CDO

hi Daniel again!

how does the result look like? What exactly did you call?

Hi!

What can you suggest to me to solve the problem.
He would be:

cdo lowpass, 1/408 sst.anomaly.nc out.nc?

Well I don't even know, what your threshold frequency should be. How could I suggest you anything reasonable? the related frequeny for a 10-year-peroid is 0.1

hth
ralf

Hi! Thank you very much.

How can I know about my maximum frequency. My time in the series is 408 times.I think your answer is correct, as applied 0.1 I suppress all frequencies less than 10 years old.

... or it would be the following: I have 408 months, proportionally, for 34 years. I would like to observe frequencies of 10 years (PDO), therefore: 120 months. Would it be 1/120?

the filters take into account the real time of the time axis. That's why the number of timestamps is not important as long as you have enough for properly doing an frequency analysis at all. Monthly output for a once-every-10-years analysis should be fine. As stated in the documentation frequency notation of these operators is per-year. This means

Another example can be found here

Thank you very much! I don't know how to thank you. I'm under a lot of pressure and your help is essential.