deletions | additions       diff --git a/section_Statistical_Analysis_label_discuss__.tex b/section_Statistical_Analysis_label_discuss__.tex  index ad43c66..da6fec3 100644  --- a/section_Statistical_Analysis_label_discuss__.tex  +++ b/section_Statistical_Analysis_label_discuss__.tex 
...
We use pseudo-distance metrics to effectively analyze differences between all synthetic observations. For each statistic, we produce a color-plot table denoting distance metric values for all simulation pairs. Section 3 identifies qualitative differences corresponding to stellar feedback; expanding upon this, we now quantify all simulation differences, and determine the sensitivities of the previously mentioned statistics (reword--Cramer wasn't mentioned, as it has no graphical output). Unlike the study carried out in Koch et al.~(2015), our simulation suite does not utilize experimental design to set the simulation parameter values. As discussed in \citet{Yeremi_2014}, comparisons between outputs in one-factor-at-a-time approaches may give a misleading signal since the statistical effects are not fully calibrated.  (note on how to analyze color plots--include???) Figure ??? displays examples of our color-plots. The rows and columns list our synthetic observations, and each square represents a simulation pair in which we calculate the distance. The colorbar denotes the distance values, which vary in range depending on the statistic. We arrange the simulations in order to easily compare strong wind models (W1)  with weaker wind models (W2)  or purely-turbulent models. models (given by t0).  Table ??? provides a summary of our findings, though we discuss their (implications???) in sec 4.1, 4.2, 4.3. (still have to bring color plots closer together--also have to redo PDF distance. Uploading incomplete plots now to produce rough outline of discussion)