Research Efforts of the UW SSG
Gadgets and widgets have become popular in social networking (e.g. iGoogle, Facebook). They provide a simple way to customize your view of data and analysis tools. In ASCOT (an AStronomical COllaborative Toolkit) we provide a framework for using gadgets in Astronomy. Unlike iGoogle, where all of the widgets are independent, our science tools can communicate with each other. This interactive framework can be customized to however you want to see the sky. To try ASCOT you can use this demo page (best viewed with Chrome and, for the GoogleSky plugin, using a Mac or Windows machine). Following are some movies of ASCOT in action:
- Explore SDSS data around Abell 1689
- Examine photometric outliers
- Histograms and Density plots
- Custom Dashboards
Please send feedback to Andrew Connolly
Cloud computing has revolutionized the way business is done on the Internet. By linking hundreds to thousands of computers together with petabyte data storage, massively intensive (in both data and computation) tasks can be addressed. This represents a departure from traditional thinking in the area of high performance computing in which the data, cpus, and memory all reside on the same physical machine with fast communication between all components. Instead the processing units only have access to a small amount of local storage and can be considered to be isolated from the other compute nodes in the cluster. This fundamental difference in cluster architecture requires a shift in how massively parallel problems are approached. The Cluster Exploration (CluE) initiative is a joint effort between NSF, Google, and IBM to allow academics access to cloud computing resources for research purposes.
Observational cosmology is addressing many fundamental questions about the nature of the universe through a series of ambitious wide-field optical and infrared imaging surveys (e.g. from the properties of dark matter to the nature of dark energy). The challenge these surveys pose is how do we analyze and interact with data that is being taken at a rate 1000 times greater than existing experiments. How do we determine the interdependencies between the observable properties of stars and galaxies in order to better understand the physics of the formation of the universe? At UW we address these questions in a number of different ways; from studying the physical properties of galaxies as a function of redshift, to reconstructing the mass distribution from weak gravitational lensing, to classifying galaxy spectra or looking for unusual sources in photometric and spectroscopic surveys.
Read more about our cosmology and galaxy reseach in the Galaxy Pages.
Read more about our DRaGONS survey DRaGONS.
Read more about the results of the projects listed below on the Moving Objects Pages.
Image subtraction techniques have enabled the vast majority of current time--domain optical surveys by allowing them to concentrate resources on only those objects that are varying in position or time. The basic method is derive a PSF--matching kernel that photometrically aligns two images taken of the same part of the sky, but at different times. In their difference lies anything that has varied between these epochs. The essence of this technique lies in estimating the Psf--matching kernel. The available degrees of freedom in this problem include the basis functions used to decompose the kernel, whether to apply a "smoothing" (convolution) or "sharpening" (deconvolution) kernel (one being the inverse operation of the other), and how to model the spatial variation of the kernel and/or its basis functions.
Read more about the results of the projects listed below on the Transients Pages.
As the progenitors to CVs, many of which eventually evolve to become Type 1a supernovae (SNe), pre-CVs are integral in understanding some of the more vexing questions associated with the current observed CV population, such as magnetic CV progenitors and the "period gap". The prolific use of Type 1a SNe for groundbreaking cosmology has also lead to an increased interest in the pre-CV environment as we have yet to fully understand the origin and nature of Type 1a progenitors. A detailed study of pre-CV systems allows us to test widely used theories on common envelope (CE) evolution and orbital angular momentum loss (magnetic breaking). Most of what we know about these theories are generally inferred from the properties of the current CV population, which is complicated by the significant effects of mass transfer.
Read more about the results of the projects listed below on the Variable Star Pages.
We frequently generate new tools and algorithms for analyzing astronomical data sets. These include dimensionality reduction tools such as LLE and Principal Component Analysis and tools for calculating magnitudes and errors for the LSST.
To find out more about tools that may be useful in your research check out our software pages.