The scientific operation modes and data products can be divided into four main areas: global helioseismology, local-area helioseismology, line-of-sight and vector magnetography and continuum intensity studies. The principal data flows and products are summarized in the proposal in Foldout 1.L. These four primary scientific analyses cover all main HMI objectives, and have the following characteristics:
· Global Helioseismology: Diagnostics of global changes inside the Sun. The normal-mode method will be used to obtain large-scale axisymmetrical distributions of sound speed, density and flow velocities throughout the solar interior from the energy-generating core to the near-surface convective boundary layer. These diagnostics will be based on frequencies and frequency splittings of modes of angular degree (l) up to 1000, obtained for intervals of several days each month and up to l=300 for each 2-month interval. These will be used to produce a regular sequence of internal rotation and sound-speed inversions to allow observation of the tachocline and the near-surface shear layer.
· Local Helioseismology: 3D imaging of the solar interior. The time-distance technique, ring-diagram analysis and acoustic holography represent powerful tools for investigating physical processes inside the Sun. These methods are based on measuring local properties of acoustic and surface gravity waves, such as travel times, frequency and phase shifts. The targeted high-level regular data products include:
§ synoptic maps of mass flows and sound-speed perturbations in the upper convection zone for each Carrington rotation with a 2-degree resolution, from averages of full disk time-distance maps;
§ synoptic maps of horizontal flows in the upper convection zone for each Carrington rotation with a 5-degree resolution from ring-diagram analyses;
§ higher-resolution maps zoomed on particular active regions, sunspots and other targets, obtained with 4-8-hour resolution for up to 10-day transits;
§ deep-focus maps covering the whole convection zone depth, 0-200 Mm, with 10-15 degree resolution;
§ farside images of travel-time perturbations associated with large active regions every 12 hours.
These observations require uninterrupted series of Dopplergrams of lengths 8 to 24 hours with the following characteristics: 50-second (or higher) cadence, spatial sampling of 2 Mm for distances up to 75 degrees from the disk center, and the noise level better than 20 m/s.
· Magnetography. Complete coverage of magnetic processes in the photosphere. The traditional line-of-sight component of the magnetic flux is produced as a co-product with the Doppler velocity. Several products will be computed with various cadence (up to 10 minutes) and resolution for use as input to coronal field and solar wind models and correlative studies. To accurately model the global fields the zero point accuracy should be better than 0.1G.
·The vector magnetic field. This is one of the most important physical observables of the active solar atmosphere. HMI will produce several standard data series of vector fields. A simple ‘magnetograph mode’ analysis will be computed continuously in real time for large scale coronal modeling and other space weather applications. With help of inversion techniques113, HMI will also provide tracked and full-disk vector magnetic field, filling factor, and thermodynamic parameters of photospheric plasma within reasonable errors. The data will be used to measure free energy, stresses and helicity of the magnetic field, providing important input to many prime science objectives and tasks of HMI and other SDO investigations. These polarimetric observations require a few minutes temporal cadence, a spatial sampling of 0.5”, and a 0.3% polarization precision to yield 5% accuracy of the magnetic field strength, a few tens of degrees in inclination and azimuth in strong fields.
· Continuum Intensity: Identification of irradiance sources. The observations of the intensity in the continuum near the HMI spectral line will give a very useful measure of spot, faculae area and other sources of irradiance. This will be important for studying the relationship between the MHD processes in the interior and lower atmosphere and irradiance variations. The continuum data will be also used for limb shape analysis, and for public information and education purposes. These measurements require calibration of system pixel-pixel gain variations to a level 0.1%, as demonstrated with MDI.
HMI Science
Data Measurements.
The data observing sequences that are used as input for
analyses are described here as a number of observation types. The Observation Types that support the
various science objectives are shown in the HMI_Objectives-Data_table. The Observation Types are
described here.
NOTE THESE
PAGES ARE STILL UNDER REVIEW
Observation Type HS-1
Global HS synoptic uses typically 72-day intervals. Repeated intervals are needed to follow solar variations with changes in solar activity. Loss of a few intervals in a three year span would still yield important results to extend the SOHO/MDI results into the next cycle. If GONG continues into the SDO mission years these observations have a somewhat lower priority since higher-noise but still very useful data can be obtained with GONG. In the past seven years GONG has (in 36-day intervals) coverage in the range of 70 to 85% (of the minutes). MDI has typically 96-97% coverage except of course for the large gaps. For low-degree (less than l=120) modes we believe most of the noise difference results from coverage completeness. We believe that coverage less than ninety something percent results in noticeable noise, probably the limit is about 90. We believe that solar noise is such that the noise in the frequency fits with coverage between about 95% and 100% is indistinguishable. Thus 95% is a good place to be sure that we are limited by solar noise rather than noise from data gaps. Where there are gaps we have found that random times of random sizes is least damaging. Gap filling methods can recover most of the signal when the gap is less than 10% of the surrounding spans of more complete data. Gaps of less than half the Nyquist sampling interval of the 5.3 mHz acoustic cutoff frequency have little if any effect when we use the MDI processing method of no-trend removal. Using the GONG method of first differences puts a tighter constraint on small gaps. 5.3 mHz is a period of 3.1 minutes. This leads to a maximum harmless gap of about 90 seconds (two HMI sample intervals) for global HS data.
Observation Type Table Entry
|
Type |
Observing Sequence Needed |
Observables |
Planned Span of Observations |
Planned Number of Observations (percent coverage) |
Minimum Span of Observations |
Minimum Number of Observations (percent coverage) |
Minimum Image resolution |
Notes |
|
HS-1 |
OS-1 |
fd_V, fd_Ic |
5-years |
25 (100%) |
3-years |
12 (80%) |
2” |
Global structure and rotation |
Observing Sequence Type Table Entry
|
Observing Sequence Type |
Nominal Interval |
Completeness in interval |
Max harmless gap |
Optimum Gap distribution |
Notes |
|
OS-1 |
72d |
95% |
90s |
random |
|
Observation Type HS-2
Global Scale Local Method Helioseismology. HS-2. This type of observation supports measurement of deep-in-the-Sun flows and possibly fields. The method uses time-distance techniques to probe the conditions in the vicinity of the tachocline and bottom of the convection zone. This is about 200Mm deep. The end-points of acoustic rays that turn at this depth are about 700Mm apart, i.e. about a solar radius. This means that only a narrow range of longitude or latitude from disk center can be probed. With end points about 60 degrees apart and limiting data to about 75 degrees from disk center (0.96R where projection factor is about 4) the distance from disk center at the turning point is at most about 45 degrees. A “point” could then be tracked for 90 degrees or about 6 days. Doppler noise near the limb may restrict the range to +-60 degrees (0.87R, factor of 2 in projection) giving a 60 degree or 4 day maximum observing sequence. Experiments have shown useful S/N for one day sequences .Due to the wavelength of waves at the tachocline (c. 70Mm) the maximum resolution achievable is about 10 degrees. Thus about 5 resolution elements are detectable for each day-long sequence and 36 elements covers a rotation. This means the minimum requirement is 8 one-day sequences per month with at most 4-day gaps.
|
Type |
Observing Sequence Needed |
Observables |
Planned Span of Observations |
Planned Number of Observations (percent coverage) |
Minimum Span of Observations |
Minimum Number of Observations (percent coverage) |
Minimum Image resolution |
Notes |
|
HS-2 |
OS-2 |
fd_V |
5-years |
1800 (98%) allows 5 lost days/yr |
3-years |
300 with max gap of 4 days (25%) |
2” |
Tachocline structure from time-distance |
Observing Sequence Type Table Entry
|
Observing Sequence Type |
Nominal Interval |
Completeness in interval |
Max harmless gap |
Optimum Gap distribution |
Notes |
|
OS-2 |
24h |
85% |
90s |
bunched |
|
Observation Type HS-3
Sub-Surface “Weather” – SSW. This type of observation supports both Time-Distance and Ring type analyses (as well as holographic techniques). The goal is to study the region from the surface to 30Mm in detail and watch global and AR scale flows as they evolve. Sequences of a day are tracked in data cubes of typically 15-degrees or 7.5 degrees (perhaps 5 degrees with HMI). The Sun’s disk is tiled with overlapping tracked cubes and flow analysis is performed on each cube. MDI has shown changes on day time scales with larger scales persistent for rotations (or years in case of zonal flows a.k.a. torsional oscillations). MDI has only allowed one 2-rotation sequence per year. This is not sufficient to determine the persistence or nature of evolution of active region scale or even radius-scale flows. We believe a minimum of three contiguous rotations at least once each six months would yield important data on these flows. We should sample over the full rising part of the cycle – i.e. from min to max. The base plan is for continual mapping.
|
Type |
Observing Sequence Needed |
Observables |
Planned Span of Observations |
Planned Number of Observations (percent coverage) |
Minimum Span of Observations |
Minimum Number of Observations (percent coverage) |
Minimum Image resolution |
Notes |
|
HS-3 |
OS-3 |
fd_V |
5-years |
20 (100%) |
Min to max |
6 – one 3-rotation sequence each 6 months |
2” |
Large scale subsurface “weather” |
Observing Sequence Type Table Entry
|
Observing Sequence Type |
Nominal Interval |
Completeness in interval |
Max harmless gap |
Optimum Gap distribution |
Notes |
|
OS-3 |
90d |
95% |
90s |
random |
|
Observation Type HS-4
Active Longitude Studies. This observation type will provide data for the examination of the source of active zones or longitudes or complexes of activity. We know a good deal statistically about the clumpiness of activity but do not know of any internal cause for such clumping. This study will use techniques nearly identical to HS-3 except that longer contiguous spans are required. We believe we need at least 4-rotation spans of synoptic maps of sub-surface flows. Both Ring and Time-Distance analyses will be applied. Time-Distance prefers clumped gaps and Rings prefer random gaps.
|
Type |
Observing Sequence Needed |
Observables |
Planned Span of Observations |
Planned Number of Observations (percent coverage) |
Minimum Span of Observations |
Minimum Number of Observations (percent coverage) |
Minimum Image resolution |
Notes |
|
HS-4 |
OS-4 |
fd_V |
5-years |
15 (100%) |
3-years |
5 four-rotation sequences (55%) |
2” |
Active Longitudes |
Observing Sequence Type Table Entry
|
Observing Sequence Type |
Nominal Interval |
Completeness in interval |
Max harmless gap |
Optimum Gap distribution |
Notes |
|
OS-4 |
120d |
95% |
90s |
random |
|
Observation Type HS-5
Active Region Studies. This observation type produces data to allow study of flows, thermal, and field structure beneath active regions. Both Time-Distance and Holographic methods will be used. Since the details in the near surface layers are important the highest resolution data will be needed. The wavelength of acoustic waves near the surface is about 2Mm and this sets a natural resolution target. Active regions have lifetimes of weeks to months. If data is used to about 0.93 R then a 1Mm pixel subtends 0.5”. This gives a 10-day span of coverage. 12-days coverage with 0.5” pixels corresponds to 2Mm pixels with data used out to 0.98R. We would hope to obtain clean data for 10-day spans with analyses extended to 12-day spans with allowances for lower resolution at the ends. This means we can observe 20-50% of the lifetime of many active regions with full coverage of many short lived small regions. With this coverage we should be able to determine the flow-evolution from prior to eruption to post-visibility for small, typical, and a few large regions. We estimate that a sample of 100 regions tracked for 12-days each will yield at least 10 large region segments and we consider this to be a minimum requirement for this study. The baseline plan calls for tracking all active regions with NOAA identification. 100 regions can be captured in about 2.5 years starting at minimum or in less than six months near a typical maximum of activity. Since Time-Distance methods will be used gaps should be clumped into as few sub-intervals as possible. Ring type analyses do not give enough resolution for these studies.
|
Type |
Observing Sequence Needed |
Observables |
Planned Span of Observations |
Planned Number of Observations (percent coverage) |
Minimum Span of Observations |
Minimum Number of Observations (percent coverage) |
Minimum Image resolution |
Notes |
|
HS-5 |
OS-5 |
fd_V |
5-years |
100% |
2.5-years starting at min or 1 year at max |
100 regions |
2” |
Active Regions, Multiple regions concurrently |
Observing Sequence Type Table Entry
|
Observing Sequence Type |
Nominal Interval |
Completeness in interval |
Max harmless gap |
Optimum Gap distribution |
Notes |
|
OS-5 |
12d |
95% |
90s |
clumped |
|