# Data for the figures #

## General definitions ##
g, r, z = Tractor magnitudes in bands g, r, z [AB mag]
W1, W2 = Tractor magnitudes in WISE channel 1 and 2 [AB mag]
G = Gaia G magnitude [AB mag]

## Figure 1 ##

-- Fig1.ecsv: G-r distributions
G_r : G_Gaia - r [AB mag]
N_star : Normalized number for SDSS class=STAR
N_gal : Normalized number for SDSS class=GALAXY
N_qso : Normalized number for SDSS class=QSO


## Figure 2 ## 

-- Fig2a_qso_bf.ecsv: G-r versus G bivariate distribution for class=QSO for BGS (grey in left-hand panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig2a_qso_agn.ecsv: G-r versus G bivariate distribution for class=QSO for Cut sample (red in left-hand panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig2b_star_bf.ecsv: G-r versus G bivariate distribution for class=STAR for BGS (grey in middle panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig2b_star_agn.ecsv: G-r versus G bivariate distribution for class=STAR for Cut sample (red in middle panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig2c_gal_bf.ecsv: G-r versus G bivariate distribution for class=GALAXY for BGS (grey in right-hand panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig2c_gal_agn.ecsv: G-r versus G bivariate distribution for class=GALAXY for Cut sample (red in right-hand panel)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin


## Figure 3 ## 

-- Fig3a_qso_bf.ecsv: W1-W2 versus (z-W1)-(g-r) bivariate distribution for class=QSO
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number of BGS BRIGHT|FAINT (BF) in the bin

-- Fig3a_qso_agn.ecsv: W1-W2 versus (z-W1)-(g-r) bivariate distribution for class=QSO
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number of BGS-AGN in the bin

-- Fig3b_star_bf.ecsv: W1-W2 versus (z-W1)-(g-r) bivariate distribution for class=STAR
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number of BGS BRIGHT|FAINT (BF) in the bin

-- Fig3b_star_agn.ecsv: W1-W2 versus (z-W1)-(g-r) scatter points for class=STAR
X: Value of G_Gaia magnitude of the point [AB mag]
Y: Value of G_Gaia-r of the point [AB mag]

-- Fig3c_gal_bf.ecsv: W1-W2 versus (z-W1)-(g-r) bivariate distribution for class=GALAXY
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number of BGS BRIGHT|FAINT (BF) in the bin

-- Fig3c_gal_agn.ecsv: W1-W2 versus (z-W1)-(g-r) scatter points for class=GALAXY
X: Value of G_Gaia magnitude of the point [AB mag]
Y: Value of G_Gaia-r of the point [AB mag]


## Figure 4 ## 

-- Fig4_agn.ecsv: G-r versus G bivariate distribution for BGS-AGN (red in figure)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin

-- Fig4_bf.ecsv: G-r versus G bivariate distribution for BGS-BF (grey in figure)
X: Value of G_Gaia magnitude of the hexbin [AB mag]
Y: Value of G_Gaia-r of the hexbin [AB mag]
Z: Number in the bin


## Figure 5 ## 

Code: https://github.com/desihub/desitarget/blob/0.57.0/bin/run_target_qa
TARG_DIR = "https://data.desi.lbl.gov/public/dr1/target/catalogs/"

Instructions to reproduce the sky map from Fig. 5 based on the above:
> run_target_qa $TARG_DIR/dr9/0.57.0/targets/sv3/resolve/bright bright.fits --prepare
> run_target_qa bright.fits output_dir -w $TARG_DIR/dr9/0.57.0/pixweight/sv3/resolve/bright/sv3pixweight-1-bright.fits


## Figure 6 ## 

-- Fig6.ecsv: Number of each spectral class for pie charts
SPECTYPE: spectral type [QSO, GALAXY, STAR] for the number of objects reported
REDROCK: Nb of objects classified in each spectral type by the Redrock pipeline alone
QN_MBII: Nb of objects classified in each spectral type by the combined pipeline with QuasarNet and MgII
VI: Nb of objects classified in each spectral type by Visual Inspection


## Figure 7 ## 

-- Fig7a_top.ecsv: Composite stacked spectrum of z<0.95 QSO from BGS-AGN
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA

-- Fig7b_bottom.ecsv: Composite stacked spectrum of z>=0.95 QSO from BGS-AGN
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA


## Figure 8 ## 

-- Fig8_missed_BGSAGN.ecsv: Composite stacked spectrum of BGS-AGN targets identified as QSO by visual inspection but missed by the Redrock pipeline
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA

-- Fig8_RR_BGSAGN.ecsv: Composite stacked spectrum of BGS-AGN targets identified as QSO by the Redrock pipeline
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA

-- Fig8_missed_QSO.ecsv: Composite spectrum of QSO targets identified as QSO by visual inspection but missed by the Redrock pipeline (Alexander et al. 2023)
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA

-- Fig8_RR_QSO.ecsv: Composite spectrum of QSO targets identified as QSO by the Redrock pipeline (Alexander et al. 2023)
WAVE: (vacuum) rest-frame wavelength in Angstroms
FLUX: flux density (f_lambda) normalized at 3000 AA


## Figure 9 ## 

-- Fig9_hist.ecsv: Histograms of [OIII]5007 velocity dispersion
LOG_SIG: log10 of the [OIII]5007 line velocity dispersion in km/s (bin center)
N_BGS_BF: Number of BGS BRIGHT|FAINT (BF) in the bin (grey in figure) (bin center)
N_BGS_AGN: Number of BGS AGN in the bin (red in figure)

-- Fig9_kex_BF.ecsv: Bivariate distribution of [OIII]/Hbeta ratio versus log10([OIII] velocity dispersion) for the BGS BRIGHT|FAINT (BF) sample
LOG_SIG: log10 of the [OIII]5007 line velocity dispersion in km/s (bin center)
LOG_OIII_HB: log10 of the [OIII]5007/Hbeta emission-line flux ratio (bin center)
N: Number of BGS BRIGHT|FAINT (BF) in the bin

-- Fig9_kex_AGN.ecsv: Values of [OIII]/Hbeta ratio versus log10([OIII] velocity dispersion) for the BGS-AGN sample, with spectral classification
LOG_SIG: log10 of the [OIII]5007 line velocity dispersion in km/s
LOG_OIII_HB: log10 of the [OIII]5007/Hbeta emission-line flux ratio
CLASS: spectral classification from this set: [QSO, QSO_NL, NLSY1, SY2, BLAZAR, GALAXY]


## Figure 10 ## 

-- Fig10_hist.ecsv: Histograms of max velocity dispersion (MgII, Hbeta, Halpha)
LOG_SIG: log10 of the maximum line velocity dispersion in km/s (bin center)
N_BGS_BF: Number of BGS BRIGHT|FAINT (BF) in the bin (grey in figure) (bin center)
N_BGS_AGN: Number of BGS AGN in the bin (red in figure)

-- Fig10_kex_BF.ecsv: Bivariate distribution of [OIII]/Hbeta ratio versus log10(max velocity dispersion) for the BGS BRIGHT|FAINT (BF) sample
LOG_SIG: log10 of the maximum line velocity dispersion in km/s (bin center)
LOG_OIII_HB: log10 of the [OIII]5007/Hbeta emission-line flux ratio (bin center)
N: Number of BGS BRIGHT|FAINT (BF) in the bin

-- Fig10_kex_AGN.ecsv: Values of [OIII]/Hbeta ratio versus log10(max velocity dispersion) for the BGS-AGN sample, with spectral classification
LOG_SIG: log10 of the maximum line velocity dispersion in km/s
LOG_OIII_HB: log10 of the [OIII]5007/Hbeta emission-line flux ratio
CLASS: spectral classification from this set: [QSO, QSO_NL, NLSY1, SY2, BLAZAR, GALAXY]


## Figure 11 ##

-- Fig11a.ecsv: Redshift distributions from SDSS sample with class=QSO
Z: redshift (center of bin)
N_BGS_BF: Number per sq. degree that satisfy the BGS BRIGHT|FAINT (BF) criteria
N_BGS_AGN: Number per sq. degree that satisfy the BGS AGN criteria
N_QSO: Number per sq. degree that satisfy the QSO targeting criteria (scaled by 1/5)

-- Fig11b.ecsv: Redshift distributions from DESI SV3 sample with class=QSO
Z: redshift (center of bin)
N_BGS_BF: Number per sq. degree that satisfy the BGS BRIGHT|FAINT (BF) criteria (scaled by 1/3)
N_BGS_AGN: Number per sq. degree that satisfy the BGS AGN criteria
N_QSO_BRIGHT: Number per sq. degree that satisfy the QSO targeting criteria (scaled by 1/5)
N_QSO_FAINT: Number per sq. degree that satisfy the QSO targeting criteria (scaled by 1/20)


## Figure 12 ## 

-- Fig12_LOIII_BF.ecsv: Bivariate distribution of [OIII] luminosity versus redshift BGS-BF (grey 2D histogram on left-hand panel)
Z: Redshift (bin center)
LOG_LOIII = log10 of [OIII]5007 luminosity in erg/s (bin center)
N: Number of BGS BIRGHT|FAINT (BF) galaxies in the bin

-- Fig12_Lbol_BF.ecsv: Bivariate distribution of AGN Bolometric luminosity versus redshift BGS-BF (grey 2D histogram on right-hand panel)
Z: Redshift (bin center)
LOG_LBOL = log10 of AGN bolometric luminosity in erg/s (bin center)
N: Number of BGS BIRGHT|FAINT (BF) galaxies in the bin

-- Fig12_LOIII_Lbol_AGN: Values of redshift, [OIII] and Bolometric Luminosities for BGS-AGN (open red circles on both panels)
Z: Redshift
LOG_LOIII = log10 of [OIII]5007 luminosity in erg/s
LOG_LBOL = log10 of AGN bolometric luminosity in erg/s


## Figure 13 ## 

-- Fig13_GALAXY.ecsv: Bivariate distribution of W1-W2 color versus redshift for SV3 galaxies with spectype=GALAXY (blue 2D histogram on top panel)
Z: Redshift (bin center)
W1_W2 = WISE W1-W2 color in AB magnitudes (bin center)
N: Number of SV3 galaxies with spectype=GALAXY in the bin

-- Fig13_QSO.ecsv: Bivariate distribution of W1-W2 color versus redshift for SV3 galaxies with spectype=QSO (orange 2D histogram on bottom panel)
Z: Redshift (bin center)
W1_W2: WISE W1-W2 color in AB magnitudes (bin center)
N: Number of SV3 galaxies with spectype=QSO in the bin

-- Fig13_BGS_AGN.ecsv: Values of W1-W2 color versus redshift for the BGS-AGN sample, with spectral classification (plotting symbols in both panels)
Z: Redshift
W1_W2: WISE W1-W2 color in AB magnitudes
CLASS: spectral classification from this set: [QSO, NLSY1, SY2, BLAZAR, GALAXY, STAR]


## Figures 14-17 ## 

DESI EDR spectrum (SPARCL), LS DR9 image (Viewer), LS DR10 or HSC DR2 as labeled (Viewer)

For images, need to input the RA and Dec in degrees for the given TARGETID of interest (ra=210.014290 and dec=5.727356 in the examples below). Then the layer name will specify if the color cutout image is made from LS DR9, LS DR10 or HSC DR2:

1) LS DR9 layer
https://legacysurvey.org/viewer/jpeg-cutout/?ra=210.014290&dec=5.727356&layer=ls-dr9&pixscale=0.262&size=38

2) LS DR10 layer
https://legacysurvey.org/viewer/jpeg-cutout/?ra=210.014290&dec=5.727356&layer=ls-dr10&pixscale=0.262&size=38

3) HSC DR2 layer
https://legacysurvey.org/viewer/jpeg-cutout/?ra=210.014290&dec=5.727356&layer=hsc_dr2&pixscale=0.25&size=40

DESI spectra are uniquely defined by TARGETID, SURVEY, PROGRAM. For this paper, SURVEY=='sv3' and PROGRAM can be 'bright' (default) or 'dark' (when available). This Python code can be used either on the Astro Data Lab platform or in a local Python environment after installing SPARCL (pip install sparclclient; with more info here: https://astrosparcl.datalab.noirlab.edu)

# Imports
import numpy as np
from sparcl.client import SparclClient

# Instantiate SPARCL client
client = SparclClient()

# Function to return the spectrum (requires TARGETID, SURVEY, PROGRAM, Z as input and 
# by default converts to rest-frame; can force observed-frame by setting rest_frame=False)
def Get_DESI_Spectra(targetid, survey, program, z, rest_frame=True):

    # Make sure Python int() format for compatibility with SPARCL
    tid = int(targetid)
    
    ## Retrieve Spectra
    inc = ['specid', 'redshift', 'flux', 'wavelength', 'ivar', 'spectype', 'survey', 'program', 'targetid', 'coadd_fiberstatus']
    res = client.retrieve_by_specid(specid_list = [tid],
                                include = inc,
                                dataset_list = ['DESI-EDR'])

    # Records returned from SPARCL for a given input ID
    records = res.records
    Nrec = res.count         # number of records were returned
    
    surveys = np.array([records[jj].survey for jj in range(Nrec)])
    programs = np.array([records[jj].program for jj in range(Nrec)])

    print('Surveys', surveys)
    print('Programs', programs)

    # Select the record with the matching survey & program from input
    select_ii = int(np.arange(Nrec)[(surveys==survey)&(programs==program)])
    
    # Extracting the spectral quantities 
    wavelength = records[select_ii].wavelength
    flux = records[select_ii].flux
    ivar = records[select_ii].ivar        # Inverse Variance
    
    if (rest_frame == True):
        # If rest_frame = True, we convert the different arrays into their rest-frame values
        wavelength = wavelength/(1+z)
        flux = flux*(1+z)
        ivar = ivar/((1+z)**2)
    
    return (wavelength, flux, ivar)

 
## Figure 18 ## 

-- Fig18.ecsv: Maximum emission-line velocity dispersion in log10(km/s) as a function of redshift
Z: redshift
LOG_SIG_B: log10 of the maximum line width for the Balmer lines (Hbeta, Halpha; top panel)
LOG_SIG_MAX: log10 of the maximum line width for these lines: MgII, Hbeta, Halpha (bottom panel)
CLASS: spectral classification from this set: [QSO, QSO_NL, NLSY1, SY2, BLAZAR, GALAXY]


## Figure 19-24 ## 

DESI EDR spectrum (SPARCL), LS DR9 image (Viewer), LS DR10 or HSC DR2 as labeled (Viewer)
Same instructions as for Figures 14-17.


## Figure 25 ## 
DESI EDR spectrum (SPARCL), LS DR9 image (Viewer), Radio from VLASS (Viewer)
Same instructions as for Figures 14-17 and Figures 19-24 except that the second images is cutout from the radio VLASS layer as follows:

VLASS (layer=vlass1.2): 
https://legacysurvey.org/viewer/jpeg-cutout/?ra=210.014290&dec=5.727356&layer=vlass1.2&pixscale=1&size=10


## Figure 26-31 ## 
## file: 
## notebook:  
DESI EDR spectrum (SPARCL), LS DR9 image (Viewer), LS DR10 or HSC DR2 as labeled (Viewer)
Same instructions as for Figures 14-17 and Figures 19-24.