We present images of the Crab synchrotron nebula obtained with the Wide Field and Planetary Camera 2 (WFPC2) on board the Hubble Space Telescope. These data are compared with ROSAT HRI images, and with 0':5 resolution Canada-France-Hawaii Telescope (CFHT) images previously published by van den Bergh & Pritchet (1989). These data strengthen the emerging picture of the Crab as a cylindrically symmetrical object with an axis running southeast to northwest, and inclined by ~20°-30° with respect to the plane of the sky. Identification of structure very near to the pulsar which shares this symmetry helps to better establish the link between the symmetry axis of the nebula and the spin axis of the pulsar. We report the discovery of a bright knot of visible emission located 0':65 to the southeast of the pulsar, along the axis of the system. This knot and a second knot 3':8 from the pulsar appear to be present but not well resolved in the 1988 CFHT image, indicating that they are persistent structures. The inner knot is interpreted as a shock in the pulsar wind ~ 1500 AU above the pole of the pulsar. No corresponding knots are seen to the northwest of the pulsar, which may indicate that the characteristics of the wind from the two poles are not symmetrical. The closest of the "wisps" to the northwest of the pulsar appear to close into a ringlike "halo" encircling the axis of the nebula. The wisps are resolved, with widths of ~ 0':2. This allows• calculation of their volumes and volume emissivities, and in turn their equipartition fields and pressures. Equipartition pressures calculated for the knots and wisps are typically 10 to as much as 80 times the equipartition pressure calculated for the nebula as a whole. The wisps show significant substructure which changed considerably between 1988 and 1994. Previous reports of relativistic motions of the wisps were probably due to changes in the unresolved substructure of these features. Comparison of the CFHT and WFPC2 images show remarkable changes in the inner nebula, but inferences about physical conditions based on this comparison are limited by the resolution of the CFHT data and the long 5 year baseline between the images. The structure of the nebula in 1994 may be inconsistent with the recent model by Gallant & Arons (1994). Very fine fibrous texture visible in the WFPC2 image follows the structure of the X-ray torus. A puzzling anticorrelation is seen between the X-ray and visible surface brightness through part of the torus. Long contiguous low contrast features with widths of ~ 1"-2" are seen to run throughout the volume of the nebula. These features are seen to move outward through the nebula at velocities in excess of homologous expansion. These features trace the magnetic structure of the nebula; they are probably due to differences in emissivity accompanying varying degrees of departure from equipartition at roughly constant total pressure. Visible fibers "drapt< over" and appear to expand away from an X-ray counterjet to the northwest of the pulsar, supporting th...

The WFPC2 was installed in the Hubble Space Telescope (HST) in 1993 December. Since then, the instrument has been providing high-quality images. A significant amount of calibration data has been collected to aid in the understanding of the on-orbit performance of the instrument. Generally, the behavior of the camera is similar to its performance during the system-level thermal vacuum test at JPL in 1993 May. Surprises were a significant charge-transfer-efficiency (CTE) problem and a significant growth rate in hot pixels at the original operating temperature of the CCDs (-76 °C). The operating temperature of the WFPC2 CCDs was changed to -88 °C on 1994 April 23, and significant improvements in CTE and hot pixels are seen at this temperature. In this paper we describe the on-orbit performance of the WFPC2. We discuss the optical and thermal history, the instrument throughput and stability, the PSF, the effects of undersampling on photometry, the properties of cosmic rays observed on-orbit, and the geometric distortion in the camera. We present the best techniques for the reduction of WFPC2 data, and describe the construction of calibration products including superbiases, superdarks, and flat fields.

We have observed SN 1987A with the optically corrected WFPC2 on the Hubble Space Telescope both in emission lines and in the UV and optical continuum. The previously observed outer nebular structure is shown to be part of two closed unresolved loops. These loops were flash-ionized by the supernova itself. They are not caused by limb brightening of an hourglass shell produced by the interaction of the winds from the progenitor. The inner ring is seen to be extended and may be connected to the new outer rings by sheets of material. However, beyond the outer rings, emission is not seen, implying a very low density (n < 10) for the outer hourglass shell if it exists. The new outer rings are unresolved, and this together with their observed brightness implies a density n > 1000. This density contrast of at least 100 is difficult to reconcile with the conventional picture of the progenitor evolution. Two models for the rings are presented, but each is deficient in important respects. A proper understanding of this system will require new physical insight.

No abstract

The protoÈplanetary nebula, CRL 2688, has been imaged through a wideband Ðlter centered at 606 nm (F606W) with the Wide Field Planetary Camera 2 onboard the Hubble Space T elescope. CRL 2688 is the prototypical bipolar reÑection nebula in which a star is surrounded by a dense, Ñattened cocoon of dust seen nearly edge-on and starlight escapes preferentially along the polar directions producing a pair of bright nebulosities, one above and one below the equatorial plane. We Ðnd a pair of radial "" searchlight beams ÏÏ emerging from within the dusty cocoon which intersect at the position expected for the central star when extrapolated inside the cocoon. The beams are crisscrossed by a large number of roughly round arcs with their center of curvature in the vicinity of the central star. The arcs are not systematically elongated along the polar axis of the nebula, as would be expected in the current model of CRL 2688 where the nebular density decreases with latitude.Our image directly shows the last D13,000 yr history of mass ejection from the central star while it was on the tip of the asymptotic giant branch (AGB). We Ðnd that the average surface brightness varies as r~3.7, implying that the average mass-loss rate or the scattering opacity of the dust grains varies as r~0.7, i.e., one or both of these parameters have steadily increased with time. The temporal resolution of B25 yr in our images has provided direct evidence for episodic increases in the mass-loss rate by factors of D2 or more, occurring every 150È450 yr, and lasting over periods of 75È200 yr. These irregularities in the mass-loss rate are roughly spherically symmetric. We have resolved the edges of the searchlight beams, as well as the peculiar structure of the inner region of the nebula Our data (1A .5 \ radius \ 6A). require a new model for CRL 2688 in which the beams result from starlight escaping through a pair of nonuniform annular holes in the dust cocoon which are coaxial with the polar axis of the nebula. The holes have probably been generated by a young (less than 200 yr) high-velocity outÑow which streams out through these holes and interacts with the surrounding dense AGB wind to produce the peculiar structure of the inner nebula. The cocoon contains dust grains of size about 0.6 km, signiÐcantly larger than those in the extended nebula.