Skip to contents

From raw gas concentration over time to clean fluxes

Source: R/stupeflux.R
stupeflux.Rd

Wrapper function for the Fluxible workflow. We recommand using the step-by-step workflow for more control over the process.

Usage

stupeflux(
  raw_conc,
  field_record,
  datetime_col,
  start_col,
  end_col,
  conc_col,
  startcrop,
  measurement_length,
  fit_type,
  temp_air_col,
  chamber_volume,
  atm_pressure,
  plot_area,
  conc_unit,
  flux_unit,
  fixed_length = TRUE,
  cols_keep = c(),
  cols_ave = c(),
  cols_sum = c(),
  cols_med = c(),
  tube_volume,
  ratio_threshold = 0.5,
  time_diff = 0,
  start_cut = 0,
  end_cut = 0,
  cz_window = 15,
  b_window = 10,
  a_window = 10,
  roll_width = 15,
  t_zero = 0,
  force_discard = c(),
  force_ok = c(),
  force_zero = c(),
  ambient_conc = 421,
  error = 100,
  pvalue_threshold = 0.3,
  rsquared_threshold = 0.7,
  rmse_threshold = 25,
  cor_threshold = 0.5,
  b_threshold = 1,
  temp_air_unit = "celsius",
  cut = TRUE,
  slope_correction = TRUE
)

Arguments

raw_conc

dataframe of CO2 concentration measured continuously. Has to contain at least a datetime column in ymd_hms format and a gas concentration column as double.

field_record

dataframe recording which measurement happened when. Has to contain at least a column containing the start of each measurement, and any other column identifying the measurements.

datetime_col

datetime column in raw_conc (dmy_hms format)

start_col

start column in field_record (dmy_hms format)

end_col

end columne in field_record (ymd_hms format)

conc_col

concentration column in raw_conc

startcrop

how many seconds should be discarded at the beginning of the measurement

measurement_length

length of the measurement (in seconds) from the start specified in the field_record

fit_type

exp_zhao18, exp_tz, quadratic or linear. exp_zhao18 is using the exponential model C(t) = C_m + a (t - t_z) + (C_z - C_m) exp(-b (t - t_z)) from Zhao et al (2018). expt_tz is a modified version which allows the user to fix t_zero: C(t) = C_m + a * t + (C_z - C_m) exp(-b * t).

temp_air_col

column containing the air temperature used to calculate fluxes. Will be averaged with NA removed.

chamber_volume

volume of the flux chamber in L, can also be a column in case it is a variable

atm_pressure

atmospheric pressure, can be a constant (numerical) or a variable (column name)

plot_area

area of the plot in m^2, can also be a column in case it is a variable

conc_unit

unit in which the concentration of gas was measured ppm or ppb

flux_unit

unit in which the calculated flux will be mmol outputs fluxes in mmol * m^(-2)*h^(-1); micromol outputs fluxes in micromol * m^(-2)*h^(-1)

fixed_length

if TRUE (default), the measurement_length is used to create the end column. If FALSE, end_col has to be provided.

cols_keep

columns to keep from the input to the output. Those columns need to have unique values for each flux, as distinct() is applied.

cols_ave

columns with values that should be averaged for each flux in the output. Note that NA are removed in mean calculation.

cols_sum

columns with values for which is sum is provided for each flux in the output. Note that NA are removed in sum calculation.

cols_med

columns with values for which is median is provided for each flux in the output. Note that NA are removed in median calculation.

tube_volume

volume of the tubing in L, can also be a column in case it is a variable

ratio_threshold

ratio of gas concentration data points over length of measurement (in seconds) below which the measurement will be considered as not having enough data points to be considered for calculations

time_diff

time difference (in seconds) between the two datasets. Will be added to the datetime column of the raw_conc dataset. For situations where the time was not synchronized correctly.

start_cut

time to discard at the start of the measurements (in seconds)

end_cut

time to discard at the end of the measurements (in seconds)

cz_window

window used to calculate Cz, at the beginning of cut window (exponential fit)

b_window

window to estimate b. It is an interval after tz where it is assumed that the model fits the data perfectly (exponential fit)

a_window

window at the end of the flux to estimate a (exponential fit)

roll_width

width of the rolling mean for CO2 when looking for tz, ideally same as cz_window (exponential fit)

t_zero

time at which the slope should be calculated (for quadratic and exp_tz fits)

force_discard

vector of fluxIDs that should be discarded by the user's decision

force_ok

vector of fluxIDs for which the user wants to keep the calculated slope despite a bad quality flag

force_zero

vector of fluxIDs that should be replaced by zero by the user's decision

ambient_conc

ambient gas concentration in ppm at the site of measurement (used to detect measurement that started with a polluted setup)

error

error of the setup, defines a window outside of which the starting values indicate a polluted setup

pvalue_threshold

threshold of p-value below which the change of gas concentration over time is considered not significant (linear and quadratic fit)

rsquared_threshold

threshold of r squared value below which the linear model is considered an unsatisfactory fit (linear and quadratic fit)

rmse_threshold

threshold for the RMSE of each flux above which the fit is considered unsatisfactory (exponential fit)

cor_threshold

threshold for the correlation coefficient of gas concentration with time below which the correlation is considered not significant (exponential fit)

b_threshold

threshold for the b parameter. Defines a window with its opposite inside which the fit is considered good enough (exponential fit)

temp_air_unit

units in which air temperature was measured. Has to be either celsius (default), fahrenheit or kelvin.

cut

if 'TRUE' (default), the measurements will be cut according to 'f_cut' before calculating fluxes. This has no influence on the flux itself since the slope is provided from flux_fitting, but it will influence the values of the columns in cols_ave.

slope_correction

logical. If TRUE, the flux will be calculated with the slope corrected according to the recommandations of the quality flags.

Value

a dataframe containing flux IDs, datetime of measurements' starts, fluxes in mmol*m^(-2)h^(-1) or micromolm^(-2)*h^(-1) (f_flux) according to flux_unit, temperature average for each flux in Kelvin (f_temp_ave), the total volume of the setup for each measurement (f_volume_setup), the model used in flux_fitting, any column specified in cols_keep, any column specified in cols_ave with their value averaged over the measurement after cuts and discarding NA.

References

Zhao, P., Hammerle, A., Zeeman, M., Wohlfahrt, G., 2018. On the calculation of daytime CO2 fluxes measured by automated closed transparent chambers. Agricultural and Forest Meteorology 263, 267–275. https://doi.org/10.1016/j.agrformet.2018.08.022 exponential is equal to exp_zhao18, for backwards compatibility

Examples

data(co2_df_short)
data(record_short)
stupeflux(
raw_conc = co2_df_short,
field_record = record_short,
datetime_col = datetime,
start_col = start,
conc_col = conc,
startcrop = 10,
measurement_length = 180,
fit_type = "exp_zhao18",
temp_air_col = temp_air,
conc_unit = "ppm",
flux_unit = "mmol",
chamber_volume = 24.5,
tube_volume = 0.075,
atm_pressure = 1,
plot_area = 0.0625
)
#> Cutting measurements...
#> Estimating starting parameters for optimization...
#> Optimizing fitting parameters...
#> Calculating fits and slopes...
#> Done.
#> 
#>  Total number of measurements: 6
#> 
#>  ok 	 5 	 83 %
#>  zero 	 1 	 17 %
#>  discard 	 0 	 0 %
#>  force_discard 	 0 	 0 %
#>  start_error 	 0 	 0 %
#>  no_data 	 0 	 0 %
#>  force_ok 	 0 	 0 %
#>  force_zero 	 0 	 0 %
#>  force_lm 	 0 	 0 %
#> Cutting data according to 'keep_arg'...
#> Averaging air temperature for each flux...
#> Calculating fluxes...
#> R constant set to 0.082057
#> Concentration was measured in ppm
#> Fluxes are in mmol/m2/h
#> # A tibble: 6 × 7
#>   f_fluxid f_slope_corr f_temp_air_ave datetime            f_volume_setup f_flux
#>   <fct>           <dbl>          <dbl> <dttm>                       <dbl>  <dbl>
#> 1 1               0.785           7.28 2022-07-28 23:43:35           24.6   48.3
#> 2 2               0.503           7.37 2022-07-28 23:47:22           24.6   30.9
#> 3 3               0.344           7.45 2022-07-28 23:52:10           24.6   21.1
#> 4 4               0.693           7.77 2022-07-28 23:59:32           24.6   42.5
#> 5 5               0               7.69 2022-07-29 00:03:10           24.6    0  
#> 6 6               0.433           7.74 2022-07-29 00:06:35           24.6   26.6
#> # ℹ 1 more variable: f_model <chr>