J-PLUS: Searching for very metal-poor star candidates using the SPEEM pipeline

We explore the stellar content of Javalambre Photometric Local Universe Survey (J-PLUS) Data Release 2 and show its potential to identify low-metallicity stars using Stellar Parameters Estimation based on Ensemble Methods (SPEEM) pipeline. SPEEM is a tool provide determinations atmospheric parameters for separate sources from quasars, unique J-PLUS photometric system. The adoption adequate selection criteria allows identification metal-poor star candidates suitable spectroscopic follow-up. consists series machine learning models which uses training sample observed by both SEGUE survey. has temperatures Teff between 4\,800 K 9\,000 K; $\log g$ 1.0 4.5, $-3.1<[Fe/H]<+0.5$. performance pipeline been tested with LAMOST survey within same parameter range. average differences J-PLUS, were obtained Parameter Pipeline SPEEM, respectively, are $\Delta Teff\sim 41$ K, \log g\sim 0.11$ dex, [Fe/H]\sim 0.09$ dex. A 177 have identified as new $[Fe/H]<-2.5$ 11 them ISIS spectrograph at William Herschel Telescope. analysis confirms that $64\%$ $[Fe/H]<-2.5$, including one $[Fe/H]<-3.0$. in combination filter system shown estimate (Teff, g$, [Fe/H]). validation shows yields success rate very $[Fe/H]<-2.5$.