Title: Estimating Stellar Atmospheric Parameters and [α/Fe] for LAMOST O-M type Stars Using a Spectral Emulator Authors: Liang et al. Content: (1) Description of file: dr10_LRS_OBAFGKM.fits; (2) Description of data cross-matching (using TOPCAT Version 4.9) and external comparison =========================================================================================================================================== (1) Description of file: dr10_LRS_OBAFGKM.fits ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Field Class Description ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- obsid Integer The number ID of this spectrum. uid String Unique source identifier. ra_obs Double Right ascension from the LAMOST DR10 catalog (J2000). dec_obs Double Declination from the LAMOST DR10 catalog (J2000). subclass String The detailed spectral type for each star. snr_u Float S/N of u band. snr_g Float S/N of g band. snr_r Float S/N of r band. snr_i Float S/N of i band. snr_z Float S/N of z band. z Double Redshift from LAMOST fits header. z_err Double Uncertainty in redshift. Teff (K) Short Effective temperature, and the value is retained as the truncated integer. Teff_err (K) Float Uncertainty in Teff, and the value is retained as truncated to two decimal places. log g (dex) Float Surface gravity, and the value is retained as truncated to three decimal places. log g_err (dex) Float Uncertainty in log g, and the value is retained as truncated to four decimal places. [Fe/H] (dex) Float Metallicity, and the value is retained as truncated to three decimal places. [Fe/H]_err (dex) Float Uncertainty in [Fe/H], and the value is retained as truncated to four decimal places. [α/Fe] (dex) Float The abundance of alpha elements with respect to iron, and the value is retained as truncated to three decimal places. [α/Fe]_err (dex) Float Uncertainty in [α/Fe], and the value is retained as truncated to four decimal places. log g_Imig (dex) Float Replace median log g with Imig et al. (2022); see Section 4.7. The value is retained as truncated to three decimal places. χ2 Float The difference between the observed spectrum and the generated spectrum. sum_ormask Integer Spectral quality flag, 0 means good quality; other values indicate issues. FLAG_S/N Short S/N flags for different types of stars, 0 for recommended, and 1 is not recommended. FLAG_χ2 Short χ2 flags for different types of stars, 0 for recommended, and 1 is not recommended. FLAG_[α/Fe] Short [α/Fe] flags for different types of stars, and 0 for recommended, 1 is not recommended. gaia_id Long Gaia DR3 Source ID obtained by TOPCAT (Taylor et al. (2005, 2024)). Only Source IDs with available Geometric distances are provided. r_geo Double Geometric distance derived from Gaia parallax in Bailer-Jones et al. (2021). G (mag) Double G-band magnitudes from Gaia DR3. BP (mag) Double BP-band magnitudes from Gaia DR3. RP (mag) Double RP-band magnitudes from Gaia DR3. G_err (mag) Double Uncertainty in G-band magnitudes from Gaia DR3. BP_err (mag) Double Uncertainty in BP-band magnitudes from Gaia DR3. RP_err (mag) Double Uncertainty in RP-band magnitudes from Gaia DR3. r_geo_16th Double Distance at 16th percentile of the probability distribution. r_geo_84th Double Distance at 84th percentile of the probability distribution. ebv_bayes19 Double E(B-V) interpolated from the 3D map of Green et al. (2019) using r_geo. ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Notes: [1] In this paper, spectral types follow LAMOST's "subclass" field: M dwarfs and giants correspond to dM and gM, respectively. Other stars are classified as giants if log g < 3.8 dex and as dwarfs if log g ≥ 3.8 dex. [2] The uncertainty represents the internal error in the predicted values of stellar parameters. [3] The ormask is a field in LAMOST fits header, which represents the spectral quality of each wavelength. A value of 0 signifies high spectral quality, while a non-zero value suggests otherwise Luo et al. (2015). The sum_ormask is the sum of ormask at each wavelength, and the value of 0 and non-zero have the same meanings as the ormask. ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- =========================================================================================================================================== (2) Description of data cross-matching (using TOPCAT Version 4.9) and external comparison ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Type Match Tables Recommended parameter range ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Our catalog cross-matches with APOGEE (STARFLAG==0) & (ASPCAPFLAG==0) M-ype Algorithm is "Sky", Max Error is "1 arcsec" (300020 FGK-type Algorithm is "Exact Value", Match Value column is "obsid" (Internal_Teff_err<200) & (Internal_log g_err<0.2) & (Internal_[Fe/H]_err<0.1) Match Section is "all matches", Join Type is "1 and 2" ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Our catalog cross-matches with LASP snr_g>20 A-type Algorithm is "Exact Value", Match Value column is "obsid" (Internal_Teff_err<500) & (Internal_log g_err<0.2) & (Internal_[Fe/H]_err<0.1) Match Section is "all matches", Join Type is "1 and 2" ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Our catalog cross-matches with HotPayne (chi2ratio<10) & (snr_g>30) OBA-type Algorithm is "Exact Value", Match Value column is "obsid" (or SpecID) A-type: (Internal_Teff_err<500) & (Internal_log g_err<0.2) & (Internal_[Fe/H]_err<0.4) Match Section is "all matches", Join Type is "1 and 2" OB-type: (Internal_Teff_err<1000) & (Internal_log g_err<0.2) & (Internal_[Fe/H]_err<0.4) (750030) OCs Algorithm is "Exact Value", Match Value column is "obsid" (Internal_Teff_err<200) & (Internal_log g_err<0.2) & (Internal_[Fe/H]_err<0.1) Match Section is "all matches", Join Type is "1 and 2" ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Notes: [1] Given that most external catalogs lack recommended parameter fields and contain extrapolated values, we have set the ranges of parameter and internal error to ensure fairness in external comparisons. [2] Considering the heterogeneity between catalogs, we visualized all the matched data and used 3σ-clipped residuals to calculate the mean and standard deviation, as outliers can bias these estimates and obscure systematic differences. We do not recommend Gaussian profile fitting of residuals due to its sensitivity to initial values. [3] Considering our catalog's homogeneity, we do not use 3σ-clipped residuals for mean and standard deviation calculations in internal statistics, as outliers help identify potential error sources and assess method stability. [4] In full-spectrum fitting, models predicting spectra from stellar parameters are particularly sensitive to flux outliers due to the χ2 function. Therefore, when analyzing OCs, setting sum_ormask==0 is useful for assessing differences between our catalog and LASP in parameter predictions for normal spectra. ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------