Blue Bump in SN Ia:
An Archive Project
During the Spring of 2022, Bernardo Santos (CS Major, class of '22) did an archival study of supernova data to explore the idea that the progenitor mechanism (single- vs double-degenerate) leaves a signature of a "blue bump" in the early light curves of SN Ia. For a more extensive description of this study, see the related paper (COMING SOON). This page just provides some extra plots and details which we could not fit in that publication.
Here is a sample of the initial list of Supernovae in the OAC that have over 5 observations in the red and blue filters. The rresult and bresult are the total change of the magnitude, initial value to final value, divided by the max.
This shows the grouping of the initial data, into group 1 (under 10 data points), group 2 (10-30 data points), and group 3 (over 30 data points).
After this, we filtered further to remove all sources with fewer than 15 B observations (down to 53). Then, since we want to find the area under the curves, we fit the light curves to a standard model. In the end, 40 SN Ia fit the standard models well enough for us to include in the data set.
The serious part is here - what kind of trigger are you going to use to identify "the blue bump"? We chose to compare a model fit (which assumes no blue bump) with the actual data. The ratio of the area under the actual data with the model fit, if greater then one, might signify the presence of the blue bump.
This image shows the ratio of the data to the fit for all the outliers (see the histogram to the right), upper and lower quartiles.
To get a better sense of the differences between them, we found the squared residuals between the data and the model:
There were only two outliers in both the ratio and the residual data sets:
In this table, we cautiously label our upper quartile as "blue bumps", N=2, out of 40 filtered sources (some did not fit the model well at all), for a ~5% of sources created by the single degenerate scenario (an upper limit, actually).
As discussed in the paper, only around 10% of sources that have non-degenerate companions are actually seen to have the blue excess, which means our actual estimate is 20 in our sample of 40.
The data and the scaled models are given below for the candidate sources.
