Heatwave
Added by Francis Tom about 1 year ago
Dear Karin,
Based on your response and codes you provided here (https://code.mpimet.mpg.de/boards/1/topics/13125), I would like to ask:
1. Can I compute a heatwave using "eca_hwfi" if my input data is "daily maximum temperature" instead of the default daily mean temperature? Like replicate the steps in your response but using daily maximum temperature as input.
2. Alternatively, if I use "eca_tx90p," whose input data is the daily maximum temperature, and I do the following:
cdo -timpctl,90 daily_tmax.nc -timselmin,5 daily_tmax.nc -timselmax,5 daily_tmax.nc tmax_90pctl.nc (90th percentile of daily maximum temperatures of a five-day window centered on each calendar day)
Do I need to add "5" in front of the percentile also (cdo -timpctl,90,5)?
"compute heatwave_event"
cdo eca_tx90p,3 daily_tmax.nc tmax_90pctl.nc heatwave.nc
Does adding "3" in front of "eca_tx90" imply I am considering at least three consecutive days? If no, how do I compute for at least three consecutive days in "eca_tx90p"
"eca_tx90p" unit is percent, can this be converted to "number of events"?
3.In your response, you wrote this "cdo -div $tmpfile -gt $rhumfile $rhum66 $HOME/Downloads/heatwave_days_rhum_gt_66pctl.nc" why did you use "div $tmpfile"?
Thank you so much
Replies (2)
RE: Heatwave - Added by Karin Meier-Fleischer about 1 year ago
Hi Francis,
1. + 2. I don't know if it makes sense to use the eca_hwfi ECA operator with the maximum data instead the mean when the 90th percentile is computed with the input of minimum and maximun data. Same for the second question.
The timselpctl operator has an additional argument npts but not eca_tx90p nor timpctl, see ECA and CDO documentation.
3. The result of using the operator -gt is a mask that has to be applied on the input data, here with -div operator to generate/keep missing values.