summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorbnewbold <bnewbold@alum.mit.edu>2010-07-20 16:15:52 +0000
committerbnewbold <bnewbold@adelie.robocracy.org>2010-07-20 16:15:52 +0000
commit414bb87d328bfc3dff3e8c920eaf3e33e1b72252 (patch)
treea60ac88bb5dcf45091da799f7bbf47eb8ac7b6bb
parent3db5a31a19973e4ff20da24aebd6f4c9489a0a2b (diff)
downloadafterklein-wiki-414bb87d328bfc3dff3e8c920eaf3e33e1b72252.tar.gz
afterklein-wiki-414bb87d328bfc3dff3e8c920eaf3e33e1b72252.zip
spherical data analysis stuff
-rw-r--r--ClassJuly20.page22
1 files changed, 22 insertions, 0 deletions
diff --git a/ClassJuly20.page b/ClassJuly20.page
index 1679954..e667e8d 100644
--- a/ClassJuly20.page
+++ b/ClassJuly20.page
@@ -1,3 +1,25 @@
# Lecture Notes
**[Lucas' Notes for Lecture 7](/SummerCourseRiemannSphere2.pdf)**
+
+# Sphere Pixelation
+
+Somewhat related to the contents of this lecture and our discussions afterwards, "good" pixelations of the sphere are an ongoing topic of math/numerical research with applications to astrophysics data analysis. Pretty pictures:
+
+![This is the most popular current scheme, developed at NASA/JPL (http://healpix.jpl.nasa.gov/)](/ClassJuly20/healpixGridRefinement.jpg)
+
+![This is an icosahedron scheme by Max Tegmark (http://space.mit.edu/home/tegmark/icosahedron.html)](/icosahedron.gif)
+
+One of the most important metrics calculated from CMB (cosmic microwave background) skymaps (like those from WMAP, or this [super pretty new map](http://4.bp.blogspot.com/_GVA115I1I8Y/TDLWxv-FiDI/AAAAAAAAAww/P_cC-oE67uQ/s1600/PLANCK_FSM_03_Black.jpg) from Plank) is the power spectrum: This is essentially the Fourier transform of the surface of the sphere using spherical harmonics (indexed by $\el$ instead of $n$; familiar from physical chemistry and the quantum mechanics of the hydrogen atom). The power spectrum of the skymap, which basically describes the relative populations of lumps in the skymap, is the established meeting ground between early universe theorists (who calculate what the power spectrum would look like under different cosmological models, such as inflationary, open/closed/flat, cosmic strings, quantum loop gravity, cold-dark-matter-dominated, and other pseudo-nonsense wink-wink) and observational scientists who try to constrain the spectrum with better and better measurements.
+
+![Power spectrum image from NASA/WMAP team](/ClassJuly20cmbpower.jpg)
+
+Just recently the number crunching of large observational datasets has become an obstacle to progress: we need better algorithms ([like the "the fastest Boltzmann code in the west"](http://arxiv.org/abs/astro-ph/9603033)) for dealing with these skymaps, particularly when trying to measure "non-Gaussianities" in the distribution of bumps. I have a feeling that some of the techniques we use in this course are applicable!
+
+More background (heh):
+
+- http://backreaction.blogspot.com/2007/12/cmb-power-spectrum.html
+- http://space.mit.edu/home/tegmark/cmb/pipeline.html
+- http://space.mit.edu/home/tegmark/strategy.html
+- http://space.mit.edu/home/tegmark/karhunen.html
+- http://www.astro.ucla.edu/~wright/cosmo_01.htm \ No newline at end of file