I checked on the number of days / leap year issue in the QDC-CMIP6 data.
The global warming level data is fine because that uses a 20 year period and by definition any 20 year period has 5 leap days.
The base period for the time slice data is 1985-2014, which has 7 leap days (or 10,957 days of data). The 2070-2099 period also has 7 leap days, so that's fine. The 2035-2064 period has 8 leap days, so when I produce 10,957 days of scaled QDC data it runs from 2035-01-01 to 2064-12-30 (i.e. no 31 Dec in 2064).
I don't think it's worth me re-running all the 2035-2064 data to produce one extra day. I think the way to think about our time slice datasets is that we take 30 years (i.e. 10,957 days; 1985-2014) of data from the historical period and scale those to produce 10,957 days of future data starting on 2035-01-01 or 2070-01-01.
I checked on the number of days / leap year issue in the QDC-CMIP6 data.
The global warming level data is fine because that uses a 20 year period and by definition any 20 year period has 5 leap days.
The base period for the time slice data is 1985-2014, which has 7 leap days (or 10,957 days of data). The 2070-2099 period also has 7 leap days, so that's fine. The 2035-2064 period has 8 leap days, so when I produce 10,957 days of scaled QDC data it runs from 2035-01-01 to 2064-12-30 (i.e. no 31 Dec in 2064).
I don't think it's worth me re-running all the 2035-2064 data to produce one extra day. I think the way to think about our time slice datasets is that we take 30 years (i.e. 10,957 days; 1985-2014) of data from the historical period and scale those to produce 10,957 days of future data starting on 2035-01-01 or 2070-01-01.