search

usda-ars-ussl / fluxpart

Python module for partitioning eddy covariance flux measurements.
Other
31 stars 13 forks source link

failure to find a partition in neutral atmospheric conditions? #8

Closed BernardHeinesch closed 3 years ago

BernardHeinesch commented 3 years ago

Hello Todd,

Bernard Heinesch, working in Belgium and PI of flux tower stations. I’m supervising a master thesis on comparing H2O flux partitioning methods on an flux tower station (Lonzée crop station in Belgium, part of ICOS). Together with the student (Clara Cravatte), we ran FLUXPART on a winter wheat crop season and are facing the problem of having only low % of successfully partitioned half-hours by the tool, hampering data analysis.

Surprisingly, for neutral daytime half-hours, FLUXPART is not able to perform the partition, outputting mostly the message “wue>Fc/Fq; Fc/Fq<sigc/sigq/pqc;” (or only one of the two criteria) and a wavelet decomposition of 1/15. Once z/L go to more unstable atmospheric conditions, the % of processed data increases dramatically. For instance, sticking to ten days on growing wheat (nice H2O and CO2 fluxes, nice weather), we see 8 days without any successful day-time partitioning and only two interspersed less windy days with successful partitioning.

The site is close to “ideal” in the sense of micrometeorology. Almost flat, good fetch, cospectra very close to those predicted by Kaimal & Finnigan, no sign of important low-frequency content. We checked carefully all data inputs and everything looks OK (proved by successful outputs on unstable days). FLUXPART parametrization seems also OK and our reading of linked papers does not give a hint of why statistics of partitioning should drop in these atmospheric conditions.

Does this mean anything to you ? We can easily produce some plots to illustrate our case or share the FLUXPART parametrization.

Many thanks in advance

Bernard

thskaggs commented 3 years ago

Hi Bernard,

I received your email on this question as well. I'll respond there.

BernardHeinesch commented 3 years ago

Great and thanks ! Indeed, we probably miss something and our question may lead to confusion on the forum so better by email. I let Clara send the needed info by email then.

Best

De : Todd Skaggs @.*** Envoyé : jeudi 27 mai 2021 19:29 À : usda-ars-ussl/fluxpart Cc : BernardHeinesch; Author Objet : Re: [usda-ars-ussl/fluxpart] failure to find a partition in neutral atmospheric conditions? (#8)

Hi Bernard,

I received your email on this question as well. I'll respond there.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/usda-ars-ussl/fluxpart/issues/8#issuecomment-849810097 , or unsubscribe https://github.com/notifications/unsubscribe-auth/AUGNS4JE5V35CRHANNKXMH3TPZ6L5ANCNFSM45NNBDCQ .Image supprimée par l'expéditeur.