J-VAR: the northern variable sky in 7 filters -- First Data Release

TL;DR

The J-VAR survey provides a comprehensive, multi-epoch, narrow-band optical variability dataset over a wide field, enabling detailed studies of astrophysical phenomena with unprecedented data volume and quality.

Contribution

This is the first survey combining wide-field imaging with multiple narrow band filters for variability studies, producing extensive light curves and transient detections.

Findings

Over 1.3 million light curves of point sources.

Detection of more than 6000 asteroids and 10 optical transients.

Discovery of 4 supernovae.

Abstract

Aims. The analysis of variability of astronomical sources is of extraordinary interest, as it allows the study of astrophysical phenomena in real time. This paper presents the Javalambre Variability Survey (J-VAR) which leverages the narrow band filters available at the Javalambre Auxiliary Survey Telescope (JAST80) at the Observatorio Astrof\'isico de Javalambre (OAJ). Methods. The JAST80 equipped with T80Cam, providing a field of view of 2\,square degrees and a pixel scale of 0.55\,arcsec/pixel has been designed for wide-field studies. The main characteristic is the availability of a variety of narrow band filters strategically located on stellar spectral features (the $J0395$ in correspondence of the Ca H+K doublet, the $J0515$ of the Mg $b$ triplet, the $J0660$ of the H$\alpha$ line, and the $J0861$ of the Ca~triplet). This project combines, for the first time, the wide-field with a variety of narrow band filters for a unique variability survey, observing each field 11~times with a standardised observing sequence. The median limiting magnitude for individual exposures are 19.1 mag in $J0395$ and $J0515$, $19.6$ mag in $J0660$ and $J0861$, $19.8$ mag in $i$, and $20.2$ in $g$ and $r$. The typical FWHM of the $r$-band images is $1.5$ arcsec. Results. This article introduces the first data release of J-VAR including more than 6000 individual asteroids, 10\,detected optical transients (4\,discovered supernovae), and 1.3 million light curves of point-sources. On average, J-VAR delivers an unprecedented $\sim$5000 light curves per square degree of 11\,epochs in 7\,bands, opening research opportunities for theoretical studies and new discoveries alike.

Figures

Tables

Table 1. Table 1: Filter names, limiting magnitudes used to establish the exposure times and the corresponding exposures times, calculated using the exposure time calculator for bright nights. The ”1exp” and ”3exp” refer to the magnitude limits for an individual image and three combined images respectively. Mean and standard deviation estimated from the individual catalogues. J-VAR values are compared with those of J-PLUS DR1 (Cenarro et al., 2019 ) from coadded images. All values referred to S ​ N ​ R > 5 SNR>5 within a circular aperture of 3 arcsec.

Filter	$m_{\text{lim}}^{\text{1exp}}$	$m_{\text{lim}}^{\text{3exp}}$	$t_{\text{exp}}$	$⟨m_{\text{lim}}^{\text{J-VAR}}⟩$	${\sigma }_{m_{\text{lim}}}^{}$	$m_{\text{lim}}^{\text{J-PLUS}}$
$J0395$	20.2	20.5	87	19.1	0.5	20.5
$g$	21.2	21.5	33	20.3	1.6	21.5
$J0515$	20.2	20.5	40	19.1	0.6	20.7
$r$	21.2	21.5	40	20.2	0.7	21.5
$J0660$	20.8	21.0	135	19.6	0.7	20.7
$i$	21.3	21.5	34	19.8	0.7	21.2
$J0861$	20.9	21.0	160	19.6	0.6	20.0

Table 2. Table 2: Number of most occurring stellar objects (i.e. does not consider AGNs) in Objects table based on Gaia DR3 variability classes.

object	number
SOLAR_LIKE	Solar-like variability	4176
ECL	Eclipsing binaries	1165
RS	RS Canum Venaticorum	1033
DSCT—GDOR—SXPHE	$\delta$ Scuti / $\gamma$ Doradus/ SX Phoenicis stars	901
S	Short-timescale	494
LPV	Long-period variables	286
RR	RR Lyrae stars	315

Table 3. Table 3: List of pointings of J-VAR DR1. The first three columns contain the identifier of the field, the right ascension and the declination (in degrees). The fourth and fifth column refer to the time of the first and last observation of the field in “JVN” (in modified Julian date). The sixth column is “True” if a field has high frequency observations and “False” otherwise. If “True”, the last two columns show the time of the first and last observation of this sequence (in modified Julian date).

field	ra	dec	earliest MJD	latest MJD	has HF	earliest MJD HF	latest MJD HF
JVAR00056	187.0911	20.6962	59621.042367	59722.940706	True	59717.875677	59717.973275
JVAR00316	256.5094	23.481	58230.099057	58666.92478	True	60031.176817	60062.036111
JVAR00317	258.0164	23.481	58230.129392	58749.835845	True	59783.879983	59783.989491
JVAR00421	256.5934	24.873	58236.111892	58756.823368	True	59014.015631	59014.111672
JVAR00540	119.8148	27.6581	58901.886291	59286.931817	True	59891.168177	59891.228559
JVAR00596	207.1414	27.658	58143.139323	58497.196262	True	59599.209716	60053.125885
JVAR00674	168.9654	29.05	58555.105307	58636.922118	True	59595.132459	59595.23022
JVAR00739	112.2042	30.4428	58143.049844	58845.957986	True	60039.84684	60039.933617
JVAR00746	123.4108	30.443	59165.156962	59249.990613	False
JVAR00749	128.2137	30.443	59173.095434	59287.906134	True	60041.87272	60041.959531
JVAR00752	133.0154	30.4428	58828.058027	58921.895787	False
JVAR00753	134.6166	30.443	58828.08445	58898.876273	True	60044.872106	60044.958848
JVAR00757	141.0197	30.4428	58494.058281	58772.191424	True	58865.123623	59944.187581
JVAR00762	149.0241	30.4428	58851.03349	59262.941447	True	59290.888466	59291.023669
JVAR00764	152.2257	30.4428	58846.194265	59630.087222	True	60023.924977	60024.011939
JVAR00840	273.8913	30.443	58242.04195	58666.95309	True	59718.086314	59718.144126
JVAR00841	275.4923	30.443	58242.072344	58725.966076	False
JVAR00842	277.0928	30.4428	58242.101175	59366.047824	True	59013.054983	59013.126389
JVAR00844	280.2944	30.443	58242.131615	58702.004595	True	59712.105955	59867.860949
JVAR00846	283.4963	30.443	58362.878108	59366.077488	False
JVAR00860	126.9228	31.8352	59611.893223	59722.902396	True	60045.857338	60045.949693
JVAR00863	131.7944	31.835	58080.076453	58624.923773	False
JVAR00864	133.4184	31.8352	58850.943513	59279.060648	True	60042.854398	60042.941152
JVAR00867	138.2901	31.8352	59620.951696	59698.866829	True	60058.836285	60058.923073
JVAR00877	154.5291	31.8352	58846.135226	59279.092847	True	59948.148119	59948.24581
JVAR00909	206.4934	31.835	58550.174149	58608.918241	True	59717.976047	59718.074954
JVAR00941	258.4574	31.835	58240.052888	58922.1689	False
JVAR01066	283.8711	33.2276	59729.008397	59843.875579	True	59782.970017	59783.067396
JVAR01276	279.5964	36.0123	59677.136765	59749.032245	False
JVAR01280	105.8673	37.4047	59499.128524	59591.896678	True	59506.110318	59506.207998
JVAR01477	270.1453	38.7971	59679.08849	59731.987593	True	60065.116447	60065.159149
JVAR01486	105.901	40.1895	58208.858142	59165.065313	True	58894.797043	58894.919722
JVAR01489	111.3042	40.1895	58081.986823	59205.105243	True	59563.912992	59564.016736
JVAR01514	156.3319	40.1895	58210.936719	59200.197072	True	58895.041302	58895.209491
JVAR01515	158.1334	40.19	58210.972899	58621.913657	True	60028.916563	60029.005064
JVAR01516	159.9342	40.19	58211.001395	59975.087396	True	59949.234635	59951.264664
JVAR01522	170.7408	40.1895	58851.122494	59015.944363	True	58857.121198	58857.254595
JVAR01527	179.7463	40.1895	58494.09673	58636.952095	True	58530.020966	58564.203843
JVAR01561	240.9837	40.1895	59616.177425	59719.933704	False
JVAR01578	271.6026	40.1895	59679.118351	59878.793079	False
JVAR01724	181.0813	42.9742	58494.133027	58621.972535	True	58531.015492	58531.161991
JVAR01725	182.9599	42.9742	58851.150584	59204.196285	True	58859.962633	58860.113854
JVAR01734	199.8667	42.9742	58494.163108	58603.930231	True	58532.013119	58532.218912
JVAR01755	239.316	42.9742	58494.197969	58626.053727	True	58533.04305	58533.165486
JVAR01756	241.1945	42.9742	58862.221684	59021.917083	True	59007.029948	59741.015729
JVAR01777	280.6438	42.9742	59323.0982	59410.138356	True	59391.934057	59829.841557
JVAR01778	282.5227	42.9742	59024.96695	59407.953056	True	59454.929647	59454.954954
JVAR01871	280.8233	44.3666	59369.080862	59419.91015	True	59500.820399	59745.025885
JVAR01872	282.7449	44.3666	59025.919554	59375.016782	True	59392.03548	59393.038189
JVAR01873	284.6663	44.3666	59396.104439	59435.910104	True	59505.842946	59836.960671
JVAR02023	223.025	47.1514	58851.206811	58999.920648	True	58898.113929	58898.233137
JVAR02024	225.0426	47.1514	58862.164057	59016.014722	True	59299.116117	59393.007199
JVAR02140	284.1486	48.5437	59396.076175	59435.939329	True	59838.859693	59838.957269
JVAR02320	133.2358	52.7209	58143.167841	59957.139132	False
JVAR02322	137.7534	52.7209	58239.93489	59937.236933	True	59513.129902	59513.166916
JVAR02355	212.2952	52.7209	57902.106117	59996.214722	True	59014.936638	59714.996065
JVAR02356	214.5538	52.7209	57904.012575	59996.125741	True	60040.028449	60040.116198
JVAR02357	216.8127	52.7209	57906.008443	58324.933866	True	58893.168247	59013.050694
JVAR02359	221.3304	52.721	57942.986626	58615.94125	True	60033.070903	60033.160492
JVAR02360	223.5894	52.721	57943.019242	58670.926632	False
JVAR02402	138.8162	54.1133	58148.110492	59165.124016	True	60015.986632	60016.092407
JVAR02434	213.4442	54.1132	57900.016406	58622.032882	True	58887.096383	58887.193924
JVAR02435	215.7763	54.113	57900.041499	58619.108843	True	59390.876742	59390.991123
JVAR02436	218.1083	54.1131	57900.066534	58619.137523	True	59605.222841	59743.976597
JVAR02437	220.4404	54.113	57900.092645	59022.989977	False
JVAR02438	222.7724	54.1133	57900.117703	58637.084606	True	59301.019659	59301.117292
JVAR02439	225.1047	54.1133	57902.07967	59285.082998	True	59301.121047	60044.129439
JVAR02503	197.8546	55.5056	58851.178721	59285.017928	True	58860.117367	58885.142645
JVAR02508	209.9134	55.506	57909.911186	58294.96794	True	58653.909635	59712.975307
JVAR02509	212.3253	55.506	57909.940584	58294.998021	True	58650.077506	59713.98864
JVAR02510	214.737	55.5059	57909.968374	58563.150764	True	60040.973322	60061.032876
JVAR02511	217.1483	55.506	57901.970365	58622.061574	True	58653.98349	60045.141771
JVAR02512	219.5603	55.506	57899.965203	58643.934826	True	58648.937598	58649.1264
JVAR02513	221.9724	55.506	57899.990758	58658.920984	True	60018.152106	60018.208281
JVAR02514	224.3844	55.506	57902.008212	59996.154942	False
JVAR02515	226.7964	55.5056	57902.029832	59999.172477	False
JVAR02520	238.8553	55.506	57944.99886	58722.964074	True	60041.065278	60041.152807
JVAR02622	132.2325	58.2904	58093.187407	58772.135498	True	59599.108918	59599.206516
JVAR02840	188.7493	62.468	58235.923594	58622.002789	True	58654.028686	58896.229792
JVAR03384	3.4766	-2.3038	58702.096348	59169.970197	True	59455.104855	59511.892095
JVAR03420	10.4249	-0.9114	58725.137147	59199.865255	True	59513.966649	59514.064178
JVAR03479	6.2563	1.8733	59801.101869	59975.795475	True	59956.764323	59956.787407
JVAR03481	9.0367	1.8733	59146.850145	59498.940475	True	59510.902008	59510.999792
JVAR03488	18.7682	1.8733	58746.063756	59884.933854	True	58856.761997	59839.191944
JVAR03489	20.1584	1.8733	59146.879369	59569.797998	True	59949.743015	59949.840532
JVAR03490	21.5487	1.873	58746.01511	59884.90647	False
JVAR03504	41.0114	1.8733	59821.082737	59956.822813	True	59909.982355	59948.981574
JVAR03539	3.4834	4.658	58079.75901	59818.136863	True	59563.765041	59563.869051
JVAR03557	28.5617	4.6581	59146.907506	59499.029977	True	59880.931591	59881.035104
JVAR03566	41.101	4.6581	59467.165828	59884.046771	True	59501.061256	59513.022083
JVAR03589	30.0145	6.0507	58750.074531	58859.950116	True	59597.767888	59597.86566
JVAR03590	31.4104	6.0505	58764.105064	59174.985023	True	59512.037737	59512.159826
JVAR03615	24.4934	7.4428	58079.850666	58781.08191	True	59183.917367	59183.993744
JVAR03624	37.0898	7.4428	58781.113571	59611.801516	True	59593.804068	59867.200012
JVAR03653	332.1144	10.228	58337.082216	58751.990266	True	59455.004392	59455.102025
JVAR03660	341.9802	10.2278	58671.048929	58842.826134	True	59183.792251	59839.090856
JVAR03661	343.3896	10.228	58671.077286	58841.819062	True	59510.78151	59510.87934
JVAR03705	343.4482	11.62	59163.771522	59909.78081	True	59512.817876	59512.915417
JVAR04246	35.1077	28.328	59146.936858	59870.033356	True	59502.135642	59503.083854
JVAR04376	34.6054	32.506	58079.925098	58769.099699	True	59882.014959	59882.11853
JVAR04377	36.2407	32.5056	58769.103443	59977.804282	True	59596.816973	59596.916956

Column Name	Data Type	Unit	Description
obj_id	object	None	The J-PLUS object ID, with format {REF_TILE}-{NUMBER}; used for internal cross-matching with OBJ_ID in the Photometry Tables
ra	float64	deg	The RA of the object in the J-PLUS reference tile
dec	float64	deg	The DEC of the object in the J-PLUS reference tile
mean_ra	float64	deg	the mean RA position of all object detections in the J-VAR images
rms_ra	float64	deg	the RMS (actually NMAD) of MEAN_RA
mean_dec	float64	deg	the mean RA position of all object detections in the J-VAR images
rms_dec	float64	deg	the RMS (actually NMAD) of MEAN_DEC
mag_auto	object	mag	array of values of the J-PLUS photometry MAG_AUTO
mag_auto_err	object	None	array of values of the ERRORS of J-PLUS photometry MAG_AUTO
mag_aper_6	object	mag	array of values of the J-PLUS photometry MAG_APER_6 (6 arcsec aperture with aperture correction factor applied)
mag_aper_6_err	object	None	array of values of the ERRORS of J-PLUS photometry MAG_APER_6
mag_aper_3	object	mag	array of values of the J-PLUS photometry MAG_APER_3 (3 arcsec aperture with aperture correction factor applied)
mag_aper_3_err	object	None	array of values of the ERRORS of J-PLUS photometry MAG_APER_3
flags	object	None	array of Sextractor flags of the J-PLUS photometry
mask_flags	object	None	array of the flags obtained after applying masks on J-PLUS tiles during UPAD processing
single_det	object	None	array indicating whether an object is detected on the J-PLUS SingleMode Sextractor catalogue (0=no, 1=yes)
cls	float32	None	ClassStar assigned by Sextractor during photometry on the J-PLUS tile
mps	float32	None	Probability of being a star using the morphological method to classify the object (López-Sanjuan et al. 2019b)
sgc	float32	None	Probability of being a star by combining the available priors and classifications (López-Sanjuan et al. 2019b)
bannjos	object	None	Object classification via the BANNJOS algorithm (del Pino et al. in prep); classes are ’STAR’,’QSO’,’APS’ (= Ambiguous Point Source)
fwhm	float32	None	Full Width at Half Maximum of the object on the J-PLUS reference (rSDSS) tile
morph_flag	int32	None	Additional Flag (0 or 1) to indicate potential issues with an object’s morphology. A value of ”1” is obtained when CLS ¡ 0.3 or MPS ¡ 0.3 or FWHM ¿ 3 and it might be worth checking the object out
snr	object	None	Signal-To-Noise ratio of the J-PLUS photometry
hpix11	int32	None	Healpix index ORDER 11 (NESTED schema) of object position
ebv	float32	None	E(B-V) colour excess estimated from the Schlafly & Finkbeiner (2011) recalibration of the Schlegel, Finkbeiner & Davis (1998) dust map at infinity
ebv_err	float32	None	Error in the above E(B-V) value
vsx_var	int32	None	Flag to indicate whether or not the object is a known variable included in the VSX catalogue; 0=no, 1=yes
vsx_type	object	None	The VSX variable type classification of known variables; value is ’N/A’ for all other objects
vsx_period	float64	None	The period (if available, otherwise period=0) of known variables; value is also 0 for all other objects
gaia_sid	int64	None	The Gaia DR3 source_id of the object; crossmatch radius is 1.1 arcsec, ID=0 if no match is found
gaia_xmulti	int32	None	Flag indicating the presence of multiple objects within the 1.1 arcsec crossmatch radius. Values are 0 if no, 1 if yes, -1 for unmatched objects. In case of multiple Gaia objects, the source_id assigned is the one corresponding to the smallest angular separation from the J-VAR object
gaia_prlx	float64	None	Gaia DR3 object parallax in mas; ’nan’ if no parallax is available
gaia_prlx_err	float64	None	Gaia DR3 object parallax error in mas; ’nan’ if no parallax is available
gaia_gmag	float64	None	The mean Gaia DR3 G magnitude (phot_g_mean_mag in gaia_source)
gaia_var	int32	None	Flag to indicate whether or not the object is classified as variable (phot_variable_flag in gaia_source). Values are 0 if no (GAIA=NOT_AVAILABLE), 1 if yes (GAIA=VARIABLE), -1 for unmatched objects
gaia_class	object	None	The Gaia classification of the variable (best_class_name in vari_classifier_result). There are Gaia objects flagged as ’VARIABLE’ but without classification (in_vari_classification_result == False in vari_summary). These objects are assigned a ’NO_CLASS’ value, while unmatched objects are assigned an ’N/A’ value
gaia_score	float32	None	The score of the Gaia classification algorithm (best_class_score in vari_classifier_result). ’VARIABLE’ objects with ’NO_CLASS’ get a score=0, unmatched objects get a score=-1

Column Name	Data Type	Unit	Description
obj_id	object	None	The J-PLUS object ID, with format REF_TILE-NUMBER
ra	float64	deg	The RA of the object in the J-PLUS reference tile
dec	float64	deg	The DEC of the object in the J-PLUS reference tile
hpix11	int32	None	Healpix index ORDER 11 (NESTED schema) of object position
field	int32	None	Number of the J-VAR field
filter	object	None	Name of the filter
mag	object	mag	The array containing the magnitudes
mag_err	object	None	The array containing the magnitude errors.
flags	object	None	array of Sextractor flags of the J-PLUS photometry; follows FORDER
fwhm	object	None	The array containing the FWHM value of the object in each image.
corr_curves	object	None	The array containing the curves used to calibrate each star’s light curve
mjd	object	None	Corresponding MJD(UTC) value of the UTC timestamps.

Column Name	Data Type	Unit	Description
field	int32	None	Number of the J-VAR field
filter	object	None	Name of the filter
utc	object	None	List with the dates and times of each of the images used for each measure of the object in a a field and filter.
mjd	object	None	Corresponding MJD(UTC) value of the UTC timestamps.
imtype	object	None	a flag to keep track of the different images.

Column Name	Data Type	Unit	Description
field	int32	None	Number of the J-VAR field
filter	object	None	Name of the filter
obj_id	object	None	The J-PLUS object ID, with format REF_TILE-NUMBER
cnum	int32	None	The number of comparison stars used to correct the object’s light curve
srad	float32	None	Search radius to encounter the above comparison stars
mag	object	mag	The array containing the magnitudes
mag_err	object	None	The array containing the magnitude errors.
flags	object	None	The array containing the photometric flags.
fwhm	object	None	The array containing the FWHM value of the object in each image.
corr_curves	object	None	The array containing the curves used to calibrate each star’s light curve

Column Name	Data Type	Unit	Description
source_number	int32	None	None
catalog_number	int32	None	None
ra	float64	deg	None
dec	float64	deg	None
epoch	float32	None	None
mag	float32	None	None
magerr	float32	None	None
mag_aper	float32	None	None
mag_aper_err	float32	None	None
flux_aper	float32	None	None
flux_aper_err	float32	None	None
flux	float32	None	None
fluxerr	float32	None	None
pm	float32	None	None
pmerr	float32	None	None
pmra	float32	None	None
pmra_err	float32	None	None
pmdec	float32	None	None
pmdec_err	float32	None	None
mid_exposure_mjd	float32	None	None
date_obs	object	None	None
exptime	float32	None	None
matched	bool	None	None
skybot_number	float32	None	None
skybot_name	object	None	None
skybot_class	object	None	None
skybot_mag	float32	None	None
skybot_ra	float32	None	None
skybot_dec	float32	None	None
skybot_pmra	float32	None	None
skybot_pmdec	float32	None	None
skybot_deltara	float32	None	None
skybot_deltadec	float32	None	None
object	object	None	None
filter	object	None	None
ra_image	float32	None	None
dec_image	float32	None	None
image_filename	object	None	None
extension	float32	None	None
xwin_image	float32	None	None
ywin_image	float32	None	None
awin_image	float32	None	None
errawin_image	float32	None	None
bwin_image	float32	None	None
errbwin_image	float32	None	None
thetawin_image	float32	None	None
errthetawin_image	float32	None	None
erra_world	float32	None	None
errb_world	float32	None	None
errtheta_world	float32	None	None
flags_extraction	float32	None	None
flags_scamp	float32	None	None
flags_ima	float32	None	None
flags_ssos	float32	None	None
appid_object	object	None	None
mag_cal	float32	None	None
mag_cal_err	float32	None	None

Column Name	Data Type	Unit	Description
ast_id	float32	None	ID of the SSO
ast_name	object	None	MPC name of the SSO
mag_j0395	float32	None	None
err_j0395	float32	None	None
mjd_j0395	float32	None	None
mag_gsdss	float32	None	None
err_gsdss	float32	None	None
mjd_gsdss	float32	None	None
mag_j0515	float32	None	None
err_j0515	float32	None	None
mjd_j0515	float32	None	None
mag_rsdss	float32	None	None
err_rsdss	float32	None	None
mjd_rsdss	float32	None	None
mag_j0660	float32	None	None
err_j0660	float32	None	None
mjd_j0660	float32	None	None
mag_isdss	float32	None	None
err_isdss	float32	None	None
mjd_isdss	float32	None	None
mag_j0861	float32	None	None
err_j0861	float32	None	None
mjd_j0861	float32	None	None
n_filters	int32	None	None

Full text

*Methods. *The JAST80 equipped with T80Cam, providing a field of view of 2 square degrees and a pixel scale of 0.55 arcsec/pixel has been designed for wide-field studies. The main characteristic is the availability of a variety of narrow band filters strategically located on stellar spectral features (the $J0395$ in correspondence of the Ca H+K doublet, the $J0515$ of the Mg $b$ triplet, the $J0660$ of the H$\alpha$ line, and the $J0861$ of the Ca triplet). This project combines, for the first time, the wide-field with a variety of narrow band filters for a unique variability survey, observing each field 11 times with a standardised observing sequence. The median limiting magnitude for individual exposures are 19.1 mag in $J0395$ and $J0515$, $19.6$ mag in $J0660$ and $J0861$, $19.8$ mag in $i$, and $20.2$ in $g$ and $r$. The typical FWHM of the $r$-band images is $1.5$ arcsec.

*Results. *This article introduces the first data release of J-VAR including more than 6000 individual asteroids, 10 detected optical transients (4 discovered supernovae), and 1.3 million light curves of point-sources. On average, J-VAR delivers an unprecedented $\sim$5000 light curves per square degree of 11 epochs in 7 bands, opening research opportunities for theoretical studies and new discoveries alike.

1 Introduction

Variability is a fundamental property of the Universe, offering key insights across all spatial and temporal scales. From the pioneering use of Cepheid pulsations to establish the extragalactic distance scale (e.g., Leavitt & Pickering 1912; Hubble 1929), to the discovery of exoplanets through periodic variations in stellar brightness (e.g., Mayor & Queloz 1995), the study of time-dependent phenomena has continually transformed the field of astrophysics.

In recent years, the field of time-domain astrophysics has entered a golden era, driven by the start of several high-cadence (one day) photometric surveys. Ground-based projects such as All-Sky Automated Survey for Supernovae (ASAS-SN; Shappee et al., 2014), Catalina Real-time Transient Survey (CRTS; Drake et al., 2009), and Zwicky Transient Facility (ZTF; Bellm et al., 2019), operating mostly in up to three broadband filters, have mapped vast swathes of the sky, revealing countless new transients and variable sources. In parallel, space missions like the COnvection ROtation et Transits satellite (CoRoT; Auvergne et al., 2009), Kepler (Borucki et al., 2010), the Transiting Exoplanet Survey Satellite (TESS; Ricker et al., 2015) , and Gaia (Gaia Collaboration et al., 2023) have provided light curves of exquisite precision for millions of sources, albeit typically in a single passband or pseudo-filter. Currently, the Vera C. Rubin Observatory is starting the Large Survey of Space and Time which is expected to deliver millions of alerts of transients per night covering the full sky every few nights with a dedicated 8 m telescope with 6 filters (Ivezić et al., 2019a).

Photometric monitoring in a single filter enables the detection and characterization of variability — amplitudes, timescales, and periodicities — but limits our understanding of the physical mechanisms at play. Spectroscopic time-domain follow-up adds the much-needed context of temperature, velocity, and chemical composition changes, but is observationally expensive and limited to small samples. An intermediate approach, increasingly explored in recent years, is time-domain photometry in multiple filters. This strategy combines the temporal cadence of imaging surveys with the spectral diagnostic power of multi-band photometry, allowing variability to be tracked not just in brightness, but across the spectral energy distribution (SED). Large-scale projects such as the Javalambre Photometric Local Universe Survey (J-PLUS; Cenarro et al., 2019) and the Javalambre-Physics of the Accelerating Universe Astrophysical Survey (J-PAS; Benitez et al., 2014; Bonoli et al., 2021) have demonstrated the power of this approach for static sky studies, using a combination of broad and narrow-band filters to recover low-resolution SEDs for millions of sources.

The Javalambre Variability Survey (J-VAR) builds on this legacy by bringing the multi-filter strategy into the time domain. Carried out with the JAST80, J-VAR employs a unique seven-filter system — three broad-band ($g$, $r$, $i$) and four narrow-band ($J0395$, $J0515$, $J0660$, and $J0861$) — to repeatedly observe each field 11 times. Designed to operate during conditions not suitable for the J-PLUS survey (e.g. poor seeing, bright sky background), J-VAR is unveiling the temporal evolution of stars, compact binaries, Solar System objects and optical transients with both cadence and spectral coverage.

The goal of this paper is to introduce J-VAR, its motivation, and to demonstrate its scientific potential. This paper is outlined as follows: Sect. 2 describes in detail the technical aspects and observing strategy of J-VAR, followed by an overview of the data reduction in Sect. 3. The First Data Release (DR1) of J-VAR is described in Sect. 4 and its main scientific drivers (including early results) in Sect. 5. Our conclusions and perspectives are outlined in Sect. 6.

A series of companion papers will offer further insights into different aspects of the survey and the data release, including the construction of calibrated light curves and variability indices (Pyrzas et al., submitted), the catalogue of minor bodies (Morate et al., submitted), and the light-curve fitting of RR Lyrae stars (Kulkarni et al., submitted).

Magnitudes are in the AB system (Oke & Gunn, 1983), unless noted otherwise.

2 Observations

J-VAR is a multi-band time-domain survey of the northern sky designed to study variability in sky regions previously observed by the J-PLUS survey. By targeting the same fields using identical instrumentation, J-VAR leverages J-PLUS’s robust photometric calibration (López-Sanjuan et al., 2024) and other survey products that have matured over time, including source classifications (López-Sanjuan et al., 2019; del Pino et al., 2024) and photometric redshifts for extragalactic objects. The current J-VAR photometric calibration (Pyrzas et al. submitted; Morate et al. submitted) directly depends on the one of the J-PLUS third data relase (J-PLUS DR3; López-Sanjuan et al., 2024). For this reason the J-VAR pointings follow the J-PLUS field definitions.

The main driver behind the choice of the filters of J-VAR is efficiency. Six filters ($g$, $r$, $i$, $J0515$, $J0660$, and $J0861$) have been chosen for being the most efficient filters (i.e. highest effective transmission of the whole system) of the J-PLUS filter-set, as it can be seen in Fig. 1. The $J0395$ has also been included to study the variation of the Ca H+K doublet, which is specially sensible to e.g. stellar chromospheric activity.

Observations are carried out at the OAJ (Cenarro et al., 2014) with the T80Cam panoramic camera attached to the Javalambre Auxiliary Survey Telescope (JAST80) (Marín-Franch et al., 2012, 2015). JAST80 is an 83 cm, f/4.5 modified Ritchey-Chrétien reflector on a German equatorial mount. The telescope and camera provide a pixel scale of $0.556\,\textrm{arcsec}/\textrm{pix}$ and a $1.4\,\textrm{deg}\times 1.4\,\textrm{deg}$ ($2\,\textrm{deg}^{2}$) filed-of-view (FoV) that, on average, allows the detection of a number of sources of the order of $10^{4}$ per individual image. J-VAR uses a subset of the J-PLUS photometric system (see Fig. 1). The latter was devised to measure some of the most prominent atmospheric stellar features with the aim of studying a wide variety of phenomena in the local Universe (for more details see Cenarro et al. 2019).

J-VAR’s observations started in May 2017, shortly after the beginning of the scientific operation of the JAST80/T80Cam, as a filler program running on non-photometric nights, whenever J-PLUS or other competitive open time programs demanding photometric conditions could not be executed. From 2019 to the end of 2022, J-VAR has been granted with competitive open time offered every semester by the OAJ as a large program. Since 2023 the survey continued, again as a filler program, as it did in the beginning. The evolution of the survey is shown in Fig. 2.

2.1 Observing strategy

The filler program nature of J-VAR necessitates consideration of potential cloud cover during observations. J-VAR targets a limiting magnitude $\sim$1 mag brighter than J-PLUS, conducted with identical instrumentation including telescope, camera and filters. Non-photometric conditions cannot be modelled reliably, so bright time is used as the closest operational proxy in the exposure time calculator111https://www.cefca.es/jplusetc/ and assume that the system will deliver observations 0.5-1 magnitudes shallower than the input value. The input limiting magnitudes and the exposure times for each band are shown in Table 1, together with actual average measures per filter from a sample of completed fields. As it can be seen, the individual exposures of J-VAR are between $0.4$ ($J0861$) and $1.4$ ($J0395$) magnitudes shallower than the stacked images of J-PLUS (in most of the cases, three images are combined). So it can be considered that the goal of the individual images of J-VAR being as deep as J-PLUS is overall fulfilled by the use of increased exposure times. It is worth noting that, in the case of J-PLUS, the observing exposure times are modulated according to the Moon phase and distance (Cenarro et al., 2019). This is possible because J-PLUS is carried out under photometric conditions. That approach was discarded for J-VAR both for simplicity, and because the survey is by construction carried out with varying atmospheric conditions that are difficult to track and model.

Regarding the selection of sky regions to be observed, two main considerations are taken into account. First, to ensure accurate photometric calibration, J-VAR targets J-PLUS fields that have already been published and properly calibrated. Second, the choice of fields is driven by the three key science cases of the survey: small bodies of the Solar System (Sect.  5.1), optical transients—particularly supernovae (Sect. 5.2), and variable stars (Sect. 5.3). The supernova case requires fields with a relatively low stellar density, which is best achieved by observing far from the Galactic plane. In contrast, the Solar System Objects case benefits from fields located as close as possible to the Galactic plane. This complementarity is illustrated in Fig. 3.

2.1.1 Time-domain design: cadence and observing sequence

Given the filler nature of the project, the cadence is primarily dictated by atmospheric conditions and the availability of other observing programs at any given time. The resulting cadence and sampling for J-VAR DR1 are shown in Fig. 4.

For supernovae, maximizing the time span of observations increases the likelihood of detecting events in different phases of their evolution. Therefore, a goal was set to span the campaign in each field for at least one year. Additionally, to facilitate the creation of template images from any observing epoch—defined as all observations taken within the same night—the basic observing block (BBO) was structured to consist of three exposures per filter across all seven bands. These triplets allow for stacking when needed, improving signal-to-noise and enabling high-quality template image generation. The exposures are not taken sequentially in the same band; instead, three full cycles through the seven filters are executed, distributing the observations in time. This strategy, illustrated in Fig. 5, ensures that each filter is sampled at different time intervals within an observing night. The full BBO takes approximately 40 minutes to complete, with a median gap of 12.7 minutes between exposures in the same filter.

This approach benefits not only the detection of transient events but also minor bodies in the Solar System, where repeated observations over short timescales help constrain orbital trajectories and facilitate rapid follow-up. Whenever possible, the same field is observed on consecutive nights to improve trajectory predictions. On longer timescales, sampling on the scale of weeks is crucial for tracking the brightness evolution of supernovae. The final distribution of visit frequencies for a given field is illustrated in Fig. 4.

Beyond transients and Solar System objects, J-VAR was also designed to provide multi-band time-domain information for variable stars, with RR Lyrae stars serving as a reference case for defining the total number of epochs. Unlike transient detections, where the absolute timing of observations is critical, RR Lyrae characterization primarily depends on adequately sampling the pulsation cycle. Given the somewhat randomized observing cadence, this condition is generally fulfilled naturally over time.

To determine the optimal number of epochs required, a simulation was conducted using real RR Lyrae light curves from the OGLE III Catalog of Variable Stars (Soszyński et al., 2010, 2011) as a reference. The test focused on a single OGLE band (I), where different scenarios with varying numbers of randomly selected epochs were simulated. Each simulated epoch consisted of three data points, mirroring the J-VAR observing sequence. The derived periods were then compared to the true values provided by OGLE, allowing us to assess the period recovery success rate. The study found that at least 11 epochs were necessary to retrieve the correct period in \sim$$75\% of cases when using a single band and a signal-to-noise ratio (SNR) above 50. Given that J-VAR includes observations in seven filters, this success rate was expected to improve when considering multi-band data. Consequently, 11 epochs were adopted as a reasonable trade-off between ensuring period determination accuracy and maximizing the survey area. This choice is consistent with other surveys, such as Pan-STARRS (Sesar et al., 2017). The actual period recovery success rate for J-VAR DR1 is analyzed in the accompanying RR Lyrae catalogue paper (Kulkarni et al. submitted). Using only the J-VAR $r$ band, Gaia DR3 (Gaia Collaboration et al., 2023; Clementini et al., 2023) periods of RR Lyrae stars in common are recovered within $1\%$ relative difference in $\gtrsim 70\%$ of the cases for $G$ magnitude from 14 to 19. Some examples of J-VAR RR Lyrae light curves are shown in Sect. 5.3.

2.1.2 The High Frequency (HF) Fields

During periods around full Moon, one of the JAST80 filter wheels may be temporarily replaced to host alternative filters required by other projects (e.g. GALANTE, Ma´ız Apellániz et al. 2021). When atmospheric conditions prevent the execution of those programs, we implement a dedicated J-VAR observing strategy, hereafter the “high-frequency mode” (HF). In this mode, a single field is monitored continuously for $\sim$3 hours, repeating the standard sequence of exposures with the set of available J-VAR filters J-VAR sequence of seven filters (Fig. 5). This configuration delivers intra-night light curves with much denser sampling than in the nominal survey, while maintaining full compatibility with the standard observing setup.

3 Data Reduction

The image reduction is performed using jype, a custom-built pipeline developed at the Centro de Estudios de Física del Cosmos de Aragón (CEFCA) for processing data from OAJ surveys (see, e.g. Cenarro et al., 2019; Bonoli et al., 2021). Written primarily in Python, jype incorporates SExtractor (Bertin & Arnouts, 1996) for source extraction and initial photometry. The pipeline has been continuously developed since 2010, and its most recent version before the release, jype-3.1.9, has been used for the processing of J-VAR DR1 images.

A key aspect of jype is its ability to handle the particular readout structure of the T80Cam CCD, which consists of 16 amplifiers. Each amplifier includes overscan and prescan regions along both the $x$ and $y$ axes, allowing for precise bias level removal. Immediately after observation, scientific images undergo an initial calibration using the latest valid flat field and illumination correction (see below). These pre-processed images are typically available within minutes and enable time-sensitive analyses, such as asteroid and transient detection, which are usually conducted the day after observation.

Beyond this real-time processing, a more refined reduction is performed at the end of each observing run, defined as a period during which the optical system remains stable (i.e., no interventions affecting the filter wheel or main mirror cleaning). Observing runs typically last about a month, at which point master flat fields and illumination corrections are generated to ensure optimal photometric accuracy.

In addition to standard image reduction steps, including bias subtraction, flat-fielding, fringing correction (when necessary), and cosmic ray and satellite trail masking, jype features an essential illumination correction step. This is particularly important for wide-field systems like JAST80, where conventional flat-fielding alone can introduce a two-dimensional photometric bias of several tens of milimagnitudes across the image due to the presence of field correctors. To mitigate this effect, an additional processing step is applied; see details in Appendix B.1 in Bonoli et al. (2021).

Finally, jype computes the aperture correction for photometry performed within a $6$-arcsecond diameter integration area, which serves as a reference for total flux measurements in J-VAR DR1, particularly for point-like sources. This standardized photometric approach ensures consistency across all reduced data products.

4 First Data Release

The First Data Release of J-VAR was made public on 19th July 2024. The number of pointings distributed in this data release is $101$, providing time-domain information for \sim$$202\,\textrm{deg}^{2} on the sky in seven photometric bands. The list of coordinates is in Table LABEL:tab:pointings, together with information on the earliest and latest time of observation (in modified Julian date) and if the field has high frequency observations. In case HF observations are available, the earliest and latest time of observation is also available (again, as modified Julian date). The sky distribution is shown in Fig. 8. From an analysis of each of the images in this data release, we derived the distribution of limiting magnitudes (Fig.6). The limiting magnitudes are rarely achieved in most filters and never achieved in the reddest filters ($J0861$ and $i$). This is a consequence of the approximations made at the time of setting the exposure times. Similarly, Fig. 7 shows the distribution of the FWHM. The lower limit of the distribution is close to 1 arcsec which is slightly smaller than 2 pixels. The peak of the distribution is consistently close to 1.5 arcsec. The quality of the data, in particular having a shallower survey in the red, does not affect the main goals of the project, as the multi-filter light curves are still deliverable.

The data of J-VAR DR1 are accessible through the CEFCA archive222https://archive.cefca.es/catalogues/jvar-dr1. The principal way to access to the data is through the public catalogue, whose structure is related to the main objectives of the project. The time of observation of each field in a given filter is in the Timestamps table (its contents are shown in Table LABEL:tab:timestamps). In this table, the observing time of each image is provided. Each image is identifiable via the identifier of the field and the identifier of the filter. The time is provided as coordinated universal time (UTC). For convenience, the modified Julian date (MJD) is also given.

Each image is also tagged according to the type of observation: regular or high frequency (see above). The regular J-VAR observations are tagged as “JVN”. The high frequency observations are marked differently, as described hereafter. The narrow-band images are marked as “HFN”. The broad-band images are marked “HFJ” if executed with the same filters as the main “JVN” survey. If they are acquired using, not the same physical filters, but the $g$, $r$, and $i$ filters initially devoted to the Pathfinder camera to conduct of the miniJPAS survey (Bonoli et al., 2021), they are named “HFG”. Those $gri$ filters were usually mounted in the auxiliary filter wheel of T80Cam, and were manufactured following the same optical design.

The photometry of each object is provided in the Photometry table (a description is made available in Table LABEL:tab:photometry). In this table, each observation of an object is identified by an ID of the object. Both the J-PLUS identifier (the ID of the J-PLUS DR3 tile and the identifier of the object within the tile) and the J-VAR identifier (the string jvar followed by the right ascension and the declination of the object) are provided. The magnitudes and associated errors are offered as arrays. The full-width at half maximum of the objects is also provided as array, as well as a list of photometric flags which can be used for quality assessment. Finally, the photometric calibration is represented by an array containing the values used to go from instrumental magnitude to calibrated magnitude. The number of comparison stars and the maximum radius within which these stars are located are also shown. For details, see Pyrzas et al. (submitted). For convenience, the previous two tables are combined in a light_curves table whose columns are described in Table LABEL:tab:lightcurves.

The table Objects conveniently gathers useful information from J-PLUS DR3 (magnitudes in 3 and 6 arcsec, different star/galaxy separation indicators, FWMM), the interstellar extinction, the type and period of variability according to the International Variable Stars Index and the parallax, G magnitude, variability class and score from Gaia DR3. The columns of this table are summarised in Tab.LABEL:tab:objects.

Two tables related to the Solar System Objects are also available. The first one (SSOs_DETECTIONS, the columns are shown in Table LABEL:tab:ssos_detections) shows the detections by the SSOS pipeline, described in Sect. 5.1 and in Morate et al. (submitted), while the second one (SSOs_MAGNITUDES, whose columns are shown in Table LABEL:tab:ssos_magnitudes) shows the magnitudes for each object.

Following the common practice, the detections of optical transients are reported through the Transient Name Service333https://www.wis-tns.org/search and therefore there is no associated table in the database.

All the tables are accessible via a web interface and via the Table Access Protocol of the Virtual Observatory, both in a programmatic way (e.g. with Python) or via programs like TOPCAT (Taylor, 2005). Similarly, The reduced images are accessible through the archive and through the Simple Image Access Protocol (SIAP) of the Virtual Observatory. An example of how to access the data of J-VAR is available at https://github.com/aederocl/J-VAR_paper1.

5 Science Cases

5.1 Small bodies of the Solar System

Small bodies (SB) in the Solar System are not usually the drivers of observational surveys, with the noticeable exceptions of the Asteroid Terrestrial-impact Last Alert System, ATLAS, (Tonry et al., 2018), the Panoramic Survey Telescope and Rapid Response System, PAN-STARRS (Hodapp et al., 2004), or the Rubin Observatory’s Legacy Survey for Space and Time, LSST (Ivezić et al., 2019b) surveys. SBs are usually regarded as contamination for the main science objectives of most large surveys. Nevertheless, the SB community has exploited this ”contamination” by developing pipelines that detect and extract their information. The Sloan Digital Sky Survey has been largely exploited for SB (e.g., Ivezić et al., 2001; DeMeo & Carry, 2013) and so was the first data release of J-PLUS (see Morate et al., 2021). Sykes et al. (2000), Popescu et al. (2016) and Mahlke et al. (2018) have extended this search to infrared wavelengths. SBs information, such as their colors, is used for mineralogical analysis, which places strong constraints on how different materials are distributed in the Solar System and, together with dynamical and collisional models, may shed light into how the Solar System as we know it today came to be.

J-VAR observational strategy is well-suited for the detection of SBs because the fields (Sect. 2.1) are selected as a balance between its three main science drivers. Within the 101 fields included in J-VAR DR1, there are 131 900 individual detections, i.e., an asteroid was detected in one of the J-VAR filters. These detections correspond to a total of 6 570 individual asteroids, some of them detected in two or more epochs.

The detection of the SBs is done using the SOSS pipeline (see details in Mahlke et al., 2019). The pipeline detects and extracts the instrumental magnitude of the SBs, but it does not produce calibrated magnitudes because the J-VAR data is taken during non-photometric nights. Thus, no zero points are provided in the images’ headers. Therefore, the calibration step is done using J-PLUS observed fields, computing the zero points for the common objects in the field and applying this secondary calibration to the SB (more details are provided in Morate et al. submitted).

Regarding the number of detections, for those J-VAR fields centered around the ecliptic plane of the Solar System ($\pm 10\deg$), there are around $\sim$400 detections per epoch (which translates roughly into $\sim$30 different objects). This number decreases as one moves away from the ecliptic: for high latitudes, i.e., above $30\deg$, the number of detections is very scarce, usually below 20 (this is, one object or none at all.) However, the number of detected objects ensures good statistics (see below).

The distribution of objects observed with a given number of filters peaks at six (with about 3 000). Almost 800 were observed with all seven filters, and only two objects were observed with one filter. We show the number of detections per filter in Fig. 9. The lower numbers in the $J0395$ and $J0515$ filters were expected as these are narrow-band filters, and, in the case of the $J0395$ filter, the reflectance of SB is known to decrease significantly at blue wavelengths.

The most efficient filters are the SDSS filters $g$, $r$, and $i$, which was expected because these are the widest filters in the J-VAR system.

Finally, we show a highlight of our results (a detailed description of the data is provided in Morate et al., submitted): Fig. 10 shows two colour-colour diagrams, zoomed into their densest regions to avoid outliers.

In Morate et al. (submitted) we present a value added catalogue (VAC) for J-VAR’s DR1: the detection catalogue with calibrated magnitudes, RA, DEC, time of observation, among other useful information. We also provide a first look into the taxonomy of the objects and rotational properties of some targets, mostly observed in HF fields. In conclusion, J-VAR works as a stand-alone survey, but, importantly, it nicely complements J-PLUS, which uses all J-VAR filters.

5.2 Optical Transients

Ranging from novae, supernovae (SNe) up to the progenitors of the most powerful gravitational waves, the transient sky is part of the most challenging puzzles in astrophysics and cosmology. Although mostly known as responsible for the first evidences of the Dark Energy dominated Universe in the late 1990s, type Ia SNe, for example, are still of paramount importance to cosmology. The so-called ”Hubble tension” (Knox & Millea, 2020), a discrepancy in the measurement of the Hubble constant in different stages of the Universe’s evolution, is becoming a decade-long problem and some of the most precise results on which it relies are based on observations of low-redshift type Ia SNe.

From an astrophysical point of view, there seems to be still no universal consensus on the nature of type Ia SNe progenitors and their explosion mechanisms (Liu et al., 2023). The same difficulties arise in finding a single model to explain superluminous SNe explosions (Gal-Yam, 2019).

As commented in Sect. 1, a variety of surveys aim at the study of the transient sky. In particular, currently ASAS-SN and ZTF which can be considered a progenitor of LSST.

We took advantage of the repeated visits from J-VAR and applied the image subtraction or differential image technique to look for optical transients within DR1 images. Our differential image pipeline consists in a sequence of open source packages and scripts to perform image subtraction and produce a catalogue of candidates to be analysed individually. The pipeline, optimized for SNe detection, also recovers variable stars, AGN, and asteroids. To reduce false positives, we apply strict shape and multi-band detection criteria, followed by SIMBAD cross-matching. The final step is a visual inspection and confirmation of the candidates.

Since our search was performed after the observations of a given field were completed, the pipeline detected some candidates which had been already observed by other surveys, which is the case for supernovae SN 2017icq (SN Ia), SN 2019obw (SN Ia), SN 2020cvy (SN II) and SN 2020amv (SN II) and transient candidates AT 2017eke, AT 2017dzn, AT 2018fds, AT 2019roh, AT 2019ioo and AT 2020abzu. Fig. 11 shows the detection, light curve and spectral energy evolution of the supernova SN 2020amv444https://www.wis-tns.org/object/2020amv. This was discovered on 23rd January 2020 and it is detected in J-VAR field 13 days after discovery. The bottom panel of Fig. 11 shows an excess in the $J0660$ filter, suggesting the presence of H$\alpha$ in emission, characteristic of type II supernovae, thus showing the potential of J-VAR to characterise low-redshift supernovae without the need of spectroscopy.

Our final lists of transient candidates contained a large number of variable sources. We recorded 252 known variable stars and 50 known AGN, along with 236 variable star candidates and 18 AGN candidates.

Finally, we have discovered by visual inspection four transient events which were submitted to the TNS database. Two of them, SN 2020admb (JVAR20a) and SN 2024slh (JVAR24b), were successively confirmed as supernovae and classified by the ZTF collaboration as a type Ia supernovae at redshifts $\leavevmode\nobreak\ 0.04$ and $0.05$, respectively. The other two, AT 2021aggv (JVAR21a) and AT 2024bxn (JVAR24a) are still classified as transient candidates at the TNS database.

5.3 Variable Stars

The J-VAR survey provides multi-band time-domain data that enables the study of various classes of variable stars across a range of timescales, amplitudes, and wavelengths. The combination of its multi-filter approach, flexible cadence, and moderate number of epochs makes it particularly well-suited for the characterization of pulsating stars, close or contact eclipsing binaries, and semi-regular variables. Moreover, the irregular spacing of the epochs mitigates aliasing effects in period determination, thereby increasing the robustness of variability classification.

The light curves for all the point-sources in common with J-PLUS have been extracted. This sums up to 1.3 million sources. The light curves have been obtained with ensemble differential photometry (see e.g., Everett & Howell 2001; Tamuz et al. 2005). For each star, at least 15 comparison stars are selected. In all cases, this is achieved in a search radius smaller than 9 arcminutes. The resulting photometric precision (RMS) is 2% down to mag $\sim$16 and 5% to mag $\sim$18 in the broad-band filters. The extraction and calibration process is described in detail in Pyrzas et al. (submitted). In total, the J-VAR DR1 catalogue has 1621 objecs classified as variables in VSX and 15221 objects classified as variables in *Gaia * DR3. The most frequent stellar *Gaia * DR3 classes are shown in Table 2.

The variety of stellar systems that show changes in luminosity is very rich (e.g. Catelan & Smith, 2015). The Hertzsprung-Russell diagram in Fig. 12 shows the locations of the objects whose light curves are shown in Fig. 13. The effective cadence of the observations favours the suitability of the study of some type of variable stars over others. In particular it fits well to the variability time-scales of RR Lyrae stars, Cepheids, high-amplitude $\delta$Sct stars, semi-detached or contact binaries (with short or no interruptions between eclipses), etc. Conversely, the phase coverage is often not good for detached binaries because their eclipses last for a relatively short time in comparison with the whole period, so the probability of capturing both adequately is relatively low. And lastly, long-period variable stars, with periods of hundreds of days, are usually covered by the temporal baseline of J-VAR that typically extends beyond 1 year, although the phase is more irregularly sampled in this case than for shorter period variables mentioned before.

5.3.1 Pulsating stars: RR Lyrae, Cepheids, and $\delta$ Scuti stars

RR Lyrae stars were a key reference in defining the J-VAR cadence and total number of epochs, as they serve as excellent standard candles for distance determination in the Milky Way and nearby galaxies. The survey’s seven-band photometry allows for refined period determinations, variability classification, and potential metallicity estimates based on color indices. The adopted 11-epoch strategy ensures that light curves sample the pulsation cycle adequately, improving period recovery rates compared to single-band surveys.

In addition to RR Lyrae stars, J-VAR data can also be used to identify and study classical Cepheids and $\delta$ Scuti stars, which provide complementary constraints on stellar evolution and pulsation theory. Given that J-VAR primarily targets the Galactic halo, the survey contains only a small number of Cepheids. The multi-band nature of the survey is particularly useful for distinguishing between different pulsation modes and for improving period-luminosity relationships.

5.3.2 Eclipsing binaries and other periodic variables

The survey’s multi-epoch observations facilitates the identification and classification of semi-detached (EB) and contact (EW) eclipsing binaries, and in fewer cases, detached (EA) systems. The broad wavelength coverage helps to constrain better the temperature ratios between the components (e.g. Kallrath & Milone, 2009), improving constraints on stellar parameters. Additionally, J-VAR’s ability to capture multiple points within an eclipse event in a single observing block increases its potential for detecting and characterizing short-period systems.

Other periodic variables, such as rotationally modulated stars, including RS CVn (RS), chromospherically active rotational binaries, and BY Dra (BY), late-type dwarf stars with starspots and rotational modulation, can also be analysed using J-VAR data. The survey’s multi-band information helps distinguish between RS and BY stars.

5.3.3 Semi-regular and cataclysmic variables

The J-VAR cadence also enables the detection of longer-term variations associated with semi-regular and Mira variables, which exhibit large-amplitude, multi-periodic brightness fluctuations. Additionally, the survey’s transient monitoring capabilities may allow for serendipitous detections of eruptive variables, including cataclysmic variables (CVs) undergoing outbursts.

The presence of short intra-night separations in the observing strategy further provides sensitivity to flickering behaviour in CVs and rapidly variable phenomena in compact stellar remnants. Combined with spectroscopic follow-up, these data could contribute to the discovery of rare or previously unknown subclasses of variable stars.

6 Conclusions and perspectives

The parameter space for time-domain wide-field photometric surveys can be summarized by six factors: (i) survey area – the amount of sky area where the light curves are distributed, (ii) cadence – the time interval between two observations of the same targets, (iii) limiting magnitude, (iv) number of filters used to characterize a light curve, (v) pixel scale of the camera, and (vi) field of view555In this analysis we are not using the etendue (the multiplication of the size of the primary mirror by the field of view) or the exposure time, since they are somewhat embedded in the values that we are considering, in particular limiting magnitude and cadence. The science cases driving the different projects are reflected in how they cover such parameter space (see Fig. 14). Since TESS is aiming at transits of extrasolar planets, its cadence is very short with respect to all other projects. TESS also employs a vast field of view and a very large pixel scale. With the noticeable exception of ZTF, ground based projects have all comparable fields of view and pixel scales. This is mostly due to the need of having an optical system which is adapted to the atmospheric conditions of the observing site. The three parameters where there is the largest scatter are sky area, limiting magnitude, and number of filters. Ground-based transient surveys like ZTF and ASAS-SN cover a very large sky area in a small number of filters as often as possible. The main difference between these two surveys is the limiting magnitude (ZTF is carried out with a larger telescope) and the survey area (ZTF is carried out with one telescope only, thus only covering the Northern celestial hemisphere, while ASAS-SN, being a network of telescopes, can survey the whole sky). The observing strategy of LSST is the result of the trade off between different projects, which makes the number of filters larger than other surveys and the cadence somewhat less frequent, although it has an unrivalled depth.

J-VAR leverages three characteristics of the JAST80 telescope: good image quality, wide field of view, and the narrow band filters. To the best of our knowledge, at the time of beginning the survey (2017), J-VAR was the only wide-field time-domain survey routinely observing with narrow-band filters. This brings J-VAR to occupy a unique position in the phase-space for wide-field time-domain surveys as the cadence is comparable with the one of LSST, the depth comparable with ZTF, and the filter set is unique.

J-VAR is not carried out with a dedicated telescope, unlike other time-domain surveys, and therefore shares observing time with other awarded proposals. Projects like ASAS-SN and ZTF can observe within a few days all the sky available from their sites and, therefore, they can, in principle, be extended indefinitely to increase the length of the light curves. In the case of J-VAR, the observation of a field is limited to 11 epochs. The number of completed fields grows at a rate of approximately 58 square degrees per year (see Fig 2). This has been roughly constant since the beginning, when J-VAR was carried out as a two-years long large project at the OAJ, and now when it is a filler among J-PLUS and the Legacy Surveys at the OAJ. It is likely that J-VAR will continue at the same pace until 2027, when the Legacy Surveys will end. At the current rate, J-VAR will reach 1000 square degrees in 2035, although this could vary in case of changes in the programs carried out on the JAST80 after 2027.

Adopting a unique strategy leveraging narrow band filters, J-VAR represents an unprecedented opportunity for time-domain astrophysics. Thanks to the use of narrow-band filters, J-VAR constitutes a bridge between spectroscopic time-domain surveys (e.g., SDSS-V; Kollmeier et al., 2025) and wide-field photometric surveys. J-VAR’s cadence and wavelength coverage make it complementary to large-scale surveys such as Gaia, ZTF, and LSST. While those surveys provide extensive temporal coverage, J-VAR adds the advantage of quasi simultaneous multi-band observations, aiding in precise classification and improving constraints on variability mechanisms. The multi-filter approach also facilitates cross-matching with existing variability catalogues, enabling further refinement of classification schemes. The high-frequency mode further extends J-VAR’s scope, enabling intra-night variability studies (asteroids, eclipsing binaries, CVs) not accessible in most wide-field surveys.

J-VAR DR1 contains 1.3 million light curves of 11 epochs in seven photometric bands with no selection bias on colour or type. By covering a diverse range of variable star types, J-VAR provides a unique dataset for time-domain astrophysics. The combination of multiple bands, strategically spaced epochs, and intra-night sampling opens the door to numerous scientific applications, from refining period-luminosity relations to characterizing new classes of variables. The survey’s contribution to variable star studies will be further enhanced through future data releases and dedicated variability catalogues, ensuring an average of 5000 new multi-band light curves per newly observed square degree.

Acknowledgements.

We ackowledge support from the Governments of Spain and Aragón through their general budgets and the Fondo de Inversiones de Teruel. The Spanish Ministry of Science and Innovation (MCIN/AEI/10.13039/501100011033) and ERDF, A way of making Europe with grant PID2021-124918NB-C42. AE acknowledges the financial support from the Spanish Ministry of Science and Innovation and the European Union - NextGenerationEU through the Recovery and Resilience Facility project ICTS-MRR-2021-03-CEFCA. DM acknowledges the financial support from the European Union - NextGenerationEU through the Recovery and Resilience Facility program Planes Complementarios con las CCAA de Astrofísica y Física de Altas Energías - LA4. SK has received financial support from the Aragonese Government through the Research Groups E16_23R and the Spanish Ministry of Science and Innovation (MCIN/AEI/10.13039/501100011033 y FEDER, Una manera de hacer Europa) with grant PID2021-124918NB-C41. AAC acknowledges financial support from the Severo Ochoa grant CEX2021-001131-S funded by MCIN/AEI/10.13039/501100011033. AAC and DM acknowledge financial support from the Spanish project PID2023-153123NB-I00, funded by MCIN/AEI. P.C. acknowledges financial support from the Spanish Virtual Observatory project funded by the Spanish Ministry of Science and Innovation/State Agency of Research MCIN/AEI/10.13039/501100011033 through grant PID2023-146210NB-I00. F.R.H. acknowledges support from FAPESP grants 2018/21661-9 and 2021/11345-5. The authors would like to acknowledge the use of the following open-source software packages that were essential in the analysis presented herein: Python, NumPy, Astropy, pandas, matplotlib, and PyVO (including TAP). Their combined functionality enabled efficient data handling, manipulation, visualization, and access to Virtual Observatory services.

Appendix A Pointings in J-VAR DR1

The coordinates of the pointings of J-PLUS DR1 are shown in Table LABEL:tab:pointings. These are the coordinates that the telescope used to point for the first exposure of each observing sequence.

1]

Appendix B Tables of the Database

This section contains the description of the tables in the database. Following J-PLUS, measurements in different filters are shown as arrays. The order of filters is GSDSS, RSDSS, ISDSS, J0395, J0515, J0660, J0861. The labels of the filters have been capitalised to indicate that they are a label.