Development and Resultant Implementation of Custom Software for Analyses of Galactic Neighborhoods

Download (PDF)

Abstract

The structure and organization of galaxies and their environments within the universe is a critical and fundamental question in the field of astrophysics. The amount of data available necessitates computer software to aid analyses. Previous research in the field utilized software built in FORTRAN for these analyses. The amount of data now available to astrophysicists is astounding and a new generation of analysis software is needed. In this thesis I present an overview of the questions posed in the field of astrophysics. I also discuss the pointwise dimension, which is the methodology used to conduct analyses of galactic environments. I discuss the development of custom software, Galactic, and the requirements which are obtained from an expert in the field. The development process including initial technology decisions as well as issues encountered and the solutions devised are presented. Particular attention is paid to improving performance relative to previous software. The Galactic software is then used to conduct scientific analysis of a newer data catalog, the 2MASS Redshift Survey (2MRS). Conclusions of this scientific research find that the environments of early-type galaxies are statistically different from late-type galaxies. The environments of barred spiral galaxies are found to be similar to the environments of unbarred spiral galaxies and to spirals for which the presence of a bar cannot be confirmed or rejected. I find that the environments of early-type galaxies are similar to each other; also all types of spirals also share similar environments. This suggests secular evolution is a larger factor than environment for evolution of elliptical vs. lenticular and barred vs. unbarred spirals. I also find that the clustering of disk-type galaxies around disk-type galaxies evolves over time. Finally, conclusions are presented regarding the benefit of collaboration between software engineers and scientists. Also discussed is the importance of domain knowledge and understanding the user's experience.

Citation Detail

Moore, Michael D. Development and Resultant Implementation of Custom Software for Analyses of Galactic Neighborhoods. Master's Thesis. West Virginia University, 2014.