Data taken before February 4, 1998

For data taken before February 4, 1998, the value of T_HOT was not recorded. Instead, T_AMB (the ambient temperature in the telescope dome) was stored in the data header. In order to reduce skydips taken prior to this date, we have attempted to determine the relationship between T_HOT and T_AMB (see Figure 8).

Figure 8: Correlation between T_AMB and T_HOT. Black data points show the recorded values of T_HOT and T_AMB for every skydip taken after February 4, 1998. The red line is the locus of T_AMB=T_HOT.

The relationship seems to be linear, and is almost a 1:1 correspondence. However, there is considerable scatter, which may be due to a thermal time lag between T_AMB and T_HOT. The extent of the scatter is more than a few degrees Kelvin, enough to make a 450-µm skydip fit or fail. Furthermore, the scatter is larger than the corrections applied to T_HOT. Thus, we cannot use the T_HOT/T_AMB correlation to determine T_HOT and accurately reduce the skydips taken during this period.

For data taken before February 4, 1998, it is suggested that people use the revised narrowband pre-upgrade CSO Tau relations. Where the CSO Tau monitor was not working, there is only one option - reduce the 850-µm skydip using T_AMB (the 850-µm skydips are more impervious to errors in T_HOT) and then use the Tau₄₅₀-Tau₈₅₀ relation to derive Tau₄₅₀.