Project

General

Profile

Forums » Support »

A question about the operator "ECAHWDI -- Heat wave duration index wrt mean of reference period"

Added by Lei Chu about 6 years ago

Hello,
I'm trying to calculate the HWDI of an exact time period(now i have the daily maximum temperature data of this period, and another max. temp. data for reference of another 30 years). Supposed infile1 is the maximum temperature of the period i want to calculate, what method should i use to build infile2? Is it compute "TXnorm" by myself according to the user guide? If my infile1 is 30 days long, how long should infile2 be?
And if you know where the HWDI calculating formula can be referenced, please get me accessed. THANK YOU VERY MUCH:)


Replies (2)

RE: A question about the operator "ECAHWDI -- Heat wave duration index wrt mean of reference period" - Added by Karin Meier-Fleischer about 6 years ago

Hi Lei,

what about ecahwdi?

Search ecahwdi in the ECA documentation: [[https://code.mpimet.mpg.de/projects/cdo/embedded/cdo_eca.pdf]]

ecahwdi[,nday[,T]] infile1 infile2 outfile

nday  INTEGER   Number of consecutive days (default:  nday = 6)
T     FLOAT     Temperature offset (unit: degC; default:  T = 5degC)

-Karin

RE: A question about the operator "ECAHWDI -- Heat wave duration index wrt mean of reference period" - Added by Lei Chu about 6 years ago

Hi Karin,
thanks for your attention! I have read it on the documentation already, but instead of how the ECA_HWDI program processes, what I'm caring about more, literally, is where this program's algorithm is from, or who came up with this algorithm. If you could tell me about this, I'll be very appreciated.

I have tested ECA_HWDI in recent days, using the same infile1 and different infile2, and the outfiles I got finally differ. here is my testing record:

| number | infile1 | timesteps of infile1 | infile2 | timesteps of infile2 | outfile |
1 test.nc 2010-04-01/2010-04-30 txn1.nc 2000-03-30/2000-05-02 out1.nc
2 test.nc 2010-04-01/2010-04-30 txn2.nc 2000-04-01/2000-04-30 out2.nc
3 test.nc 2010-04-01/2010-04-30 txn3.nc 2000-03-30/2000-04-28 out3.nc

Interestingly, the info shows out1.nc is all the same as out3.nc. Can we interpret this as that eca_hwdi cannot read infile2's timestep information? And if yes, does it mean that I have to build the TXnorm set by myself as infile2, to make the program calculate in its correct way?

    (1-2/2)