Project

General

Profile

Issues with monmin not being calculated correctly (or how I expect it to)

Added by Stone Chen about 2 years ago

Hi all,

I'm attempting to calculate the monpctl, of CMIP6 climate data.

The file is CMIP6.CMIP.AS-RCEC.TaiESM1.historical.r1i1p1f1.day.pr.gn.v20200626.nc

I've selected out June 1850 as an example.

-1 :       Date     Time   Level Gridsize    Miss :     Minimum        Mean     Maximum : Parameter ID
1 : 1850-06-01 12:00:00 0 55296 0 : -8.3448e-18 2.7635e-05 0.0020175 : -1
2 : 1850-06-02 12:00:00 0 55296 0 : -1.3227e-17 2.8597e-05 0.0013684 : -1

However when I compute the monmin the minimum above (-8.3448e-18) does not appear in the minimum below (-1.8718e-17)

system("cdo info -monmin june_1850_cmip.nc")

cdo(1) monmin: Process started
-1 : Date Time Level Gridsize Miss : Minimum Mean Maximum : Parameter ID
1 : 1850-06-16 00:00:00 0 55296 0 : -1.8718e-17 1.1401e-06 0.00012135 : -1

I'd expect these to match, and I suspect this is causing missing values to occur downstream.

Is this correct?

image.png (145 KB) image.png

Replies (2)

RE: Issues with monmin not being calculated correctly (or how I expect it to) - Added by Stone Chen about 2 years ago

Sorry I just realized this could be issues with floating point errors?

RE: Issues with monmin not being calculated correctly (or how I expect it to) - Added by Stone Chen about 2 years ago

Okay, never mind, I realized after getting some sleep I was going insane.

    (1-2/2)