1 / 38

Large Scale Structure Systematics : weights from observing condition maps

This study focuses on characterizing the observing conditions of the Dark Energy Survey and their impact on galaxy clustering measurements. Survey property maps are used to identify spatial correlations and quantify the significance of these maps. An iterative process is implemented to mitigate systematic effects and assign weights to the galaxy catalogue.

marquezj
Download Presentation

Large Scale Structure Systematics : weights from observing condition maps

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. LargeScaleStructureSystematics: weightsfromobservingconditionmaps Martín Rodríguez Monroy in collaborationwith Jack Elvin-Poole and Aurelio Carnero VII Meeting on Fundamental Cosmology, Madrid, Spain. 11 September 2019

  2. Dark Energy Survey • DECam: 62 CCDs in an hexagonal pattern () • 5 photometricbands: () • Depth • Photometricredshifts, up to Victor Blanco Telescope, 4m, Cerro Tololo, Chile

  3. Dark Energy Survey • DECam: 62 CCDs in an hexagonal pattern () • 5 photometricbands: () • Depth • Photometricredshifts, up to Victor Blanco Telescope, 4m, Cerro Tololo, Chile • SV: • Y1: • Y3/Y5-Y6: • Observingfrom 2013 to 2019 • SurveyingDark Energy with 4 probes: • LargeScaleStructure • Weaklensing • Supernovae • ClusterCounts 1/8 oftheskyobserved

  4. SurveyPropertymaps (SPs) (M. Crocce et al., arXiv:1507.05360) Effectiveexposure time, z-band Effectiveexposure time, z-band Observingconditions(seeing, exposure time, …) haveanimpact in theclusteringsignalofthegalaxies This non-cosmologicalsignal can biasourmeasurements How can wecharacterizetheobservingconditions? -> SurveyPropertymaps (SPs) SP = healpixmapthatkeepstrackofthespatialvariationof a certainstatisticconcerningtheimagingconditionsofthesurveyacrossthesky Assume

  5. SurveyPropertymaps (SPs) • DES Y3 -> 4 photometric bands x 25 SPs + 1 extinction map + 1 stellar density map = 102 SPs • Some of these maps correspond to the same physical quantity derived using different statistics for the stacked images: minimum, maximum, weighted averaged mean or sum of the quantity among all the contributing images • In other cases, a map is a function of other SPs, e.g., the depth maps • This means that many of these SP maps will be strongly correlated -> we can reduce their number by studying their spatial correlations at first • This previous study is performed per photometric band separately. Correlations between bands will be dealt with later by our pipeline

  6. SurveyPropertymaps (SPs) • Pearson’s correlationcoefficient, , helpstoidentifythese “families” ofSPs • Italsoallowstoidentifycorrelationsbetweenothermaps • Set differentcutsfor • Wefixto define a family. When posible, wetaketheweightedaveraged mean as representative • 102 SPs -> 34 SPs 0.7 -0.7

  7. Systematicsmitigation: definitions • How can wequantifytheimpactoftheSPsonthe data? -> Relationbetweenthegalaxy numberdensity () at eachregionofthesky and theSP mapvaluethere • Goal: determine whethertheimpactissignificant and mitigateit • Ifsignificant, how do wemitigateit? -> Weightingthe data Simulations Data Beforeweighting Beforeweighting

  8. Systematicsmitigation: definitions • How can wequantifytheimpactoftheSPsonthe data? -> Relationbetweenthegalaxy numberdensity () at eachregionofthesky and theSP mapvaluethere • Goal: determine whethertheimpactissignificant and mitigateit • Ifsignificant, how do wemitigateit? -> Weightingthe data Simulations Data After weighting After weighting

  9. Systematicsmitigation: definitions Determine thesignificanceofthe SP signal: : fit to : fit to Model: Define

  10. Systematicsmitigation: definitions Effectiveexposure time, z-band

  11. Systematicsmitigation: definitions Effectiveexposure time, z-band However, thistellsnothingaboutthesignificanceofthissignal. WeneedrealizationsoftheUniversetocheckthis

  12. Systematicsmitigation: definitions 1000 Effectiveexposure time, z-band Define oursignificance:

  13. Systematicsmitigation: iterativeprocess (J. Elvin-Poole et al., arXiv:1708.01536) • In ordertoremovethesystematicimpact, aniterativeprocessis run. Firstofall, wemust: • Calculatethesignificanceofeach SP • Set a thresholdforthesignificance • SorttheSPsfromhighesttolowestsignificance and identifythe SP withhighestoneoverthethreshold

  14. Systematicsmitigation: iterativeprocess Mostsignificant SP = Eff. exp. time-z

  15. Systematicsmitigation: iterativeprocess • Fit to • Define theweight • Applytheweightmaptothe Galaxy catalogue: • Iterative process: • Compute againthe SP significances and sortthemfromhighesttolowest • SP with? -> Go back to 1, 2 and 3 • RepeatuntilallSPshavesignificanceunderthresholdormax. numberofiterationsreached

  16. Systematicsmitigation: iterativeprocess • Fit to • Define theweight • Applytheweightmaptothe Galaxy catalogue: • Iterative process: • Compute againthe SP significances and sortthemfromhighesttolowest • SP with? -> Go back to 1, 2 and 3 • RepeatuntilallSPshavesignificanceunderthresholdormax. numberofiterationsreached

  17. Systematicsmitigation: iterativeprocess • Fit to • Define theweight • Apply theweightmaptothe Galaxy catalogue: • Iterative process: • Compute againthe SP significances and sortthemfromhighesttolowest • SP with? -> Go back to 1, 2 and 3 • RepeatuntilallSPshavesignificanceunderthresholdormax. numberofiterationsreached

  18. Systematicsmitigation: iterativeprocess • Fit to • Define theweight • Applytheweightmaptothe Galaxy catalogue: • Iterativeprocess: • Compute againthe SP significances and sortthemfromhighesttolowest • SP with? -> Go back to 1, 2 and 3 • RepeatuntilallSPshavesignificanceunderthresholdormax. numberofiterationsreached

  19. Systematicsmitigation: iterativeprocess • Fit to • Define theweight • Applytheweightmaptothe Galaxy catalogue: • Iterative process: • Compute againthe SP significances and sortthemfromhighesttolowest • SP with? -> Go back to 1, 2 and 3 • RepeatuntilallSPshavesignificanceunderthresholdormax. numberofiterationsreached

  20. Systematicsmitigation: iterativeprocess SPswehavetoweightfor ( ): Eff. exp. time-z Skybrightness-i FWHM-i

  21. Systematicsmitigation: testingtheimpact Resultsfrom DES Y1 0.3 < z < 0.45 0.45 < z < 0.6 0.15 < z < 0.3 (J. Elvin-Poole et al., arXiv:1708.01536) Y3 resultscomingsoon

  22. THANK YOU

  23. POST-CREDITS SCENES

  24. SurveyPropertymaps (SPs) • Pearson’s correlationcoefficient, , helpstoidentifythese “families” ofSPs • Italsoallowstoidentifycorrelationsbetweenothermaps

  25. SurveyPropertymaps (SPs) • Pearson’s correlationcoefficient, , helpstoidentifythese “families” ofSPs • Italsoallowstoidentifycorrelationsbetweenothermaps • Set differentcutsfor

  26. SurveyPropertymaps (SPs) • Pearson’s correlationcoefficient, , helpstoidentifythese “families” ofSPs • Italsoallowstoidentifycorrelationsbetweenothermaps • Set differentcutsfor • Wefixto define a family. When posible, wetaketheweightedaveraged mean as representative • 102 SPs -> 34 SPs 0.7 -0.7

  27. Systematicsmitigation: testingtheimpact Resultsfrom DES Y1 Y3 resultscomingsoon (J. Elvin-Poole et al., arXiv:1708.01536)

  28. Systematicsmitigation: testingtheimpact Resultsfrom DES Y1 0.3 < z < 0.45 0.45 < z < 0.6 0.15 < z < 0.3 (J. Elvin-Poole et al., arXiv:1708.01536) Y3 resultscomingsoon

  29. Systematicsmitigation: testingtheimpact Resultsfrom DES Y1 0.6 < z < 0.75 0.75 < z < 0.9 (J. Elvin-Poole et al., arXiv:1708.01536) Y3 resultscomingsoon

  30. Systematicsmitigation: weightsvalidation • How do wevalidatetheseweights? • Selectionof a threshold: • Do wecorrectforthe 102 SPsusingonlythe 34 representatives? • Doestheselectionof a significancethresholdimpactourconstraints? • Isthere a thresholdthatistoohigh, so systematiccontaminationisleftwithoutcorrecting? • Biases: • Doesthe pixel estimatorfor introduce a biasduetotheweightingmethoditself? • Isthere a thresholdthatistoostrict, so weovercorrect? • Increaseuncertainty: • Istheweightingmethodaffectingthecovariance?

  31. Systematicsmitigation: weightsvalidation • Do wecorrectforthe 102 SPsusingonlythe 34 representatives? SPswehavetoweightfor ( ): Eff. exp. time-z Skybrightness-i FWHM-i

  32. Systematicsmitigation: weightsvalidation • Doestheselectionof a significancethresholdimpactourconstrains? 0.3 < z < 0.45 0.45 < z < 0.6 0.15 < z < 0.3 (J. Elvin-Poole et al., arXiv:1708.01536)

  33. Systematicsmitigation: weightsvalidation • Does theour pixel estimatorfor introduce a biasduetotheweightingmethoditself? Landy-Szalayestimator (pixel version): Define estimatorbias: • Isthere a thresholdthatistoostrict, so weovercorrect? Define false correctionbias:

  34. Systematicsmitigation: weightsvalidation • Does theour pixel estimatorfor introduce a biasduetotheweightingmethoditself? • Isthere a thresholdthatistoostrict, so weovercorrect? 0.3 < z < 0.45 0.45 < z < 0.6 0.15 < z < 0.3 (J. Elvin-Poole et al., arXiv:1708.01536)

  35. Systematicsmitigation: weightsvalidation • Istheweightingmethodaffectingthecovariance? 0.3 < z < 0.45 0.45 < z < 0.6 0.15 < z < 0.3 (J. Elvin-Poole et al., arXiv:1708.01536)

More Related