Finding Gas in Space Clouds

These posts provide documentation of my research process in order to characterize massive clouds of hydrogen in our galaxy. They demonstrate my ability to track progress in context of previous research. I wrote the posts laden with astronomy jargon, but feel free to peruse to get a feel for the diversity of data visualization.


I compare model consistencies predicting radiation fields.
I present results showing the dust emissivity index of the Taurus-California-Perseus Region
In this post I show results from calculating HI transition surface densities.
I outline the steps to derive the incident radiation field on dust grains given the dust grain temperature.

Comparison of cumulative distribution functions of observed and predicted quantities.

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Map of dust of three molecular clouds in our own galaxy. Adopted Bayesian priors to identify the dense cores shown as white crosses.

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Table of Contents

Omitting Variance Estimation of Binned Images

Variability in Between Code Edits

Faint Masking Results

Faint Masking

Test with mock data

We are continuing to test the derived intercept as outlined in the previous post

The following analysis was performed with Lee+12 map: /d/bip3/ezbc/perseus/data/av/perseus_av_lee12_iris_regrid_planckres.fits

These results are completed after reorganizing the parameter estimation code. We should be more confident in these results than previous ones, seeing as each step of the new code was modularized and tested to some extent.

Below are background subtracted images of California, Using a simple mean offset, and a 2D spline fit. The regions outlined in white in the top panel represent the background regions used to fit the intercept and spline.

I have investigated the dependence of derived parameters, width, DGR, and intercept based on the selection of region in the California cloud.

I am continuing the discussion on using intercepts.

I’ve been attending the Penn State University Summer School for Statistics. Here is the program. I’ve highlighted a few key points from some of the lectures.

Determine relevance of including an intercept in the model. This means we would solve for the DGR, width and an intercept for each cloud. Previous attempts to include an intercept led to large intercepts, on order of negative several magnitudes in . See this post for more details.

Below are action items already described in Paper Comments 2 and Paper Comments.

Comments 1

Comments 2

Adding Cores

The K+09 derived their column density calculations from the analytic model from krumholz08 of formation and photodissociation of a spherical cloud bathed in a uniform ISRF.\@ See krumholz09 and krumholz08 for details in the derivation, and a summary of the results in lee12. Here we quickly summarize the important assumptions and results.

I successfully fit the Sternberg model to the vs. relationship. I assumed that our case is a two-sided irradiation by an isotropic field, where they predict an threshold given by

I successfully fit the Sternberg model to the vs. relationship. I assumed that our case is a two-sided irradiation by an isotropic field, where they predict an threshold given by

This post is a continuation of this post, but don’t bother with the previous post, it’s a poor, short, summary.

Below are results from using different maps. See yesterday’s post for how regions are chosen and comparison with Lee+12. Each map leads to similar DGRs, however the K+09 map yields a slightly larger HI width. This leads to thresholds similar to Lee+12.

Below are results from Planck . We can see that the thresholds are quite low compared to those found in Lee+12.

I have found systematically negative in California. Below is an example histogram of the calculated during a monte carlo simulation for a single core. These values are obviously unphysical.

Jouni Kainulainen shared 2MASS Av image with us today. He provided the following warning:

Summary of Sternberg et al. (2014)

Iterative MLE binning

Iterative MLE binning

Application of Iterative MLE binning

Here is my first image