flux: Estimate gas flux rates using dynamic closed chamber data

Description

The functions are used to prepare and carry out gas flux rate estimation with dynamic closed chamber data for one gas at a time. Either you can do the calculation in two steps or in just one single step. flux is a convenience wrapper for the other two functions that should be the best for most users. fluxes can be used to estimate gas flux rates for all three commonly measured greenhouse gases (CO2, CH4, N2O) at once.

Usage

flux(x, var.par, ghg = "CH4", co2ntrol = CO2.control(), min.allowed = 3, 
max.nrmse = 0.1, nrmse.lim = 0.2, r2.qual = 0.8, range.lim = 30, 
out.unit = "auto", elementar = FALSE, hardflag = list(range = TRUE), 
p.values = TRUE)
fluxes(x, var.par, co2ntrol = list(leak = TRUE, relay = FALSE), 
min.allowed = 3, max.nrmse = 0.1, nrmse.lim = 0.2, r2.qual = 0.8, 
range.lim = 30, out.unit = "auto", elementar = FALSE, 
hardflag = list(range = TRUE), p.values = TRUE)
flux.odae(dat, var.par, min.allowed = 3, max.nrmse = 0.1, rl = NULL)
flux.conv(fl.dat, ghg = "CH4", r2.qual = 0.8, nrmse.lim = 0.2, 
out.unit = "auto", elementar = FALSE, hardflag = list(range = TRUE))

Arguments

A list with data tables as returned by conz.prep. Each table contains data for one chamber measurement. Usually there will be identifying columns, date, time and time sequence columns, and columns containing the concentrations of the greenhou

var.par

A named list specifying the variables and parameters that are used in the estimation process and variables that should be handed through the function so that they are easily available for further analysis. Some of the names are obligatory, others are free

ghg

Character string identifying the greenhouse gas for which concentration measurements are handled. Can be "CH4", "N2O" or "CO2". Defaults to "CH4". This argument determines the molar weight that is used i

co2ntrol

Options for doing CO2 concentration measurement controlled estimation of flux rates. This means that the outliers and the slope of the CO2 concentration measurements influence the estimated flux rate. Obviously this only makes sense for N2O and CH4. With

min.allowed

Minimum number of concentration measurements that are allowed for the determination of the best model fit during the outlier detection and elimination procedure via odae. Integer between 3 and the number of concentration measurements during o

max.nrmse

The maximum acceptable normalized root mean square error for configurations with higher numbers of concentration measurements. Numeric value between 0 and 1. Defaults to 0.1. Above that value lower numbers of concentration measurements down to min.a

nrmse.lim

The main quality parameter for the model fit. The maximum acceptable normalized root mean square error. Numeric between 0 and 1. Default is 0.2. If the best fit for one chamber measurement is above this value, the function reports FALSE in th

range.lim

The minimum range of the concentration measurements during one chamber placement. You can either give one numerical value, a numeric vector with the same length as x giving different range limits for each chamber placement (for instance obtai

r2.qual

Alternative quality parameter for the model fit. The minimum acceptable r2 of the best fitted model. Numeric between 0 and 1. Defaults to 0.8. A chamber measurement is not defined invalid when the value is below the quality setting. Instead, a quality fla

out.unit

Output unit of the flux rate mass part. Character string. Defaults to "auto". In this case, the function tries to find a unit that ranges the output value between 0.01 and 10 which is fine for figures. Can be "ng", "mug", "mg", or "g". "mug" stands for "$

elementar

When the fluxes are wanted as element values set elementar = TRUE. Defaults to FALSE.

hardflag

Named list that controls which of the quality flags are to be hard flagged (the value is changed according to the quality flag). Default is that the range.lim is hard flagged. So when the range of the actual concentration values for a chamber

p.values

Logical. If p-values (come as asterisks in standard console output) for the chosen regression models are wanted in the output, set to TRUE.

dat

One data table for one chamber placement. See x and var.par for details.

Specifies range.lim in the low level function flux.odae. As with range.lim several options are allowed. Defaults to NULL. In this case, the function looks for a column rl in dat. If it doesn

fl.dat

An object with the same structure as returned by flux.odae. See details and value.

Value

flux returns an object of class fluss that is a 4 entry list. When the object is printed to the console only the third entry is displayed in a modified form that is meant to maximize information display with small footprint for easy inspection. This third entry of the list is a data.frame with the estimated flux rates. The columns represent the entries in fluss of the single chamber measurements (including quality flags, see below) plus naming information according to the settings in the nmes argument of conz.prep.
In case of an object returned by fluxes there is information regarding the average range limit and its spread per gas. A table is printed to the console with three columns per gas. The first contains the quality flags (e.g. 111.02). The order is: nrmse.f, r2.f, range.f, nomba.f, leak.f. The first three are considered more important, if '1' everything is fine. For the two behind the dot its a bit different: The first just gives the number of measurements below ambient, the second is '2' when no co2ntrol has been carried out, '0' when leaking occurred, and '1' when no leaking occurred.
The first entry contains the call to the function.
The second entry is itself a list of lists and data tables. It is called flux.res and is comprised of objects that are returned by flux.conv. Each first level entry in this list contains the information for one chamber measurement. It is named according to the nmes-setting in conz.prep and contains the elements fluss (which is itself a list with the elements given below), fl.dat (equals the object returned by flux.odae; see below), and unit which provides information on the output mass unit of the flux rate that is handed over to the function plot.fluss and to the table output.
The elements of fluss:
ghgCharacter. The gas species for which the flux has been estimated.
fluxNumeric. Calculated flux rate in mass unit per m2 and hour.
r2The actual r2 of the best fitted model that has been used for flux caclulation.
nrmseThe actual nrmse of the best fitted model that has been used for flux caclulation.
r2.fLogical. r2 quality flag telling whether the r2 quality setting given in r2.qual is fulfilled.
range.fLogical. Range quality flag telling whether the range of the concentration measurements exceeded the quality range of the measurement device that has been specified in range.lim.
nrmse.fLogical. nrmse quality flag telling whether the nrmse quality setting given in nrmse.lim is fulfilled (i.e. if the nrmse of the best model <= nrmse.lim).
nomba.fInteger. Reports the number of measurements below ambient. When one observes concentrations below ambient that might make the measurements unstable, it is possible to filter the results later and allow only a maximum acceptable number of measurements below ambient. The ambient concentration is build into the function with data from Mace Head Ireland (N2O, CH4) and global average (CO2) obtained from http://cdiac.ornl.gov/pns/current_ghg.html as of August 1st, 2011.
leak.fLogical. When co2ntrol was applied with leak = TRUE, possible chamber leakage as represented by decreasing CO2 concentrations over time is shown by a FALSE / 0.
The elements of fl.dat that is also the object returned by flux.odae are:
lm4fluxComplex object. The best fitting model as reported by lm. It builds the basis for the calculation of the flux rate via flux.conv.
row.selectInteger vector giving the indices of the rows of the data table that have been used to construct the best fitting model. This information is later used in the plotting functions plot.flux and plot.fluss.
orig.datdata.frame with the original data provided according to arguments x and columns.
out.datData to be handed through. Per default area and volume of the chamber are relayed but these values are not part of the table output whereas all additionally relayed data are part of the table output.

encoding

UTF-8

Details

Typically it will be most convenient to use flux on objects returned by conz.prep (i.e. on lists of data tables that contain all necessary data per chamber measurement including supporting information). flux simply wraps flux.odae and flux.conv applied on lists of chamber measurement data tables into one function. Thus, the data of a one day field campaign or a year of chamber measurements can easily handled by simply running two functions (conz.prep and flux) consecutively per gas species or by running first conz.prep and then fluxes to estimate ghg fluxes for all three common ghg gases. See example.

Probably the most important argument is var.par. It specifies the variables (by referring to the names of the data columns) from x and parameters (fixed values that are constant for all chamber placements) that are used for the flux estimations. For simple handling it is expecting a named list.

For flux the obligatory list items are ghg – ghg concentrations; gc.qual – quality flag of the GC; time – chamber closing time in minutes; volume – chamber volume during placement (in cbm); area – chamber area (in sqm); t.air – air temperature inside chamber during concentration measurements (in °C); p.air – air pressure during concentration measurements (in Pa).

For fluxes the obligatory list items are the same as for flux with the exceptions that all ghg columns and the respective gc quality columns have to be specified by using the named list items CH4 – CH4 concentrations; CH4.gcq – CH4 GC quality flag; CO2 – CO2 concentrations; CO2.gcq – CO2 GC quality flag; N2O – N2O concentrations; N2O.gcq – N2O GC quality flag.

All these list items can either be given as a variable (name of a column in x) or as a fixed parameter. This makes no sense for ghg and time, but in many cases chamber volume and area will be constant across measurements. Another likely candidate for a fixed parameter is p.air because not all teams log air pressure during measurements. All additional list items should be of type ‘variable‘ and refer to further columns in x if you want those data handed through the function and be part of the result tables (for having all data in one place for further analyses). You are free to choose appropriate names. Fixed parameters will not be relayed.

If the flux estimation is carried out in two steps it will typically be carried out on a list structure as returned by conz.prep. Therefore, it is used within a lapply call. For details see examples. However, the functions flux.odae and flux.conv are designed to be carried out on single data tables (data.frame) per chamber measurement.

First flux.odae is run. It simply tries to find the best model fit for the series of concentration measurements that are given in dat. This data.frame has to consist of five columns that give (in that order): gas concentration, closing time of the chamber in minutes, gas concentration quality flag, chamber volume, temperature within the chamber headspace during measurements (may change during chamber placement). See example data.

At the moment the optimization bases on linear regression. All possible models with n (= total number of concentration measurements per chamber placement) to min.allowed number of concentration measurements are fitted and the best fit is evaluated in a stepwise procedure. The normalized root mean square error is used as the quality criterion for the outlier detection and elimination procedure. All model fits with a nrmse <= max.nrmse are extracted and ranked according to the number of concentration measurements (decreasing) and to the nrmse (increasing). The first ranked model is stored along with the original data table and some other information. Therefore a model with e.g. a nrmse of 0.081 constructed from 5 concentration measurements wins against a model with a nrmse of 0.07 with only 4 concentration measurements. This reflects the idea that models with nrmse <= max.nrmse already represent a sufficient fit and do not have "outliers" that must be eliminated.

In case no model has a nrmse <= max.nrmse, the models are simply ranked according to their nrmse and the model with the lowest nrmse wins and is stored. In that way outliers are detected and exluded. flux.odae returns a complex object that contains most of the necessary information for the flux.conv function and also carries information that is later needed for the plot functions (plot.flux and plot.fluss).

The flux calculation is then carried out with the function flux.conv. It takes the object returned by flux.odae and additional information (chamber area, gas species, several quality settings and in- as well as output units) and calculates the flux rates. Further several quality checks (r2 check, range check, nrmse check, nomba check; for details see Value) are carried out and quality flags are reported along with the fluxes in the output. It is best when all quality flags are returned TRUE. Depending on the application quality requirements might vary. Therefore, per defaults the functions rather report quality flags than excluding data. However, this can be changed via hardflag.

The idea behind co2ntrol in flux and fluxes is that the CO2 concentration measurements might serve as a further check on the integrity of the chamber measuremnt in the field. When co2ntrol (for details see CO2.control) is set, the function first carries out an outlier procedure on the CO2 concentration data (the respective columns have to be in x of course). Further, the slope of the CO2 concentration change over time is checked. When it is negative, chamber leakage is assumed and a respective quality flag (leak.flag) is reported FALSE. The leak.flag cannot be hard flagged.

References

Nakano T (2004) A comparison of regression methods for estimating soil-atmosphere diffusion gas fluxes by a closed-chamber technique. Soil Biology and Biochemistry 36: 107-113.

Forbrich I, Kutzbach L, Hormann A, Wilmking M (2010) A comparison of linear and exponential regression for estimating diffusive CH4 fluxes by closed-chambers in peatlands. Soil Biology and Biochemistry 42: 507-515.

Examples

Run this code

## load example data
data(tt.pre)

## extract field concentration measurements
gcd <- tt.pre[tt.pre$sampletype_a=="P",]

## partition the data into data tables per chamber measurement
# first tell which columns of the data you want to carry on
columns <- c("date", "veg", "spot", "time_min", "temp_dC", "cham_area", 
"cham_vol", "CO2ppm", "CO2Code", "CH4ppb", "CH4Code", "N2Oppb", "N2OCode")
# then do the partitioning
gcd.parts <- conz.prep(gcd, columns, factors = c("date", "spot", "veg"), 
nmes = c("date", "veg", "spot"))

## calculate flux rates for methane
# first define a global CH4 range limit
CH4.lim <- 30
# do the flux rate estimation (it will often be best to define
# var.par separately, note that p.air is given as a parameter)
vp.CH4 <- list(ghg = "CH4ppb", time = "time_min", gc.qual = "CH4Code", 
volume = "cham_vol", t.air = "temp_dC", area = "cham_area", p.air = 101325)
flux.CH4 <- flux(gcd.parts, var.par = vp.CH4, ghg = "CH4", co2ntrol = NULL)
# look at the results table
flux.CH4

# now we extract range limits from the calibration gas measurements
# and attach them to gcd.parts. first we get the calibration gas
# measurements from tt.pre (we change the date because it is in 
# a strange format and has to be the same as the dates in gcd.parts
cgm <- tt.pre[tt.pre$sampletype_a=="E",c("date_gc", "CH4ppb", "CH4Code", 
"CO2ppm", "CO2Code", "N2Oppb", "N2OCode")]
names(cgm)[1] <- "date"
cgm$date <- "2011-03-16"
# now we can do the flux.calib
gcd.parts.cal <- flux.calib(gcd.parts, columns = c("date", "CH4ppb"), 
calib = cgm, format="%Y-%m-%d")
# do the flux rate estimation (we use the same var.par as before)
flux.CH4 <- flux(gcd.parts.cal, var.par=vp.CH4, ghg = "CH4", 
co2ntrol = NULL, range.lim=NULL)
# look at the results table
flux.CH4

## plot the concentration-change-with-time-plots as kind of diagnostic
plot(flux.CH4, subs = NULL, dims = c(3,6))

## do the flux rate estimation whilst using CO2 concentrations to
## control for possible chamber leakage
# first set the desired controls (without outlier transfer)
cntrl <- CO2.control(relay = FALSE, range.lim = 40, min.allow = 3, 
max.nrmse = 0.2)
# do the flux rate estimation
flux.CH4.b <- flux(gcd.parts, var.par=vp.CH4, ghg = "CH4", co2ntrol = cntrl)
# look at the results table
flux.CH4.b
# plot the concentration-change-with-time-plots as kind of diagnostic
plot(flux.CH4.b, subs = NULL, dims = c(3,6))

## do the flux rate estimation whilst using CO2 concentrations to
## control for outliers and possible chamber leakage
# first set the desired controls
cntrl <- CO2.control(relay = TRUE, range.lim = 40, min.allow = 4, 
max.nrmse = 0.2)
# do the flux rate estimation
flux.CH4.c <- flux(gcd.parts, var.par=vp.CH4, ghg = "CH4", 
co2ntrol = cntrl)
# look at the results table
flux.CH4.c
# plot the concentration-change-with-time-plots as kind of diagnostic
plot(flux.CH4.c, subs = NULL, dims = c(3,6))

Run the code above in your browser using DataLab