distClean: Distributional cleaning

In FREddyPro: Post-Processing EddyPro Full Output File

Description

Function to clean and de-spike gas flux data based on a distribution produced for each half hour and removing values lower and higher than the 5th and 95th percentile respectively

Usage

distClean(var, hour, df)

Arguments

var	The variable to clean/de-spike
hour	The hour variable created by the createTimestamp function
df	The data frame

Author(s)

Georgios Xenakis

Examples

## Load the data
data(fluxes)

## Clean data
fluxes=cleanFluxes(fluxes,sdCor=TRUE,sdTimes=3,timesList=3,distCor=TRUE,
                   thresholdList=list(H=c(-100,1000),LE=c(-100,1000)))	

## Create timestamp
fluxes<-createTimestamp(fluxes)

## Find which rows have positive and which negative CO2 flux values 
positive<-which(fluxes$co2_flux>0)
negative<-which(fluxes$co2_flux<0)

## Clean data for positive and negative separately
fluxes$co2_flux[positive]<-distClean(fluxes$co2_flux[positive],
                                     fluxes$hour[positive],fluxes[positive,])
fluxes$co2_flux[negative]<-distClean(fluxes$co2_flux[negative],
                                     fluxes$hour[negative],fluxes[negative,])

FREddyPro documentation built on May 29, 2017, 7:22 p.m.