Tycho-2 sorted counts ====================================================================== This is a short description of the Tycho-2 sorted (identified) counts. These data have been recovered from a set of DVDs held by C. Fabricius and V. Makarov and have been reformatted before being served here. Introduction: The original data set consisted of 2176 binary files, each containing (up to) 2000 stars. The files were named s_1000 up to s_3175 and each file contained the raw, folded counts for a number of stars, as follows: File s_1000: star 1-1999; File s_1001: star 2000-3999; .... File s_3175: star 4350000-4351323. In total, the files contained 574,915,849 records for 3,325,148 stars. Stars included Tycho stars, Hipparcos stars (Hp < 12.5 mag), empty fields, marginal cases, etc. The records in the original binary files had a fixed length of 88 bytes: 12 bytes for the header subrecord; 12 bytes for the parasites subrecord; 64 bytes for the data subrecord. The reformatted data that is being served here have been decoded from the original binary format to ASCII integer format and the various subrecords have been split. The subrecords can be joined using the unique keys "StarNumberInTIC2" as well as "TransitIndex". TIC2 refers to the second version of the Tycho Input Catalogue that has been used for the Tycho-2 processing. This catalogue is (as yet) unpublished but is described in Section 3.1 of Høg et al. (2000, A&A, 357, 367; https://ui.adsabs.harvard.edu/abs/2000A%26A...357..367H/abstract). Explanatory remarks: For each TIC2 star, up to 3 parasites are recorded (if both fields of view of the spacecraft are at galactic latitude b = 0 deg, there is a probability of only about 0.001 for having more than 3 parasites; at star densities twice as high, this probability is larger but still only about 0.01). The parasites are the (at most) 3 stars, other than the main star, with the brightest magnitude in V_T. Parasites are recorded with the brightest one first and in sequence of magnitude. Any parasite subrecords not containing a parasite are set to zero. The parasites have been selected among the predicted group crossings (PGCs) of those high-quality stars from HIP, TYC, or PPM that have been included in the TIC2 catalogue used in this work, i.e., from the about 1 million brightest stars in the sky. PGCs within an interval of 31 + 8 = 39 samples = 11.0 arcsec have been taken into account, centred on the 31 photon-count samples in the record. The 31 samples correspond to 51.7 ms = 8.72 arcsec. A transit can be fully utilised for the estimation if the transit time occurs about 5 samples inside the record due to the widths of the single-slit response functions (SSRFs). At a distance of 5 samples = 1.4 arcsec from the centre, the SSRFs of the vertical and inclined slits are, respectively, less than 0.03 and 0.10. The remaining interval of 31 - 10 = 21 samples = 5.9 arcsec implies that stars at a distance of 3.0 arcsec from the TIC2 position will be fully measured by all transits of the vertical slits and by 82 percent of the transits of the inclined slits. At a distance of 3.0 / sqrt(2) = 2.1 arcsec, the star will be fully measured by all transits. This is expected to be sufficient for at least 99 percent of the GSC stars since the precision of the GSC-1.2 catalogue at the plate epoch and at the Tycho epoch (J1991) is about 0.2 to 0.5 arcsec, according to a comparison with the Tycho Catalogue. A parasite subtraction can be implemented, based on the information of parasitic stars given in the parasite subrecords. Only parasites with non-extreme transits as well as reliable astrometry (position uncertainties less than about 200 mas) and reliable photometry (magnitudes uncertainties less than about 0.3 mag) are suited for subtraction; this corresponds to, respectively, objects with parasite flags F04 = 0 and F05 = 0. Parasitic subtraction shall be done after the sorting of transits so that all photon counts for a given TIC2 star are available for processing in the following sequence: Parasite subtraction: the effect of a recorded parasite may be subtracted in the B_T and V_T photon counts using the data in the parasite subrecord and the corresponding SSRF. These 'cleaned counts' may be produced for the sorted counts belonging to a given TIC2 star, just before they shall be used for the estimation of position and amplitude of the star. Parasite detection and recognition: a detection process should be carried out on the T counts (B_T + V_T) derived from the cleaned, sorted counts record. The detection process should be trimmed to detect peaks higher than about SNR = 2.0 and to estimate the transit time and approximate amplitude. The estimated values from all transits for the star should be used in a recognition process in order to locate the TIC2 star and possible other components within about 2 arcsec from the TIC2 position. The treatment of components remains TBD. Since subtraction of a parasite with a too bright magnitude will generate a negative peak in the counts and since the detection is carried out in the cleaned counts, the detection should be designed to detect both positive and negative peaks. Parasite rejection: by means of an approximate estimate of the position of the TIC2 star, it is possible to compute the transit time corresponding to this position within every photon record for the star. By means of the parasite records, it can be decided whether a given transit contains a parasite sufficiently close to the star and sufficiently bright that the transit shall be rejected before a new estimate of position and amplitude is carried out. Time intervals and magnitude limits are TBD, but may initially be adapted from those used in the first Tycho processing (1997). Transits having an uncertain parasite (F04 = 1) close to the actual transit of the TIC2 star should be rejected. Signal estimation: a background value is required for (1) the non-linear 4-peak folding, (2) the weighting of observations, and (3) the amplitude estimation. Two backgrounds are distinguished: a large-scale, format background and a local background. The format background, which has been determined from 6400 samples by a median method, is useful for the 4-peak folding. The local background has been determined by a similar method from 192 samples centred on the PGC and is useful for the weighting of observations (see ESA SP-1200 "The Hipparcos and Tycho Catalogues", Volume 4 "Construction of the Tycho Catalogue", Section 4.2 "Background Determination"; https://cosmos.esa.int/web/hipparcos/catalogues). In the first Tycho data reduction (1997), the local background was used mainly near very bright stars. However, the local background should preferentially be used (over the format background) as the 'actual' background for amplitude estimation because the uncertainty of a background from so many samples contributes very little to the uncertainty of a star amplitude which is in fact determined from about 4 * 4 = 16 samples representing the main peak. Therefore, nothing would be gained in precision by using the format background for the amplitude estimation although it is formally more precise because of the 6400 samples used for its determination. Header subrecord: "StarNumberInTIC2": Star number in Tycho Input Catalogue version 2 (TIC2; original label = SN). "PredictedGroupCrossingEpoch": Epoch of initial, predicted group crossing (TT-J1992.0; unit = 0.0001 yr; original label = PGCTT). "PredictedGroupCrossingPhase": Phase of refined epoch of predicted group crossing, PGCEP (0 = PGC at beginning of 16th (central) sample, 0.5 = PGC centred on central sample, 1.0 = PGC at end of central sample; unit = 0.005 sample; original label = PGCPH). "PredictedGroupCrossingFlags": Flags for predicted group crossing (original label = PGCFL). See details below. "PositionAngleWDirection": Position angle of scanning direction (w) for vertical slit (unit = 40 arcsec; original label = PANGW). "Scale": Scale, with 1 sample = 0.28125 arcsec * scale for vertical slits and 1 sample = 0.19887 arcsec * scale for inclined slits (unit = 0.0001; original label = Scale). Parasites subrecord: "StarNumberInTIC2": Star number in Tycho Input Catalogue version 2 (TIC2; original label = SN). "PredictedGroupCrossingParasiteFlag1": Flags for predicted group crossing of first parasite (original label = PARFL). See details below. "PredictedGroupCrossingParasiteFlag2": Flags for predicted group crossing of second parasite (original label = PARFL). "PredictedGroupCrossingParasiteFlag3": Flags for predicted group crossing of third parasite (original label = PARFL). "PredictedGroupCrossingParasite1DeltaEpoch": Relative epoch of first parasite coded as 150*(deltaEpoch+0.78), with epoch in units of 0.04 samples. PARX is relative to a time which is 19.0 samples before the PGC epoch of the TIC2 star, which is about 1.1 arcsec before the centre of the first sample in the counts record (unit = 150 * (deltaEpoch + (39/25)/2), with epoch in filing-key (FK) units with 1 FK = 25 samples; the range [0, 39/25 = 1.56] becomes [0, 234] when multiplying by 150; original label = PARX). "PredictedGroupCrossingParasite2DeltaEpoch": Relative epoch of second parasite coded as 150*(deltaEpoch+0.78), with epoch in units of 0.04 samples (original label = PARX). "PredictedGroupCrossingParasite3DeltaEpoch": Relative epoch of third parasite coded as 150*(deltaEpoch+0.78), with epoch in units of 0.04 samples (original label = PARX). "PredictedGroupCrossingParasite1BTmag": Tycho B magnitude B_T of first parasite (unit = 0.1 mag; original label = PARB). "PredictedGroupCrossingParasite2BTmag": Tycho B magnitude B_T of second parasite (unit = 0.1 mag; original label = PARB). "PredictedGroupCrossingParasite3BTmag": Tycho B magnitude B_T of third parasite (unit = 0.1 mag; original label = PARB). "PredictedGroupCrossingParasite1VTmag": Tycho V magnitude V_T of first parasite (unit = 0.1 mag; original label = PARV). "PredictedGroupCrossingParasite2VTmag": Tycho V magnitude V_T of second parasite (unit = 0.1 mag; original label = PARV). "PredictedGroupCrossingParasite3VTmag": Tycho V magnitude V_T of third parasite (unit = 0.1 mag; original label = PARV). Counts subrecord: "StarNumberInTIC2": Star number in Tycho Input Catalogue version 2 (TIC2; original label = SN). "FormatBackgroundBTchannel": Sqrt of Tycho B_T format background (determined from 6400 samples by a median method; unit = 0.01 sample; original label = BB). "FormatBackgroundVTchannel": Sqrt of Tycho V_T channel format background (determined from 6400 samples by a median method; unit = 0.01 sample; original label = BV). "CompressedSamplesBTchannel": Compressed Tycho B_T channel samples (b_samples = 1.035**SB - 1.0; original label = SB). "CompressedSamplesVTchannel": Compressed Tycho V_T channel samples (v_samples = 1.035**SV - 1.0; original label = SV). "PredictedGroupCrossingFlags" (original label = PGCFL): F00 = 0 always (most significant bit). F01 = 0 for lower branch of slits (z<0); = 1 for upper branch of slits (z>=0). F02 = 0 for following field of view; = 1 for preceding field of view. F03 = 0 for vertical slit group; = 1 for inclined slit group. F04 = 0 for non-extreme transit on the slits; = 1 for extreme transit on the slits. A transit is deemed extreme either if it is close to the ends of the vertical slits, i.e., if (F03 = 0 AND abs(z) > 1200 - 3 arcsec), or if it is close to the apex of the inclined slits, i.e., if (F03 = 1 AND abs(z) < 3 arcsec). This condition is met for about 1.5 * 6 / 1200 = 0.75 per cent of the recorded parasites. F05 = 0 if the star is a TYC star suited to be subtracted as parasite (~980,000 entries); = 1 if it is not suited to be subtracted as parasite since being dubious and/or photometrically variable and/or unresolved double (= F09 in TIC2; ~70,000 entries). The following TYC flags have been used: T47 = non-blank, previously known variables (~6600); T48 = U or V, variability indicated in Tycho measurements (~33,000); T49 = D, duplicity clearly indicated in Tycho measurements (~8400); and T57 = non-blank, a dubious entry (~37,000). F06 = 0 if the source of stellar position is <3 (=F17-F19 in TIC2), i.e., high-quality data from HIP, TYC, or PPM; = 1 otherwise (low-quality data). F07 = 0 always (least significant bit). "PredictedGroupCrossingParasiteFlag1" (original label = PARFL): F00 = PGCFL F00 of the parasite. F01 = PGCFL F01 of the parasite. F02 = PGCFL F02 of the parasite. F03 = PGCFL F03 of the parasite. F04 = PGCFL F04 of the parasite. F05 = PGCFL F05 of the parasite. F06 = 0 always. F07 = 0 if no parasite is recorded; = 1 if a parasite is recorded. ==========================================================================================