Spectropolarimetery of umbral fine structures from Hinode: evidence for magnetoconvection

Mon. Not. R. Astron. Soc. 393, 65–70 (2009) doi:10.1111/j.1365-2966.2008.14203.x Spectropolarimetery of umbral fine structures from Hinode: evidence for magnetoconvection Lokesh Bharti,1 † Chandan Joshi,1 S. N. A. Jaaffrey1 and Rajmal Jain2 1 Department 2 Physical of Physics, University College of Science, Mohanlal Sukhadia University, Udaipur 313 001, India Research Laboratory, Department of Space, Government of India, Navrangpura, Ahmedabad 380 009, India Accepted 2008 November 5. Received 2008 October 28; in original form 2008 February 12 We present spectropolarimetric analysis of umbral dots and a light bridge fragment that show dark lanes in G-band images. Umbral dots show upflow as well as associated positive Stokes V area asymmetry in their central parts. Larger umbral dots show downflow patches in their surrounding parts that are associated with negative Stokes V area asymmetry. Umbral dots show weaker magnetic field in central part and higher magnetic field in peripheral area. Umbral fine structures are much better visible in total circularly polarized light than in continuum intensity. Umbral dots show a temperature deficit above dark lanes. The magnetic field inclination shows a cusp structure above umbral dots and a light bridge fragment. We compare our observational findings with 3D magnetohydrodynamic simulations. Key words: Sun: granulation – Sun: magnetic fields – Sun: photosphere – sunspots. 1 I N T RO D U C T I O N Two models of the umbral dots (UDs) are under discussion these days. The first is the cluster model (Parker 1976; Choudhuri 1986) that suggests that UDs are the top of the intrusion of the field free material between the flux tubes beneath the sunspot. The second model is known as the monolithic model (Weiss 2002, and reference therein) and suggests that UDs show up because of magnetoconvection in monolithic flux tube. Recent simulations by Schüssler & Vögler (2006) with grey radiative transfer show UDs appearing due to magnetoconvection in strong background magnetic field. The knowledge of the nature of UDs is essential to understand the energy transport from below the sunspot (see reviews from Solanki 2003; Thomas & Weiss 2004, and reference therein on the subject). Bharti, Joshi & Jaaffrey (2007a) analysed Dopplergrams obtained from filtergraph data, and found a correlation between intensity and velocity in UDs, which suggests a magnetoconvective origin. Using high-quality G-band images from Hinode, Bharti, Jain & Jaaffrey (2007b) reported on dark lanes in UDs. These separate observational findings are compatible with some aspects of simulations by Schüssler & Vögler (2006). Socas-Navarro et al. (2004) analysed peripheral UDs in detail from spectropolarimetric data and found higher temperature (∼1 kK), weaker field (∼500 G), small upflow (∼100 m−1 ) and more inclined field (∼10◦ ) in UDs. In this article, we present spectropolarimetric analysis of dark laned umbral fine structure from Hinode spectropolarimetric data.  E-mail: †Present address: Max-Planck Institute for Solar System Research, 37191 Katlenburg-Lindau, Germany.  C C 2008 RAS 2008 The Authors. Journal compilation  The high polarimetric sensitivity and spatial resolution achieved by Hinode spectropolarimeter now it became possible to compare observational results directly with predictions of numerical simulations (Rezaei et al. 2007; Sainz Dalda & Bellot Rubio 2008). 2 O B S E RVAT I O N S A N D I N V E R S I O N TECHNIQUE We used spectropolarimeteric data obtained by the spectropolarimeter onboard the Hinode (Kosugi et al. 2007) on 2006 December 12. The four Stokes profiles of the two iron line pairs at 630.15 nm (the Lánde factor g = 1.67) and 630.25 nm (g = 2.5) were recorded for the active region 10 930 close to the disc centre (μ = 0.99). We used fast map. The integration time for fast map was 3.2 s. The field of view comprises an area of 295 × 162 arcsec2 . The spatial sampling for the fast map was 0.316 arcsec along the slit and 0.295 arcsec in the scanning direction. The spatial resolution of the resulting spectropolarimetric map is approximately 0.6 arcsec for the fast map with the spectral sampling at 2.15 pm. The calibration of the spectropolarimetric data is described by Ichimoto et al. (2007). We used the solar-soft pipeline to calibrate the spectropolarimetric data. To derive accurate photospheric height stratification of the temperature (T), magnetic field strength (B), line-of-sight (LOS) velocity (Vlos ) and inclination (γ ), we employ the SIR code (Ruiz Cobo & del Toro Iniesta 1992). This code presumes hydrostatic equilibrium and local thermodynamic equilibrium. By solving numerically the radiative transfer equation for polarized light, the inversion code SIR computes the synthetic Stokes profiles. The optimal parameters for the model were determined iteratively. The difference between the observed and synthetic Stokes profiles was minimized using a non-linear, least-square Marquardt’s algorithm. The values of the ABSTRACT 66 L. Bharti at al. physical parameters are computed at only a few grid points called nodes instead of computing at all optical depths of the model. For rest of depths, they are approximately computed by the cubic-spline interpolation between the equidistantly distributed grid points. We perform the SIR inversion with only one magnetic component, for which we allow five nodes in T(τ ), three for B(τ ), V los (τ ) and γ (τ ). G-band time series obtained in the broad-band filter were used to follow the evolution of the sunspot fine structure as seen in the spectropolarimetric maps (see Bharti et al. 2007b). Wiener filtering was applied to the G-band images for the point spread function correction of telescope. Understanding of the evolution of the umbral fine structure is necessary as they may have common physical origin (Bharti et al. 2007c; Katsukawa et al. 2007; Rimmele 2008). Here, we would like to mention that it is our aim to investigate dark lane in UDs as reported by Bharti et al. (2007b) and the spectropolarimetric fast scan at 0.6 arcsec spatial resolution cover similar features. The calibrated Stokes profiles were used to create maps of total circular polarization and Stokes V area asymmetry (Bellot Rubio et al. 2007). Figure 3. Maps showing continuum intensity (a) and total circular polarization (b) for the emerging sunspot that rises below a developed sunspot in active region 10 930. Marked in white in panel (a) are the locations where stratifications of various physical parameters along cut were measured. All maps are up scaled two times using cubic-spline interpolation. The outer rectangle shows the field of view that was subjected to the SIR inversion and the inner rectangle shows the region that was subjected to area asymmetry measurement. 3 R E S U LT S We have chosen G-band images, whose timing was close to the spectropolarimetric scans. Fig. 1 shows one of the G-band images taken close to the fa (...truncated)