As shown in earlier sections, the merging of 2MASS data with data from the POSS-I gives a long baseline in color useful for object discrimination. To study the photometric characteristics of a typical piece of 2MASS sky, we have selected a random field from the protocamera data. This field was selected to satisfy two basic requirements: (1) The area was surveyed in J, H, and K; (2) the area is far enough from the galactic plane so that confusion is not an overwhelming concern. A piece of sky was selected having RA between and and Dec between and (, ). This areal coverage (0.05 square degree) was chosen to be small to keep the source count low; the band merging and APM matches for this preliminary data set were carried out by hand. (This process is now being automated.)
Figure 4: Pseudo-color plot designed to pull the quasar locus even farther from the stellar locus. Symbols and photometry are the same as in Figures 2 and 3.
In total there were 148 repeated sources (observed in at least two bands) in this piece of 2MASS data. Of these, 138 matched with objects found in the APM data base. The histogram of positional differences between the APM and 2MASS data sets is shown in Figure 5. Note that the majority of objects matched to within 1 arcsecond, with a tail extending out to 4 arcseconds. The combined J, H, K, and R data is plotted in Figures 2, 3, and 4 for comparison to results discussed earlier.
Of the 10 2MASS objects which did not match with sources in the APM database, 8 of these were found to lie within 5 to 20 arcseconds of a brighter source on the 2MASS images. Although both sources were properly extracted by the pipeline software, these same pairs were blended in the APM data set. Therefore, the difficulty in matching these objects is attributable to an APM, not a 2MASS, limitation. We are currently seeking a way to detect and flag such blends during the APM/2MASS data merge.
The other 2 objects which did not have a counterpart in the APM data set are presumably very red objects invisible at R but red enough to appear at K in the 2MASS survey. Inspection of the APM finder charts shows no single or blended source at the position of either object. The first source has K = 14.95 and H-K = 0.69. Assuming that the APM is complete to a magnitude of R = 19.00, this sets a limit of R-K > 4.05 for this object. The second source was detected in all three bands --- K = 14.96, H-K = 0.75, J-K = 1.20 --- with a limit of R-K > 4.04. A comparison to Figures 2 and 3 shows that the colors of these objects are consistent with both a stellar and a quasi-stellar interpretation due to their weak limit on R-K. Nonetheless, these objects appear to be interesting, bona fide sources.
Figure 5: Difference between the APM and 2MASS positions for all 138 matched objects in the random field.