INNAME UV data (name). INCLASS UV data (class). INSEQ 0.0 9999.0 UV data (seq. #). 0 => high INDISK 0.0 9.0 Disk unit #. 0 => any STOKES Stokes' parameter - only 1 BIF 0.0 100.0 First IF to consider, 0=>1. EIF 0.0 100.0 Last IF to consider BCHAN 0.0 First channel to average ECHAN 0.0 Last channel in average FLAGVER Flag table version, 0->high ANTENNAS Antenna numbers to include or omit. OPCODE 'INDE' -> independent triangles/quadrangles else all. SOLINT Time interval (minutes) over which to average to form closure phases and amplitudes

EVACL Type: Task Use: Examines the statistics in a UV data set which should be a single-source only. The task forms closure phases and closure amplitudes over averages of SOLINT minutes and then looks at the mean and rms of these over all times and all triangles/quadrangles. The results are printed as a function of IF. Tasks CLPLT and CAPLT will do this for selected triangles/quadrangles plotting as a function of time. Closure amplitude in the quadrangle of antennas a, b, c, and d is defined as LN [ (amp_ab * amp_cd) / (amp_ac * amp_bd) ] which should be independent of any antenna-based instrumental gains. Closure phase in the triangle of antennas a, b, and c is defined as Phase_ab - Phase_ac + Phase_bc which should be independent of any antenna-based instrumental phases. Since these apply to the RR and LL, closure amplitude and phase are only meaningful in STOKES 'RR' or 'LL', not in STOKES 'I' or any other compund Stokes parameter or cross-polarization. EVACL will allow you to compute these, but they are not expected to be meaningful. Closure amplitude and phase are not zero when there is real source structure. Therefore, you might wish to divide your data by your model with OOSUB or UVSUB, and then compute the closure amplitudes and phases from the resulting "gains" which ideally act like a point source at the origin (if your model is really wonderful). EVAUV does this. Closure information is produced by the following tasks: CLPLT Plots closure phases for individual triangles as a funtion of time. CAPLT Plots closure amplitudes for individual quadrangles as a funtion of time. EVACL Computes the RMS of closure phase and amplitude as a function of IF for an individual Stokes. EVAUV Computes two data sets: data-model and data/model. The closure RMSes of the latter may be computed as functions of polarization and IF. Other statistics are computed and plotted. CLOSE Computes the RMS closure phase or amplitude as a function of spectral and IF channel. A plot is produced and a text file may be written. Adverbs: INNAME.....UV file name (name). Standard defaults. INCLASS....UV file name (class). Standard defaults. INSEQ......UV file name (seq. #). 0 => highest. INDISK.....Disk unit #. 0 => any. STOKES.....Specifies the single Stokes type to be computed. Recognized values are 'I', 'Q', 'U', 'V', 'RR', 'LL', 'LR', 'RL', 'VV', 'HH', 'VH', 'HV'. ' ' => 'I'. BIF........Lowest IF to consider. 0 -> 1 EIF........Highest IF to consider. 0 -> highest FLAGVER....specifies the version of the flagging table to be applied. 0 => highest numbered table. <0 => no flagging to be applied. ANTENNAS...A list of the antennas which if all > 0 are to be included or if, any one or more are < 0, are to be omitted. All 0 -> all included. OPCODE.....'INDE' -> select only independent triangles, e.g. 1,2,3 then 2,3,4 then 3,4,5, etc. else -> select all triangles/quadrangles SOLINT.....The time interval (minutes) over which to average the data in order to form the closure phase and amplitude. It is possible, with editing and especially on correlators which only correlate a subset of the available stations simultaneously that the time tags associated with visibility values on different baselines differ. Therefore one has to specify a time interval over which to average the data to form the closure phase and amplitude. If this is not necessary with your data, set SOLINT <= 0.9 * data integration time. You should experiment a little with this parameter. If the fringe rate for the source is fairly rapid, then it is very important to have the sampling be the same on all baselines in each triangle. Otherwise, samples at slightly different times will be combined but the phases and amplitudes will not cancel properly. Use of a very small value of SOLINT (but > 0) will cause inadequately aligned data to be discarded in order to avoid misleadingly high closure phases and amplitudes. If SOLINT < 0, the task will form a closure phase/amplitude as soon as it has accumulated at least one sample for each of the three/four baselines in the triangle/quadrangle.