search

cmip6dr / Request

Request specification (which variables for which experiments, etc)
2 stars 0 forks source link

DAMIP 42 year and core 35 year time slices #49

Closed martinjuckes closed 7 years ago

martinjuckes commented 7 years ago

Hello Karl (@taylor13),

DAMIP have asked for data from historical and perturbed historical simulations (e.g. hist-sol) for the period 1979 to 2020. For the historical simulation this is split into a request for 36 years (1979 to 2014) from historical and 6 years from historical-ext. There is also a CMIP request for 35 years of data (1980 to 2014), applied to a partially overlapping set of variables.

I suspect that the DAMIP request for data from 1979 is associated with the start of a significant range of observation datasets.

Should we adjust the CMIP request to use a 36 year slice instead of 35?

taylor13 commented 7 years ago

Not sure I understand. These runs are supposed to all be initiated in 1850, I think. Is certain data requested only for the last years? Which frequencies? For CMIP-requested variables for years 1980-2014, I don't mind modifying that to 1979-2014.

martinjuckes commented 7 years ago

@taylor13 : Sorry, my mistake: the request for 35 years of data comes from C4MIP, LS3MIP and PMIP (it is for 3 hourly, daily and some monthly model level data), the 36 year request from CFMIP is for 3 hourly data.

Yes, the experiments start in 1850 in all cases.

I'll check with the MIP co-chairs, but recommend standardising on 36 years.

taylor13 commented 7 years ago

Sounds good. They were probably trying to follow the CMIP5 example to start in 1980 and may not mind adding a year.

martinjuckes commented 7 years ago

CFMIP request extended to 1979-2015