To start off the research institute I learned how to use the computer programming language called Matlab. We have been using Matlab to plot spectra and perform many mathematical functions, which came from various journal literature by researchers who have been experimenting with the expanding photosphere method (EPM) before. EPM is a method to determine the distance to a supernova by measuring how fast the photosphere is expanding by measuring temperature of the photosphere and combining these two measurements to find the radius of the photosphere and therefore, the luminosity. From this the distance to the galaxy that the supernova occurred in can be calculated.
So far we have looked at Spectra from 3 separate supernovas from 1994, 2002, 2010. We have also started looking at one from 1999. To determine the distance, we have to calculate the apparent size of the supernova photosphere divided by the expansion velocity, and then graph this against the dates on which the spectra were taken.
The slope of the graph indicates the distance, and the y-intercept indicates when the supernova actually exploded (they’re usually not discovered until at least several days after the explosion). Once the technique is perfected we can not only determine how far away the supernova occurred, but when it occurred. But to perfect this method, we need several high-quality spectra over several different observations of a supernova, and not every supernova is observed that closely. We’re hoping that by looking at different combinations of the data available for each supernova, we’ll be able to narrow down which sets give more reliable results.
The first two supernovae we’ve looked at, from 2010 and 2002, gave distances for them pretty close to what other methods say are the distances to their home galaxies. So far, the uncertainties in the times of these explosions are a little high (up to plus or minus several days). This could be due to high uncertainties in some of the data points, especially for later spectra due to greater signal noise and line blanketing. Line blanketing makes it harder to estimate the photosphere’s temperature and to properly identify which line is which (for determining velocities).
The 1994 supernova’s calculated distance is further off than for the first two; we’ll be looking at our data again to see if any measurements need to be retaken. Signal noise is a concern with this supernova as well; so far, only two separate spectra have been usable. This further stresses the need for more comprehensive observations of supernovae.