| Title: | Tools to Calibrate and Work with NEON Atmospheric Isotope Data |
|---|---|
| Description: | Functions for downloading, calibrating, and analyzing atmospheric isotope data bundled into the eddy covariance data products of the National Ecological Observatory Network (NEON) <https://www.neonscience.org>. Calibration tools are provided for carbon and water isotope products. Carbon isotope calibration details are found in Fiorella et al. (2021) <doi:10.1029/2020JG005862>, and the readme file at <https://github.com/lanl/NEONiso>. Tools for calibrating water isotope products have been added as of 0.6.0, but have known deficiencies and should be considered very experimental currently. |
| Authors: | Rich Fiorella [aut, cre] |
| Maintainer: | Rich Fiorella <[email protected]> |
| License: | GPL-3 |
| Version: | 0.7.0 |
| Built: | 2024-11-12 05:57:52 UTC |
| Source: | https://github.com/lanl/neoniso |
calculate_12CO2
calculate_12CO2(total_co2, delta13c, f = 0.00474)calculate_12CO2(total_co2, delta13c, f = 0.00474)
total_co2 |
Vector of CO2 mole fractions. |
delta13c |
Vector of d13C values. |
f |
Fraction of CO2 that is not 12CO2 or 13CO2. Assumed fixed at 0.00474 |
Vector of 12CO2 mole fractions.
Rich Fiorella [email protected]
calculate_12CO2(total_co2 = 410, delta13c = -8.5)calculate_12CO2(total_co2 = 410, delta13c = -8.5)
calculate_13CO2
calculate_13CO2(total_co2, delta13c, f = 0.00474)calculate_13CO2(total_co2, delta13c, f = 0.00474)
total_co2 |
Vector of CO2 mole fractions. |
delta13c |
Vector of d13C values. |
f |
Fraction of CO2 that is not 12CO2 or 13CO2. Assumed fixed at 0.00474 |
Vector of 13CO2 mole fractions.
Rich Fiorella [email protected]
calculate_13CO2(total_co2 = 410, delta13c = -8.5)calculate_13CO2(total_co2 = 410, delta13c = -8.5)
calibrate_ambient_carbon_Bowling2003
calibrate_ambient_carbon_Bowling2003( amb_data_list, caldf, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, gap_fill_parameters = FALSE, r2_thres = 0.9 )calibrate_ambient_carbon_Bowling2003( amb_data_list, caldf, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, gap_fill_parameters = FALSE, r2_thres = 0.9 )
amb_data_list |
List containing an ambient d13C dataset. Will include all variables in 000_0x0_xxm. (character) |
caldf |
Calibration data frame containing gain and offset values for 12C and 13C isotopologues. |
site |
Four-letter NEON code corresponding to site being processed. |
filter_data |
Apply median absolute deviation filter from Brock 86 to
remove impulse spikes? Inherited from
|
force_to_end |
In given month, calibrate ambient data later than last calibration, using the last calibration? (default true) |
force_to_beginning |
In given month, calibrate ambient data before than first calibration, using the first calibration? (default true) |
gap_fill_parameters |
Should function attempt to 'gap-fill' across a bad calibration by carrying the last known good calibration forward? Implementation is fairly primitive currently, as it only carries the last known good calibration that's available forward rather than interpolating, etc. Default FALSE. |
r2_thres |
Minimum r2 value for calibration to be considered "good" and applied to ambient data. |
Depends on write_to_file argument.
If true, returns nothing to environment;
but returns calibrated ambient observations to the output file.
If false, returns modified version of amb_data_list that include
calibrated ambient data.
Rich Fiorella [email protected]
Function called by calibrate_carbon_bymonth() to apply
gain and offset parameters to the ambient datasets (000_0x0_09m and
000_0x0_30m). This function should generally not be used independently,
but should be used in coordination with
calibrate_carbon_bymonth().
calibrate_ambient_carbon_linreg
calibrate_ambient_carbon_linreg( amb_data_list, caldf, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, gap_fill_parameters = FALSE, r2_thres = 0.9 )calibrate_ambient_carbon_linreg( amb_data_list, caldf, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, gap_fill_parameters = FALSE, r2_thres = 0.9 )
amb_data_list |
List containing an ambient d13C dataset. Will include all variables in 000_0x0_xxm. (character) |
caldf |
Calibration data frame containing gain and offset values for 12C and 13C isotopologues. |
site |
Four-letter NEON code corresponding to site being processed. |
filter_data |
Apply median absolute deviation filter from Brock 86 to
remove impulse spikes? Inherited from
|
force_to_end |
In given month, calibrate ambient data later than last calibration, using the last calibration? (default true) |
force_to_beginning |
In given month, calibrate ambient data before than first calibration, using the first calibration? (default true) |
gap_fill_parameters |
Should function attempt to 'gap-fill' across a bad calibration by carrying the last good calibration forward? Implementation is fairly primitive currently, as it only carries the last known good calibration that's available forward rather than interpolating, etc. Default FALSE. |
r2_thres |
Minimum r2 value for calibration to be considered "good" and applied to ambient data. |
Nothing to environment; returns calibrated ambient observations to the function orchestrating calibration (calibrate_carbon). This function is not designed to be called on its own, and is not exported to the namespace.
Rich Fiorella [email protected]
Function called by calibrate_ambient_carbon_linreg to apply
gain and offset parameters to the ambient datasets (000_0x0_09m and
000_0x0_30m). This function should generally not be used independently,
but should be used with calibrate_ambient_carbon_linreg.
calibrate_ambient_water_isotopes
calibrate_ambient_water_linreg( amb_data_list, caldf, outname, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, r2_thres = 0.9 )calibrate_ambient_water_linreg( amb_data_list, caldf, outname, site, filter_data = TRUE, force_to_end = TRUE, force_to_beginning = TRUE, r2_thres = 0.9 )
amb_data_list |
List containing ambient d18O/d2H datasets. Will include all variables in 000_0x0_xxm. (character) |
caldf |
Calibration data frame containing slope and intercept values for d18O and d2H values. |
outname |
Output variable name. Inherited from
|
site |
Four-letter NEON code corresponding to site being processed. |
filter_data |
Apply a median filter to output ambient data? inherited. |
force_to_end |
In given month, calibrate ambient data later than last calibration, using the last calibration? (default true) |
force_to_beginning |
In given month, calibrate ambient data before than first calibration, using the first calibration? (default true) |
r2_thres |
Minimum r2 value for calibration to be considered "good" and applied to ambient data. |
Nothing to environment; returns calibrated ambient observations to the output file. This function is not designed to be called on its own.
Rich Fiorella [email protected]
Function called by calibrate_ambient_water_linreg to apply
slope and intercept parameters to the ambient datasets (000_0x0_09m and
000_0x0_30m) to correct to the VSMOW scale.
This function should generally not be used independently,
but should be used with calibrate_ambient_water_linreg.
Note that in this version NO CORRECTION FOR HUMIDITY is performed.
Use with caution.
![[Experimental]](data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHhtbG5zOnhsaW5rPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5L3hsaW5rIiB3aWR0aD0iMTM2IiBoZWlnaHQ9IjIwIj48bGluZWFyR3JhZGllbnQgaWQ9ImIiIHgyPSIwIiB5Mj0iMTAwJSI+PHN0b3Agb2Zmc2V0PSIwIiBzdG9wLWNvbG9yPSIjYmJiIiBzdG9wLW9wYWNpdHk9Ii4xIi8+PHN0b3Agb2Zmc2V0PSIxIiBzdG9wLW9wYWNpdHk9Ii4xIi8+PC9saW5lYXJHcmFkaWVudD48Y2xpcFBhdGggaWQ9ImEiPjxyZWN0IHdpZHRoPSIxMzYiIGhlaWdodD0iMjAiIHJ4PSIzIiBmaWxsPSIjZmZmIi8+PC9jbGlwUGF0aD48ZyBjbGlwLXBhdGg9InVybCgjYSkiPjxwYXRoIGZpbGw9IiM1NTUiIGQ9Ik0wIDBoNTN2MjBIMHoiLz48cGF0aCBmaWxsPSIjZmU3ZDM3IiBkPSJNNTMgMGg4M3YyMEg1M3oiLz48cGF0aCBmaWxsPSJ1cmwoI2IpIiBkPSJNMCAwaDEzNnYyMEgweiIvPjwvZz48ZyBmaWxsPSIjZmZmIiB0ZXh0LWFuY2hvcj0ibWlkZGxlIiBmb250LWZhbWlseT0iRGVqYVZ1IFNhbnMsVmVyZGFuYSxHZW5ldmEsc2Fucy1zZXJpZiIgZm9udC1zaXplPSIxMTAiPjx0ZXh0IHg9IjI3NSIgeT0iMTUwIiBmaWxsPSIjMDEwMTAxIiBmaWxsLW9wYWNpdHk9Ii4zIiB0cmFuc2Zvcm09InNjYWxlKC4xKSIgdGV4dExlbmd0aD0iNDMwIj5saWZlY3ljbGU8L3RleHQ+PHRleHQgeD0iMjc1IiB5PSIxNDAiIHRyYW5zZm9ybT0ic2NhbGUoLjEpIiB0ZXh0TGVuZ3RoPSI0MzAiPmxpZmVjeWNsZTwvdGV4dD48dGV4dCB4PSI5MzUiIHk9IjE1MCIgZmlsbD0iIzAxMDEwMSIgZmlsbC1vcGFjaXR5PSIuMyIgdHJhbnNmb3JtPSJzY2FsZSguMSkiIHRleHRMZW5ndGg9IjczMCI+ZXhwZXJpbWVudGFsPC90ZXh0Pjx0ZXh0IHg9IjkzNSIgeT0iMTQwIiB0cmFuc2Zvcm09InNjYWxlKC4xKSIgdGV4dExlbmd0aD0iNzMwIj5leHBlcmltZW50YWw8L3RleHQ+PC9nPiA8L3N2Zz4=)
This function drives a workflow that reads in NEON carbon isotope data
of atmospheric CO2, calibrates it to the VPDB scale, and (optionally)
writes the calibrated data to a new HDF5 file. Two different approaches
are possible: a) a calibration on 12CO2 and 13CO2 isotopologues
independently, after Bowling et al. 2003 (Agr. For. Met.), or b) a direct
calibration of d13C and CO2 values using linear regression. Most of the time
the results generated are extremely similar to each other.
Wen et al. 2013 compared several different carbon
isotope calibration techniques and found this to be the superior method
under most circumstances. We also found this to be the case for NEON data
(Fiorella et al. 2021; JGR-Biogeosciences).
calibrate_carbon( inname, outname, site, method = "Bowling_2003", calibration_half_width = 0.5, force_cal_to_beginning = TRUE, force_cal_to_end = TRUE, gap_fill_parameters = FALSE, filter_ambient = TRUE, r2_thres = 0.95, correct_ref_data = TRUE, write_to_file = TRUE, remove_known_bad_months = TRUE, plot_regression_data = FALSE, plot_directory = NULL, avg = 6, min_nobs = NA, standards = c("co2Low", "co2Med", "co2High") )calibrate_carbon( inname, outname, site, method = "Bowling_2003", calibration_half_width = 0.5, force_cal_to_beginning = TRUE, force_cal_to_end = TRUE, gap_fill_parameters = FALSE, filter_ambient = TRUE, r2_thres = 0.95, correct_ref_data = TRUE, write_to_file = TRUE, remove_known_bad_months = TRUE, plot_regression_data = FALSE, plot_directory = NULL, avg = 6, min_nobs = NA, standards = c("co2Low", "co2Med", "co2High") )
inname |
Input file(s) that are to be calibrated. If a single file is given, output will be a single file per site per month. If a list of files corresponding to a timeseries at a given site is provided, will calibrate the whole time series. |
outname |
Name of the output file. (character) |
site |
Four letter NEON site code for site being processed. (character) |
method |
Are we using the Bowling et al. 2003 method ("Bowling_2003") or direct linear regression of d13C and CO2 mole fractions ("linreg")? |
calibration_half_width |
Determines the period (in days) from which reference data are selected (period is 2*calibration_half_width). |
force_cal_to_beginning |
Extend first calibration to the beginning of the file? (default true) |
force_cal_to_end |
Extend last calibration to the end of the file? (default true) |
gap_fill_parameters |
Should function attempt to 'gap-fill' across a bad calibration by carrying the last good calibration forward? Implementation is fairly primitive currently, as it only carries the last known good calibration that's available forward rather than interpolating, etc. Default FALSE. |
filter_ambient |
Apply the median absolute deviation filter (Brock 86) to remove impulse spikes in output ambient data? (logical; default true) |
r2_thres |
Minimum r2 threshold of an "acceptable" calibration. Acts to remove calibration periods where a measurement error makes relationship nonlinear. Default = 0.95 |
correct_ref_data |
NEON has indicated there are a few instances where reported d13C or CO2 reference values are wrong. If set to true, correct known incorrect values. This argument will (hopefully, eventually) go away after NEON has fixed the reference database. Users will be warned prior to removal of this argument. |
write_to_file |
Write calibrated ambient data to file? (Mostly used for testing) |
remove_known_bad_months |
There are a few site months with known spectral issues where the isotope ratios are likely unrecoverable. This parameter allows removal of these files, but allows them to remain in archive. |
plot_regression_data |
Default false; this is useful for diagnostics. |
plot_directory |
Only used if plot_regression_data is TRUE, but specify where to write out diagnostic plot of regression data. |
avg |
The averaging interval to extract, in minutes. Default 6. |
min_nobs |
Minimum number of high-frequency observations to define a peak. |
standards |
Which reference gases (standards) to use? Default is all, but can pass a subset of "co2Low", "co2Med", and "co2High" as a vector to this argument as well. |
The 'linreg' method simply takes measured and reference d13C and CO2 values
and generates a transfer function between them using lm(). For the
gain-and-offset method, d13C and CO2 values are converted to 12CO2 and 13CO2
mole fractions. Gain and offset parameters are calculated for each
isotopologue independently, and are analogous to regression slope and
intercepts, but jointly correct for CO2 concentration dependence
and place d13C values on the VPDB scale.
The gain and offset parameters are defined by:
Calibrated ambient isotopologues are then given as:
Measurements of reference materials were considered "good" if the following conditions were met:
Measured CO2 concentrations were within 10 ppm of known "reference" concentrations.
Variance of the CO2 concentration in standard peak was < 5 ppm.
Measured d13C value must be within 5 per mil of known "reference" d13C value.
The first two criteria are intended to filter out periods where there is a clear issue with the gas delivery system (i.e., nearly empty gas tank, problem with a valve in the manifold, etc.); the third criterion was adopted after visual inspection of data timeseries revealed that often the first standard measurement following an instrument issue had higher-than-expected error. This criterion clips clearly poor values. Selection of these criteria will become a function argument, and therefore customizable, in a future release.
The behavior of this function will be a bit different depending on what
is supplied as inname. If a single file is provided, the output will be
monthly. However, a list of files corresponding to a site can also be
provided, and then a single output file per site will be generated.
Returns nothing to the environment, but creates a new output HDF5 file containing calibrated carbon isotope values.
Rich Fiorella [email protected]
## Not run: fin <- system.file('extdata', 'NEON.D15.ONAQ.DP4.00200.001.nsae.2019-05.basic.20201020T211037Z.packed.h5', package = 'NEONiso', mustWork = TRUE) calibrate_carbon_bymonth(inname = fin, outname = 'out.h5', site = 'ONAQ', write_to_file = FALSE) calibrate_carbon_bymonth(inname = fin, outname = 'out.h5', site = 'ONAQ', method = 'linreg', write_to_file = FALSE) ## End(Not run)## Not run: fin <- system.file('extdata', 'NEON.D15.ONAQ.DP4.00200.001.nsae.2019-05.basic.20201020T211037Z.packed.h5', package = 'NEONiso', mustWork = TRUE) calibrate_carbon_bymonth(inname = fin, outname = 'out.h5', site = 'ONAQ', write_to_file = FALSE) calibrate_carbon_bymonth(inname = fin, outname = 'out.h5', site = 'ONAQ', method = 'linreg', write_to_file = FALSE) ## End(Not run)
This function uses NEON validation data to apply drift corrections to
measured ambient water isotope ratios. In brief, ambient water isotope
ratios are calibrated by generating regressions using reference water
measurements bracketing an ambient period. Three reference waters are
measured once per day, with several injections per reference water.
Due to memory effects, only the last three are used currently to generate
calibration equations. Regressions between measured d18O and d2H values
and NEON-provisioned known reference values are generated, and used to
calibrate the period of ambient measurements between them if the r2 of
the regression is greater than a threshold value (by default, this is 0.95).
Most of this function deals with selecting the appropriate calibration data
and determining calibration quality. This function also contains a wrapper
for calibrate_ambient_water_linreg, which calibrates the ambient
water data using the calibration parameters generated in this function.
This function also copies over data in the qfqm and ucrt hdf5 data groups.
calibrate_water( inname, outname, site, calibration_half_width = 14, filter_data = TRUE, force_cal_to_beginning = FALSE, force_cal_to_end = FALSE, r2_thres = 0.95, slope_tolerance = 9999, correct_ref_data = TRUE, write_to_file = TRUE )calibrate_water( inname, outname, site, calibration_half_width = 14, filter_data = TRUE, force_cal_to_beginning = FALSE, force_cal_to_end = FALSE, r2_thres = 0.95, slope_tolerance = 9999, correct_ref_data = TRUE, write_to_file = TRUE )
inname |
Input file(s) that are to be calibrated. If a single file is given, output will be a single file per site per month. If a list of files corresponding to a timeseries at a given site is provided, will calibrate the whole time series. |
outname |
Name of the output file. (character) |
site |
Four-letter NEON code for site being processed. |
calibration_half_width |
Determines the range of standard measurements
to use in determining the calibration regression dataset. Creates
a moving window that is |
filter_data |
Apply median absolute deviation filter from Brock 86 to remove impulse spikes? |
force_cal_to_beginning |
Extend first calibration to the beginning of the file? |
force_cal_to_end |
Extend last calibration to the end of the file? |
r2_thres |
Minimum r2 threshold of an "acceptable" calibration. Acts to remove calibration periods where a measurement error makes relationship nonlinear. Default = 0.95 |
slope_tolerance |
How different from 1 should we allow 'passing' regression slopes to be? Experimental parameter, off by default (e.g., default slope parameter = 9999) |
correct_ref_data |
There are a few instances where the reference d18O and d2H values may have been switched, causing very anomalous d-excess values. If TRUE, implement a switch that corrects this issue. |
write_to_file |
Write calibrated ambient data to file? (Mostly used for testing) |
IMPORTANT NOTE Currently this function does not apply a correction for humidity dependence of Picarro isotopic measurements. This is because the data to implement these corrections is not yet publicly available. Caution is suggested when analyzing data at low humidities, below ~5000 ppm, with likely higher biases at lower humidity values.
Additionally, please note that this function is meant to work on all files for a given site at the same time. A more flexible version that can handle all files or monthly files will be added to a future release.
nothing to the workspace, but creates a new output file of calibrated water isotope data.
Rich Fiorella [email protected]
carbon_regression_plots
carbon_regression_plots(caldata, plot_filename, method, mtitle)carbon_regression_plots(caldata, plot_filename, method, mtitle)
caldata |
Data frame corresponding to a specific calibration period. |
plot_filename |
What should the output file name for diagnostic plot be? |
method |
Which method are we using? Currently works for gain/offset. |
mtitle |
Fed from above routine - what should the plot title be? |
Nothing to the environment, but a pdf plot to a file.
Rich Fiorella [email protected]
convert_NEONhdf5_to_POSIXct_time
convert_NEONhdf5_to_POSIXct_time(intime)convert_NEONhdf5_to_POSIXct_time(intime)
intime |
Vector of datetimes in NEON data files (as string) to convert to POSIXct class |
Vector of datetimes from NEON data file now in POSIXct format.
Rich Fiorella [email protected]
convert_NEONhdf5_to_POSIXct_time("2019-06-01T12:00:00.000Z")convert_NEONhdf5_to_POSIXct_time("2019-06-01T12:00:00.000Z")
Converts a POSIXct object back to the character format used by NEON in their HDF eddy covariance files. Output format, using strptime syntax, is %Y-%m-%dT%H:%M:%OSZ.
convert_POSIXct_to_NEONhdf5_time(intime)convert_POSIXct_to_NEONhdf5_time(intime)
intime |
POSIXct vector to convert to NEON time format. |
Returns character version of POSIXct object matching NEON time variable format.
Rich Fiorella [email protected]
convert_POSIXct_to_NEONhdf5_time(Sys.time())convert_POSIXct_to_NEONhdf5_time(Sys.time())
copy_qfqm_group
copy_qfqm_group(data_list, outname, site, file, species)copy_qfqm_group(data_list, outname, site, file, species)
data_list |
List of groups to retrieve qfqm data from. |
outname |
Output filename. |
site |
Four-letter NEON site code. |
file |
Input filename. |
species |
CO2 or H2O? Same function used for both CO2 and H2O isotopes. |
Nothing to the workspace, but copies qfqm group from input file to output file.
Rich Fiorella [email protected]
copy_ucrt_group
copy_ucrt_group(data_list, outname, site, file, species)copy_ucrt_group(data_list, outname, site, file, species)
data_list |
List of groups to retrieve ucrt data from. |
outname |
Output file name. |
site |
NEON 4-letter site code. |
file |
Input file name. |
species |
H2O or CO2. |
Nothing to the workspace, but copies ucrt group from input file to output file.
Rich Fiorella [email protected]
This ugly function is present out of necessity, and will only exist for as long as it is necessary. It is an internal correction within the NEONiso calibration routines that is required as there are some mismatches between the 'true' isotope reference values and those in the NEON HDF5 files. NEON is working on correcting this, and after it has been corrected, this function has no need to exist and will be immediately deprecated. As a result, this function is fairly messy but there is little incentive to improve it.
correct_carbon_ref_cval( std_frame, site, omit_already_corrected = TRUE, co2_tol = 5, d13c_tol = 0.25 )correct_carbon_ref_cval( std_frame, site, omit_already_corrected = TRUE, co2_tol = 5, d13c_tol = 0.25 )
std_frame |
Standard data frame to perform swap on. |
site |
NEON four letter site code. |
omit_already_corrected |
Should we attempt correction, if it's already been corrected in the raw files. |
co2_tol |
Tolerance to use to select co2 values that need to be replaced, in ppm. Default = 5 ppm. |
d13c_tol |
Tolerance to use to select d13C values that need to be replaced, in ppm. Default = 0.25 per mil. |
Current sites and time periods affected:
A data.frame, based on std_frame, where NEON-supplied
reference values have been corrected if a mismatch has previously
been identified.
Rich Fiorella [email protected]
Correct carbon ref output
correct_carbon_ref_output( std_list, site, omit_already_corrected = TRUE, co2_tol = 5, d13c_tol = 0.25, ref_gas )correct_carbon_ref_output( std_list, site, omit_already_corrected = TRUE, co2_tol = 5, d13c_tol = 0.25, ref_gas )
std_list |
List containing reference/validation gas measurements. |
site |
Four-letter NEON site code. |
omit_already_corrected |
Skip correction if the reference gas values have already been corrected in the files (default TRUE) If you have older versions of the files, you may want to set this to FALSE. |
co2_tol |
Tolerance used to identify a mismatch in CO2 values. Will correct measured CO2 values within +/- co2_tol within time period identified as having incorrect reference values. |
d13c_tol |
Tolerance used to identify a mismatch in d13C values. Will correct measured d13C values within +/- d13c_tol within time period identified as having incorrect reference values. |
ref_gas |
Which reference gas is being corrected? Expects "co2High", "co2Med", or "co2Low" |
A version of std_list with corrected reference values.
Rich Fiorella [email protected]
delta_to_R
delta_to_R(delta_values, element)delta_to_R(delta_values, element)
delta_values |
A vector of isotope ratios in delta notation. |
element |
Which element to return R values - carbon, oxygen, or hydrogen. |
Vector of isotope ratios (R values).
Rich Fiorella [email protected]
delta_to_R(delta_values = 0, element = 'oxygen') # 2005.2e-6 for VSMOW.delta_to_R(delta_values = 0, element = 'oxygen') # 2005.2e-6 for VSMOW.
estimate_calibration_error
estimate_calibration_error(formula, data)estimate_calibration_error(formula, data)
formula |
Formula to pass to caret::train to perform cross validation. |
data |
Data frame to perform cross-validation on. |
Rich Fiorella [email protected]
extract_carbon_calibration_data.R
extract_carbon_cal_data( data_list, standards = c("co2Low", "co2Med", "co2High") )extract_carbon_cal_data( data_list, standards = c("co2Low", "co2Med", "co2High") )
data_list |
List containing data, from the /*/dp01/data/ group in NEON HDF5 file. |
standards |
Which reference gases (standards) to use? Default is all, but can pass a subset of "co2Low", "co2Med", and "co2High" as a vector to this argument as well. |
Returns data frame of required variables.
Rich Fiorella [email protected]
extract_water_calibration_data
extract_water_calibration_data(data_list)extract_water_calibration_data(data_list)
data_list |
List containing data, from the /*/dp01/data/ group in NEON HDF5 file. |
Returns data frame of required variables.
Rich Fiorella [email protected]
Median absolute deviation filter of Brock 1986.
filter_median_brock86(data, width = 7, threshold = 5)filter_median_brock86(data, width = 7, threshold = 5)
data |
Vector to filter. |
width |
Width of filter, in rows. |
threshold |
Only filter values that are |
Returns filtered vector.
Rich Fiorella [email protected]
fit_carbon_regression
fit_carbon_regression( ref_data, method, calibration_half_width, plot_regression_data = FALSE, plot_dir = "/dev/null", site, min_nobs = NA )fit_carbon_regression( ref_data, method, calibration_half_width, plot_regression_data = FALSE, plot_dir = "/dev/null", site, min_nobs = NA )
ref_data |
Reference data.frame from which to estimate calibration parameters. |
method |
Are we using the Bowling et al. 2003 method ("Bowling_2003") or direct linear regression of d13C and CO2 mole fractions ("linreg")? |
calibration_half_width |
Determines the period (in days) from which reference data are selected (period is 2*calibration_half_width). |
plot_regression_data |
True or false - should we plot the data used in the regression? Useful for debugging. |
plot_dir |
If plot_regression_data is true, where should the plots be saved? |
site |
Needed for regression plots. |
min_nobs |
Minimum number of high-frequency observations to define a peak. |
Returns a data.frame of calibration parameters. If
method == "Bowling_2003", then data.frame includes
gain and offset parameters for 12CO2 and 13CO2, and r^2
values for each regression. If method == "linreg",
then data.frame includes slope, intercept, and r^2 values
for d13C and CO2 values.
Rich Fiorella [email protected]
fit_water_regression
fit_water_regression( ref_data, calibration_half_width, slope_tolerance, r2_thres, plot_regression_data = FALSE, plot_dir = "/dev/null", site, min_nobs = NA )fit_water_regression( ref_data, calibration_half_width, slope_tolerance, r2_thres, plot_regression_data = FALSE, plot_dir = "/dev/null", site, min_nobs = NA )
ref_data |
Reference data.frame from which to estimate calibration parameters. |
calibration_half_width |
Determines the period (in days) from which reference data are selected (period is 2*calibration_half_width). |
slope_tolerance |
Allows for filtering of slopes that deviate from 1 by slope_tolerance. |
r2_thres |
What is the minimum r2 value permitted in a 'useful' calibration relationship. |
plot_regression_data |
True or false - should we plot the data used in the regression? Useful for debugging. |
plot_dir |
If plot_regression_data is true, where should the plots be saved? |
site |
Needed for regression plots. |
min_nobs |
Minimum number of high-frequency observations to define a peak. |
Returns a data.frame of calibration parameters. Output data.frame includes slope, intercept, and r^2 values for d13C and CO2 values.
get_Rstd
get_Rstd(element)get_Rstd(element)
element |
Which element to return standard ratio - carbon, oxygen, or hydrogen. |
Heavy-to-light isotope ratio of most common stable isotope standard. VSMOW for water, VPDB for carbon.
Rich Fiorella [email protected]
get_Rstd("carbon") # returns 0.0111797 get_Rstd("oxygen") # returns 2005.20e-6get_Rstd("carbon") # returns 0.0111797 get_Rstd("oxygen") # returns 2005.20e-6
ingest_data
ingest_data(inname, analyte, name_fix = TRUE, amb_avg, ref_avg)ingest_data(inname, analyte, name_fix = TRUE, amb_avg, ref_avg)
inname |
A file (or list of files) to extract data from for calibration. |
analyte |
Carbon (Co2) or water (H2o)? |
name_fix |
Fix to data frame required for next-generation calibration functions, but breaks old 'by_month()' functions. This parameter provides a necessary work around until these functions are removed. |
amb_avg |
The averaging interval of the ambient data to extract. |
ref_avg |
The averaging interval of the reference data to extract. |
List of data frames, taken from files specified in inname
Rich Fiorella [email protected]
loocv
loocv(mod)loocv(mod)
mod |
Fitted model to estimate leave-one-out CV on. |
Rich Fiorella [email protected]
helper function for the leave-one-out cross variance
Utility function to help retrieve new EC data and/or prune duplicates, as NEON provisions new data or re-provisions data for an existing site and month.
manage_local_EC_archive( file_dir, get = TRUE, unzip_files = TRUE, trim = FALSE, dry_run = TRUE, sites = "all", release = "RELEASE-2023" )manage_local_EC_archive( file_dir, get = TRUE, unzip_files = TRUE, trim = FALSE, dry_run = TRUE, sites = "all", release = "RELEASE-2023" )
file_dir |
Specify the root directory where the local EC store is kept. |
get |
Pull down data from NEON API that does not exist locally? |
unzip_files |
NEON gzips the hdf5 files, should we unzip any gzipped files within file_dir? (Searches recursively) |
trim |
Search through local holdings, and remove older file where there are duplicates? |
dry_run |
List files identified as duplicates, but do not actually delete them? Default true to prevent unintended data loss. |
sites |
Which sites to retrieve data from? Default will be all sites with available data, but can specify a single site or a vector here. |
release |
Download data corresponding to a specific release? Defaults to "RELEASE-2023." To download all data, including provisional data, set to NULL. |
Returns nothing to the environment, but will download new NEON HDF5
files for selected sites (if get = TRUE), unzip them in the local
file directory (if unzip_files = TRUE), and identify and remove
suspected duplicate files (if trim = TRUE and dry_run = FALSE).
Rich Fiorella [email protected]
R_to_delta
R_to_delta(R_values, element)R_to_delta(R_values, element)
R_values |
A vector of isotope ratios (e.g., R values). |
element |
Which element to return delta values - carbon, oxygen, or hydrogen. |
Vector of isotope ratios in delta notation.
Rich Fiorella [email protected]
R_to_delta(R_values = 2005.20e-6, element = 'oxygen') # returns 0.R_to_delta(R_values = 2005.20e-6, element = 'oxygen') # returns 0.
restructure_carbon_variables
restructure_carbon_variables(dataframe, varname, mode, group)restructure_carbon_variables(dataframe, varname, mode, group)
dataframe |
Input data.frame, from |
varname |
Which variable are we applying this function to? There's a list of ~10 common ones to write to the hdf5 file. |
mode |
Are we fixing a reference data frame or an ambient data frame? |
group |
Data, ucrt, or qfqm? |
data.frame formatted for output to hdf5 file.
Wrapper function around restructure_carbon_variables and restructure_water_variables.
restructure_variables(dataframe, varname, mode, group, species)restructure_variables(dataframe, varname, mode, group, species)
dataframe |
Input data.frame, from |
varname |
Which variable are we applying this function to? There's a list of ~10 common ones to write to the hdf5 file. |
mode |
Are we fixing a reference data frame or an ambient data frame? |
group |
Data, ucrt, or qfqm? |
species |
Set to 'Co2' for carbon; 'H2o' for water |
data.frame formatted for output to hdf5 file.
Rich Fiorella [email protected]
restructure_water_variables
restructure_water_variables(dataframe, varname, mode, group)restructure_water_variables(dataframe, varname, mode, group)
dataframe |
Input data.frame, from |
varname |
Which variable are we applying this function to? There's a list of ~10 common ones to write to the hdf5 file. |
mode |
Are we fixing a reference data frame or an ambient data frame? |
group |
Data, ucrt, or qfqm? |
data.frame formatted for output to hdf5 file.
select_daily_reference_data
select_daily_reference_data(standard_df, analyte, min_nobs = NA)select_daily_reference_data(standard_df, analyte, min_nobs = NA)
standard_df |
Input reference data.frame. |
analyte |
Are we calibrating CO2 and H2O? (Use argument 'co2' or 'h2o', or else function will throw error) |
min_nobs |
Minimum number of high-frequency
observations to define a peak. If not supplied,
defaults are 200 for |
Smaller data.frame where only the reference data selected to use in the calibration routines is returned. Assumes that we are calibrating on a daily basis, and not on a longer time scale. Data are selected based on two criteria: cannot be missing, and must be at least a certain number of high-frequency observations in order to qualify as a valid measurement. For the water system, this function also keeps only the last three injections for each reference water per day.
Creates a skeleton hdf5 file for the calibrated data.
setup_output_file(inname, outname, site, analyte)setup_output_file(inname, outname, site, analyte)
inname |
Input file name. |
outname |
Output file name. |
site |
NEON 4-letter site code. |
analyte |
Carbon ('Co2') or water ('H2o') system? |
Nothing to the environment, but creates a new data file with the most basic output HDF5 structure consistent with NEON's data files.
Rich Fiorella [email protected]
There are a few suspected instances where the water isotope ratios for oxygen and hydrogen have been flipped in the reference data. This function corrects them until they are corrected in the NEON database using a d-excess filter.
swap_standard_isotoperatios(std_frame, dxs_thres = 500)swap_standard_isotoperatios(std_frame, dxs_thres = 500)
std_frame |
Standard data frame to perform swap on. |
dxs_thres |
d-excess threshold to indicate when to swap. |
A data.frame based on std_frame, where d18O and
d2H values have been swapped from NEON input files if
determined to have a reference value mismatch. Mismatch
is determined based on the d-excess of the standard (=
d2H - 8*d18O), using a value of 500 by default.
Rich Fiorella [email protected]
terrestrial_core_sites
terrestrial_core_sites()terrestrial_core_sites()
A vector listing NEON core terrestrial sites.
Rich Fiorella [email protected]
terrestrial_core_sites()terrestrial_core_sites()
terrestrial_gradient_sites
terrestrial_gradient_sites()terrestrial_gradient_sites()
A vector listing NEON gradient terrestrial sites.
Rich Fiorella [email protected]
terrestrial_gradient_sites()terrestrial_gradient_sites()
validate_analyte
validate_analyte(analyte)validate_analyte(analyte)
analyte |
Co2 or H2o? |
Standardized string for the water ('H2o') or carbon ('Co2') systems to make sure strings are standardized across package functions.
Rich Fiorella [email protected]
validate_output_file
validate_output_file(inname, outname, site, analyte)validate_output_file(inname, outname, site, analyte)
inname |
Input file name. |
outname |
Output file name. |
site |
NEON 4-letter site code. |
analyte |
Carbon ('Co2') or water ('H2o') system? |
Nothing to environment, simply checks to make sure expected groups are in output.
Rich Fiorella [email protected]
water_isotope_sites
water_isotope_sites()water_isotope_sites()
A vector listing NEON sites measuring water vapor isotope ratios.
Rich Fiorella [email protected]
water_isotope_sites()water_isotope_sites()
Write out ambient observations from the NEON EC towers where the isotope data (either H2O or CO2) have been calibrated using this package.
write_carbon_ambient_data(outname, site, amb_data_list, to_file = TRUE)write_carbon_ambient_data(outname, site, amb_data_list, to_file = TRUE)
outname |
Output file name. |
site |
NEON 4-letter site code. |
amb_data_list |
Calibrated list of ambient data - this is the output from one of the calibrate_ambient_carbon* functions. |
to_file |
Write to file (TRUE) or to environment (FALSE). |
Nothing to the environment, but writes data in amb_data_list to file.
Rich Fiorella [email protected]
write_carbon_calibration_data
write_carbon_calibration_data(outname, site, cal_df, method, to_file = TRUE)write_carbon_calibration_data(outname, site, cal_df, method, to_file = TRUE)
outname |
Output file name. |
site |
NEON 4-letter site code. |
cal_df |
Calibration data frame - this is the output from fit_carbon_regression |
method |
Was the Bowling et al. 2003 or the linear regression method used in fit_carbon_regression? |
to_file |
Write to file (TRUE) or to environment (FALSE). |
Nothing to the environment, but writes out the calibration parameters (e.g., gain and offset or regression slopes and intercepts) to the output hdf5 file.
Rich Fiorella [email protected]
Write NEON's qfqm data for an isotope species to output file. Wraps copy_qfqm_group.
write_qfqm(inname, outname, site, analyte)write_qfqm(inname, outname, site, analyte)
inname |
Input file name. |
outname |
Output file name. |
site |
NEON 4-letter site code. |
analyte |
Carbon ('Co2') or water ('H2o') system? |
Nothing to the environment, but writes qfqm data to file.
Rich Fiorella [email protected]
Write NEON's ucrt data for an isotope species to output file. Wraps copy_ucrt_group.
write_ucrt(inname, outname, site, analyte)write_ucrt(inname, outname, site, analyte)
inname |
Input file name. |
outname |
Output file name. |
site |
NEON 4-letter site code. |
analyte |
Carbon ('Co2') or water ('H2o') system? |
Nothing to the environment, but writes ucrt data to file.
Rich Fiorella [email protected]
Write out ambient observations from the NEON EC towers where the isotope data have been calibrated using this package.
write_water_ambient_data(outname, site, amb_data_list)write_water_ambient_data(outname, site, amb_data_list)
outname |
Output file name. |
site |
NEON 4-letter site code. |
amb_data_list |
Calibrated list of ambient data - this is the output from one of the calibrate_ambient_water* functions. |
Nothing to the environment, but writes data in amb_data_list to file.
Rich Fiorella [email protected]
write_water_calibration_data
write_water_calibration_data(outname, site, cal_df)write_water_calibration_data(outname, site, cal_df)
outname |
Output file name. |
site |
NEON 4-letter site code. |
cal_df |
Calibration data frame - this is the output from fit_water_regression |
Nothing to the environment, but writes out the calibration parameters (e.g., regression slopes and intercepts) to the output hdf5 file.
Rich Fiorella [email protected]