A short report on: joint thwg-cswg WP on Definition and selection - PowerPoint PPT Presentation

A short report on: joint thwg-cswg WP on “Definition and selection of non- standard cosmological models” CSWG WP5 “numerical tools for non- standard cosmologies” (Marco Baldi & Kazuya Koyama) Euclid CosmoSim Meeting - Barcelona, 4 XII 2014

JOINT WP: GOALS AND ORGANISATION The joint WP has been established at the last CM in Marseille, and is supposed to conclude its activities by the next CM in Lausanne common mailing list for whole WP discussions: extended-cosmology-group@port.ac.uk The key documents are available on Redmine wiki: http://euclid.roe.ac.uk/projects/swg-th-wp12/wiki Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

JOINT WP: GOALS AND ORGANISATION AIMS : select a list of non-standard cosmologies for which non-linear observables are essential to fully exploit Euclid data and match the broad Euclid science goals ➡ deliver a prioritised list of Reqs . 0 . 6 2.0 EXP003 z=0.00 LCDM − 1 | ¯ f R 0 | = 10 − 4 P(k)/P Λ CDM (k) 1.8 Λ CDM, σ 8 =0.967 1.6 | ¯ f R 0 | = 10 − 5 0 . 5 1.4 | ¯ f R 0 | = 10 − 6 power spectrum enhancement P f ( R ) / P 1.2 0 . 4 Li et al. 2012 2.0 z=1.00 1.8 P(k)/P Λ CDM (k) 0 . 3 1.6 1.4 0 . 2 1.2 2.0 z=3.02 0 . 1 1.8 P(k)/P Λ CDM (k) 1.6 1.4 0 . 0 1.2 1.0 − 0 . 1 0.01 0.10 1.00 10.00 10 − 2 10 − 1 10 0 10 1 k [h/Mpc] k [ h / Mpc ] MB 2012 Puchwein, MB, Springel 2013 Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

JOINT WP: GOALS AND ORGANISATION PARTICIPANTS : more than 60 people expressed interest. Too many for a focused discussion: the group has been divided into 9 sub-groups . SUB-GROUPS : each SG corresponds to one of the main classes of non- standard models identified by the THWG, plus an additional group “others”. Each group is coordinated by 2 SG leads , and is supposed to deliver a list of selected models for the respective class of NSC. SG 1 - Quintessence and EDE SG 5 - Massive neutrinos Leads: E. Linder, A. Fuzfa Leads: J. Lesgorgues, C. Carbone SG 2 - Inhomogeneous cosmologies SG 6 - Self-interacting Dark Matter Leads: V. Marra, J. Garcia-Bellido Leads: T. Kitching, C. Germani SG 3 - Warm Dark Matter SG 7 - Linear DE fluctuations Leads: J. Lesgorgues, M. Viel Leads: P . Ferreira, V. Pettorino SG 4 - Primordial non-Gaussianity SG 8 - Nonlinear DE fluctutuations Leads: V. Desjacques, S. Matarrese Leads: P . Brax, B. Li SG 9 - Others Leads: S. Camera, D. Bertacca Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

JOINT WP: GOALS AND ORGANISATION MEMBERSHIP : the list of participants has been divided into SG leads, active members, and “observing members” SCHEDULE : activities should end by June 2015 - June 15, 2014 : Definition of the WP members list and of the composition of the 9 sub-groups; - December 15, 2014 : First draft of the models lists from the 9 WP sub-groups; - February 15, 2015 : Final document of the models lists from the 9 WP sub-groups; - April 15, 2015 : Prioritization of the list of models based on a general WP discussion; - May 31, 2015 : Final Document on the Redmine wiki page. Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: GOALS AND ORGANISATION AIMS : Develop , Validate , and Optimise numerical tools for non- standard cosmological models (i.e. anything beyond plain “vanilla” Λ CDM) Numerical tools include N-body codes, ICs generators, post- processing pipelines PARTICIPANTS : about 15-20 participants (number of active participants fluctuates) ORGANISATION : Most of the collaborative work is done through discussion sessions at meetings and conferences , much less by email or telecons. MEMORANDUM OF CODES : The first release (v1.0) is available on RedMine. An updated release is planned for mid 2015 Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: DEVELOPMENT PHASE Development phase almost “concluded” (does that actually mean anything?) ✓ At least one 1. Quintessence and Early Dark Energy working code , if not (quintessence, k-essence, phantom, quintom, EDE, etc...) ✓ several, for each class 2. Inhomogeneous large-void models (LTB, swiss-cheese, backreaction, etc...) ✓ 3. Warm Dark Matter (WDM, mixed Hot+Cold DM, etc...) ✓ 4. Non-Gaussian initial conditions (local, equilateral, orthogonal, running, etc...) ✓ 5. Massive Neutrinos (normal & inverted hierarchy) Not from EC members ✓ 6. Self-Interacting Dark Matter (M. Vogelsberger & J. Zavala) (velocity-dependent/independent X section) ✓ 7. Linear spatial Dark Energy fluctuations (clustering DE, coupled quintessence, extended quintessence) ✓ 8. Non-linear spatial Dark Energy fluctuations (massive coupled quintessence, Modified Gravity, etc...) Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: VALIDATION PHASE VALIDATION : presently ongoing code-comparison initiative for various types of Modified Gravity cosmologies. INITIAL CONDITIONS : same for all runs, same IC code as Λ CDM code comparison (R. Smith) Model Participating Codes Reference People M. Baldi, E. Puchwein mg-gadget S. Bose ecosmog f(R) isis H. Winther mg-enzo G. Zhao B. Li ecosmog Symmetron isis C. Llinares F. Schmidt dgpm DGP/Galileon ecosmog A. Barreira ANALYSIS : all runs analysed by the same person (H. Winther) with the same code ( POWMES for the power spectrum, AHF for the mass function) Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: VALIDATION PHASE VALIDATION : some preliminary results from the code-comparison power spectrum (plots from Hans Winther) 1.06 1.06 F5 z=0.0 F5 z=0.0 F5 z=0.5 F5 z=0.5 F5 z=1.0 F5 z=1.0 1.04 1.04 1.02 1.02 P ECOSMOG /P MGGADGET 1 1 P ISIS /P MGGADGET 0.98 0.98 0.96 0.96 0.94 0.94 0.92 0.92 0.9 0.9 0.1 1 10 0.1 1 10 k (h/Mpc) k (h/Mpc) 1% convergence at z = 0 up to k~1 isis seems to agree better with mg-gadget than ecosmog Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: VALIDATION PHASE VALIDATION : some preliminary results from the code-comparison mass function (plots from Hans Winther) 1.1 F5 n ISIS /n MGGADGET z=0.0 F5 MGGADGET z=0.0 F5 n ECOSMOG /n MGGADGET z=0.0 F5 ISIS z=0.0 F5 n ISIS /n MGGADGET z=0.5 F5 ECOSMOG z=0.0 100000 F5 n ECOSMOG /n MGGADGET z=0.5 F5 MGGADGET z=0.5 F5 n ISIS /n MGGADGET z=1.0 F5 ISIS z=0.5 F5 n ECOSMOG /n MGGADGET z=1.0 F5 ECOSMOG z=0.5 F5 MGGADGET z=1.0 1.05 F5 ISIS z=1.0 10000 F5 ECOSMOG z=1.0 n code 1 / n code 2 1000 1 100 0.95 10 1 0.9 1e+12 1e+13 1e+14 1e+12 1e+13 1e+14 M (Msun/h) M (Msun/h) 15-20% scatter among the codes, especially at low masses mg-gadget has more small structures than the other codes Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama

CSWG-WP5: FUTURE... OPTIMISATION : We will need strategies to optimise the non-standard codes for large runs (not as large as EFS but as large as possible...) Most codes are extensions to the standard Λ CDM codes already under optimisation in Euclid: we should have access to optimised Λ CDM codes and port the modified modules into them. POST-PROCESSING : We will need to interface the modified codes with the post-processing pipelines built within Euclid (as e.g. light-cones on-the-fly generation, mock catalogues generation, etc...). So a common effort with the other CSWG WPs will be needed in the near future. Euclid CosmoSim Meeting - Marco Baldi & Kazuya Koyama