sMDSGapFillAfterUStarDistr: GapFilling for several filters of estimated friction velocity...

In REddyProc: Data processing and plotting utilities of (half-)hourly eddy-covariance measurements

Description

sEddyProc$sMDSGapFillUStarDistr - calling sMDSGapFillAfterUstar for several filters of friction velocity Ustar

Usage

sMDSGapFillAfterUStarDistr(..., UstarThres.df, UstarSuffix.V.s = colnames(UstarThres.df)[-1])

Arguments

...	other arguments to sMDSGapFillAfterUstar and sMDSGapFill
UstarThres.df	data.frame with first column, season names, and remaining columns different estimates of uStar Threshold. If the data.frame has only one row, then each uStar threshold estimate is applied to the entire dataset. Entries in first column must match levels in argument seasonFactor.v
UstarSuffix.V.s	String vector to distinguish result columns for different ustar values. Its length must correspond to column numbers in UstarThres.m.n.

Details

The eddy covariance method does not work with low turbulence conditions. Hence the periods with low turbulence indicated by a low friction velocity u* needs to be filtered out and gapfilled (see sMDSGapFill. The threshold value of a sufficient u* causes one of the largest uncertainty components within the gap-filled data. Hence, it is good practice to compare derived quantities based on gap-filled data using different u* threshold values.

For example a user could provide the the following collumns in argument UstarThres.df and corresponding suffixes in argument UstarSuffix.V.s

• season: identifier for which season this row is used.

• Ustar: estimate on original series

• U05: 5% of bootstrap

• U50: median of bootstrap

• U95: 95% of bootstrap)

Then the difference between output columns NEE_U05_f and NEE_U95_f corresponds to the uncertainty introduced by the uncertain estimate of the u* threshold.

Value

Matrix (columns correspond to u*Scenarios) with quality flag from filtering ustar (0 - good data, 1 - filtered data)

Gap filling results in sTEMP data frame (with renamed columns), that can be retrieved by sExportResults. Each of the columns is calculated for several u*r-estimates and distinguished by a suffix after the variable. E.g. with an first entry "U05" in UstarSuffix.V.s corresponding to the first column in UstarThres.m.n, the corresponding filled NEE can be found in output collumn "NEE_U05_f".

Author(s)

TW (Department for Biogeochemical Integration at MPI-BGC, Jena, Germany)

See Also

Advanced Example 1b in sEddyProc.example

REddyProc documentation built on May 2, 2019, 5:19 p.m.