STEP 3: Results
STEP home Simulations Data Mailing List

Shear TEsting Programme 3: Weak lensing analysis of simulated space-based images

Jason Rhodes, Catherine Heymans, Will High, Joel Berge, Gary Bernstein, Sarah Bridle, Thomas Erben, Christopher Hirata, Henk Hoekstra, Mike Jarvis, David Johnston, Rachel Mandelbaum, Yannick Mellier, Reiko Nakajima, Stephane Paulin-Henriksson, Molly Peeples, Alexandre Refregier, Tim Schrabback, Uros Seljak, Ludovic Van Waerbeke.

The third STEP project tests shear measurement methods on simulations of high resolution space-based data. Here we present the results of the blind analysis.

Summary of results
In summary, we find that:

  • The methods that have been used to analyse space-based data fare well, to better than 7% accuracy (TS (~1% accuracy), CH, JR). CH and TS have a similar accuracy with STEP1/2.
  • The methods that have not been applied to space-based data do not perform as well in STEP3 as in STEP2.
  • The methods that fare well (TS, CH, JR) show no significant trend of measurement accuracy with pixel scale, galaxy profile type or PSF type.
  • JB, SP and RM find more accurate results with a larger pixel scale. RN finds more accurate results with a smaller pixel scale. RN and RM have more accurate results with the SNAP PSF. JB finds slightly better results with shapelet galaxy profiles.
  • The number density of galaxies used in these analyses ranged from ~45 galaxies per sq arcmin, to ~175 galaxies per sq arcmin.
Shear measurement methods and lensers
The following people completed the STEP3 analysis.

ID LenserMethod
JB Joel BergeShapelets (Refregier 03)
CH Catherine HeymansKSB+
RM Rachel MandelbaumREGLENS (Hirata & Seljak 03 +)
RN Reiko NakajimaBJ02 (Deconvolution method)
SP Stephane Paulin-HenrikssonKSB+
JR Jason RhodesRRG
TS Tim SchrabbackKSB+
T2 Tim SchrabbackKSB+ using high resolution PSF

STEP3 simulations
STEP3 has 12 different simulations that have the following properties:

PSF ID Pixel scale (arcsec)PSF type galaxy type
C0.101.4m SNAPShapelet

STEP3 accuracy analysis; multiplicative and additive errors
The STEP shear catalogues have undergone the usual STEP analysis, quantifying errors in terms of a multiplicative error (m) and an additive error(c). You can download a plot of STEP m and c parameters for each lensers analysis of each PSF simulation, as listed above. A table of average m and c parameters, is also available, listing the average number density of galaxies (per sq arcmin) used in each analysis. Averages are taken over data with the same pixel scale.

We thank Caltech for hosting the STEP3 simulations, and for providing the extensive computing resources required to produce them. We also thank the Jet Propulsion Laboratory for financial and administrative support of STEP workshops and teleconferences, and Marie Curie research grant for financial support of STEP teleconferences.

Last modified 21st April.
STEP pages maintained by Catherine Heymans: heymans[at]