The use of type Ia supernovae ( SNe Ia ) as cosmological standard candles is fundamental in modern observational cosmology .
In this letter , we derive a simple empirical photometric redshift estimator for SNe Ia using a training set of SNe Ia with multiband ( griz ) light-curves and spectroscopic redshifts obtained by the Supernova Legacy Survey ( SNLS ) .
This estimator is analytical and model-independent ; it does not use spectral templates .
We use all the available SNe Ia from SNLS with near maximum photometry in griz ( a total of 40 SNe Ia ) to train and test our photometric redshift estimator .
The difference between the estimated redshifts z _ { phot } and the spectroscopic redshifts z _ { spec } , ( z _ { phot } - z _ { spec } ) / ( 1 + z _ { spec } ) , has rms dispersions of 0.031 for 20 SNe Ia used in the training set , and 0.050 for 20 SNe Ia not used in the training set .
The dispersion is of the same order of magnitude as the flux uncertainties at peak brightness for the SNe Ia .
There are no outlyers .

This photometric redshift estimator should significantly enhance the ability of observers to accurately target high redshift SNe Ia for spectroscopy in ongoing surveys .
It will also dramatically boost the cosmological impact of very large future supernova surveys , such as those planned for Advanced Liquid-mirror Probe for Astrophysics , Cosmology and Asteroids ( ALPACA ) , and the Large Synoptic Survey Telescope ( LSST ) .