linkages between the atmosphere on the Earth and sun Phần 9 potx

31 226 0
linkages between the atmosphere on the Earth and sun Phần 9 potx

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

Thông tin tài liệu

434 G.R. Gupta et al. have measured oscillations in coronal holes in the polar off-limb regions of the Sun. All these studies point to the presence of compressional waves thought to be slow magnetoacoustic waves as found by Deforest and Gurman (1998); O’Shea et al. (2006, 2007). Recently, Gupta et al. (2009) have reported the detection of these waves in the disk part of the polar coronal hole (hereafter PCH). They also find a difference in nature of the compressional waves between bright (network) and dark (internetwork) regions in the PCH. In this contribution, we extend such analysis to another dataset. More detail is given in Gupta et al. (2009). 2 Observations and Data Analysis The data used in this analysis were taken on 25 February 1997, during 00:00– 13:59UT with the 1  300 00 slit of SUMER and an exposure time of 60 s in the N IV 765 ˚ A and Ne VIII 770 ˚ A lines in a southern PCH. Details of the data reduc- tion are given in Gupta et al. (2009). The chromosphere and transition region show enhanced-intensity network boundaries and darker internetwork cells. Presumably, the magnetic field is pre- dominantly concentrated at the network boundaries and, within coronal holes, the footpoints of coronal funnels emanate from these network boundaries. As the ob- serving duration of this dataset is very long, the locations of bright and dark pixels along the slit change with time. For this reason, we have analyzed the whole dataset pixel by pixel and timeframe by timeframe. For example, for one given moment, we first determined the average intensity along the slit. All pixels having an intensity higher than 1.25 times this average intensity were chosen as bright pixels. If such pixels are bright for at least 60 min (or 60 timeframes), then these are considered to be a bright network location over that time interval. The bright pixel identification is done only for the low-temperature N IV line; the network pixels obtained from it are assumed to be the same in the higher-temperature Ne VIII line. 3 Results and Discussion Figure 1 shows a representative example of the oscillations measured in a bright region of the PCH. We use wavelet analysis to provide information on the temporal signal variation (Torrence and Compo 1998). Further details on this wavelet anal- ysis are found in Gupta et al. (2009); O’Shea et al. (2001) and references therein. Figure 1 shows oscillations of about 18 min periodicity in both lines at the same lo- cation. This suggests that these two layers are linked by a propagating wave passing from one layer to the other. To test this hypothesis and to ascertain the nature of the propagating waves, we measured phase delays in intensity and in Dopplershift be- tween the two lines at each of the measurable pixels along the slit for a full frequency Network Loop Oscillations with EIS/Hinode A.K. Srivastava, D. Kuridze, T.V. Zaqarashvili, B.N. Dwivedi, and B. Rani Abstract We analyze a time sequence of He II 256.32 ˚ A images obtained with EIS/Hinode, sampling a small magnetic loop in magnetic network. Wavelet anal- ysis indicates 11-min periodicity close to the loop apex. We interpret this oscillation as forcing through upward leakage by the fundamental acoustic eigenmode of the underlying field-free cavity. The observed loop length corresponds to the value pre- dicted from this mechanism. 1 Introduction Field-free cavities under bipolar magnetic canopies (Centeno et al. 2007)inthe vicinity of magnetic network are likely to serve as resonators for fast magnetoacous- tic waves (Kuridze et al. 2007). Srivastava et al. (2008) have studied the properties of the fundamental fast magnetoacoustic mode in brightened magnetic network. It leaks through the magnetic network into the upper solar atmosphere. Recently, Martin´ez Gonz´alez et al. (2007) found evidence for low-lying loops in magnetic internetwork. In EIS/Hinode observations of bright magnetic network, we found a small loop located near the south pole. We search for magnetoacoustic oscillations in this loop through wavelet analysis. 2Observations The observations were acquired on 11 March 2007 during 19:04–19:54UT in the study HPW005 QS Slot 60m. The slot-center position was .X D 118 00 ; Y D973 00 /, with a 40 00  512 00 field of view (Fig. 1).Thedatawerebinned A.K. Srivastava (  ) and B. Rani Aryabhatta Research Institute of Observational Sciences, Manora Peak, Nainital, India D. Kuridze and T.V. Zaqarashvili Abastumani Astrophysical Observatory, Tbilisi, Georgia B.N. Dwivedi Department of Applied Physics, Institute of Technology, Banaras Hindu University, Varanasi, India S.S. Hasan and R.J. Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007/978-3-642-02859-5 50, c  Springer-Verlag Berlin Heidelberg 2010 437 Dynamical Evolution of X-Ray Bright Points with Hinode/XRT R. Kariyappa, B.A. Varghese, E.E. DeLuca, and A.A. van Ballegooijen Abstract We analyzed a 7-h long time sequence of soft X-ray images obtained on 14 April 2007 from a quiet region using the X-Ray Telescope (XRT) onboard Hinode. The aim was to observe intensity oscillations in coronal XBPs of differ- ent brightness and to study differences, if any, in the periodicity of the intensity variations and the heating mechanism during their dynamical evolution. We have compared the XRT images with GONG magnetograms using Coronal Modeling Software (CMS), and found that some of the XBPs are located at magnetic bipoles. The coronal XBPs are highly dynamic and oscillatory in nature, showing a wide variety of time scales in their intensity variations. 1 Introduction Coronal X-ray bright points (XBPs) were discovered using a soft X-ray telescope on a sounding rocket in the late 1960s (Vaiana et al. 1978). Their nature remained enigmatic. Later, using Skylab and Yohkoh/SXT X-ray images, XBPs were studied in detail (Golub et al. 1974; Longcope et al. 2001; Hara and Nakakubo-Morimoto 2003). The number of XBPs that is daily present on the visible hemisphere of the Sun varies from several hundred to a few thousand (Golub et al. 1974), with 800 on the entire solar surface at any given time (Zhang et al. 2001). The number of coronal bright points varies inversely with the solar activity cycle (Sattarov et al. 2002; Hara and Nakakubo-Morimoto 2003). The XBP diameters are about 10–20 00 (Golub et al. 1974). Their lifetime ranges from a few hours to a few days (Zhang et al. 2001; Kariyappa and Varghese 2008). In this contribution, we report the analysis of XBPs on soft X-ray images ob- tained from Hinode/XRT and on magnetograms from GONG. We briefly discuss the dynamical evolution of the XBPs in relation to the magnetic field. R. Kariyappa (  ) and B.A. Varghese Indian Institute of Astrophysics, Bangalore, India E.E. DeLuca and A.A. van Ballegooijen Harvard-Smithsonian Center for Astrophysics, Cambridge, USA S.S. Hasan and R.J. Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007/978-3-642-02859-5 51, c  Springer-Verlag Berlin Heidelberg 2010 440 Dynamical Evolution of X-Ray Bright Points with Hinode/XRT 441 2 Results and Discussion We use a 7-h (17:00–24:00UT) time sequence of soft X-ray images obtained on 14 April 2007 with the X-Ray Telescope (XRT) onboard the Hinode mission, using the Ti poly filter for a quiet region near the center of the solar disk. We selected 14 XBPs foranalysis,markingthemasXBP1,XBP2, ,XBP14,andtwobackground, very dark coronal comparison regions as XBP15 and XBP16. The XRT images have been calibrated using the SSW subroutine xrt prep.pro (Kariyappa and Varghese 2008). We also obtained the full-disk magnetograms obtained with GONG during the XRT observing period. These magnetograms have been co-registered with the XRT images using the Coronal Modeling Software (CMS) developed by the fourth author. The XBPs, defined as the sites where intense brightness enhancement is seen, are highly dynamic in nature. We derived light curves for the XBPs by summing their brightness over small square image cut-outs covering the selected XBPs. Fig. 1 GONG magnetogram overlayed on an XRT image using CMS modeling. The magnetic field lines are computed from a potential-field extrapolation of the magnetogram Helicity at Photospheric and Chromospheric Heights S.K. Tiwari, P. Venkatakrishnan, and K. Sankarasubramanian Abstract In the solar atmosphere, the twist parameter ˛ has the same sign as magnetic helicity. It has been observed using photospheric vector magnetograms that negative/positive helicity is dominant in the northern/southern hemisphere of the Sun. Chromospheric features show dextral/sinistral dominance in the north- ern/southern hemisphere and sigmoids observed in X-rays also have a dominant sense of reverse-S/forward-S in the northern/southern hemisphere. It is of interest whether individual features have one-to-one correspondence in terms of helicity at different atmospheric heights. We use UBF H˛ images from the Dunn Solar Tele- scope (DST) and other H˛ data from Udaipur Solar Observatory and Big Bear Solar Observatory. Near-simultaneous vector magnetograms from the DST are used to es- tablish one-to-one correspondence of helicity at photospheric and chromospheric heights. We plan to extend this investigation with more data including coronal intensities. 1 Introduction Helicity is a physical quantity that measures the degree of linkage and twistedness in the field (Berger and Field 1984). It is derived from a volume integral over the scalar product of the magnetic field B and its vector potential A. Direct calculation of helicity is not possible due to the nonuniqueness of the vector potential A and the limited availability of data sampling different layers of the solar atmosphere. The force-free parameter ˛ estimates one component of helicity, that is, twist, the other component being writhe, which cannot be derived from the available data. This ˛ is a measure of degree of twist per unit axial length. It has the same sign as magnetic helicity (Pevtsov et al. 2008, Pevtsov 2008). It is now well known that negative/positive helicity dominates in the northern/southern hemisphere. For S.K. Tiwari (  ) and P. Venkatakrishnan Udaipur Solar Observatory, Physical Research Laboratory, Udaipur, India K. Sankarasubramanian Space Astron. & Instrument. Div., ISRO Satellite Center, Bangalore, India S.S. Hasan and R.J. Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007/978-3-642-02859-5 52, c  Springer-Verlag Berlin Heidelberg 2010 443 444 S.K. Tiwari et al. active regions the hemispheric helicity rule holds in the photosphere, see Hagino and Sakurai (2005) and references therein. Similarly for the chromospheric and coronal helicity rules, see Bernasconi et al. (2005) and references therein, and Pevtsov et al. (2001) and references therein. The topology of chromospheric and coronal features decide the sign of the associated helicity. Chirality is the term used for the sign of the helicity in these features. Thus, helicity is a physical measure of chirality. The chirality of active region features shows correspondence with the sign of the helic- ity in the associated lower/upper atmospheric features. For example, the chirality of X-ray features with S (inverse-S) shapes are associated with sinistral (dextral) filaments (Martin 2003, Rust 2003). Chae (2000) reported for a few cases that ac- tive filaments showing dextral/sinistral chirality are related with negative/positive magnetic helicity. Pevtsov et al. (2001) demonstrated correspondence between pho- tospheric and coronal chirality for a few active regions. However, this needs to be confirmed. We have reported similar helicity signs at photospheric, chromospheric, and coronal heights for a few active regions (Tiwari et al. 2008). Comparison between magnetic helicity signs at different heights in the solar at- mosphere may be a useful tool to predict solar eruptions leading to interplanetary events. Also, it may help to constrain modeling chromospheric and coronal fea- tures taking the photosphere as boundary condition. However, the data required to do this are not directly available and are often nonconclusive. Vector magnetic fields are not available as routinely as is necessary to derive photospheric twist val- ues. Chromospheric H˛ images may be available most of the time by combining data from different telescopes, but are not always conclusive due to lack of angular resolution. Analysis of coronal loop observations is required to determine coronal helicity signs, but these are also not available routinely. Above all, it is hard to find data taken simultaneously at different heights in the solar atmosphere. In this work we combine photospheric and chromospheric data from multiple solar observatories and telescopes. They were often not taken at precisely the same time. We therefore assume that the sign of the magnetic helicity does not change within a few hours. 2 Sign of Magnetic Helicity The sign of helicity in the photosphere is usually found from the force-free param- eter ˛, for example, ˛ best (Pevtsov et al. 1995), averaged ˛, for example,<˛ z > = < J z =B z > (Pevtsov et al. 1994) with current density J z DrB z ,whereB z is the vertical component of the magnetic field. Some authors have used the current helicity density H c D B z  J z (Bao and Zhang 1998; Hagino and Sakurai 2005). A good correlation was found between ˛ best and h˛ z i by Burnette et al. (2004)and Leka et al. (1996). The force-free parameter ˛ has the same sign as magnetic helicity (Pevtsov et al. 2008). Also, the current helicity (which is not a conserved quantity like magnetic helicity) has the same sign as that of magnetic helicity (Seehafer 1990; Hagyard and Pevtsov 1999; Pevtsov 2008; Sokoloff et al. 2008). In this study, we use the sign of the global ˛ parameter as the sign of magnetic helicity, giving the twist present in the active region. Helicity at Photospheric and Chromospheric Heights 445 Numerical measurement of the sign of the chromospheric magnetic helicity is not possible due to non-availability of vector magnetic field observations at these heights. However, the twist present in morphological intensity features were reported already long ago (Hale 1925; Richardson 1941) to tend to follow the hemi- spheric helicity rule, independent of the solar cycle. Later, many researchers studied the chirality of different chromospheric features such as filaments, fibrils, filament channels etc. (Martin 1998, 2003). We use the chirality of these chromospheric fea- tures, mostly whirls observed in H˛, to derive its association with the photospheric sign of magnetic helicity. 3Data Apart from a few data sets, most are obtained from different solar observatories and telescopes due to the unavailability of all required data from the same place. The vector magnetic field data were obtained from the Advanced Stokes Polarime- ter (ASP, Elmore et al. 1992) as well as the Diffraction Limited Spectropolarimeter (DLSP, Sankarasubramanian et al. 2004, 2006), both mounted at the DST. Near- simultaneous H˛ images from the Universal Bi-refringent Filter (UBF) at the DST are used whenever obtained along with the ASP and DLSP. For the vector field observations that do not have corresponding UBF data, H˛ images from Udaipur Solar Observatory (USO) and Big Bear Solar Observatory (BBSO) were used. We made sure that in these cases the H˛ images were collected within less than a day. Standard and well-established processing was done to derive vector fields. The pro- cedure is described in the references given above. 4 Results and Discussion Table 1 shows how the sign of helicity at the photosphericlevel and of the chirality in associated features at chromospheric heights are related with each other. Figure 1a, b clearly show that the H˛ whirls follow the transverse magnetic field vectors mea- sured at photospheric heights. The positive/negative helicity derived from the global twist in this sunspot in the photospheric vector data is directly associated with the sinistral/dextral chirality derived from the chromospheric H˛ data. In this preliminary analysis, we thus conclude that the sign of helicity (posi- tive/negative) derived from global twist present around sunspots in the photosphere has one-to-one correspondence with the (sinistral/dextral) sense of chirality ob- served in the associated chromospheric data. We mostly use the chirality of chromo- spheric whirls to derive the chromospheric helicity sign. It is known (Martin 1998, 2003) that filaments, filament channels, etc. have the same sense of chirality as the whirls above the associated active regions. The chirality of filaments associated with an active region can therefore be used to determine the chromospheric sense of chi- rality when high resolution H˛ data are not available. Evolution of Coronal Helicity in a Twisted Emerging Active Region B. Ravindra and D.W. Longcope Abstract Active-region magnetic fields are believed to be generated near the shear layer of the convection zone by dynamo processes. These magnetic fields are con- centrated into fluxtubes, which rise, due to buoyancy, through the convection zone to appear in the form of bipoles at the photosphere. Thin-fluxtube simulations sug- gest that active regions require twist to emerge. All regions are observed to emerge with some twist; some of them show larger twist than others. A theoretical model [Longcope and Welsch 2000, ApJ, 545, 1089] predicts that an emerging fluxtube in- jects helicity into the corona for one or two days after its initial emergence through rotation of its footpoints driven by magnetic torque. There have been very few ob- servational studies of helicity injection into the corona by emerging flux. This paper presents a study of helicity during the emergence of active region NOAA 8578. The time history of spin helicity injection, related to footpoint rotation, suggests that an Alfv´en wave packet crossed the apex of the emerging fluxtube. 1 Introduction Active regions often emerge as bipolar regions on the solar surface resembling ˝ shaped fluxtubes. These newly emerged active regions appear as sunspots, concentrated with intense magnetic field strength. Longcope and Welsch (2000)pre- dicted, through a simplified dynamicalmodel, that a fraction of the current generated in the twisted emerging fluxtube will enter the corona. Any initial current mismatch between the photosphere and the corona results in a magnetic torque that drives rotation of the photospheric footpoints. This rotation is part of a torsional Alfv´en wave propagating downward along the fluxtube and injecting helicity upward into the corona. B. Ravindra (  ) Indian Institute of Astrophysics, Bangalore, India D.W. Longcope Department of Physics, Montana State University, Bozeman, USA S.S. Hasan and R.J. Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007/978-3-642-02859-5 53, c  Springer-Verlag Berlin Heidelberg 2010 448 Evolution of Coronal Helicity in a Twisted Emerging Active Region 449 By decomposing the photospheric helicity flux into spin and braiding compo- nents, it is possible to interpret an observed emergence in terms of the model of Longcope and Welsch (2000). In this paper we perform such a decomposition on high-resolution MDI magnetograms (0.6 00 pixel size) taken at 1-min cadence during the emerging of active region NOAA 8578. 2Results Co-aligned portions of MDI magnetograms were corrected for magnetic field under- estimation (Berger and Lites 2003) and then 5-min boxcar-averaged to reduce noise. The horizontal velocities were computed and partitioning of the magnetogram se- quence was performed as in Longcope et al. (2007). The emergence of AR 8578 is shown as a magnetogram time sequence in Fig. 1. The active region started to emerge at 07:00 UT on 08 June 1999 at a rapid rate of 10 20 Mx h 1 . As the active region emerged, the bipoles moved away from each other at a rate of 292 m s 1 . The active region emerged with a tilt away from the East-West direction, and after 40 h of emergence aligned itself along the East-West direction as can be seen in Fig. 1. The two largest regions resulting from the partition, P01 and N01, are shown in the left-hand panel of Fig. 2. The right-hand part of Fig. 2 shows the rotation rate of the regions P01 and N01, computed with the method of Longcope et al. (2007). As soon as the active region started emerging, region P01 rotated anticlockwise, but it gradually changed its direction of rotation to clockwise. The region then changed Fig. 1 Emergence of a bipole in active region NOAA 8578 on the solar surface close to disk center at latitude 19 ı . The date and time of the magnetogram snapshots are specified on each image Power-Law Nanoflare Heating L. Prasad and V.K. Joshi Abstract Nanoflares are small impulsive events in the corona that dissipate magnetic energy in the range 5  10 23 to 10 26 ergs. We model their character- istics through assuming a power-law event distribution. 1 Introduction The nanoflare hypothesis of Parker (1988) is that the corona contains a large assembly of high-temperature elemental magnetic filaments or loops, created to- gether with the coronal magnetic field through randomly distributed impulsive heating events. They heat the loop plasma to 5 10 6 K(Shimizu 1995) and suppos- edly are one of the main agents generating the high coronal temperature (Yoshid a and Tsuneta 1996; Watanabe 1995). Nanoflare coronal heating in terms of power- law distributions has been discussed by Dennis (1985). Kopp and Poletto (1993) extended this work to power-law indices ˛>2. 2 Power-Law Estimates We express the total energy release by nanoflaring as the power law dN dE D AE ˛ ; (1) with ˛ the power law index, dN the number of small impulsive events be- tween energy range E and E C dE,andA a normalization factor (Hudson 1991; Kreplin 1997). The nanoflare rate over the whole solar surface is R total D R E max E min .dN=dE/dE D A=.˛ 1/ E 1˛ min . L. Prasad (  ) and V.K. Joshi Kumaun University, Nainital, India S.S. Hasan and R.J. Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007/978-3-642-02859-5 54, c  Springer-Verlag Berlin Heidelberg 2010 452 [...]... a function of energy in solar flares: the time-of-flight dispersion of free-streaming electrons, the collisional trapping of electrons, and the Neupert effect (Hudson 199 1; Dennis and Zarro 199 3) The characteristic times for heating and cooling of the X-ray emitting plasma in solar flares are estimated from the time profile and temperature and the emission measure of the thermal X-ray burst and the over-all... anomalies in plumes and inter-plume regions, and discuss their implications for better understanding of these structures in the Sun s atmosphere 1 Introduction Polar coronal plumes are ray-like structures aligned along open magnetic field lines in polar coronal holes A total eclipse of the Sun shows these rays in white light, depicting the magnetic configuration of the Sun in a coronal hole Many studies... and R.J Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007 /97 8-3-642-028 59- 5 55, c Springer-Verlag Berlin Heidelberg 2010 454 Spectroscopic Diagnostics of Polar Coronal Plumes 455 environment, such as electron densities, ne , and electron temperatures, Te , the effective ion temperatures and non-thermal motions, the. .. region NOAA 501 on 20 November 2003 The H˛ and magnetogram measurements show interaction between two filaments, which produced a slowly rising flare event, corresponding to two stages of magnetic reconnection The relative clockwise rotation between the two sunspot systems caused filament destabilization The cusp-shaped magnetic field in the main phase of the second flare and its evolution in correlation... disappearance of these plumes The fact remains that we know little about them, probably because we have no direct knowledge of the coronal magnetic field The identification of the sources that produce coronal plumes and their contribution to the fast solar wind is still a matter of investigation (DeForest et al 199 7; Wang et al 199 7; Wilhelm et al 199 8; Gabriel et al 2003; Teriaca et al 2003; Antonucci et al 2004;... scale of the flare-heated plasma at thermal X-ray maximum The heating is assumed to be due to magnetic reconnection, the cooling to heat conduction, and radiation As the conductive losses increase rapidly with temperature, several authors have concluded that conduction dominates radiation at least during the initial phase when T 107 K (Culhane et al 197 0; Moore and Datlowe 197 5) Culhane et al ( 197 0) treated... 10 MK and ı reduces the demand for nonthermal electrons by up to 85% Our paper with these results will be submitted to ApJ Letters In related work on diagnosing ion acceleration and relativistic electron acceleration in flares, we have also been reconsidering the role of Inverse Compton (IC) scattering of photospheric photons Gamma-ray observations clearly show the presence of 100 MeV electrons and positrons... acceleration for two CMEs that occurred on 9 April 2008 and 19 May 2007, respectively, and were observed with the coronagraphs onboard SOHO as well as the STEREO A and B spacecrafts 2 Observations The CME on 9 April 2008 first appeared in the STEREO A COR1 field of view (FOV) at 10:05 UT It showed a classical three-part structure, and was seen near the south-west limb of the Sun Data from STEREO A and B... shows the projected height, speed, and acceleration for the event on 19 May 2007 EUVI 304 images were used to measure the separation between the two footpoints of the prominences preceding the two CMEs 3 Discussion and Summary Based on a statistical study, Zhang and Dere (2006) reported that most CMEs exhibit bimodal acceleration profiles: a main acceleration phase occurring close to the Sun s surface and. .. example, the corona Using the P.C.V Mallik ( ), J.C Brown, and A.L MacKinnon Department of Physics and Astronomy, University of Glasgow, Scotland, UK S.S Hasan and R.J Rutten (eds.), Magnetic Coupling between the Interior and Atmosphere of the Sun, Astrophysics and Space Science Proceedings, DOI 10.1007 /97 8-3-642-028 59- 5 57, c Springer-Verlag Berlin Heidelberg 2010 463 Time-Varying Thermal Emission in . events. They heat the loop plasma to 5 10 6 K(Shimizu 199 5) and suppos- edly are one of the main agents generating the high coronal temperature (Yoshid a and Tsuneta 199 6; Watanabe 199 5). Nanoflare. line ratios Fig. 1 Contribution functions of the observed lines and the FIP values of the corresponding el- ements, based on ionic fractions from Mazzotta et al. ( 199 8). Neon and oxygen have high. Coronal Plumes 455 environment, such as electron densities, n e , and electron temperatures, T e , the effec- tive ion temperatures and non-thermal motions, the plume cross-section relative to the

Ngày đăng: 08/08/2014, 01:20

Tài liệu cùng người dùng

  • Đang cập nhật ...

Tài liệu liên quan