The total intensity of a source should appear at in transform space. Since, however, two antennas cannot be placed with zero separation, the "zero spacing flux" cannot be measured. It can be estimated analytically for simple shapes (by fitting the uv distance versus amplitude) or, less reliably, by the total amount of flux put into the map by the CLEAN algorithm. CLEAN can be given a zero spacing flux which it will attempt to distribute over the image. Beware of using a ZEROSP for TB sorted data in MX, however, since the flux is always inserted as a Gaussian at the phase center.
ZEROSP is greatly limited in its use. Therefore, be sure to use a model containing all the expected flux. Do not rely on ZEROSP to insert a large amount of flux.
Choosing the "weight" for ZEROSP is difficult. The best estimate seems to be the number of missing cells around the origin of the gridded uv plane (typically 10-20). Provided that the CLEAN goes deep enough, however, results are fairly insensitive to the weight.
References
Cornwell, T. and Braun, R. "Deconvolution." Ch. 8 in Synthesis Imaging in Radio Astronomy: Third NRAO Summer School, 1988 (Ed. R. A. Perley, F. R. Schwab, and A. H. Bridle). San Francisco, CA: Astronomical Society of the Pacific, 1989.
