Wavelet Space-Scale-Decomposition Analysis of Spectrum of Density Perturbations from QSO's Lyalpha Forests
A method for measuring the spectrum of a density field by a discrete wavelet space-scale decomposition (SSD) has been studied. We show how the power spectrum can effectively be described by the father function coefficients (FFC) of the wavelet SSD. We demonstrate that the features of the spectrum, such as the magnitude, the index of a power law, and the typical scales, can be determined with high precision by the FFC reconstructed spectrum. This method does not require the mean density, which normally is poorly determined. The problem of the complex geometry of observed samples can also be easily solved because the basis are always orthogonal, regardless the geometry of the samples. Using this method, we examine the spectra inferred from Lyalpha forests of both simulated and real samples. We find that 1.) the magnitude of the 1-D spectra is significantly different from a Poisson process; 2.) the 1-D spectra are flat on scales less than about 5 h(-1) Mpc, and show a slow increase with the scale in a range larger than 5 h(-1) Mpc; 3.) the reconstructed 3-D spectra have about the same power as the COBE normalized linear spectrum of the SCDM model on scales less than 40 h(-1) Mpc, but the larger than the SCDM model on scales larger than 40 h(-1) Mpc; 4) the magnitudes of high redshift (z>2.51) spectra generally are larger than those of low redshift (z<2.51) results. Points 3) and 4) are probably caused by large geometric biasing on large scales and high redshifts.
American Astronomical Society Meeting Abstracts
- Pub Date:
- December 1995