[Reddyproc-users] inspecting uStar threshold estimation

[Thomas Wutzler]

Hello Efrén,

You can inspect whats going on in the u-Star threshold estimation by
plotting the season-temperature classes (relationship NEE ~ uStar).
See: ?sPlotNEEVersusUStarForSeason (also used in the REddyProc Example)

If there is not enough data or if there is no clear threshold in the
relationship, the procedure returns NA.
You can try changing seasons:
- Longer periods if data scarcity is a problem
- shorter periods, if you have temporal changes in the NEE-uStar
relationship, such as after changing the surface roughness at harvest

Yours,
Thomas

On 10.05.2016 18:19, Efren Lopez Blanco wrote:

> I have attached in the txt file 4 years, all are working except 2011. If
> I only take 2011:
> 
>     ######2011#####
> 
>     ##+++ Load data with one header and one unit row from
>     (tab-delimited) text file
>     Dir.s <- paste(system.file(package='REddyProc'), 'examples', sep='/')
>     EddyData_all.F <- fLoadTXTIntoDataframe('hourly_partition_all.txt',
>     Dir.s)
> 
>     EddyData_2011.F<-subset(EddyData_all.F,Year==2011) # <<<<<<<<<<
>     Subsetting the 2011 year.
> 
>     #+++ If not provided, calculate VPD from Tair and rH
>     EddyData_2011.F <-
>     cbind(EddyData_2011.F,VPD=fCalcVPDfromRHandTair(EddyData_2011.F$rH,
>     EddyData_2011.F$Tair))
> 
>     #+++ Add time stamp in POSIX time format
>     EddyDataWithPosix_all.F <- fConvertTimeToPosix(EddyData_2011.F,
>     'YDH', Year.s='Year', Day.s='DoY', Hour.s='Hour')
> 
>     #+++ estimate different u* thresholds from the data (see also
>     function sEstUstarThresholdDistribution)
>     # for real application, use larger sample size (it takes longer)
>     EddyProc_2011.C <- sEddyProc$new('2011', EddyDataWithPosix_all.F ,
>     c('NEE','H','LE','Rg','Tair','Tsoil','rH','VPD', 'Ustar'), DTS.n=24)
>     # it is in hourly basis! At least it works for previous versions
>     #+++ specify seasonal subsets of the data
>     seasonFactor.v <-
>     usCreateSeasonFactorMonth(EddyProc_2011.C$sDATA$sDateTime,
>     startMonth=c(3,6,9,12));
> 
>     (uStarRes <- EddyProc_2011.C$sEstUstarThresholdDistribution
>     (seasonFactor.v=seasonFactor.v, nSample=10L))
> 
>     season uStar U05 U50 U95
>     1 2010012 NA NA NA NA
>     2 2011003 NA NA NA NA
>     3 2011006 NA NA NA NA
>     4 2011009 NA NA NA NA
>     5 2011012 NA NA NA NAWhy ??!!!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4895 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.r-forge.r-project.org/pipermail/reddyproc-users/attachments/20160512/b933fe3a/attachment.bin>