How to measure metallicity from five-band photometry with supervised machine learning algorithms
arXiv:1510.08076 · doi:10.1093/mnras/stv2703
Abstract
We demonstrate that it is possible to measure metallicity from the SDSS five-band photometry to better than 0.1 dex using supervised machine learning algorithms. Using spectroscopic estimates of metallicity as ground truth, we build, optimize and train several estimators to predict metallicity. We use the observed photometry, as well as derived quantities such as stellar mass and photometric redshift, as features, and we build two sample data sets at median redshifts of 0.103 and 0.218 and median r-band magnitude of 17.5 and 18.3 respectively. We find that ensemble methods, such as Random Forests of Trees and Extremely Randomized Trees, and Support Vector Machines all perform comparably well and can measure metallicity with a Root Mean Square Error (RMSE) of 0.081 and 0.090 for the two data sets when all objects are included. The fraction of outliers (objects for which |Z_true - Z_pred| > 0.2 dex) is 2.2 and 3.9%, respectively and the RMSE decreases to 0.068 and 0.069 if those objects are excluded. Because of the ability of these algorithms to capture complex relationships between data and target, our technique performs better than previously proposed methods that sought to fit metallicity using an analytic fitting formula, and has 3x more constraining power than SED fitting-based methods. Additionally, this method is extremely forgiving of contamination in the training set, and can be used with very satisfactory results for training sample sizes of just a few hundred objects. We distribute all the routines to reproduce our results and apply them to other data sets.
arXiv commentMinor revisions, matching version published in MNRAS
References in corpus (16)
- Scikit-learn: Machine Learning in Python
- Maps of Dust IR Emission for Use in Estimation of Reddening and CMBR Foregrounds
- The Origin of the Mass--Metallicity Relation: Insights from 53,000 Star-Forming Galaxies in the SDSS
- CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey
- A fundamental relation between mass, SFR and metallicity in local and high redshift galaxies
- LSST Science Book, Version 2.0
- The 3D-HST Survey: Hubble Space Telescope WFC3/G141 grism spectra, redshifts, and emission line measurements for galaxies
- An Analytic Model for the Evolution of the Stellar, Gas, and Metal Content of Galaxies
- A fundamental plane for field star-forming galaxies
- Relative merits of different types of rest-frame optical observations to constrain galaxy physical parameters
- SED fitting with Markov Chain Monte Carlo: Methodology and Application to z=3.1 Lyman Alpha Emitting Galaxies
- The rise and fall of the star formation histories of blue galaxies at redshifts 0.2<z<1.4
- The Relationship between Stellar Mass, Gas Metallicity, and Star Formation Rate for Halpha-selected Galaxies at z~0.8 from the NewHalpha Survey
- Star formation histories from multi-band photometry: A new approach
- Using Colors to Improve Photometric Metallicity Estimates for Galaxies
- SED fitting with MCMC: methodology and application to large galaxy surveys
Cited by in corpus (8)
- Separation of pulsar signals from noise with supervised machine learning algorithms
- A machine learning approach to galaxy properties: joint redshift-stellar mass probability distributions with Random Forest
- Using convolutional neural networks to predict galaxy metallicity from three-color images
- Panchromatic SED modelling of spatially-resolved galaxies
- COSMOS2020: Manifold Learning to Estimate Physical Parameters in Large Galaxy Surveys
- Constraining stellar population parameters from narrow band photometric surveys using convolutional neural networks
- Euclid preparation. LXVIII. Extracting physical parameters from galaxies with machine learning
- Interpreting deep learning-based stellar mass estimation via causal analysis and mutual information decomposition