The 2MASS level 1 documentation requires measurement of bright stars
with the following accuracy
Photometric bias (zeropoint drift) <2% for stars with K > 4.0 magnitude
Average repeatability better than 5% for sources at K=8.0 magnitude.
Night to night zeropoint calibrations should agree to within +/-4% of the
mean calibration.
Several areas of the sky have been scanned on two or more occasions, by both North and South telescopes, as a test of photometric repeatability
The above comparisons are based on the psf magnitudes, which become unreliable at brighter than 8th magnitude. The plot shown below compiles default magnitude data for the brightest stars in tiles 900-907, including read 1 magnitudes for < 8th magnitude. The comparison involves observations from 13 Jan, 1998 (North) and 18 Nov, 1998 (South). The table gives the mean offset and rms as a function of magnitude. Given that these data originate from a single night (each), the values are within the formal specification.
| Mag | J | J |
NJ | H | H |
NH | K | K |
NK |
| 6.25 | 0 | 0 | 0 | 0.036 | 0.039 | 3 | 0.019 | 0.018 | 8 |
| 6.75 | 0 | 0 | 0 | 0.040 | 0.031 | 18 | 0.015 | 0.017 | 20 |
| 7.25 | 0.048 | 0.026 | 14 | 0.028 | 0.028 | 39 | 0.021 | 0.030 | 44 |
| 7.75 | 0.052 | 0.025 | 34 | 0.027 | 0.035 | 54 | 0.016 | 0.045 | 69 |
| 8.25 | 0.044 | 0.043 | 55 | 0.017 | 0.051 | 93 | 0.017 | 0.034 | 106 |
| 8.75 | 0.046 | 0.032 | 83 | 0.021 | 0.039 | 148 | 0.013 | 0.035 | 163 |
mag is the mean offset, in 0.5 magnitude bins
mag is the rms dispersion about the mean. The three plots illustrated below show the residuals between 2MASS
observations and literature photometry (from Leggett, 1992, ApJS 82, 351)
for a subset of nearby stars from the Gliese/Jahreiss CNS3.
The rms uncertainties are <0.03 magnitude, within specification, with no
indication of systematic variation with magnitude.
The same data are used to check for colour terms at this reference. There is no evidence for a significant colour term in any of J, H or KS.