As someone who has used SAS for many, many different projects: it is terrible, vastly inferior to Pandas or R, and the only reason to ever use it is when you're forced to. Even simple stuff like functions that operate on data have to be hacked on with macros.

Case in point, your production SAS code could be replaced with this Pandas code (and the R code would look very similar):

  temperatures[[TEMPERATURE_COLUMNS]].apply(lambda t: (t > 160).sum(), axis=1)

or if your data is in proper long form

  data.groupby('time').temperature.gt(160).sum()