of 1000 m/s followed by an expanding ring with amplitude of several tens of m/s accelerating outward at from 10 km/s to greater than 110 km/s. The ring shown here is from sound waves that were refracted back to the surface from a depth of about 17000 km.
    

• Magnetic Evolution
  • Magnetic Fields-Active and quiet magnetic field regions evolve and interact to produce the larger-scale field patterns. MDI's observations at solar minimum clearly showed how emerging flux interacts with existing structures, both locally and globally. Closer to solar maximum, MDI-T will observe with sufficient spatial and temporal resolution to sufficiently characterize the changes that result in solar flares, Coronal Mass Ejections (CMEs) and other transient activity, irradiance variations, development of coronal holes, and ultimately the reversal of the polar fields. Magnetic fields also affect the photospheric characteristics of solar p modes; these effects must be understood to determine what happens below the surface. There is evidence from SOHO/EIT that low frequency (millihertz band) acoustic waves exist in the corona and contribute to coronal heating. Detailed joint studies of the quiet sun with both high resolution imager and MDI-T magnetograph observations are needed to understand the "background" heating mechanisms of the solar corona.

    ps version. Figure MDI_4 shows a section of an MDI magnetogram (left) with an EIT 195Å image (center) and an overlay of the field on the EIT image (right). This figure shows the strong correspondance between fields and coronal brightenings. These images are extracted from a time series that shows dramatic expansion and brightening of coronal loops as new flux emerges from below the photosphere.
    
    

  • Magnetic Carpet- MDI confirmed that small magnetic bipolar flux element pairs are continually emerging at random locations within supergranules. This provides a mechanism to refresh the dominant polarity, while at the same time changing the photospheric location of flux elements at a speed comparable to the horizontal flow speed, rather than the much slower random walk time. The entire network flux is rearranged in less than a day by this mechanism, and the total flux in the quiet Sun is replaced in two to four days. There are profound implications for coronal heating.

    jpeg version., ps version.
    Figure MDI_5 shows magnetic field lines computed from MDI observations overlain on a simultaneous image from SOHO/EIT. Bright spots on the EIT image correspond to locations above merging magnetic fields. Because of the lack of 3-D intensity information, the EIT image has been rendered on the photosphere, beneath the extrapolated magnetic field lines.
    
    

  • Prominences, Filaments, and CMEs-MDI-T magnetograms will allow us to determine the magnetic field configuration at the foot points of prominences. Prominences remain enigmatic features after decades of study. Prominences are complex structures that demonstrate chirality and strongly twisted magnetic fields. While models can explain the forces needed to suspend the core of each prominence, there are still major holes in our understanding of prominences. The secret is likely to be found in the detailed magnetic structure of the legs and "backbone" of the prominence. A major result from SOHO is that CMEs appear frequently to arise from disappearing filaments (aka prominences). Hence understanding prominence evolution and behavior is crucial to predicting CMEs and the interplanetary storms that accompany them.

MDI-T Operations and Data Access

• MDI-T Operations will be managed in the same manner as SOHO/MDI. A small staff located at the science planning center will engage in joint planning with other Triana experiments and other space and ground observations. The main MDI-T science planning, health monitoring, and data processing will continue at Stanford University. We expect raw data to be delivered to the Stanford processing center from the ground stations in much the same way that the SOHO Virtual Channel 2 telemetry dedicated to MDI science is currently delivered to the SOHO Experiment Operations Facility.
• MDI-T Data Access will be quick and unrestricted. The data will be available in both raw and calibrated form as soon as they are received and processed. For SOHO/MDI, the processing typically lagged receipt of data by several days. The MDI-T pipeline will make data available at level 0 after one day, and at level 1 (calibrated) after one week. ALL MDI-T data will be accessible via the World Wide Web forms as SOHO/MDI data are now. Data will be distributed directly via the Internet and on various hard media depending on the quantity requested. The MDI data analysis and processing center at Stanford presently contains over 30,000 gigabytes that are searchable and exportable in this form, and MDI-T data can be integrated seamlessly into the existing system. MDI-T magnetogram and campaign data will also reside in the GSFC and European SOHO data archives.

Science Analysis Team

The MDI-T science objectives require a significant analysis effort to be accomplished. During the SOHO development phase and two years of operations, significant progress was made in the field of helioseismology. Much remains to be done. We have only now begun to exploit the potential for local-area helioseismology. These techniques will help provide an understanding of how the surface, interior, and convective transport interact with magnetic fields and evolve during the solar cycle. The long time series of oscillation mode frequencies will enable us to determine the global-scale thermal and dynamic variations with the phase of the cycle. Regular high-cadence magnetic field observations will allow detailed tracking of magnetic flux from eruption to the large-scale remnant field patterns for the first time.

To achieve these goals we have two components to the MDI-T science team. The instrument science team consists of scientists at Stanford University and Lockheed Martin Solar and Astrophysics Laboratory. This team has responsibility for MDI-T operations, data processing, calibration, access, and many of the prime science goals. The table lists the MDI-T investigators and indicates their areas of responsibility.

The extended science team is responsible for important aspects of the prime science goals and many associated science goals. The extended science team Co-Investigators were SOHO/MDI Co-Is during the prime SOHO mission who were to obtain support via the SOHO Guest Investigator program during the extended mission phase. The efforts of these investigators are critical to completion of the MDI-T science goals. Their efforts could be supported via the prime Triana program, or via competitive selection via the NASA SEC Guest Investigator program if that program will contain support for Triana GIs.