Previous Article
Next Article
Table of Contents

"New" Photometry of an "Old" Supernova Gives...(1Jun94)
a Large Hubble Constant
(from NOAO HIGHLIGHTS!, NOAO Newsletter No. 38, 1 June 1994)

The major stumbling block in measuring the Hubble Constant (Ho) is that
different techniques produce inconsistent values.  Although George
Jacoby and collaborators (1992) have demonstrated that most modern
techniques produce consistent distance estimates to galaxies, the use
of type Ia supernovae (SN Ia) as standard candles stands as a major
exception.  Sandage and collaborators (1993) recently used the Hubble


                         [Figure not included]

          Figure 1:  A V-band CCD image of IC 4182 (left), the host
	  galaxy to SN 1937C, taken with the KPNO 2.1-m telescope, and
	  a small region of the first epoch film (right) taken by Baade
	  and Zwicky.  The two bright stars form a right triangle with
	  SN 1937C at the vertex.  The supernova is not visible in the
	  2.1-m image.

Space Telescope to find Cepheids in the nearby galaxy IC 4182,
determining the absolute magnitude of the historical supernova, SN
1937C, hence calibrating the SN Ia distance scale for the first time.
Distances to galaxies using this scale, however, are inconsistent with
those from H I line width-Luminosity relationship (Tully-Fisher),
Surface Brightness Fluctuations, Planetary Nebula Luminosity Functions,
and the Expanding Photosphere Method applied to type II SNe.  The
inescapable conclusion is that either SN Ia or the four other methods
are miscalibrated.

The recent discovery of a strong correlation between the absolute
luminosity of Type Ia SNe and their rates of decline by Phillips (1993)
has led Mike Pierce and George Jacoby to re-examine the photometry of
SN 1937C as a potential source of the discrepancy.  Unfortunately, the
photographic magnitudes available for SN 1937C published by Baade and
Zwicky (1938) and others are too uncertain to derive an accurate rate
of decline.  Consequently, Pierce and Jacoby acquired the original
18-inch Schmidt films taken by Baade and Zwicky of SN 1937C from the
Caltech plate archive (Figure 1).


		          [Figure not included]

          Figure 2:  A typical transformation curve used to convert PDS
          photographic density to m(pg). The vertical dotted line,
	  representing the PDS density of SN 1937C at ~19 days after
	  maximum, intersects the transformation curve at the position
	  of the horizontal dotted line to yield m(pg).


                          [Figure not included]

          Figure 3:  B (solid points) and "V" (open circles) light
	  curves for SN 1937C.  The template represents SN 1991T, a
	  SN Ia similar to SN 1937C.  The lower panel shows the color
	  evolution of SN 1937C.  Templates are shown for SN 1991T
	  (dotted), SN 1992bc (dot-dash), and SN 1992A (dashes).  The
	  latter has a V-band excess similar to that seen in SN 1937C.

A region of the 76 photographic and the previously unpublished 50
"visual" bandpass films surrounding SN 1937C was digitized using the
KPNO PDS.  This region contains a sequence of 26 photometric standards
measured by Schaefer (1994).  Excellent relations can be derived
between the total integrated photographic density of the stellar images
on each film and their m(pg) or "V" mags (Figure 2).  (The "V" bandpass
is not strictly a Johnson V but extends into the R band.  The SN
spectrum generally is not a stellar continuum, especially at later
times, and so classic transformation methods may incur an added
uncertainty.  We refer to these magnitudes as "V.") Typical RMS errors
of the transformation curve are 0.06 and 0.04 mags for m(pg) and "V,"
respectively.  The standard m(pg) to B transformation (Arp 1961; Hamuy
et al. 1991) was used to convert the photographic data to the modern
system.  The B and "V" light curves and color evolution are shown in
Figure 3.  The scatter around a low-order fit is smaller than 0.05 mag
at B and 0.03 at "V," or less than half that derived from the Baade and
Zwicky photometry.  This improvement implies that significant gains in
the precision of photographic photometry are possible with modern
techniques and a sequence of local standards.

Since SN 1937C was discovered after maximum light, its peak brightness
is not well constrained by the Baade and Zwicky data, particularly
given the variety in the shapes of individual SN Ia light curves.
Fortuitously, Comet Finsler (1937f) passed within 8 degrees of IC 4182
7-8 days prior to the discovery of SN 1937C.  A plate containing images
of both the comet and SN 1937C was obtained by Leutenegger (1937).
This plate was kindly lent to Pierce and Jacoby by Herrn Mohr.  This
pre-maximum B mag provides a strong constraint to both the peak
brightness and the time of maximum for SN 1937C.  Pierce and Jacoby
find B(max) = 9.03 +- 0.05 and "V"(max) = 9.06 +- 0.05 for SN 1937C,
values 0.25 and 0.47 mag, respectively, fainter than Saha et al. (1994)
assumed in their calibration.  Furthermore, the light curve shape is
intermediate between that of SN 1991T and SN 1992bc (see Figure 3), two
of the slowest declining SN Ia ever observed (e.g. Phillips 1993).
This implies that SN 1937C was ~ 0.4 mag more luminous than a typical
SN Ia event.

Combined with the Sandage et al. (1993) distance to IC 4182, the
revised photometry and the correction for the slow rate of decline give
Ho = 75 +- 12 km/s/Mpc from SN Ia.  This value is now consistent with
that estimated using the four other methods and eliminates much of the
discrepancy between the different methods of estimating the Hubble
Constant.

This program is indebted to Alain Porter for locating and acquiring the
original Baade and Zwicky films of SN 1937C.  Unfortunately, Alain
passed away before he could further contribute to this effort.  A
detailed report of this work is about to be submitted to the
Astronomical Journal.  The 136 digitized scans will be made available
to interested parties shortly.

Previous Article
Next Article
Table of Contents