Photometric redshifts for the S-PLUS Survey: is machine learning up to the task?
Authors:
E. V. R. Lima,
L. Sodré Jr.,
C. R. Bom,
G. S. M. Teixeira,
L. Nakazono,
M. L. Buzzo,
C. Queiroz,
F. R. Herpich,
J. L. Nilo Castellón,
M. L. L. Dantas,
O. L. Dors,
R. C. T. Souza,
S. Akras,
Y. Jiménez-Teja,
A. Kanaan,
T. Ribeiro,
W. Schoennell
Abstract:
The Southern Photometric Local Universe Survey (S-PLUS) is a novel project that aims to map the Southern Hemisphere using a twelve filter system, comprising five broad-band SDSS-like filters and seven narrow-band filters optimized for important stellar features in the local universe. In this paper we use the photometry and morphological information from the first S-PLUS data release (S-PLUS DR1) c…
▽ More
The Southern Photometric Local Universe Survey (S-PLUS) is a novel project that aims to map the Southern Hemisphere using a twelve filter system, comprising five broad-band SDSS-like filters and seven narrow-band filters optimized for important stellar features in the local universe. In this paper we use the photometry and morphological information from the first S-PLUS data release (S-PLUS DR1) cross-matched to unWISE data and spectroscopic redshifts from Sloan Digital Sky Survey DR15. We explore three different machine learning methods (Gaussian Processes with GPz and two Deep Learning models made with TensorFlow) and compare them with the currently used template-fitting method in the S-PLUS DR1 to address whether machine learning methods can take advantage of the twelve filter system for photometric redshift prediction. Using tests for accuracy for both single-point estimates such as the calculation of the scatter, bias, and outlier fraction, and probability distribution functions (PDFs) such as the Probability Integral Transform (PIT), the Continuous Ranked Probability Score (CRPS) and the Odds distribution, we conclude that a deep-learning method using a combination of a Bayesian Neural Network and a Mixture Density Network offers the most accurate photometric redshifts for the current test sample. It achieves single-point photometric redshifts with scatter ($σ_\text{NMAD}$) of 0.023, normalized bias of -0.001, and outlier fraction of 0.64% for galaxies with r-auto magnitudes between 16 and 21.
△ Less
Submitted 1 February, 2022; v1 submitted 26 October, 2021;
originally announced October 2021.
Data Release 2 of S-PLUS: accurate template-fitting based photometry covering $\sim$1000 square degrees in 12 optical filters
Authors:
F. Almeida-Fernandes,
L. Sampedro,
F. R. Herpich,
A. Molino,
C. E. Barbosa,
M. L. Buzzo,
R. A. Overzier,
E. V. R. de Lima,
L. M. I. Nakazono,
G. B. Oliveira Schwarz,
H. D. Perottoni,
G. F. Bolutavicius,
L. A. Gutiérrez-Soto,
T. Santos-Silva,
A. Z. Vitorelli,
A. Werle,
D. D. Whitten,
M. V. Costa Duarte,
C. R. Bom,
P. Coelho,
L. Sodré Jr.,
V. M. Placco,
G. S. M. Teixeira,
J. Alonso-García,
T. C. Beers
, et al. (4 additional authors not shown)
Abstract:
The Southern Photometric Local Universe Survey (S-PLUS) is an ongoing survey of $\sim$9300 deg$^2$ in the southern sky in a 12-band photometric system. This paper presents the second data release (DR2) of S-PLUS, consisting of 514 tiles covering an area of 950 deg$^2$. The data has been fully calibrated using a new photometric calibration technique suitable for the new generation of wide-field mul…
▽ More
The Southern Photometric Local Universe Survey (S-PLUS) is an ongoing survey of $\sim$9300 deg$^2$ in the southern sky in a 12-band photometric system. This paper presents the second data release (DR2) of S-PLUS, consisting of 514 tiles covering an area of 950 deg$^2$. The data has been fully calibrated using a new photometric calibration technique suitable for the new generation of wide-field multi-filter surveys. This technique consists of a $χ^2$ minimisation to fit synthetic stellar templates to already calibrated data from other surveys, eliminating the need for standard stars and reducing the survey duration by $\sim$15\%. We compare the template-predicted and S-PLUS instrumental magnitudes to derive the photometric zero-points (ZPs). We show that these ZPs can be further refined by fitting the stellar templates to the 12 S-PLUS magnitudes, which better constrain the models by adding the narrow-band information. We use the STRIPE82 region to estimate ZP errors, which are $\lesssim10$ mmags for filters J0410, J0430, $g$, J0515, $r$, J0660, $i$, J0861 and $z$; $\lesssim 15$ mmags for filter J0378; and $\lesssim 25$ mmags for filters $u$ and J0395. We describe the complete data flow of the S-PLUS/DR2 from observations to the final catalogues and present a brief characterisation of the data. We show that, for a minimum signal-to-noise threshold of 3, the photometric depths of the DR2 range from 19.9 mag to 21.3 mag (measured in Petrosian apertures), depending on the filter. The S-PLUS DR2 can be accessed from the website: https://splus.cloud}{https://splus.cloud.
△ Less
Submitted 31 March, 2021;
originally announced April 2021.
