load_nz_fits

Load a number density n(z) from a FITS file

File	number_density/load_nz_fits/load_nz_fits.py
Attribution	CosmoSIS Team
URL

This module is designed to work with the number density part of the FITS files described in: http://github.com/joezuntz/2point/

High-precision analyses of two-point measurements require knowing precisely what is meant by a given n(z) analysis. Most of the CosmoSIS standard library modules downstream of this one assume that the n(z) specifies sample points in a smooth curve, since that seems most physical. Most photo-z codes, on the other hand, provide data in the form of histograms, with small bins in z having a constant assumed n(z) within them.

The actual difference between these two forms is usually well within the systematic errors associated with photometric redshift estimation, so if you’re doing a realistic analysis the difference should be washed out. But for code comparison exercises where you are trying to ensure 0.1% level differences they are very large.

The parameter ‘upsampling’ in this module is designed to address this. Higher upsampling values add new sample points in between the existing ones, so that the two forms look much closer.

A proper solution to this, where we use splines the parts of the code that actually use the n(z) that understand the histogram form, is in our roadmap.

Assumptions

• Data is provided in FITS extensions NZ_{NAME}

• There are various subtleties to do with the sampling and form of the n(z) data

Setup Parameters

Name	Type	Default	Description
nz_file	str		Absolute or relative path to an n(z) file
data_sets	str		Space separated names of the extensions from the FITS files to load and save to the block
prefix_extension	bool	True	Add the prefix NZ_ to the names in data_sets when looking in the FITS file
prefix_section	bool	True	Add the same NZ_ prefix to the section names used in the block.
upsampling	int	1.0	The number of sample points output for each one in the file. n(z) is assumed flat between them. See notes above.

Input values

None

Output values

Output values

Section	Name	Type	Description
wl_number_density	nz	int	Number of redshift samples
	nbin	int	Number of bins
	z	real 1d	Redshift sample values
	bin_{i}	real 1d	n(z) at redshift sample values. bin_1, bin_2, …