Optimizing supernova search - Which Galaxies?

[madbadgalaxyman (Robert)]

Originally Posted by PeterM

Indeed with 6 supernova in the last 100 years M83 would be a great target.

G'day there, Mr "SN cracker",

Thanks for the invite to Leyburn. I haven't seen the light of galaxies with my eyeballs for so long that perhaps I need to go and check that those galaxies are still there!

This post, and thread, continues from the recent Science thread by faaarwest2K, which asks what science projects can be undertaken by a person with an 8 inch Dobsonian. Visual search for supernovae is a "possible" with an 8 inch telescope, though, as you know, the target galaxies for eyeball searching with small apertures should be :
(1) nearby (so that the SN is visually detectable with this small aperture).
(2) preferably those galaxies which have a demonstrably high rate of star formation, as these galaxies will also have a high rate of occurring core-collapse supernovae.
(3) galaxies in which a very prominent O & B star population is in the low extinction regions of the disk component of their host galaxy. (Virtually any meaningful extinction from the host galaxy of a supernova can quickly make the SN too faint for visual detection )

In this post, there is a simple discussion of how to maximize the success-rate of a limited visual search for supernovae, with only small numbers of target galaxies.
(( this contrasts with the "scattergun" approach taken by people who use automated imaging and tracking systems..... potentially imaging very large numbers of galaxies in order to find that "supernova in a haystack".))

The right choice of galaxies to survey is the key to success in small-galaxy-number Supernova search......

The current and total Star Formation Rate (SFR) in a galaxy (by convention, this is measured in Units of solar masses per annum) correlates well with the rate of occurring core-collapse supernovae in that galaxy. This is because of the (approximate) constancy of the Initial Mass Function of stars;
which states, in the form of a mathematical function, that the relative numbers of stars formed of various masses are (nearly) always in the same ratios.
[[ As mentioned in the other thread, the LMC has a total Star Formation Rate of about 0.4 solar masses per Annum, and our own Galaxy is currently thought to have an SFR of about 2 solar masses per annum (about normal for a big spiral galaxy) ]]

There are several ways to estimate the SFR of a galaxy.
For instance :
- the SFR of a galaxy correlates well with its total Far-ultraviolet Luminosity.
- the SFR of a galaxy also correlates well with its total Infrared Luminosity.
- SFR can be calculated from the H-alpha luminosity of a galaxy.
There are abundant numerical formulae in the literature that enable the calculation of the SFR of a galaxy from its luminosity in certain bandpasses.

In general, a galaxy with a high SFR will host a lot of core-collapse supernovae, though merely saying that supernovae are expected to frequently occur in a particular galaxy says nothing about whether or not these occurring supernovae can actually be detected with a specific telescope, because many galaxies with a very high SFR have a significant fraction of their massive and luminous stars concentrated in regions of high extinction from the dust within a galaxy (the biggest known OB star concentrations are found in the high-extinction centres of non-dwarf spiral galaxies)

In this post, I use an indicator of the Total Mass of a Galaxy (its Blue luminosity), together with how blue the U-B colour of a galaxy is, as a "quick and easy" method to estimate how much ultraviolet light is coming from a galaxy, and therefore to estimate how large is its population of those OB stars that are not subject to high foreground extinction from their host galaxy.

High mass (and, therefore, high luminosity) spiral galaxies that also have very blue (more negative) U-B colours...... usually do have high rates of detectable supernovae at visible wavelengths, because the characteristically blue optical colour of these galaxies is usually caused by Massive and Luminous O & B stars that occupy relatively low-extinction regions of the disk component of a galaxy.

So here is my attempt to extract, from a comprehensive Galaxy Catalog, the names and identities of the most massive and luminous spiral galaxies within about 78 million Light Years which also have the following properties:
- a high rate of star formation, minimally equivalent to the typical Star Formation Rate in a first-ranked (-21 to -22 absolute blue magn. ) spiral galaxy of Hubble type Sc.
- much of the current star formation is occurring in the low extinction regions of the galaxy

I have extracted (using SQL), from those galaxies contained in the Hyperleda online galaxy catalog, a sample of Very Luminous and Very Strongly Star-forming spiral galaxies, which have all of the following numerical properties :::

- a Blue absolute magnitude of less than -20.7 (in other words, these are massive & luminous spiral galaxies)
AND
- a Hubble type of Sb or later, in the Hubble Sequence
AND
- a recession velocity of less than 1800 km/s. In other words, I have selected only nearby galaxies, those within a maximum distance of approximately 24 Megaparsecs (= 78 million light years )
AND
- a total U-B colour of less than 0.1 (so these are blue galaxies, with a substantial ultraviolet flux which will mainly come from little-obscured OB stars.....unless of course the galaxy hosts a bright AGN! )

This should be a very profitable sample of galaxies for supernova patrol, especially for visual SN search with modest apertures, as these galaxies are nearby and they each have a large total stellar mass, and they will (most of them, anyway) have a large population of not very dust-obscured OB stars on account of their blue U-B colours. In other words, this list is an attempt to name those particular nearby galaxies which should, on theoretical grounds, have the highest rates of visually very evident core-collapse supernovae, as I made a cut in colour, such that only optically Very Blue galaxies were included.

cheers,
Robert

Here are the selected galaxies, in a Word 2000 format (.doc) document:
LEDA srch for bigbluespirals_.doc

[madbadgalaxyman (Robert)]

Hello there, Supernova searchers,

In order to create a more reliable list of nearby high-supernova-rate Spiral Galaxies that are suitable for visual Supernova search, it will be necessary for me to remove a few galaxies from the above list, for reasons such as ::
- unfavourable galaxy orientation
- Observed strong excess of UV light coming mainly from an Active Galactic Nucleus rather than from OB stars
- Bad Data leading to erroneous conclusions regarding the luminosity and/or colour of a galaxy

In other words, I need to check the accuracy of the catalog data on the "High Supernova Rate" galaxies in the above Draft List, and I also need to assess the truth or otherwise of the tentative conclusions that I have drawn from this data.

Sadly, it remains very difficult to come to robust conclusions as to the values of even the most important physical parameters of many prominent southern galaxies, as data is too often scarce and of poor quality. Hopefully the SkyMapper and VISTA southern sky surveys will soon remedy this shocking situation!!
In order to refine my conclusions, I need to thoroughly check the literature, in case there are more accurate magnitudes and distances and luminosities available for these galaxies.

Also, there are several more good candidates for nearby and very luminous and High Supernova Rate Galaxies, which did not make it into the draft list of "High Supernova Rate" galaxies.......


Some more Nearby Spiral Galaxies which are "possibles or probables" for having a high rate of "easily detectable" core-collapse supernovae ::

(1) NGC 2997
Last time I considered the properties of NGC 2997, I concluded that this galaxy is arguably very comparable to other big Strongly Star-forming spiral galaxies like M100 and M61 and M83, but oddly, NGC2997 seems to have had relatively few reported supernovae. Perhaps this is due only to the fact that people were not looking...... at the time that the supernovae occurred in this galaxy.
If the reddening-corrected U-B and B-V colours of NGC 2997 are comparable to that of the other big "highly star forming" spirals on the list, the size of the O & B star population of this galaxy could be similar to that found in the other vigorously star-forming massive spirals. This galaxy is certainly very large in angular size and physical size;
for instance, the ESOLV photographic surface photometry measures a major axis diameter of 9.9 arcminutes to the standard D25 isophote, which corresponds to 110,000 Light Years across if the (very uncertain) distance of this galaxy is Distance Modulus 30.4
On the other hand, Hyperleda gives some Rather Red effective U-B and B-V colours for the inner parts of this galaxy, but these reddish optical colours might merely be the effect of reddening from the foreground dust screen of our Milky Way Galaxy.
I will have another look at the physical properties of NGC2997, to see if its properties are in accord with my initial impression. There are significant uncertainties as to its distance and luminosity, and as to how much its true optical colours are reddened by foreground dust from our own Galaxy.

(2) NGC 1097
NGC 1097 has a high blue luminosity, and it also has a High total infrared luminosity (5 x 10E10 solar luminosities) as given in the IRAS Revised Bright Galaxy sample, indicating that its total star forming rate is somewhat higher than that of a typical first-ranked Type Sbc or Sc spiral galaxy. A good fraction of the star formation in NGC 1097 could be in the tight ring of star-forming regions that exists near to its centre; so it is worth keeping in mind that occurring SNe could be in the crowded and high-surface-brightness central part of this galaxy. In the case of N1097, the star forming ring is only modestly obscured by dust, which is usually very heavy at the centre of a non-dwarf spiral galaxy.
This galaxy is selected as a very good prospect for having a high Star Formation Rate and a high Supernova Rate, on the grounds of its high infrared luminosity and its high total luminosity. ( I may find other good prospects in the aforementioned IRAS catalog, which gives the total infrared luminosity of many galaxies.).

(3) Messier 101
M101 is very luminous , and nearby, and very blue. Therefore, Supernovae should be frequent and bright.
This is a very blue (observed B-V = 0.45) spiral galaxy of similarly high luminosity to many other galaxies in the table. (it did not make it into the table because there was no total U-B value for it in Hyperleda)
Therefore M101 will have a high rate of occurring and observed core-collapse supernovae.

(4) NGC 1365
NGC 1365 did not make it into table because its colours are apparently a little too red , but it is a Very Luminous spiral galaxy which is large and massive. Its distance estimate is well constrained, and, from the recent published literature distance estimates, one can derive with high confidence the fact that NGC 1365 is undeniably one of the most luminous spiral galaxies found within a recession velocity of 2000 km/s.
(the luminosity of NGC 1365 is usually numerically expressed as Blue Absolute Magnitude of minus 21.0 or brighter. It could even be -21.5 . The catalog luminosity of NGC 1365 has to be regarded as a lower limit, in the absence of CCD surface photometry that includes its very extended outer regions )
Furthermore, the total infrared luminosity of this galaxy is much higher than that of one of the most luminous Virgo Cluster spirals like M61 and M100; this strongly suggests that this galaxy has a much higher star formation rate than "normal" high-luminosity spiral galaxies.
A high supernova rate is therefore very likely in NGC 1365; but it is conceivable that a large fraction of the OB stars and supernovae could be located in the highly-obscured central 'starburst' region.

(5) NGC 1084
Hyperleda gives a blue absolute magnitude of -20.69 for NGC 1084, which is similar to the luminosity of some of the less luminous galaxies in the table. The U-B colour index, as given in Hyperleda, is a very blue -0.09 . This very blue U-B colour would be an indicator that this galaxy has a very prominent population of O and B stars; I hope that this U-B value is correct, as good numbers of catalog U-B values have a very low accuracy. The IRAS Revised Bright Galaxy Catalog gives a Large total infrared luminosity of 3.5x10E10 solar luminosities for this galaxy. The general properties of N1084 are probably comparable to those of the other very luminous and strongly-star-forming spiral galaxies in the table, and the infrared luminosity indicates that there is particularly strong ongoing Star Formation. However, a very large distance uncertainty for this galaxy casts significant doubt on many of the conclusions in this paragraph.

(6) NGC 1087
The very blue optical colours given in the galaxy catalogs are very plausible, but there is still a high degree of uncertainty attaching to the luminosity and distance of this galaxy.

[madbadgalaxyman (Robert)]

EVEN MORE FUN WITH GALAXY DATA !

Firstly, I must warn readers that my notes on various galaxies throughout this thread are still in a somewhat tentative and incomplete state, as I may not have used the best available data to draw my conclusions. It is time to hit the literature and improve on what I have written!

In the below New Query, I relaxed the luminosity criterion somewhat (lowered it by 0.2 magnitudes) and also I made the colour criterion considerably more stringent ;
only those galaxies with a very strong Ultraviolet excess of U-B < -0.04 were isolated.

So here is my new SQL query of the Hyperleda catalog of galaxies:
select objname, de2000, mabs, bt, ubt, t, vlg where t>2.7 and vlg<1800 and mabs<-20.5 and mabs>-20.7 and ubt<-0.04

Here I am asking the galaxy database to print the following parameters:
Object name,
declination,
blue absolute magnitude,
blue apparent magnitude,
observed U-B colour,
Hubble type (expressed as a numerical stage),
recession velocity (in the Local Group reference frame)

for :
all t > 2.7 (Hubble type Sb and later) galaxies

with Recession Velocity less than 1800 km/s

and Between Blue absolute magnitude -20.5 and -20.7

and with a U-B colour index of less than -0.04 (very strong ultraviolet excess)

This query isolates the identities of some more galaxies, possibly of somewhat lower luminosity, but many of them with very very strong excesses of ultraviolet light, indicating the presence of extreme concentrations of O and B stars.


Notes on some of the candidates for "high supernova rate galaxies" in this new query:

NGC 3621
According to most of the standard catalog data, this strongly star-forming galaxy has a relatively modest Luminosity and Size compared to many of the other Very Big and Very Luminous spirals under consideration for "high success rate" visual supernova search. For instance R. Brent Tully in the Nearby Galaxies Catalog gives a rather modest Blue absolute magnitude of minus 19.87.

However, for NGC 3621, the photographic surface photometry in the ESOLV Catalog measures a truly enormous Major Axis angular dimension (out to the faint 26th B magn. per sq. arcsec isophotal ellipse) of 22.2 arcminutes !!
This is a plausible diameter measurement, as it corresponds rather well to the extent of the region of strong star formation seen in the GALEX satellite's Ultraviolet view of this galaxy:


Adopting this value as a reasonable measure of the angular diameter of the stellar distribution of this galaxy, and a physical scale of 2 kiloparsecs per arcminute of angular measure, the true diameter of NGC 3621 is at least 140,000 light years , thus contradicting the idea that this is not a really big spiral galaxy.

In the absence of CCD magnitude measurements that include all of the light of the vastly extended outer regions of NGC 3621, I would have to regard any conclusions as to the total magnitude and the total luminosity of this galaxy as being very tentative. For galaxies of such large angular size, it can happen that CCD photometry adds half a magnitude to a galaxy's brightness as quoted in the catalogs!

This galaxy has a very blue U-B colour of -0.08 according to Hyperleda, and its distance is well-constrained (using Cepheid observations) at near to 22 million light years. The B-V optical colour of 0.46, as measured in the Carnegie-Irvine Galaxy Survey, is also notably blue. The rather nearby distance of this galaxy implies that supernovae should be bright and easily detectable.

Incidentally, the outer spiral arms of NGC 3621 are not in the same plane as the inner region of the galaxy, as they are warped or bent in some way, so I do not anticipate that extinction will be so high that supernovae will become too faint for easy detection.

In summary, the nearness of this galaxy makes it a good prospect for visual supernova search. But I will have to do some more work before I can say for sure how big its population of OB stars is and how high its supernova rate is.


NGC 3359
This galaxy is an excellent candidate for having a high rate of supernovae. The optical colours given in Hyperleda are very very blue ::
B-V = 0.46 and U-B = -0.20
If this data is good, these values indicate that there is a very large population of O and B stars on this galaxy. This galaxy is probably undergoing an intense burst of current star formation. I should check the Far-ultraviolet luminosity to confirm this conclusion!

"Just for fun and profit", here is the Far ultraviolet + Near Ultraviolet view of NGC 3359 from GALEX, showing the intense far ultraviolet emission from the massive and luminous stars of the spiral arms:



NGC 4654
This interesting Virgo Cluster spiral is strangely distorted in its outer regions, and the galaxy catalogs give very blue optical colours for it. The admittedly uncertain distance measurements in the catalogs indicate that this galaxy is at least absolute Blue magnitude -20 in luminosity. R. Brent Tully, based on his surprisingly good redshift distances in the Nearby Galaxies Catalog, gave -20.3 in that catalogue.