arXiv subjects
Publications and source records attributed to Robert A. Fesen.
The luminosity, light curve, post--maximum spectrum, and lack of a progenitor on deep pre-outburst images suggest that SN 2010U was a luminous, fast nova. Its outburst magnitude is consistent with that for a fast nova using the Maximum Magnitude-Rate of Decline relationship for classical novae.
We present Chandra ACIS X-ray observations of the Galactic supernova remnant Cassiopeia A taken in December 2007. Combining these data with previous archival Chandra observations taken in 2000, 2002, and 2004, we estimate the remnant's forward shock velocity at various points around the outermost shell to range between 4200 and 5200 +/- 500 km/s. Using these results together with previous analyses of Cas A's X-ray emission, we present a model for the evolution of Cas A and find that it's expansion is well fit by a rho_ej ~ r^{-(7-9)} ejecta profile running into a circumstellar wind. We further find that while the position of the reverse shock in this model is consistent with that measured in the X-rays, in order to match the forward shock velocity and radius we had to assume that ~ 30% of the explosion energy has gone into accelerating cosmic rays at the forward shock. The new X-ray images also show that brightness variations can occur for some forward shock filaments like that seen for several nonthermal filaments seen projected in the interior of the remnant. Spectral fits to exterior forward shock filaments and interior nonthermal filaments show that they exhibit similar spectra. This together with similar flux variations suggests that interior nonthermal filaments might be simply forward shock filaments seen in projection and not located at the reverse shock as has been recently proposed.
We discuss the findings of a comprehensive imaging and spectroscopic survey of the optical emission associated with the supernova remnant 3C 58 (Fesen et al. 2007) as they relate to the topic of pre-SN mass loss. Spectroscopically measured radial velocities of ~450 emission knots within the remnant show two distinct kinematic populations of optical knots: a high-velocity group with radial velocities in the range of 700 - 1100 km/s and a lower velocity group exhibiting radial expansion velocities below ~250 km/s. We interpret the high-velocity knots as ejecta from the SN explosion and the low-velocity knots as shocked circumstellar material likely resulting from pre-SN mass loss. The chemical signatures of the two populations also show marked differences. The high velocity group includes a substantial number of knots with notably higher [N II]/H-alpha ratios not seen in the lower velocity population, suggesting greater nitrogen enrichment in the SN ejecta than in the CSM. These results are compared with evidence for pre-SN mass loss in the Crab Nebula, perhaps the SNR most similar to 3C 58. These SNRs may comprise two case studies of pre-SN mass loss in relatively low mass (~8 - 10 solar masses) core-collapse SN progenitors.
We present HST images of the remnant of SN 1885 seen in absorption against M31's bulge via resonance lines of Ca I, Ca II, Fe I, and Fe II. Viewed in CaII H & K line absorption, the remnant appears as a nearly black circular spot with an outermost angular radius of 0.40" +/- 0.025" implying r = 1.52 pc and a 120 yr average expansion velocity of 12400 +/-1400 km/s. The strongest Ca II absorption is organized in a broken ring structure with a radius of 0.20" (=6000 km/s) with several apparent absorption `clumps' of an angular size near the pixel scale of 0.05" (= 1500 km/s). The detection of Ca II clumps is the first direct evidence for some instabilities and the existence of a deflagration phase in SNe Ia or, alternatively, mixing induced by radioactive decay of 56^Ni over time scales of seconds or days. However, the degree of mixing allowed by the observed images is much smaller than current 3D calculations for Rayleigh-Taylor dominated deflagration fronts. The images also require a central region of no or little Ca but iron group elements indicative of burning under sufficiently high densities for electron capture taking place, i.e., burning prior to a significant pre-expansion of the WD.
We present wide-field Halpha images of the Galactic supernova remnant G156.2+5.7 which reveal the presence of considerable faint Halpha line emission coincident with the remnant's X-ray emission. The outermost Halpha emission consists largely of long and thin (unresolved), smoothly curved filaments of Balmer-dominated emission presumably associated with the remnant's forward shock front. Patches of brighter Halpha emission along the western, south-central, and northeastern regions appear to be radiative shocked ISM filaments like those commonly seen in supernova remnants, with relatively strong [O I] 6300,6364 and [S II] 6716,6731 line emissions. Comparison of the observed Halpha emission with the ROSAT PSPC X-ray image of G156.2+5.7 shows that the thin Balmer-dominated filaments lie along the outermost edge of the remnant's detected X-ray emission. Brighter radiative emission features are not coincident with the remnant's brightest X-ray or radio regions. Areas of sharply weaker X-ray flux seen in the ROSAT image of G156.2+5.7 appear spatially coincident with dense interstellar clouds visible on optical and IRAS 60 and 100 micron emission images, as well as maps of increased optical extinction. This suggests significant X-ray absorption in these regions due to foreground interstellar dust, especially along the western and southern limbs. The close projected proximity and alignment of the remnant's brighter, radiative filaments with several of these interstellar clouds and dust lanes hint at a possible physically interaction between the G156.2+5.7 remnant and these interstellar clouds and may indicate a smaller distance to the remnant than previously estimated.
Several commercial telecommunication ventures together with a well funded US military program make it a likely possibility that an autonomous, high-altitude, light-than-air (LTA) vehicle which could maneuver and station-keep for weeks to many months will be a reality in a few years. Here I outline how this technology could be used to develop a high-altitude astronomical observing platform which could return high-resolution optical data rivaling those from space-based platforms but at a fraction of the cost.
Absorption-line spectroscopy is an effective probe for cold ejecta within an SNR, provided that suitable background UV sources can be identified. For the SN 1006 remnant we have identified four such sources, in addition to the much-studied Schweitzer-Middleditch (SM) star. We have used STIS on HST to obtain UV spectra of all four sources, to study "core samples" of the SN 1006 interior. The line of sight closest to the center of the SNR shell, passing only 2.0 arcmin away, is to a V = 19.5 QSO at z = 1.026. Its spectrum shows broad Fe II absorption lines, asymmetric with red wings broader than blue. The similarity of these profiles to those seen in the SM star, which is 2.8 arcmin from the center in the opposite direction, confirms the existence of a bulge on the far side of SN 1006. The Fe II equivalent widths in the QSO spectrum are ~ 50% greater than in the SM star, suggesting that somewhat more iron may be present within SN 1006 than studies of the SM star alone have indicated, but this is still far short of what most SNIa models require. The absorption spectrum against a brighter z = 0.337 QSO seen at 57% of the shell radius shows broad silicon absorption lines but no iron other than narrow, probably interstellar lines. The cold iron expanding in this direction must be confined within v <~ 5200 km/s, also consistent with a high-velocity bulge on the far side only. The broad silicon lines indicate that the silicon layer has expanded beyond this point, and that it has probably been heated by a reverse shock. Finally, the spectra of two ~ A0V stars near the southern shell rim show no broad or unusually strong absorption lines, suggesting that the low-ionization ejecta are confined within 83% of the shell radius, at least at the azimuths of these background sources.
We report a high resolution, far-ultraviolet, STIS E140M spectrum of the strong, broad Si II, III, and IV features produced by the ejecta of SN1006 seen in absorption against the background Schweizer-Middleditch star. The spectrum confirms the extreme sharpness of the red edge of the redshifted Si II 1260 A feature, supporting the idea that this edge represents the location of the reverse shock moving into the freely expanding ejecta. The expansion velocity of ejecta at the reverse shock is measured to be 7026 +-3(relative) +-10(absolute) km/s. If the shock model is correct, then the expansion velocity should be decreasing at the observable rate of 2.7 +-0.1 km/s per year. The pre-shock velocity, post-shock velocity, and post-shock velocity dispersion are all measured from the Si II 1260 A feature, and consistency of these velocities with the shock jump conditions implies that there is little or no electron heating in this fast (2680 km/s) Si-rich shock.
We present results from observations of SN 1979C with XMM-Newton in X-rays and in the UV, archival X-ray and HST data, and follow-up ground-based optical imaging. The XMM-Newton MOS spectrum shows two-temperature thermal plasma emission characteristics of both the forward (4.1 keV) and reverse shock (0.78 keV) with no intrinsic absorption. The long-term X-ray lightcurve of SN 1979C shows no sign of a decline over 16-23 yrs after its outburst. The high X-ray luminosity (8 x 10^38 ergs/s) is caused by the interaction of the shock with dense circumstellar matter, likely deposited by a strong stellar wind from the progenitor with a high mass-loss rate of 1.5 x 10^-4 M_sun/yr. The X-ray data support a strongly decelerated shock and show a mass-loss rate history which is consistent with a constant progenitor mass-loss rate and wind velocity over the past >16,000 yrs in the evolution of the progenitor. We find a CSM density profile of rho_CSM \propto r^-s with index s<1.7 and high CSM densities (>10^4 cm^-3) out to large radii from the site of the explosion (> 4 x 10^17 cm). Using XMM-Newton Optical Monitor data we further detect optical/UV emission from SN 1979C, with B, U, and UVW1-band lum. of 5,7, and 9 x 10^36 ergs/s. The young stellar cluster in the vicinity of the SN, as imaged by the HST and follow-up ground-based optical imaging, can only provide a fraction of the total observed flux, so that a significant contribution to the output likely arises from the strong interaction of SN 1979C with dense CSM.
Late-time near-infrared and optical spectra are presented for the normal-bright SN2003du. At about 300 days after the explosion, the emission profiles of well isolated [FeII] lines (in particular that of the strong 1.644mu feature) trace out the global kinematic distribution of radioactive material in the expanding. In SN2003du, the 1.644 mu [FeII] line shows a flat-topped, profile, indicative of a thick but hollow-centered expanding shell, rather than a strongly-peaked profile that would be expected from a ``center-filled'' distribution.Based on detailed models for exploding Chandrasekhar mass white dwarfs, we show that the feature is consistent with spherical explosion models.Our model predicts central region of non-radioactive electron-capture elements up to 2500--3000km/s as a consequence of burning under high density, and an extended region of 56Ni up to 9,000--10,000km/s. Furthermore our analysis indicates that the 1.644mu [FeII] profile is not consistent with strong mixing between the regions of electron- capture isotopes and the 56Ni layers as is predicted by detailed 3D models for nuclear deflagration fronts. We discuss the possibility that the flat-topped profile could be produced as a result of an infrared catastrophe and conclude that such an explanation is unlikely. We put our results in context to other SNeIa and briefly discuss the implications of our result for the use of SNe Ia as cosmological standard candles.
We present K-band (1.9 -- 2.5 micron) spectra of the Type Ic SN 2000ew observed with IRCS on the Subaru Telescope. These data show the first detection of carbon monoxide (CO) emission in a Type Ic supernova. The detection of CO in SN 2000ew provides further evidence that molecule formation may be a common occurrence in core-collapse supernova ejecta. The spectrum also contains narrow emission lines of [Fe II] and He I probably from dense clumps of hydrogen-poor circumstellar gas surrounding SN 2000ew. Our spectrum of SN 2000ew shows no trace of an unidentified feature seen near 2.26 micron, just blueward of the CO emission, in the spectrum of SN 1987A and we discuss proposed detections of this feature in other Type II supernovae.
The kinematic and morphological properties of a string of [Ar III] bright knots in the Crab Nebula are examined using 1994 - 1999 HST WFPC-2 images of the remnant. We find that five southern [Ar III] bright knots exhibit ordinary radial motions away from the nebula's center of expansion with magnitudes consistent with their projected radial displacements. These results do not support the suggestion by MacAlpine et al.(1994) that these knots might be moving rapidly away from the Crab pulsar due to a collimated wind. The HST images also do not show that the [Ar III] knots have unusual morphologies relative to other features in the remnant. Our proper motion results, when combined with radial velocity estimates, suggest these knots have relatively low space velocities implying relatively interior remnant locations thus placing them closer to the ionizing radiation from the Crab's synchrotron nebula. This might lead to higher knot gas temperatures thereby explaining the knots' unusual line emission strengths as MacAlpine et al.(1994) suspected.
The young SNR Cassiopeia A was imaged with WFPC-2 through four filters selected to capture the complete velocity range of the remnant's main shell in several important emission lines. Primary lines detected were [O III] 4959,5007, [N II] 6583, [S II] 6716,6731 + [O II] 7319,7330 + [O I] 6300,6364, and [S III] 9069,9532. About 3/4th of the remnant's main shell was imaged in all four filters. Considerable detail is observed in the reverse-shocked ejecta with typical knot scale lengths of 0.2"-0.4" (1 - 2 x 10^16 cm). Both bright and faint emission features appear highly clumped. Large differences in [S III] and [O III] line intensities indicating chemical abundance differences are also seen, particularly in knots located along the bright northern limb and near the base of the northeast jet. A line of curved overlapping filament in the remnant's northwestern rim appears to mark the location of the remnant's reverse shock front in this region. Finger-like ejecta structures elsewhere suggest cases where the reverse shock front is encountering the remnant's clumped ejecta. Narrow-band [N II] images of the remnant's circumstellar knots ("QSFs") reveal them to be 0.1"-0.6" thick knots and filaments, often with diffuse edges facing away from the center of expansion. Three color composite images of the whole remnant and certain sections along with individual filter enlargements of selected regions of the bright optical shell are presented and discussed.
Optical and ultraviolet observations of the Type IIn supernova 1995N at epochs between 321 and 1799 days after the explosion show three distinct velocity components. The narrow lines come from circumstellar gas and show both low and high ionization. This component has a low filling factor, and is photoionized by X-rays from the shock. The intermediate component, which is dominated by newly processed oxygen, originates in a shell with velocity of 2500-5000 km/s, and most likely comes from the ejecta. The hydrogen- and helium-dominated gas has a low ionization, a high density, and velocities that extend out to ~10,000 km/s. Strong signatures of Ly-alpha-pumped fluorescence lines of Fe II are seen in the near-infrared and ultraviolet. The He/H ratio, ~ 0.3 by number, and the nitrogen overabundance provide strong evidence for CNO burning products. The fluxes of the broad hydrogen and helium lines decrease considerably faster than the oxygen lines. The H-alpha line profile shows strong evolution, with the red wing decreasing faster than the blue. Possible scenarios, involving either a clumpy circumstellar medium, or an aspherical distribution of the surrounding gas, are discussed based on the line profiles and physical conditions. Finally, we propose that Type IIn supernovae have their origin in red supergiants in a superwind phase.
We present proper motions for 21 bright main shell and 17 faint, higher-velocity, outer ejecta knots in the Cas A supernova remnant and use them to derive new estimates for the remnant's expansion center and age. Our study included 1951 - 1976 Palomar 5 m prime focus plates, 1988 - 1999 CCD images from the KPNO 4 m and MDM 2.4 m telescopes, and 1999 HST WFPC2 images. Measurable positions covered a 23 to 41 yr time span for most knots, with a few outer knots followed for almost 48 yr. We derive an expansion center of alpha = 23h 23m 27s.77 +- 0s.05, delta = 58d 48' 49".4 +- 0".4 (ICRS), with little difference between centers derived using outer or main shell knots. This position is 3.0 arcsec due north of that estimated by van den Bergh and Kamper (1983). It also lies 6.6 +- 1.5 arcsec almost due north (PA = 354 deg) of the remnant's recently-detected central X-ray point source, implying a transverse velocity for the X-ray point source of about 330 km/s at a distance of 3.4 kpc. Using the knots which lie out ahead of the remnant's forward blast wave, we estimate a knot convergent date of A.D. 1671.3 +- 0.9, assuming no deceleration. However, a deceleration of just approximately 1.6 km/(s yr) over a 300 yr time span would produce an explosion date A.D. 1680, consistent with the suspected sighting of the Cas A supernova by J. Flamsteed.
Near-infrared spectra (0.95 -- 2.4 micron) of the peculiar Type IIn supernova 1998S in NGC 3877 from 95 to 355 days after maximum light are presented. K-band data taken at days 95 and 225 show the presence of the first overtone of CO emission near 2.3 micron, which is gone by day 355. An apparent extended blue wing on the CO profile in the day 95 spectrum could indicate a large CO expansion velocity (~2000 -- 3000 km/s). This is the third detection of infrared CO emission in nearly as many Type II supernovae studied, implying that molecule formation may be fairly common in Type II events, and that the early formation of molecules in SN 1987A may be typical rather than exceptional. Multi-peak hydrogen and helium lines suggest that SN 1998S is interacting with a circumstellar disk, and the fading of the red side of this profile with time is suggestive of dust formation in the ejecta, perhaps induced by CO cooling. Continuum emission that rises towards longer wavelengths (J -> K) is seen after day 225 with an estimated near-infrared luminosity >~ 10^40 erg/s. This may be related to the near-infrared excesses seen in a number of other supernovae. If this continuum is due to free-free emission, it requires an exceptionally shallow density profile. On the other hand, the shape of the continuum is well fit by a 1200 +- 150 K blackbody spectrum possibly due to thermal emission from dust. Interestingly, we observe a similar 1200 K blackbody-like, near-infrared continuum in SN 1997ab, another Type IIn supernova at an even later post-maximum epoch (day 1064+). A number of dust emission scenarios are discussed, and we conclude that the NIR dust continuum is likely powered by the interaction of SN 1998S with the circumstellar medium.
Ultraviolet imaging of the remnant of Supernova 1885 in M31 with the Hubble Space Telescope using the F255W filter on the WFPC2 reveals a dark spot of Fe II absorption at the remnant's known position in the bulge of M31. The diameter of the absorbing spot is 0"55 +- 0"15, slightly smaller than, but consistent with, the 0"70 +- 0"05 diameter measured in the higher quality WFPC2 Ca II absorption image previously reported by us. The measured ratio of flux inside to outside SNR 1885 in the Fe II image is 0.24 +- 0.17, consistent with the ratio 0.33 +- 0.04 expected on the basis of a model fit to the previously obtained near-UV FOS spectrum. The observed depth of Fe II absorption suggests that Fe II is fully saturated, with an iron mass in the range M_Fe = 0.1-1.0 Msun. Besides Fe, ion species Mg I, Mg II, and Mn I probably make some contribution to the absorption from the SN 1885 remnant in the F255W image.
A low-dispersion Keck I spectrum of SN 1980K taken in August 1995 (t = 14.8 yr after explosion) and a November 1997 MDM spectrum (t = 17.0 yr) show broad 5500 km s^{-1} emission lines of Hα, [O I] 6300,6364 A, and [O II] 7319,7330 A. Weaker but similarly broad lines detected include [Fe II] 7155 A, [S II] 4068,4072 A, and a blend of [Fe II] lines at 5050--5400 A. The presence of strong [S II] 4068,4072 A emission but a lack of [S II] 6716,6731 A emission suggests electron densities of 10^{5-6} cm^{-3}. From the 1997 spectra, we estimate an Hαflux of 1.3 \pm 0.2 \times 10^{-15} erg cm^{-2} s^{-1} indicating a 25% decline from 1987--1992 levels during the period 1994 to 1997, possibly related to a reported decrease in its nonthermal radio emission.