HAO 2012 Profiles In Science: Dr. Matthias Rempel

Contact:

Area of expertise: Sun and Upper Atmosphere

Specialties: computer modeling of sunspots and other solar features

Dr. Matthias Rempel is a Scientist III in the High Altitude Observatory of the National Center for Atmospheric Research. He received his PhD in Astronomy and Astrophysics in 2001 from the University of Göttingen, Germany. His initial work with HAO began in 2002 as an ASP Fellow. He focuses his research on modeling of MHD processes in the solar interior, structure of the overshoot region at the base of the solar convection zone, coupled models of the differential rotation, meridional flow, and large-scale dynamo, addressing non-kinematic effects and cycle variations of the solar differential rotation (torsional oscillations). Rempel uses computer models to study key features of the Sun, including regions that cannot be directly observed. He led an international team of scientists to create the first-ever comprehensive model of sunspots, the dark patches on the Sun's surface that are associated with the 11-year solar cycle and solar storms. His work can help scientists better understand the Sun's magnetic fields and the transport of energy from the Sun's interior into the solar system.

Detailed description of individual projects and links to work published during the last year:

1. High resolution simulations of sunspot fine structure:
A detailed comparison of sunspot models with different numerical grid resolutions was published in "M. Rempel, Numerical Sunspot Models: Robustness of Photospheric Velocity and Magnetic Field Structure", 2012, ApJ 750, 62. The study concluded that most aspects of photospheric sunspot structure are well captured with the resolution that is affordable by current simulations. While certain details such as the Evershed flow speed or the width of individual filaments still show some resolution dependence, the overall underlying magneto-convection process is found robust. Nevertheless higher resolution is desirable in the future for a direct comparison with spectropolarimetric observations of sunspots. Such comparisons are currently in progress.

2. Helioseismology of realistic magneto-convective sunspot simulations:
Numerical simulations of photospheric magneto-convection contain oscillation-modes that are very similar to those observed on the sun through instruments like SOHO/MDI and SDO/HMI. Synthetic Doppler data from sunspot simulations have been analyzed in detail an compared to observations. The results published in Braun et al. 2012, ApJ 744, 77 show that simulated sunspots are seismically very similar to observed sunspots in terms of the frequency dependence of travel time shifts. At the same time it is found that standard methods used to infer the subsurface structure of sunspots (through inversion) lead to a substantial disagreement when applied to simulation data. Overall this indicates that a new approach is needed for structure inversions in and around sunspots.

3. Simulations of flux emergence and sunspot formation:
In collaboration with M. Cheung (LMSAL, Lockheed Martin) numerical simulations of flux emergence in 147×74×16 Mm3 sized domains were performed. It was found that results from previous simulations in smaller domains are robust. Furthermore results do not show a strong dependence on the initial field strength, pointing toward a dominant role of photospheric processes in the determination of sunspot field strengths. In addition the role of horizontal flows on active region asymmetries was investigated. A retrograde flow (as expected from angular momentum conservation during the rise from base of convection zone) leads to a more coherent leading sunspot as observed in most active regions on the sun. Results from this study are currently being analyzed and a publication is in preparation.

4. High latitude solar torsional oscillations:
The extended last solar minimum and slower than usual rise of the current solar cycle is also manifest in the torsional oscillation pattern on the sun. Helioseismic studies have indicated so far little evidence for a new poleward directed branch of fast rotation as it has been observed in the previous cycle. Using a non-kinematic dynamo model it was shown that variations in the magnetic cycle strength lead to variations of the mean solar rotation rate, which is most significant in high latitudes. During phases of reduced magnetic activity it is expected to see a drop in the high latitude rotation rate. It was shown that such changes of the mean rotation rate can temporarily mask the torsional oscillation pattern, which is a possible explanation for the weak evidence of a new polar branch on the sun.

Highlighted Publications

Sunspot fine structure at the τ = 1 level

Figure 1: High Resolution

(1) Rempel, M. 2012: Numerical Sunspot Models: Robustness of Photospheric Velocity and Magnetic Field Structure, ApJ, 750, 62.

Abstract: MHD simulations of sunspots have successfully reproduced many aspects of sunspot fine structure as a consequence of magneto-convection in inclined magnetic field. We study how global sunspot properties and penumbral fine structure depend on the magnetic top boundary condition as well as on grid spacing. The overall radial extent of the penumbra is subject to the magnetic top boundary condition. All other aspects of sunspot structure and penumbral fine structure are resolved at an acceptable level starting from a grid resolution of 48 [24] km (horizontal [vertical]). We find that the amount of inverse polarity flux and the overall amount of overturning convective motions in the penumbra are robust with regard to both resolution and boundary conditions. At photospheric levels Evershed flow channels are strongly magnetized. We discuss in detail the relation between velocity and magnetic field structure in the photosphere and point out observational consequences.

Figure 1 caption: Influence of the numerical grid resolution on the properties of the penumbra. Left: bolometric intensity. Right: magnetogram at τ = 1. In each panel the four quadrants correspond to simulations performed with different resolution as indicated in the corners. Note that all simulations were performed in a 49 Mm wide domain; we show here only subsections.

Figure 2: High resolution

(2) Rempel, M. 2012: High-latitude Solar Torsional Oscillations during Phases of Changing Magnetic Cycle Amplitude, ApJ, 750, L8.

Abstract: Torsional oscillations are variations of the solar differential rotation that are strongly linked to the magnetic cycle of the Sun. Helioseismic inversions have revealed significant differences in the high-latitude branch of torsional oscillations between cycle 23 and cycle 24. Here we employ a non-kinematic flux-transport dynamo model that has been used previously to study torsional oscillations and simulate the response of the high-latitude branch to a change in the amplitude of the magnetic cycle. It is found that a reduction of the cycle amplitude leads to an increase in the amplitude of differential rotation that is mostly visible as a drop in the high-latitude rotation rate. Depending on the amplitude of this adjustment the high-latitude torsional oscillation signal can become temporarily hidden due to the unknown changing mean rotation rate that is required to properly define the torsional oscillation signal.

Figure 2 caption: High-latitude torsional oscillation pattern (color shades) together with the toroidal magnetic field near the base of the convection zone as proxy for a magnetic butterfly diagram. The torsional oscillation amplitude is clipped off at 1.5% of the core rotation rate, corresponding to an amplitude of about 6.5 nHz. The peak toroidal magnetic field strength is about 14 kG, the contours indicate the 30%, 60%, and 90% levels, with solid and dashed line styles separating the polarity. Top to bottom we show three dynamo solutions that differ in the value of the α–effect after t = 7.5 years. Panel (a) presents the reference solution with a constant α–effect of 0.15 m s–1 throughout the simulation run. In panels (b) and (c) the α–effect was reduced to 75% and 50% after t = 7.5 years, respectively. In panels (b) and (c) the high latitude branch of torsional oscillations becomes hidden since the mean rotation rate drops in high latitudes