Continue to investigate n(z) accuracy

[tms-epcc]

From Oct'22 to Dec' 22 quarterly plan FY22 Annual Evaluation: Continue to investigate n(z) accuracy as a function of sky position and extend study to additional effects, including local large-scale structure and blends

[tms-epcc]

17/NOV/22

• looked at 1 year data and have predictions
• need to find out more from DESC on depth and binning.
• addressed large scale structure and blends summarily but need to look at these more specifically as suggested by recipient group

[tms-epcc]

13/DEC/22

• this is still to be tackled

[tms-epcc]

@joachimi reported

• Encouraged by recipients to put this on the back burner for the moment since the major focus is on clustering.

[tms-epcc]

11/MAY/23

• encouraged to write this up as a paper
• plans need to adjust as more is learnt about the datasets
• infrastructure is there but need to be able to propagate to cosmology

[tms-epcc]

29/JUN/23

• have a draft of the paper ready
• some more discussion on contents required
• salmo software package being investigated to analyse results from n(Z) and impact on cosmology
• once tested and working, it will be used to generate the simulated catalogue, do 3x2 with a standard DESC pipeline and then fit with a standard DESC analysis pipeline to get the cosmological parameters

[tms-epcc]

20/JUL/23

• testing the salmo software but not yet able to generate a mock

[tms-epcc]

10/AUG/23

• still learning how to use salmo
• this is referred to in the Jun'23 QU as "We will generate GLASS+Salmo full-sky mocks with varying depth according to the LSST survey strategy"
• Note paper (referred to above) is on hold until the last steps (see #49 and #50) are finished

[tms-epcc]

05/OCT/23

• updated previous results based on comments from DESC meeting regarding

  • survey strategy simulations

  • Photoz quality parameter,

  • machine learning algorithm in additiono to previous PhotoZ estimator

  • will discuss and decide how to proceed

[tms-epcc]

11/JAN/24

• machine learning algorithm has been added to the PhotoZ estimator
• now generating the mock data for full analysis on the cosmological analysis parameters. Expect data vectors in 2 weeks. A further 2 weeks to get the impact on the parameters. Once done, can then start wrap this into a draft paper.

[tms-epcc]

01/FEB/24

• may not look at parameter impact after all since data vector difference is small
• instead may look at impact of small calibration field vs whole wide field survey

[tms-epcc]

06/MAR/24

• looked at impact of small calibration field vs whole wide field survey
• small calibration fields effect is still dominated by cosmic variance
• so may not be sensible to use n(z) from the small calibration fields
• this will be discussed further and mentioned in the D3.6.2 paper
• also given the change to to use Roman-Rubin simulation for the paper the level of impact will have to be clarified

[tms-epcc]

27/MAR/24

• updated all to the new simulation already.
• no major change from previous cases
• this is now finished