Measuring Sizes & Shapes of Galaxies


Wed, 25 Oct 2017, 12:00 pm - 01:00 pm
Natural Science Bldg. 312 - Louisville, KY
Astronomy Lunch
Samir Kusmic
University of Louisville, Department of Physics and Astronomy


Software is how galaxy morphometrics are calculated, cutting down on time needed to categorize galaxies. However, new surveys coming in the next decade is expected to count upwards of a thousand times more galaxies than with current surveys. This issue would create longer time consumption just processing data. In this research, we looked into how we can reduce the time it takes to get morphometric parameters in order to classify galaxies, but also how precise we can get with other findings. The software of choice is Source Extractor, known for taking a short amount of time, as well as being recently updated to get compute morphometric parameters. This test is being done by running CANDELS data, five fields in the J and H filters, through Source Extractor and then cross-correlating the new catalog with one created with GALFIT, obtained from van der Wel et al. 2014, and then with spectroscopic redshift data. With Source Extractor, we look at how many galaxies counted, how precise the computation, how to classify morphometry, and how the results stand with other findings. The run-time was approximately 10 hours when cross-correlated with GALFIT and approximately 8 hours with the spectroscopic redshift; these were expected times as Source Extractor and already faster than GALFIT's run-time by a large factor. As well, Source Extractor's recovery was large: 79.24% of GALFIT's count. However, the precision is highly variable. We have created two thresholds to see which would be better in order to combat this;we ended up picking an unbiased isophotal area threshold as the better choice. Still, with such a threshold, spread was relatively wide. However, comparing the parameters with redshift showed agreeable findings, however, not necessarily to the numerical value. From the results, we see Source Extractor as a good first-look, to be followed up by other software.


Natural Science Bldg. 312