坂尾 太郎

To understand the physical mechanisms for activity and heating in the solar atmosphere, the magnetic coupling from the photosphere to the corona is an important piece of information from the Hinode observations, and therefore precise positional alignment is required among the data acquired by different telescopes. The Hinode spacecraft and its onboard telescopes were developed to allow us to investigate magnetic coupling with co-alignment accuracy better than 1 ''. Using the Mercury transit observed on 2006 November 8 and co-alignment measurements regularly performed on a weekly basis, we have determined the information necessary for precise image co-alignment, and have confirmed that co-alignment better than 1 '' can be realized between Solar Optical Telescope (SOT) and X-Ray Telescope (XRT) with our baseline co-alignment method. This paper presents results from the calibration for precise co-alignment of CCD images from SOT and XRT.

The X-Ray Telescope (XRT) aboard Hinode has revealed the fine structure of solar X-ray jets. One of the fine structures observed by XRT is an expanding loop. The loop appeared near the footpoint of the jet when footpoint brightening was observed. Additionally, we have found that the X-ray jets began just after the expanding loop "breaks". Other fine structures discovered by XRT are thread-like features along the axis of the jets. XRT has shown that these thread structures compose the cross-section of jets. The fine structures and their motions strongly support an X-ray jet model based on magnetic reconnection, and also suggest that we must consider the three-dimensional configuration of the magnetic field to understand the jet phenomenon. We also investigated the reverse jet associated with the X-ray jet in the quiet Sun, and propose that the reverse jet is produced by heat conduction, or a MHD wave subsequent to the main jet.

We used Hinode X-Ray Telescope (XRT) and Solar Optical Telescope (SOT) filtergraph (FG) Stokes-V magnetogram observations, to study the early onset of a solar eruption that includes an erupting filament that we observe in TRACE EUV images. The filament undergoes a slow rise for at least 20min prior to its fast eruption and strong soft X-ray (SXR) flaring; such slow rises have been previously reported, and the new Hinode data elucidate the physical processes occurring during this period. XRT images show that during the slow-rise phase, an SXR sigmoid forms from apparent reconnection low in the sheared core field traced by the filament, and there is a low-level intensity peak in both EUV and SXRs during the slow rise. MDI and SOT FG Stokes-V magnetograms show that the pre-eruption filament is along a neutral line between opposing-polarity enhanced network cells, and the SOT magnetograms show that these opposing fields are flowing together and canceling for at least six hours prior to eruption. From the MDI data we measured the canceling network fields to be similar to 40 G, and we estimated that similar to 10(19) Mx of flux canceled during the five hours prior to eruption; this is only similar to 5% of the total flux spanned by the eruption and flare, but apparently its tether-cutting cancellation was enough to destabilize the sigmoid field holding the filament and resulted in that field's eruption.

We present multi-wavelength observations of the evolution of the sheared magnetic fields in NOAA Active Region 10930, where two X-class flares occurred on 2006 December 13 and December 14, respectively. Observations made with the X-ray Telescope (XRT) and the Solar Optical Telescope (SOT) aboard Hinode suggest that the gradual formation of the sheared magnetic fields in this active region is caused by the rotation and west-to-east motion of an emerging sunspot. In the pre-flare phase of the two flares, XRT shows several highly sheared X-ray loops in the core field region, corresponding to a filament seen in the TRACE EUV observations. XRT observations also show that part of the sheared core field erupted, and another part of the sheared core field stayed behind during the flares, which may explain why a large part of the filament is still seen by TRACE after the flare. About 2 - 3 hours after the peak of each flare, the core field becomes visible in XRT again, and shows a highly sheared inner and less-sheared outer structure. We also find that the post-flare core field is clearly less sheared than the pre-flare core field, which is consistent with the idea that the energy released during the flares is stored in the highly sheared fields prior to the flare.

The Hinode X-ray Telescope provides unprecedented observations of the solar corona in X-rays, due in part to its fine resolution. The X-ray point spread function (PSF) was measured before launch at the NASA X-ray Calibration Facility to have a FWHM of 0 ''.8. This paper describes the work to verify the PSF measurements using on-orbit observations of planetary transits and solar eclipses. Analysis of a Mercury transit gives a PSF FWHM 1 '' 0 +/- 0.'' 12.

The X-ray Telescope (XRT) of the Hinode mission provides an unprecedented combination of spatial and temporal resolution in solar coronal studies. The high sensitivity and broad dynamic range of XRT, coupled with the spacecraft's onboard memory capacity and the planned downlink capability will permit a broad range of coronal studies over an extended period of time, for targets ranging from quiet Sun to X-flares. This paper discusses in detail the design, calibration, and measured performance of the XRT instrument up to the focal plane. The CCD camera and data handling are discussed separately in a companion paper. © 2007 Springer.

The Hinode satellite (formerly Solar-B) of the Japan Aerospace Exploration Agency's Institute of Space and Astronautical Science (ISAS/JAXA) was successfully launched in September 2006. As the successor to the Yohkoh mission, it aims to understand how magnetic energy gets transferred from the photosphere to the upper atmosphere and results in explosive energy releases. Hinode is an observatory style mission, with all the instruments being designed and built to work together to address the science aims. There are three instruments onboard: the Solar Optical Telescope (SOT), the EUV Imaging Spectrometer (EIS), and the X-Ray Telescope (XRT). This paper provides an overview of the mission, detailing the satellite, the scientific payload, and operations. It will conclude with discussions on how the international science community can participate in the analysis of the mission data. © 2007 Springer Science+Business Media B.V.

We present H alpha observations from ARIES (Nainital) of a compact and impulsive solar flare that occurred on March 10, 2001 and which was associated with a CME. We have also analyzed HXT, SXT/Yohkoh observations as well as radio observations from the Nobeyama Radio Observatory to derive the energetics and dynamics of this impulsive flare. We coalign the H alpha, SXR, HXR, MW, and magnetogram images within the instrumental spatial-resolution limit. We detect a single HXR source in this flare, which is found spatially associated with one of the H alpha bright kernels. The unusual feature of HXR and H alpha sources, observed for the first time, is the rotation during the impulsive phase in a clockwise direction. We propose that the rotation may be due to asymmetric progress of the magnetic reconnection site or may be due to the change of the peak point of the electric field. In MW emission we found two sources. The main source is at the main flare site and another is in the southwest direction. It appears that the remote source is formed by the impact of accelerated energetic electrons from the main flare site. From the spatial correlation of multiwavelength images of the different sources, we conclude that this flare has a three-legged structure.

On the basis of the Yohkoh Hard X-Ray Telescope (HXT) data, we present a statistical study of different types of the hard X-ray ( HXR) source motions during solar flares. A total of 72 flares that occurred from 1991 September to 2001 December have been analyzed. In these flares, we have found 198 intense HXR sources that are presumably the chromospheric footpoints of flare loops. The average velocity V and its uncertainty σ were determined for these sources. For 80% of them, the ratio of V to 3 σ is larger than 1, strongly suggesting that (1) the moving sources are usually observed rather than stationary ones and (2) the regular displacements of HXR sources dominate their chaotic motions. After co-alignment of the HXT images with the photospheric magnetograms, we have conducted an additional analysis of 31 flares out of 72 and distinguished between three main types of the footpoint motions. Type I consists of the motions preferentially away from and nearly perpendicular to the neutral line (NL). About 13% of flares (4 out of 31) show this pattern. In type II, the sources move mainly along the NL in antiparallel directions. Such motions have been found in 26% of flares (8 out of 31). Type III involves a similar pattern as type II, but all the HXR sources move in the same direction along the NL. Flares of this type constitute 35% (11 out of 31). In 26% of flares (8 out of 31) we observed more complicated motions that can be described as a combination of the basic types or some modification of them. For the most interesting flares, the results of analysis are illustrated and interpretation is suggested. © 2005. The American Astronomical Society. All rights reserved.

The X-Ray Telescope (XRT) on Solar-B is designed to provide high resolution, high cadence observations of the X-ray corona through a wide range of filters. The XRT science team has identified four general problems in coronal physics as the primary science goals for our instrument. Each of these goals will require collaborative observations from the other Solar-B instruments: EUV Imaging Spectrograph (EIS) and Solar Optical Telescope Focal Plane Package (SOT). We will discuss the science goals and observations needed to address those goals. © 2005 COSPAR. Published by Elsevier Ltd. All rights reserved.

Using the Yohkoh Hard X-Ray Telescope (HXT) data, we have examined motions of the hard X-ray (HXR) sources during 72 solar flares occurred from 1991 September to 2001 December. In these flares, we have found 198 intense sources that are presumably the chromospheric footpoints (FPs) of flare loops. The average velocity V and the velocity dispersion σ were determined by a linear regression for these sources. For 80% of them, the ratio of V to 3σ is larger than 1, strongly suggesting that the regular motions of the HXR sources dominate their chaotic motions. For 43 of 72 flares, coalignment of the HXT images with the photospheric magnetograms allows us to consider the HXR sources located on the both sides of the photospheric neutral line (NL) as the FP sources, and to distinguish between three main types of the FP motions. The type I is the motions of the HXR sources preferentially away from and nearly perpendicular to the NL. Less than 5% of the flares show this pattern of motion. In the type II, the sources move mainly along the NL in anti-parallel directions. Such motions have been found in 26% of flares. The type III involves a similar pattern of motions as the type II but all the HXR sources move in the same direction along the NL. Flares of this type constitute 30% of the flares. About 19% of flares can be described as a combination of these basic types. The remaining 20% of flares seem to be more complicated or less regular in the motion scale under consideration. An interpretation of results is suggested. © 2005 COSPAR. Published by Elsevier Ltd. All rights reserved.

Coronal hard X-ray (HXR) sources were discovered by the Yohkoh HXT telescope in about two dozen limb flares: Impulsive and gradual ones. On the basis of HXT data, we investigated the spatial evolution of coronal sources. Slow ascending motions of sources are seen in several flares. In five events, it was possible to estimate the velocity of the upward motion with values between 10 and 30 km/s. We present these observational results and conclude that coronal source motions should be studied statistically using the RHESSI high-resolution HXR imaging data. We discuss the possibility that coronal HXR emission is generated as bremsstrahlung of the fast electrons accelerated in collapsing magnetic traps due to joint action of the Fermi-type first-order mechanism and betatron acceleration. © 2005 COSPAR. Published by Elsevier Ltd. All rights reserved.

We present the results of a detailed analysis of multi-wavelength observations of a very impulsive solar flare 1B/M6.7, which occurred on 10 March, 2001 in NOAA AR 9368 (N27 W42). The observations show that the flare is very impulsive with a very hard spectrum in HXR that reveal that non-thermal emission was most dominant. On the other hand, this flare also produced a type II radio burst and coronal mass ejections (CME), which are not general characteristics for impulsive flares. In Hα we observed bright mass ejecta (BME) followed by dark mass ejecta (DME). Based on the consistency of the onset times and directions of BME and CME, we conclude that these two phenomena are closely associated. It is inferred that the energy build-up took place due to photospheric reconnection between emerging positive parasitic polarity and predominant negative polarity, which resulted as a consequence of flux cancellation. The shear increased to >80° due to further emergence of positive parasitic polarity causing strongly enhanced cancellation of flux. It appears that such enhanced magnetic flux cancellation in a strongly sheared region triggered the impulsive flare. © Springer 2005.

The Soft X-ray Telescope (XRT) aboard Solar-B is a grazing incidence X-ray telescope equipped with 2k x 2k CCD. XRT has I arcsec resolution with wide field-of-view of 34 x 34 arcmin. It is sensitive to < 1 MK to 30 MK, allowing us to obtain TRACE-like low temperature images as well. Co-alignment with SOT and EIS is realized through the XRT visible light telescope and with temperature overlap with EIS. Spacecraft mission data processor (MDP) controls XRT through the sequence tables with versatile autonomous functions such as exposure control, region-of-interest tracking, flare detection and flare location identification. Data are compressed either with DPCM or JPEG, depending on the purpose. This results in higher cadence and/or wider field-of-view for given telemetry bandwidth. With focus adjust mechanism, higher resolution of Gaussian focus may be available on-axis.

The tip-tilt mirror (TTM) system was developed for the extreme ultraviolet (XUV) Cassegrain telescope aboard the Institute of Space and Astronautical Science sounding rocket. The spatial resolution of the telescope is about 5 arcsec, whereas the rocket pointing is only controlled to be within ± 0. 5 deg of the target (sun) without additional stability control. To stabilize the XIV image within about 5 arcsec on the focal plane, the TTM system controls the tilt of the secondary mirror with two-axis fixed-coil magnetic actuators. The TTM system has a wide tilt angle and can drive the large secondary mirror at high frequency. The two position-sensitive detector-, one placed in the telescope and the other in the TTM mechanical structure, are used for closed-loop control of toe TTM. The closed-loop control system, which has command and telemetry, is executed by the flight software on the digital signal processor. The TTM has a launch-lock mechanism to protect against launch vibrations up to about 16G. The sounding rocket was launched from the Kagoshima Space Center on 31 January 1998. The TTM worked perfectly during the flight and achieved better than the expected 5-arcsec stability on the focal plane during CCD charge-coupled device exposures.

We present scientific as well as engineering overview of the X-Ray Telescope (XRT) aboard the Japanese Solar-B mission to be launched in 2006, with emphasis on the focal plane CCD camera that employs a 2k×2k back-thinned CCD. Characterization activities for the flight CCD camera made at the National Astronomical Observatory of Japan (NAOJ) are discussed in detail with some of the results presented.

A suite of images from the XUV Doppler Telescope (XDT) the Yohkoh Soft X-ray Telescope (SXT), and the Extreme-Ultraviolet Imaging Telescope (EIT) on the Solar and Heliospheric Observatory (SOHO) allow us to see the whole (T > 1 MK) temperature evolution of coronal loops. The detailed morphological comparison of an active region shows that hot loops seen in SXT (T > 3 MK) and cool loops seen in the the EIT 195 Å band (T ∼ 1.5 MK) are located in almost alternating manner. The anticoincidence of the hot and the cool loops is conserved for a duration much longer than the estimated cooling timescale. However, both hot and cool loops have counterparts in the intermediate-temperature images. The cross-correlation coefficients are higher for neighboring temperature pairs and lower for pairs with larger temperature differences. These results suggest that loops are not isothermal but rather have a differential emission measure distribution of modest but finite width that peaks at different temperatures for different loops.

The basic ideas to model the large solar flares are reviewed and illustrated. Some fundamental properties of potential and non-potential fields in the solar atmosphere are recalled. In particular, we consider a classification of the non-potential fields or, more exactly, related electric currents, including reconnecting current layers. The so-called 'rainbow reconnection' model is presented with its properties and predictions. This model allows us to understand main features of large flares in terms of reconnection. We assume that in the two-ribbon flares, like the Bastille-day flare, the magnetic separatrices are involved in a large-scale shear photospheric flow in the presence of reconnecting current layers generated by a converging flow. © 2003 COSPAR. Published by Elsevier Ltd. All rights reserved.

On the basis of Yohkoh Hard X-Ray Telescope data and the magnetograms taken by the SOHO Michelson Doppler Imager and the Solar Magnetic Field Telescope at Huairou Solar Observing Station, we suggest an interpretation of the well-observed "Bastille Day 2000" flare. The large-scale structure and dynamics of the flare, as seen in hard X-rays, can be explained in terms of the three-dimensional reconnection at a separator in the corona. More specifically, we suggest that before occurrence of two-ribbon flares with significant decrease of a distance between the hard X-ray (HXR) footpoints, like the Bastille Day flare, the bases of magnetic field separatrices are moved by the large-scale photospheric flows of two types. First, the shear flows, which are parallel to the photospheric neutral line, increase the length of field lines in the corona and produce an excess of magnetic energy. Second, the converging flows, i.e., the flows directed to the neutral line, create the preflare current layers in the corona and provide an excess of energy sufficient to produce a large flare. During the flare, both excesses of magnetic energy are released completely or partially. In the Bastille Day flare, the model describes two kinds of apparent motions of the HXR kernels. One is an increase of a distance between the flare ribbons in which the HXR kernels appear. The effect results from fast reconnection in a coronal current layer. The second effect is a decrease of the distance between the kernels moving to each other as a result of relaxation of magnetic tensions generated by the photospheric shear flows.

We developed a unique telescope to obtain simultaneous XUV images and the velocity maps by measuring the line-of-sight Doppler shifts of the Fe XIV 211Å line (T = 1.8 MK): the Solar XUV Doppler Telescope (hereafter XDT). The telescope was launched by the Institute of Space and Astronautical Science with the 22nd S520 rocket on January 31, 1998, and took 14 XUV whole sun images during 5 minutes. Simultaneous observations of XDT with Yohkoh (SXT), SOHO (EIT, CDS, LASCO and MDI) were successfully carried out. The images taken with EIT, XDT and SXT are able to cover the wide temperature ranging from 1 to 10 MK, and clearly show the multi-temperature nature of the solar corona. Indeed, we notice that both the cool (1-2 MK) loops observed with EIT and XDT, and the hot (>3 MK) loops observed with SXT exist in the same active regions but in a spatially exclusive way. The XDT red-blue ratio between longer-and shorter-wavelength bands of Fe XIV 211Å line indicates a possible down-flow of 1.8 MK plasma near the footpoints of multiple cool loops. © 2000 COSPAR. Published by Elsevier Science Ltd.

New results concerning magnetic reconnection in the solar corona are reviewed. We apply the collisionless 3D reconnection theory - more exactly, the model of a high-temperature turbulent-current non-neutral current sheet - to the coronal conditions derived from the observations with the HXT and SXT onboard Yohkoh. New interpretations are discussed of the Yohkoh data on the site and mechanism of magnetic energy transformation into kinetic and thermal energies of superhot plasmas and accelerated high-energy particles. Open issues are the focus of our attention. © 2000 COSPAR. Published by Elsevier Science Ltd.

It is observed in impulsive flares with the Hard X-ray Telescope (HXT) onboard Yohkoh that a hard X-ray source is located above the apex of a soft X-ray flaring loop in addition to double footpoint sources (Masuda et al. 1994, 1995). This observation suggests that the flare energy-release, probably magnetic reconnection, takes place not in the soft X-ray loop, but above the loop. It is important to derive the hard X-ray spectrum of this source accurately in order to understand how electrons are energized there. Using an improved image-synthesis technique with new modulation patterns (Sato 1997), we obtained hard X-ray images of the 13 January 1992 flare in the four HXT energy bands with a common photon-accumulation time. The spectrum of the above-the-looptop source in the energy range of 14-93 keV seems to be well fit by emission from an isothermal plasma whose temperature is about 100 MK, rather than by a non-thermal, single power-law spectrum. Although this is consistent with the observational result that the density of the ambient plasma is not high enough to stop high energy electrons, we need to study other events, especially intense events from which more accurate spectra can be derived, in order to confirm this result. © 2000 COSPAR. Published by Elsevier Science Ltd.

We present hard X-ray imaging observations of impulsive solar flares with Yohkoh HXT, focusing on the evolution of hard X-ray double footpoint sources seen in the energy range above 30 keV. In 7 out of the 14 events studied, the separation between the double sources increases (at the 3σ confidence level) as the flare progresses, with the double sources moving nearly anti-parallel to each other in most cases. Implications of these results to the global magnetic field structure responsible for solar flare energy release / particle acceleration are discussed. © 2000 COSPAR. Published by Elsevier Science Ltd.

The imaging capability of the Hard X-ray Telescope (HXT) on board Yohkoh has been drastically improved by (1) adopting new instrumental response functions (modulation patterns) derived from a self-calibration procedure that makes use of solar flares themselves as calibration sources, (2) revising the Maximum Entropy (MEM) image synthesis procedure for better total flux estimation, and (3) incorporating in MEM properly estimated observation errors. One of the most intense flares so far observed with HXT, the 1997 November 6, X9 flare, has been analyzed with the new HXT imaging program. Two footpoint sources are clearly seen and show systematic motions during the impulsive phase. This may provide a new clue to understand the coronal magnetic field structure involved in the energy release process of flares. © 2000 COSPAR. Published by Elsevier Science Ltd.

The XUV Doppler Telescope (XDT) is a sounding rocket experiment designed to detect flows in the solar corona using filter ratios. The XDT, successfully launched on 1998 January 31, is a normal incidence telescope composed of narrow-bandpass multilayer mirrors and capable of obtaining images 2 Å above and 2 Å below the Fe XIV 211 Å (T = 1.7 MK) emission line. It has the potential to make a velocity map of the entire solar disk with just a few minutes of observation. The image ratio maps show features that translate to Doppler shifts of 200 km s-1 or more, including several 'redshift' features located near footpoints of coronal loops. However, no corresponding velocity features were seen by the Solar and Heliospheric Observatory (SOHO) Coronal Diagnostic Spectrometer (CDS) in the Mg IX 368 Å (T = 1 MK) line, suggesting that the features are not caused by Doppler shift. Instead, the features seem to be related to contamination of lower temperature (T < 1 MK) emission lines and the nearby density-sensitive Fe XIII lines. We conclude that while no flows were positively identified by the XDT, this observing technique is capable of detecting flows of 1000 km s-1 independently, and 200 km s-1 when combined with simultaneous plasma diagnostic observation.

The X-ray observations from the Yohkoh SXT provided the greatest step forward in our understanding of the solar corona in nearly two decades. We believe that the scientific objectives of the Solar-B mission can best be achieved with an X-ray telescope (XRT) similar to the SXT, but with significant improvements in spatial resolution and in temperature response that take into account the knowledge gained from Yohkoh. We present the scientific justification for this view, discuss the instrumental requirements that flow from the scientific objectives, and describe the instrumentation that will meet these requirements. XRT is a grazing-incidence (GI) modified Wolter I X-ray telescope, of 35 cm inner diameter and 2.7 m focal length. The 2048×2048 back-illuminated CCD has 13.5 μ pixels, corresponding to 1.0 arcsec and giving full Sun field of view. This will be the highest resolution GI X-ray telescope ever flown for Solar coronal studies, and it has been designed specifically to observe both the high and low temperature coronal plasma.

We have developed narrow-bandpass MoSi/Si multilayer mirrors for a Japanese sounding-rocket program. A high spectral resolution λ/Δλ exceeding 40 was achieved by a two-mirror telescope with a multilayer coating. The single telescope had two bandpasses in the extreme-UV range for detecting a coronal high-velocity flow; the wavelength at peak reflectance (hereafter peak wavelength) in one of the bandpasses was 210.2 å, situated on the blue side of the target’s Fe XIV 211.3-å coronal emission line, and the peak wavelength in the other was 213.3 å on the red side. A high uniformity in a peak wavelength of less than 1 å was achieved over a primary (secondary) mirror surface 158 (96) mm in diameter. The ratio of the reflectance for the Fe XIV line at 211 å to that for an intense He II line as a contaminant at 304 å in the telescope system became 2 × 105 owing to a wave trap consisting of a single Si layer on the MoSi/Si multilayer. The narrow-bandpass (;5-å) telescope was launched on 31 January 1998 by sounding rocket S520CN-22, and images of the whole-Sun corona at Fe XIV 211 å were successfully obtained. © 1999 Optical Society of America.

We analyze the footpoint separation d and flux asymmetry A of magnetically conjugate double footpoint sources in hard X-ray images from the Yohkoh Hard X-Ray Telescope (HXT). The data set of 54 solar flares includes all events simultaneously observed with the Compton Gamma Ray Observatory (CGRO) in high time resolution mode. From the CGRO data we deconvolved the direct-precipitation and trap-precipitation components previously (in Paper II). Using the combined measurements from CGRO and HXT, we develop an asymmetric trap model that allows us to quantify the relative fractions of four different electron components, i.e., the ratios of direct-precipitating (qP1, qP2) and trapprecipitating electrons (qT1, qT2) at both magnetically conjugate footpoints. We find mean ratios of qP1 = 0.14 ± 0.06, qP2 = 0.26 ± 0.10, and qT = qT1 + qT2 = 0.60 ± 0.13. We assume an isotropic pitchangle distribution at the acceleration site and double-sided trap precipitation (qT2/qT1 = qP2/qP1) to determine the conjugate loss-cone angles (α1 = 42° ± 11° and α2 = 52° ± 10°) and magnetic mirror ratios at both footpoints (R1 = 1.6, ..., 4.0 and R2 = 1.3, ..., 2.5). From the relative displacement of footpoint sources we also measure altitude differences of hard X-ray emission at different energies, which are found to decrease systematically with higher energies, with a statistical height difference of hL0 - hM1 = 980 ± 250 km and hM1 - hM2 = 310 ± 300 km between the three lower HXT energy channels (Lo, M1, M2).

We present an overview and instrumental details of the solar XUV Doppler Telescope (XDT) launched in January 1998 with the S520CN-22 sounding rocket of the Institute of Space and Astronautical Science. The XDT observes nearly single-temperature solar corona at 1.8 MK with angular resolution of ≈ 5″ pixel size, together with the ability to detect the coronal velocity field with a full-Sun field of view. By use of normal incidence optics whose primary and secondary mirrors are coated with multilayer materials in two sectors, the XDT takes images of the Sun in a set of shorter and longer wavelength bands around the FeXIV 211.3 Å emission line. Summation of a pair of images in the two bands provides an image of the 1.8 MK-corona while the difference between the two provides velocity images of the Fe XIV-emitting plasma. A brief description on the observation sequence together with the flight result is also given.

We have analysed two Long Duration solar Events (LDEs) which produced large systems of Post Flare Loops (PFLs) and which have been observed by Yohkoh and ground-based observatories. Using the Maximum Entropy Method (MEM) image synthesis technique with new modulation patterns we were able to make hard X-ray (HXR) images of the post flare loops recorded in the L Channel (13.9-22.7 keV) of the Yohkoh Hard X-ray Telescope. We obtained co-aligned 2-D maps in Hα (104 K), in soft X-rays (5 × 106 K) and in hard X-rays (20 × 106 K). We conclude that the soft X-ray (SXR) loops lie higher than the Ha loops and the former are overlaid by HXR emission. This is suggestive of the magnetic reconnection process. However some details are not consistent with the standard models. Firstly the separation between the HXR source and the SXR loop increases with time, with the HXR source being approximately a factor of five larger than the equivalent source in impulsive flares. Secondly the cooling times deduced from observations are longer than the theoretically expected ones and the discrepancy increases with time. We review the current models in view of these results.

We present an overview of an ongoing Japanese sounding rocket project with the Solar XUV Doppler Telescope (XDT) scheduled to be launched in January, 1998. The telescope employs a pair of high-wavelength-resolution multilayer mirrors and a back-thinned CCD. It is designed to observe the coronal velocity field of the whole Sun by measuring line-of-sight Doppler shifts of the Fe XIV 211 Angstrom line (1.8 MK). The detection limit is estimated to be about 60 km/s.

Two subclasses of impulsive solar flares, observed with the Hard X-Ray Telescope (HXT) onboard Yohkoh, have been discovered by Sakao et al. The two subclasses can be characterized as more impulsive (MI) and less impulsive (LI) flares, the former having a shorter total duration of the impulsive phase in the hard X-ray emission than the latter. We assume that in both subclasses, the collisionless three-dimensional reconnection process occurs at the separator with a longitudinal magnetic field. The high-temperature turbulent-current sheet (HTTCS), located along the separator, generates accelerated particles and fast outflows of "superhot" (T ≥ 30 MK) plasma. Powerful anomalous heat-conductive fluxes along the reconnected field lines maintain a high temperature in the superhot plasma. The difference between the LI and MI flares presumably appears because the footpoint separation (the distance between two brightest hard X-ray sources) increases in time in the LI flares, but decreases in the MI flares. According to our model, in the LI flares the three-dimensional reconnection process accompanies an increase in the longitudinal magnetic field at the separator. In contrast, in the MI flares the reconnection proceeds with a decrease of the longitudinal field; hence, the reconnection rate is higher in the MI flares. Since reconnection in the MI flares proceeds with a decrease of the longitudinal field, the reconnected field lines become shorter in this process. As the reconnected lines become shorter, accelerated electron beams arrive at the upper chromosphere faster. So, in the MI flares chromospheric evaporation begins earlier than in the LI flares. The evaporation process driven by accelerated electron beams generates upflows of "warm" (T ≤ 10 MK) plasma that interacts with downflows of superhot plasma and can switch off the accumulation of superhot plasma in the MI flares during the impulsive phase. In the LI flares, however, an observable amount of superhot plasma is accumulated even during the impulsive phase. Moreover, since the cooling timescales increase with the length of the reconnected field lines, our argument for the association of superhot plasma with longer lines may remain valid to a reasonable extent even if the chromospheric evaporated plasma mixes with the reconnected outflow and superhot temperatures are reached in this mixture. Further analysis of the Yohkoh data obtained simultaneously with the Hard and Soft X-Ray Telescopes and the bent crystal spectrometer (BCS) is necessary to distinguish the superhot components of chromospheric and coronal origins in different classes of flares as well as at different phases of their development. © 1998. The American Astronomical Society. All rights reserved.

This paper describes the design and prelaunch performance of the tip-tilt mirror (TTM) system developed for the XUV Cassegrain telescope aboard the ISAS sounding rocket experiment. The spatial resolution of the telescope is about 5 arcsec, whereas the rocket pointing is only controlled to be within +1-0.5 degree around the target (Sun) without stability control. The 'ITM is utilized to stabilize the XUV image on the focal plane by tilting the secondary mirror with two-axes fixed-coil magnetic actuators. The actuator used here is robust, has wide tilt angle, and is able to drive the large secondary mirror (10cm dia., 300g) with high frequency, as compared with conventional Piezo-electric or moving-coil type actuators. The two position-sensitive detectors in the telescope optics and in the TFM mechanical structure form the normal and local closed-loop modes. The Tl'M has four gain modes with automatic transition among the modes. The low gain mode is used in the initial acquisition, and in case the TTM loses the tracking. The high gain mode is used in the normal tracking mode. This arrangement provides us with the wide initial acquisition angle with single T1'M system as well as the high pointing accuracy once the tracking is established. The TFM has a launch-lock mechanism against the launch vibration of 160. The closed-loop control with command and telemetry interface is done by the flight software on the DSP processor. The use of the fast processor brings in the significant reduction in the weight and size of the control-electronics, more flexible control system, and shorter design and testing period.

Multilayer mirrors with a system wavelength resolution (λ/Δλ) as high as 30-50 are required for the diagnostics of cosmic plasmas with temperatures of 1-20 MK. Such a high wavelength resolution can be realized by increasing the number of layer pairs contributing to the reflectance, by selecting lessabsorbing materials for both the reflector and the spacer, and by decreasing the thickness of the reflector. We have fabricated a multilayer mirror tuned to 284 Å with a silicon carbide reflector (20% thickness of the layer period) and an aluminum spacer and achieved λ/Δλ ; 26.8 with a peak reflectivity of ∼13.0%. This wavelength resolution is close to the value obtained with a numerical simulation and is considerably higher than the value obtained with the conventional MoySi multilayer. © 1997 Optical Society of America.

We present the development status of the normal incidence XUV multilayer mirrors for XUV Doppler Telescope, which observes coronal velocity fields of the whole Sun. The telescope has two narrow band-pass multilayer mirrors tuned to slightly longer and shorter wavelengths around the Fe XIV line at 211.3 Å. From the intensity difference of the images taken with these two bands, we can obtain Dopplergram of 1.8 MK plasma of the whole Sun. It is required that the multilayer has high wavelength-resolution (λ/Δλ ∼ 30 per mirror), anti-reflection coating for intense He II 304 Å emission line and high d-spacing uniformity of ∼ 1%.

A XUV Doppler telescope was developed using multilayer optics. A high-spectral resolution is achieved by adopting the multilayer materials. Whole-sun images were obtained in an XUV emission line, Fe XIV 211Å to carry out the velocity-field measurement with detection limit of 100 km/s.

© 1996 SPIE. All rights reserved. We present an overview of an ongoing Japanese sounding rocket project with the Solar XUV Doppler telescope. The telescope employs a pair of normal incidence multilayer mirrors and a back-thinned CCD, and is designed to observe coronal velocity field of the whole sun by measuring line-of-sight Doppler shifts of the Fe XIV 211 A line. The velocity detection limit is estimated to be better than 100 km/s. The telescope will be launched by the Institute of Space and Astronautical Science (ISAS) in 1998, when the solar activity is going to be increasing towards the cycle 23 activity maximum. Together with the overview of the telescope, the current status of the development of each telescope components including multilayer mirrors, telescope structure, image stabilization mechanism, and focal plane assembly, are reviewed. The observation sequence during the flight is also briefly described.

A small flare (C4.3 in the GOES X-ray class) was well observed by all of the instruments on board Yohkoh. The X-ray light curves have double peaks which are about 5 min apart. Until the first peak from flare onset, four compact areas brighten up in the soft X-ray region, which are aligned almost on one straight line. We regard them as being footpoints of two sets of loops, which are identifiable in soft X-ray images, since their locations match those of hard X-ray sources. Indeed, after the second peak, the temporal behavior of the temperature and emission measure at each point is consistent with the existence of two such loops. Comparing our results with recent MHD simulations, we propose a possible scenario for this flare that is based on the coalescence of two loops.

Imaging observations of solar flare hard X-ray sources with the Hard X-ray Telescope aboard the Yohkoh satellite have revealed that hard X-ray emissions (〉30 keV) originate most frequently from double sources. The double sources are located on both sides of the magnetic neutral line, suggesting that the bulk of hard X-rays is emitted from footpoints of flaring magnetic loops. We also found that hard X-rays from the double sources are emitted simultaneously within a fraction of second and that the weaker source tends to be located in the stronger magnetic field region, showing a softer spectrum. Physical implications on the observed characteristics of the hard X-ray double sources are discussed. © 1995, All rights reserved.

We present multifrequency spectra of the Seyfert 1 galaxy H2106-099, from radio to hard X-rays, spanning over a decade of observations. The hard X-ray (2-20 keV) spectrum measured with Ginga was not unusual, with a log slope (Fν ∝ Eα), of -0.80 ± 0.02 on 1988 May 18 and -1.02 ± 0.10 on 1988 May 22/23 UT, and with no significant observed variations in total flux. Other measurements showed variability and unusual spectral features: The V band flux was observed to change by a factor of 1.8 (> 10 σ) in 6 weeks. Only moderate optical Fe II emission is present, but strong [Fe VII] and [Fe X] epochs. The Balmer lines show greater than 25% variations in flux relative to the mean, and He I changed by more than 100% relative to the mean in 6 yr. The most surprising finds came from the composite UV through near-IR spectrum: If the spectrum is dereddened by the galactic extinction value derived from 21 cm observations, a residual 2175 Å absorption feature is present. Additional dereddening to correct the feature yields E(B-V) = 0.07 mag due to material outside our Galaxy, most probably associated with the active galactic nucleus (AGN) or its host galaxy. No other clear indications of reddening are observed in this object, suggesting that blue bump strength measurements in low and intermediate redshift AGNs could be incorrect if derived without UV observations of the region near 2175 A in the AGN frame. After all reddening corrections are performed, the log slope of H2106-099 from the near-IR (∼ 12500 Å) to the UV (∼1400 Å), -0.94 ± 0.05, is steep compared to other AGNs, suggesting that the blue bump in this object is intrinsically weak. Weak blue bumps are, therefore, not always an artifact caused by reddening. The spectral indices of this object in the optical-UV region are steeper than those predicted by optically thin free-free emission models; therefore, some other mechanism must dominate the continuum in this region. © 1996. The American Astronomical Society. All rights reserved.

We discuss the dynamics of four soft X-ray flares as revealed by the X-ray resonance line profiles of the He-like ions, Fe XXV, Ca XIX, and S XV. The flares chosen for analysis belong to a subclass of events that have intense Doppler blueshifted spectral signatures during the rise phase. The spectra were obtained from the Bragg crystal spectrometer (BCS) experiment flown on the Japanese Yohkoh spacecraft. We deconvolve the line profiles into a flare decay phase component (stationary component) produced by nonmoving plasma with little or no turbulent motions, and a flare rise phase component (dynamic component) produced by bulk plasma motion and turbulence. The dynamic component is further deconvolved into a high-speed component (greater than 400 km s-1) and a lower speed, more turbulent component. We find that the dynamic and stationary components have approximately the same electron temperature. The dynamic and stationary component fluxes are compared with hard X-ray fluxes obtained from the hard X-ray telescope (HXT) on Yohkoh. We find that the onset time of the high-speed dynamic component corresponds closely to the onset of hard X-rays. We discuss the X-ray morphology of the flares as revealed by the soft X-ray telescope (SXT) and hard X-ray telescope (HXT) on Yohkoh. From analysis of the X-ray light curves of individual pixels in the SXT images, we believe we have been able to determine at least some of the locations of the dynamic component in the images. Although the observations support many aspects of chromospheric evaporation models, there are significant problems. No obvious rising fronts of plasma are observed for any of the flares. There are apparent footpoint hard X-ray sources that cannot be physically related in an obvious manner with the main sources of soft X-ray emission. The flares are complex, and therefore there are some ambiguities in interpretation of the morphology. © 1996. The American Astronomical Society. All rights reserved.

Using observations from the Yohkoh Bragg Crystal Spectrometer, hard X-ray telescope, and soft X-ray telescope, we have examined eight limb flares. Four of the flares have the footpoints occulted by the solar limb. We find that the occulted flares generally have softer hard X-ray spectra and smaller peak values of the nonthermal broadening velocity than nonocculted flares. All other physical parameters show no differences between occulted flares and nonocculted flares. The hard X-ray spectra support a model in which the footpoint emission is due to thick-target sources, while the looptop emission is due to thin-target sources. High spectral resolution hard X-ray observations should thus show a break in the hard X-ray spectrum of the looptop source. We can find no obvious explanation for the differences in nonthermal broadening velocity. © 1996. The American Astronomical Society. All rights reserved.

We investigate radio and X-ray imaging data for two solar flares in order to test the idea that asymmetric precipitation of nonthermal electrons at the two ends of a magnetic loop is consistent with the magnetic-mirroring explanation. The events we present were observed in 1993 May by the HXT and SXT X-ray telescopes on the Yohkoh spacecraft and by the Nobeyama 17 GHz radioheliograph. The hard X-ray images in one case show two well-separated sources; the radio images indicate circularly polarized, nonthermal radio emission with opposite polarities from these two sources, indicating oppositely directed fields and consistent with a single-loop model. In the second event there are several sources in the HXT images which appear to be connected by soft X-ray loops. The strongest hard X-ray source has unpplarized radio emission, whereas the strongest radio emission lies over strong magnetic fields and is polarized. In both events the strongest radio emission is highly polarized and not coincident with the strongest hard X-ray emission. This is consistent with asymmetric loops in which the bulk of the precipitation (and hence the X-ray emission) occurs at the weaker field footpoint.

We compare 1-14 GHz microwave images observed at the Owens Valley Radio Observatory (OVRO), 16-and 256-channel hard X-ray spectra obtained by the Burst and Transient Source Experiment (BATSE) onboard the Compton Gamma Ray Observatory (CGRO), soft and and hard X-ray images obtained by Yohkoh, and Hα images and magnetograms observed at the Big Bear Solar Observatory (BBSO) for the 1992 June 25 M1.4 flare. We find the following unique properties for this flare: (1) Soft X-ray emissions connect two footpoints, the primary microwave source is located at one footpoint, and hard X-ray emissions are concentrated in the other footpoint. The radio footpoint is associated with an umbra and may have stronger magnetic field. (2) During the period that 256-channel BATSE data are available, the hard X-ray photon spectrum consists of two components: a superhot component with a temperature of 8.4 × 107 K and emission measure of 2.5 x 1046 cm-3 and a power-law component with a photon index of 4.2. This is the first time that such a high temperature is reported for the hard X-ray thermal components. It is even more interesting that such a superhot component is identified before the peak of the flare. The microwave brightness temperature spectra during the same period also demonstrate two components: a thermal component near the loop top and a nonthermal component at the footpoint of the loop. The microwave thermal component has the similar temperature as that of the hard X-ray superhot component. These measurements are consistent with the theory that the microwaves and hard X-rays are due to the same group of electrons, despite the fact that they are separated by 35,000 km. (3) The soft X-ray emissions brighten the existing loops and co-align with Hα emissions throughout the entire duration of the flare.

Particle acceleration is intrinsic to the primary energy release in the impulsive phase of solar flares, and we cannot understand flares without understanding acceleration. New observations in soft and hard X-rays, γ-rays and coherent radio emissions are presented, suggesting flare fragmentation in time and space. X-ray and radio measurements exhibit at least five different time scales in flares. In addition, some new observations of delayed acceleration signatures are also presented. The theory of acceleration by parallel electric fields is used to model the spectral shape and evolution of hard X-rays. The possibility of the appearance of double layers is further investigated. © 1994 Kluwer Academic Publishers.