Color Dependence

Another interesting issue to address is whether the difference between North and South includes any systematic color dependence. To answer this question I analyze the distribution of residuals as the function of color (e.g. J-K).

The residuals as the function of Northern color are shown in Figure 9. Linear fit to the relation gives the slope of $-0.001\pm 0.006$. Since color of a source does not correlate with the cross-scan position of the source (see e.g. Figure 10), i.e. all three bands are similarly affected by the non-optimal PSF photometry, this result is not contaminated by the systematic difference at the left edge of the calibration scans. Thus, there seems to be no systematic difference in the North and South solutions due to source colors.