What is dark energy?
What is dark energy?
I/DARK ENERGY UNIVERSE SIMULATION SERIES (DEUSS)
Rasera, Y., Alimi, J.-M., Courtin, J., Roy, F., Corasaniti, P.-S., Fuzfa, A., Boucher, V., Introducing the Dark Energy Universe Simulation Series (DEUSS), 2010, American Institute of Physics Conference Series, 1241, 1134
Deschamps, R. et al, 2011
Imprints of realistic dark energy models on cosmic structure formation
According to current SNIa Hubble diagrams and CMB anisotropies power spectrum, the universe energy budget is dominated by dark energy. However the nature of dark energy is still unknow. What are the properties and origin of dark energy? How do density and equation of state evolve across cosmic times? What is its spatial distribution? Several physical interpretations to the observed recent acceleration of the expansion of the universe have been proposed. For instance, cosmological constant is the “standard” interpretation. However, it suffers from the coincidence problem (why the energy density of dark matter and dark energy are so close only today?) and fine tuning problem (why its value is so low compared to vaccuum energy?). Another interpretations are backreaction, quintessence, deviations from general relativity, among many others. In order to discriminate between these models one has to refine current observations at the homogeneous and linear level (SNIa, CMB), or to find new observables at the non-linear level (mass function, power spectrum, profile, angular momentum, ?, etc.). We focus on the latter idea by investigating the imprints of dark energy on cosmic structure formation using very high-resolution N-body simulations. To this aim we have performed a Grand Challenge N-body simulation suite for various realistic dark energy models. These dark energy models are degenerated since we built them so as to be in agreement with current constraints at the homogeneous and linear level (CMB, SNIa). How can we break these degeneracies? What new observables can we find to constrain dark energy properties? What is the role of dark energy on structuration at all scale from kpc to Gpc? How much do present-day structure record the past expansion rate of the universe?
Grand Challenge N-body simulation suite with billion particles
To tackle these problems, we have run the Dark Energy Universe Simulations (DEUSS) which represents the largest dynamical dark energy simulation suite to date in term of spatial dynamics (see picture above). It consists in nine Grand Challenge simulations with 1 billion particles each (and up to 7 billion cells). And, it resolves scales from 4 Gpc down to 3 kpc for the three considered realistic dark energy models (ie calibrated on SNIa and CMB): LCDM, quintessence with Ratra-Peebles potential, quintessence with Sugra potential. For each cosmology, we have run three simulation volumes: 225 Mpc, 900 Mpc and 3600 Mpc. The MPI code RAMSES has been run on 4096 Blue Gene/P cores at IDRIS supercomputing center. It represents 5 millions mono-cpu hours (that is to say about 600 years).
We have backuped three kinds of data during the runs. First we have saved 24 snapshots for each simulations. Second, wa have constructed 9 lighcones around an observer at the center of the simulation. Third we have saved 6 samples of particules at every coarse time step. The total amount of data is about 40 TB.
Parallel post-processing to deal with the amount of data
Unfortunately the output files are organized along a Peano-Hilbert curve which could be non-trivial to manipulate and contain a lot of non-useful information. We have therefore run a Slicer in order to clean and split the data in cubic regions or “fields”. Fabrice Roy has also developped “parallel FoF”, a parallel halo finder. On output, the code sorts particles halo per halo and backups theses “halo” files. Overall halo masses range from 4.10^10 Msun to 10^16 Msun, there are up to 500 000 halos per snapshot and up to 3 millions particles per halos. Finally the total number of halos is about 40 millions. Thanks to this heavy post-processing, data become user-friendly and much easier to analyze.
Angular momentum and shape of dark matter halos
In progress
Links: Websites, Database, Data, Codes, Presentations, Videos, Images
All of this is in progress (collaborations are welcome)!!!
Websites:
Database following the IVOA standard for interoperability:
DEUVO (Dark Energy Universe Virtual Observatory)
Download light data:
FTP
IDL code:
DEUSS mass function
Very nice (but huge) movies:
Halo formation in LCDM (red) and quintessence Sugra (green) and quintessence Ratra-Peebles (blue)
Views of the nine simulations at various scale
All of this is in progress (collaborations are welcome)!!!
II/MATTER POWER SPECTRUM AND REALISTIC DARK ENERGY MODELS
Alimi, J.-M., Fuzfa, A., Boucher, V., Rasera, Y., Courtin, J., Corasaniti, P.-S.,
Imprints of dark energy on cosmic structure formation - I. Realistic quintessence models and the non-linear matter power spectrum, 2010, MNRAS, 401, 775
Structuration in an accelerated-expansion universe
Website in progress
III/CLUSTER MASS FUNCTION AS A FOSSILE RECORD OF PAST EXPANSION HISTORY OF THE UNIVERSE
Courtin, J., Rasera, Y., Alimi, J.-M., Corasaniti, P.-S., Boucher, V., Fuzfa, A., Imprints of dark energy on cosmic structure formation: II) Non-Universality of the halo mass function, 2010, arXiv:1001.3425, accepted for publication in MNRAS
Courtin, J., Alimi, J.-M., Rasera, Y., Corasaniti, P.-S., Fuzfa, A., Boucher, V., Imprints of dark energy on structure formation : no universality in mass functions?, 2010, American Institute of Physics Conference Series, 1241, 804
Dark matter halo count and the nature of dark energy
Website in progress
What is dark energy?
Fields of research
Projected dark matter density from 3 simulations of the Dark Energy Universe Simulation Series (DEUSS). Each color represents the density field for one dark energy model (red=LambdaCDM, green=quintessence with SUGRA potential, blue=quintessence with Ratra-Peebles potential). On large scale, the difference are small (black and white image) but imprints of dark energy show up on small non-linear scale (multi-color image).
Credits: Rasera et al, 2010