1 COSMOS Weak Lensing With COSMOS: An Overview Jason Rhodes (JPL) May 24, 2005 For the COSMOS WL team : (Justin Albert, Richard Ellis, Alexie Leauthaud, Richard Massey, Jean-Paul Kneib, Yannick Mellier, Alexandre Refregier, Ludovic van Waerbeke, Satoshi Miyazaki, James Taylor, Anton Koekemoer)

2 If there is any intervening large-scale structure, light follows the distorted path (exaggerated). Background images are magnified and sheared by ~2%, mapping a circle into an ellipse. Like glass lenses, gravitational lenses are most effective when placed half way between the source and the observer. z observer =0 z galaxy ≈1 z lens ≈0.3–0.5 Weak lensing effect cannot be measured from any individual galaxy. Must be measured statistically over many galaxies Weak Gravitational Lensing

3 How Gravitational Lensing Works UnlensedLensed Statistical measurement on many galaxies Lensing induced ellipticities 1-2% Telescope Point Spread Function (PSF) is the primary systematic concern –it changes e!

4 COSMOS Cosmology with WL Dark matter maps (Richard) Galaxy properties as a function of environment (James) Properties of galaxy DM haloes (Alexie) Statistical measures of cosmological parameters (Ludo)

5 COSMOS WL Pipeline

6 COSMOS That Pesky PSF Problem The PSF pattern is time dependent

7 COSMOS Time Dependence Why a three week time scale

8 COSMOS The GEMS Solution Unique combination of high data volume is a short time Divide data into two time bins of 5 days Can do internal correction with stars in each time bin Empirically derived correction using stars in the field- the typical weak lensing methodology Why won’t it work for COSMOS?? Data too spread out in time Not enough stars per field (or time bin) for internal correction Usual method of using stars in the fields will not work We need more stars!!

9 COSMOS Our Solution Not enough stars→ Make our own! Use TinyTim PSF modeling software Create stars without geometric distortion (but with diffraction and other PSF effects Required modification of TinyTim program Distortion removal (aka multidrizzle) introduces complications that are stochastic Create dense stellar grids (up to 50x50 across field) Create at range of focus positions (-10 to +5 μm) Use stars in each COSMOS field (~10 to 20) to pick the best focus value Use the template at that focus value to perform correction Eliminates need for interpolation between stars

10 COSMOS TT PSF Models

11 COSMOS Our Solution Works… Pretty Well

12 COSMOS Diagnostic Plots From alexie

13 COSMOS The Need for Re-reduction

14 “Aliasing” from pixellization Unfortunately, a star’s sub-pixel position affects its observed shape.

15 “Aliasing” from second pixellization Then it happens again during DRIZZLE! This at least must be avoidable…

16 Solution: adjust pixel scale & DRIZZLE kernel …it is!

17 COSMOS Recommended Multidrizzle Parameters Created TinyTim models with same diffraction everywhere, only distortion is different Run through multidrizzle w/ various parameters Large improvement going from 0.05 to 0.03 arcsecond pixels Gaussian kernel has lower RMS(e) and is more stable

18 COSMOS Conclusions We have a workable PSF correction model Continuing to improve it Better star selection Better focus selection Getting science results now (see next few talks) Will see large improvement w/re-reduction Our pipeline is ready to incorporate all cycle 13 data Can fold in better photo-zs easily