1. EGEE-II INFSO-RI-031688 Enabling Grids for E-sciencE www.eu-egee.org EGEE and gLite are registered trademarks The ZEN project A. Tilquin Centre de Physique des Particules de Marseille

2. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 2 Determination of cosmological parameters Scientific Goal: Address open questions of fundamental cosmology: dark matter/dark energy sector and primordial universe Main people involve: ANR program CPPM Alain Bonnissent, Anne Ealet, Dominique Fouchez, Lei Sun, Diane Talon-Esmieu, Charling. Tao CEA Philippe.Brax, Jean Batiste Melin, Christophe.Yèche, Dominique.Yvon, Nathalie Palanque- Delabrouille, Alexandre.Réfrégier, Jim.Rich CPT Christian. Marinoni, Pierre Taxil, Jean-Marc Virey, Stefan. Linden LAL Francois Couchot, Olivier Perdereau, L. Perotto, S. Plaszczynski, C. Rosset FCPPL: Franco-China-Particle-Physics-Laboratory program IHEP/PKU/NAOC/Tsinghua U/Beijing Normal U. ZHANG XinMin, ZHAO Gongbo, QIAN Zuxuan, XIA Junqing, FAN Zuhui, LI Hong, Zhu Zhonghong, QIN Bo, DENG Jinsong, CHAO Wu, ZHOU Xu, WANG Xiaofeng

3. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 3 Plan Few words about cosmology and dark energy How to characterize dark energy Statistical method and datagrid Results already obtained by the Chinese and French groups (ESR and Euchina) Prospective

4. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 4 Some words about cosmology Cosmology based on: Homogenous and isotropic Universe General relativity: G  = 8  GT  Energy content:  =  /  c Equation of state: w(z)=p/  Matter (  m ) : w =0 Radiation (  r ) : w = 1/3 Cosmological cste (   ): w=-1 Dark energy (  X ) : w(z)  T =  m +  r +  X (  c =10 -29 g/cm 3 )

5. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 5 Some words about dark energy In 1998, the Supernova Cosmology Project and High-z team shown that high red-shift supernovae are fainter than expected: a new energy component is needed. Definition:  c (  c =10 -29 g/cm 3 ) Z Dark energy or cosmological constant caracterized by reduce density:      c For a flat Universe:

6. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 6 From observations to the Concordance Model CMB(WMAP) SNe Ia LSS Universe is mainly dark ~70% DE and ~25% DM. a 2006 status Our Univers is accelerating

7. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 7 Phenomenological aspect From where this acceleration is coming from ? Two main classes of theory/model : Particle physics or Gravity ?  Matière standard gg gg gg g (i)  General relativityModification of gravity New dominant fields over matter at z=0 Matière standard Matière standard Matière standard  AA aa New fields are non dominant ->expansion rate is unchanged. Coupling with photons-> apparent modif. of the dynamic: oscillations photon/axions in B ->   New non dominant fields. Long range force induced by light field  modification of the gravity+violation of the équivalence principle: Tensor- scalar theory, cameleon model, runaway dilaton …(w<-1) Differents type of gravitons, masseless or not : extra dimensions, multi brane, multigravite, brane induced gravity, simulated gravity… négative energy… Acceleration is due to new fields weekly couple to gravitation:  CDM, quintessence, topological default, Chaplygin Gaz…. Particle physics Gravity In all cases: New fields are needed Dark energy can always been described by an effective equation of state: w=p/  = w 0 +w a z/(1+z)

8. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 8 How to extract w 0,w a ? Degeneracies between parameters imply multi-probes analysis sensitive to complementary quantities: Cluster counts Supernovae Baryon Wiggles Cosmic Shear Angular diameter distance Growth rate of structure Evolution of dark matter perturbations Angular diameter distance Standard candle Luminosity distance Evolution of dark matter perturbations Angular diameter distance Growth rate of structure CMB Snapshot at ~400,000 yr, viewed from z=0 Angular diameter distance to z~1000 Growth rate of structure (from ISW)

9. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 9 Physical parameters Important number of parameters, cosmological and astrophysical –  b /  m density for baryon/matter –  density for neutrino’s –  T curvature density –HHubble constant, –n s spectral index, –  reionisation optical depth –  8 normalization for CMB, WL and BAO. –m s normalization for SNIa. –yhe Helium fraction –w 0,w a Equation of state…. Efficient statistical tools needed: –Bayesian statistic and MCMC : Chinese and astrophysicist choice –Frequentist statistic and datagrid : French and particle physicist choice

10. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 10 Statisitcal method Statistic based on  2 (  i,w 0,w a,…) Minimum using the gradient method:   2 /   i =0 Numerical resolution and iterative: Error computation: Contour: Solving the equation  2 =  2 min +s 2 Marginalization obtained by minimization:  2 (w 0,w a ) =  2 min (  i | w 0,w a ) The contour is constructed by minimizing the  2 on a grid of points (minimum 20*20) and iso-  2 are constructed using interpolation. Each point (20 hours of computing) is calculated on a CE. A simple contour requires about 1 year of CPU on a single CPU. Thanks to datagrid (result in few days)

11. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 11 The Zen program Minuit (ROOT library) Zen.cc Minimizeur Configurator Global Configuration file Fichier de Configuration CMB Fichier de Configuration BAO CMB code BAO code CMB  2 BAO  2 Minimum  2 and parameters

12. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 12 Submission graphical interface A run is a set of n*n jobs (400) send in parallel. A graphical interface has been developed (thanks to Zuxuan Qian) to construct and submit them. Global configuration files Zen programs configuration Submit jobs on datagrid

13. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 13 Job control graphical interface Jobs monitoring and output data manager Get jobs statusKill or submit jobsGet log and error filesManage output file Check output results and update jobs status

14. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 14 Cosmological constraint using BAO First results using datagrid within: –ESR (Earth Science Research) VO (thanks to M. Petit-Didier) and Euchina Virtual organization –Using SN+CMB+BAO with frequentist statistic and (BAOfit from Sun Lei, PKU-CPPM Join PhD) Data  CDM ESR:40000 hours of CPU ESR+Euchina: 10000 hours of CPU Data  CDM Diane Talon-Esmieu Riess 2006: SNLS1+HST+new calib.

15. Enabling Grids for E-sciencE EGEE-II INFSO-RI-031688 To change: View -> Header and Footer 15 Summary and perspectives The ZEN program is now running on datagrid Thanks to ESR virtual organization The two graphical interfaces are very powerfull tools to submit jobs and to debug. Major problem is still the linux version (SL3,SL4) Our first results on multi-probes analysis are compatible with competitive analysis (Bayesian) Perspectives: –Add new probes (WL) and wait for new data (PLANCK,SNLS3) –Investigate new theoretical models and implement them in ZEN –Add new statistics (MCMC, Bayesian…) –Optimize the speed –Start systematic analysis