Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
tic_id
float64
613k
766M
⌀
time
float64
3.99k
4.05k
⌀
flux
float64
0.36
4.16
⌀
flux_err
float64
0
0.17
⌀
quality
int32
0
0
⌀
613,256
3,995.336252
0.992485
0.022298
0
613,256
3,995.338567
0.975109
0.022298
0
613,256
3,995.340882
0.978205
0.022298
0
613,256
3,995.343197
1.021895
0.022298
0
613,256
3,995.345512
1.006796
0.022298
0
613,256
3,995.347826
1.011268
0.022298
0
613,256
3,995.350141
1.024017
0.022298
0
613,256
3,995.352456
0.985918
0.022298
0
613,256
3,995.354771
0.993548
0.022298
0
613,256
3,995.357086
1.019267
0.022298
0
613,256
3,995.359401
0.97667
0.022298
0
613,256
3,995.361716
0.990872
0.022298
0
613,256
3,995.36403
0.967829
0.022298
0
613,256
3,995.366345
1.018124
0.022298
0
613,256
3,995.36866
0.975848
0.022298
0
613,256
3,995.370975
0.980322
0.022298
0
613,256
3,995.37329
1.002008
0.022298
0
613,256
3,995.375605
0.975428
0.022298
0
613,256
3,995.377919
1.022518
0.022298
0
613,256
3,995.380234
1.023849
0.022298
0
613,256
3,995.382549
1.005579
0.022298
0
613,256
3,995.384864
1.018855
0.022298
0
613,256
3,995.387179
1.017992
0.022298
0
613,256
3,995.389494
1.029933
0.022298
0
613,256
3,995.391809
0.989916
0.022298
0
613,256
3,995.394123
1.013413
0.022298
0
613,256
3,995.396438
0.986457
0.022298
0
613,256
3,995.398753
1.016771
0.022298
0
613,256
3,995.401068
1.017436
0.022298
0
613,256
3,995.403383
0.9981
0.022298
0
613,256
3,995.405698
0.987286
0.022298
0
613,256
3,995.408013
0.974936
0.022298
0
613,256
3,995.410327
0.986135
0.022298
0
613,256
3,995.412642
1.025801
0.022298
0
613,256
3,995.414957
1.022066
0.022298
0
613,256
3,995.417272
1.008774
0.022298
0
613,256
3,995.419587
1.001483
0.022298
0
613,256
3,995.421902
0.98315
0.022298
0
613,256
3,995.424217
1.009679
0.022298
0
613,256
3,995.426531
0.973045
0.022298
0
613,256
3,995.428846
0.998179
0.022298
0
613,256
3,995.431161
0.985635
0.022298
0
613,256
3,995.433476
0.983571
0.022298
0
613,256
3,995.435791
0.984468
0.022298
0
613,256
3,995.438106
0.968445
0.022298
0
613,256
3,995.44042
0.9877
0.022298
0
613,256
3,995.442735
1.013567
0.022298
0
613,256
3,995.44505
0.981957
0.022298
0
613,256
3,995.447365
1.010419
0.022298
0
613,256
3,995.44968
0.994561
0.022298
0
613,256
3,995.451995
0.999007
0.022298
0
613,256
3,995.45431
0.998414
0.022298
0
613,256
3,995.456624
1.000545
0.022298
0
613,256
3,995.458939
0.997068
0.022298
0
613,256
3,995.461254
0.988944
0.022298
0
613,256
3,995.463569
1.018381
0.022298
0
613,256
3,995.465884
1.010306
0.022298
0
613,256
3,995.468199
1.000153
0.022298
0
613,256
3,995.470514
1.01679
0.022298
0
613,256
3,995.472828
0.990897
0.022298
0
613,256
3,995.475143
0.989846
0.022298
0
613,256
3,995.477458
0.980603
0.022298
0
613,256
3,995.479773
1.030359
0.022298
0
613,256
3,995.482088
0.968982
0.022298
0
613,256
3,995.484403
0.999314
0.022298
0
613,256
3,995.486718
1.019621
0.022298
0
613,256
3,995.489032
0.985237
0.022298
0
613,256
3,995.491347
1.011489
0.022298
0
613,256
3,995.493662
0.9752
0.022298
0
613,256
3,995.495977
1.007661
0.022298
0
613,256
3,995.498292
1.006159
0.022298
0
613,256
3,995.500607
0.998571
0.022298
0
613,256
3,995.502921
0.989271
0.022298
0
613,256
3,995.505236
0.994316
0.022298
0
613,256
3,995.507551
1.006387
0.022298
0
613,256
3,995.509866
1.01813
0.022298
0
613,256
3,995.512181
1.006197
0.022298
0
613,256
3,995.514496
0.998011
0.022298
0
613,256
3,995.516811
0.994748
0.022298
0
613,256
3,995.519125
1.015806
0.022298
0
613,256
3,995.52144
0.98419
0.022298
0
613,256
3,995.523755
1.009459
0.022298
0
613,256
3,995.52607
1.001645
0.022298
0
613,256
3,995.528385
0.997322
0.022298
0
613,256
3,995.5307
1.004741
0.022298
0
613,256
3,995.533015
1.0182
0.022298
0
613,256
3,995.535329
0.990035
0.022298
0
613,256
3,995.537644
1.003699
0.022298
0
613,256
3,995.539959
1.009946
0.022298
0
613,256
3,995.542274
0.97732
0.022298
0
613,256
3,995.546904
1.009815
0.022298
0
613,256
3,995.549219
0.983373
0.022298
0
613,256
3,995.551533
0.997596
0.022298
0
613,256
3,995.553848
0.989871
0.022298
0
613,256
3,995.556163
1.016361
0.022298
0
613,256
3,995.558478
0.988547
0.022298
0
613,256
3,995.560793
1.00853
0.022298
0
613,256
3,995.563108
0.9864
0.022298
0
613,256
3,995.565422
1.012795
0.022298
0
613,256
3,995.567737
0.999765
0.022298
0
End of preview. Expand in Data Studio

TESS QLP Sector 98 prototype

This is a prototype. It is a random sample of 1,000 stars from TESS (Transiting Exoplanet Survey Satellite) Sector 98, taken from the Quick-Look Pipeline (QLP) full-frame-image light curves, run through a transit search to test how a whole sector's results can be hosted and read. It makes no claim of planet detections: the search peaks (signals) are candidates from an automated search and are not vetted. The threshold and calibration tables show how noisy that search is.

What is in it

Only cadences with QUALITY == 0 and finite time, flux and flux error are kept. 821 of the 1,000 stars were searched; 179 had no usable rows and appear in stars and manifest only. Flux is the detrended DET_FLUX column, not normalised. Row groups are 98,304 rows, zstd-compressed, so a reader can fetch one star's rows without the whole file.

File Rows Row groups Meaning
stars.parquet 1,000 1 one row per sampled star: TIC id, TESS magnitude, radius, temperature, camera, CCD, position, search status, noise level, best signal-to-noise ratio (SNR), number of signals
signals.parquet 2,463 1 one row per star and BLS peak (top peaks of the box least squares search): period, mid-transit time, duration, depth, SNR, power, odd and even depths
events.parquet 28,392 1 one row per single transit of a signal: time, depth, points in transit, SNR
manifest.parquet 1,000 1 one row per sampled star: download and parse status, rows read and kept, timings
calibration.parquet 1,123 1 positive and negative controls and injection-recovery trials (synthetic transits added to real light curves)
threshold.parquet 14 1 summary of the best SNR found in inverted (flipped) light curves, where any dip is noise
quality_bits.parquet 36,318 1 rows per star, orbit and QUALITY flag bit (counts only; bit meanings are not stated)
cadences.parquet 9,908,365 101 full-resolution light curves of the stars with usable rows (200 s cadence), sorted by tic_id then time
cadences_10min.parquet 3,463,467 36 the same light curves averaged into 10-minute bins (n_points = cadences averaged), same ordering

No column held a local file path, so none was removed. MANIFEST.csv lists each file with its bytes, rows, row groups, SHA-256 and the sum of each numeric column (col_sums).

Schema

Column types are read from the files in this repository, not typed by hand. Times are BTJD (BJD - 2457000, days). "Empty" means a missing value.

stars.parquet

Column Type Meaning
tic_id DOUBLE TESS Input Catalog (TIC) identifier of the star
status VARCHAR download and parse status of the star's light-curve file (ok = read)
tessmag DOUBLE TESS magnitude (brightness in the TESS band; smaller is brighter)
radius DOUBLE stellar radius in solar radii, from the QLP file header (empty when unknown)
teff DOUBLE effective temperature in kelvin, from the QLP file header (empty when unknown)
camera DOUBLE TESS camera (1 to 4) that observed the star
ccd DOUBLE detector (1 to 4) within the camera
ra DOUBLE right ascension in degrees
dec DOUBLE declination in degrees
search_status VARCHAR outcome of the transit search: ok, no_good_rows (no usable rows), or the reason it was skipped
n_binned INTEGER number of binned points the search used (empty when not searched)
noise_ppm DOUBLE scatter of the binned flux in parts per million (robust standard deviation; empty when not searched)
flipped_best_snr DOUBLE best SNR found on the flipped (inverted) light curve, where any dip is noise
best_snr DOUBLE SNR of the strongest search peak (empty when no peak)
n_signals INTEGER number of search peaks kept for the star (rows in signals; 0 when none)
cad_offset DOUBLE byte position where the star's file starts in cadences_by_star.bin (empty when it has none)
cad_bytes DOUBLE length in bytes of the star's file in cadences_by_star.bin (empty when it has none)
cad_rows DOUBLE number of rows (cadences) in the star's file in cadences_by_star.bin (empty when it has none)

signals.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
rank INTEGER peak rank within the star, 1 = strongest
period DOUBLE orbital period of the peak in days
t0 DOUBLE mid-time of a transit, in BTJD (BJD - 2457000, days)
duration DOUBLE transit duration in days
depth DOUBLE fractional dip in flux during transit
depth_err DOUBLE uncertainty of depth
snr DOUBLE signal-to-noise ratio (SNR) of the depth
power DOUBLE box least squares (BLS) power at the peak
depth_odd DOUBLE depth from odd-numbered transits only
depth_even DOUBLE depth from even-numbered transits only
odd_even_sigma DOUBLE difference of odd and even depths in standard errors (empty when either is undefined)

events.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
rank INTEGER rank of the signal this transit belongs to (matches signals.rank)
time DOUBLE predicted mid-time of the single transit, in BTJD (days)
depth DOUBLE fractional dip in flux measured in that transit (empty when no data fall in it)
n_in INTEGER number of binned points inside the transit window
snr DOUBLE SNR of that single transit's depth (empty when no data fall in it)

manifest.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
status VARCHAR ok, download_failed or parse_failed
bytes DOUBLE size in bytes of the downloaded light-curve file
rows_total INTEGER rows in the light-curve file
rows_kept INTEGER rows kept (QUALITY == 0 and finite time, flux and flux error)
seconds DOUBLE total seconds spent on the star
download_seconds DOUBLE seconds spent downloading
reduce_seconds DOUBLE seconds spent reading and reducing
tessmag DOUBLE TESS magnitude
radius DOUBLE stellar radius in solar radii (empty when unknown)
teff DOUBLE effective temperature in kelvin (empty when unknown)
camera DOUBLE TESS camera (1 to 4)
ccd DOUBLE detector (1 to 4) within the camera
message VARCHAR error text when the download or parse failed

calibration.parquet

Column Type Meaning
kind VARCHAR positive_control, negative_control, injection_recovery (synthetic transits added to a real light curve) or flipped (search on the inverted light curve)
tic_id DOUBLE TIC identifier of the star
inj_period DOUBLE period of the injected transit in days (empty for flipped rows)
inj_depth DOUBLE fractional depth of the injected transit (empty for flipped rows)
best_period DOUBLE period of the strongest peak the search then found, in days
best_snr DOUBLE SNR of that peak
recovered BOOLEAN TRUE when the found period matched the injected one within tolerance (empty for flipped rows)
seconds DOUBLE seconds the search took

threshold.parquet

Column Type Meaning
metric VARCHAR name of the summary statistic (counts, quantiles and maxima of best SNR)
value DOUBLE value of the statistic

quality_bits.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
orbit_id INTEGER orbit of the spacecraft the rows belong to
orbit_source VARCHAR where the orbit came from: ORBITID (file column) or time_gap (derived from gaps in time)
kind VARCHAR what is counted: rows_total, bit (one QUALITY flag bit), or another dropped-row category
bit INTEGER QUALITY flag bit number (empty unless kind is bit)
n_rows_with INTEGER rows in the orbit counted under this kind (for a bit: rows with that bit set)
n_rows_only INTEGER for a bit: rows where it is the only bit set (empty otherwise)
bit_meaning VARCHAR meaning of the bit (empty: the QLP file does not state one)
meaning_source VARCHAR source of the stated meaning (empty when none)

cadences.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
time DOUBLE time of the observation in BTJD (BJD - 2457000, days)
flux DOUBLE detrended QLP flux (DET_FLUX column) as delivered, not re-normalised
flux_err DOUBLE uncertainty of flux (DET_FLUX_ERR column)
quality INTEGER QLP QUALITY flag; always 0 here, because only QUALITY == 0 rows are kept

cadences_10min.parquet

Column Type Meaning
tic_id DOUBLE TIC identifier of the star
time DOUBLE mean time of the cadences in the bin, in BTJD (days)
flux DOUBLE mean flux of the cadences in the bin
flux_err DOUBLE uncertainty of the binned flux
n_points INTEGER number of cadences averaged into the bin

cadences_by_star.bin

Not a parquet file itself: each star is its own zstd-compressed parquet file with the columns time, flux, flux_err (same meaning as in cadences.parquet), laid end to end in ascending tic_id order. In stars.parquet, cad_offset, cad_bytes and cad_rows give a star's byte range and row count; the 179 stars with search_status no_good_rows have no range (those three columns are empty). Rows within a star are ordered by time.

Column Type Meaning
time DOUBLE time of the observation in BTJD (BJD - 2457000, days)
flux DOUBLE detrended QLP flux (DET_FLUX column) as delivered, not re-normalised
flux_err DOUBLE uncertainty of flux (DET_FLUX_ERR column)

Read one star

-- DuckDB
SELECT time, flux, flux_err
FROM 'hf://datasets/JohnGavin/tess-qlp-s0098-prototype/cadences.parquet'
WHERE tic_id = 201374538
ORDER BY time;

Read one star in a single request

cadences_by_star.bin holds each of the 821 stars' light curves (time, flux, flux_err) as its own small parquet file, laid end to end in tic_id order. In stars.parquet, cad_offset is where a star's file starts, cad_bytes its length in bytes and cad_rows its row count (all three are empty for a star with no usable rows). A reader takes those two numbers from the stars table it already has and sends one HTTP request, Range: bytes=<cad_offset>-<cad_offset + cad_bytes - 1>, then opens the returned bytes as a parquet file. It saves the several round trips a reader needs to find one star inside cadences.parquet. The rows are the same as in cadences.parquet; the file is 80,988,209 bytes.

Sources and credit

License

The source light curves are public data from NASA's Mikulski Archive for Space Telescopes (MAST). This derived prototype adds no restriction of its own; follow the citation request on the MAST page above.

Downloads last month
67