Skip to contents

Analyse SAR CW-OSL measurements

Source: R/analyse_SAR.CWOSL.R
analyse_SAR.CWOSL.Rd

The function performs a SAR CW-OSL analysis on an RLum.Analysis object including growth curve fitting.

Usage

analyse_SAR.CWOSL(
  object,
  signal.integral.min = NA,
  signal.integral.max = NA,
  background.integral.min = NA,
  background.integral.max = NA,
  OSL.component = NULL,
  rejection.criteria = list(),
  dose.points = NULL,
  trim_channels = FALSE,
  mtext.outer = "",
  plot = TRUE,
  plot_onePage = FALSE,
  plot.single = FALSE,
  onlyLxTxTable = FALSE,
  ...
)

Arguments

object

RLum.Analysis (required): input object containing data for analysis, alternatively a list of RLum.Analysis objects can be provided. The object should contain only curves considered part of the SAR protocol (see Details.)

signal.integral.min

integer (required): lower bound of the signal integral. Can be a list of integers, if object is of type list. If the input is vector (e.g., c(1,2)) the 2nd value will be interpreted as the minimum signal integral for the Tx curve. Can be set to NA, in this case no integrals are taken into account.

signal.integral.max

integer (required): upper bound of the signal integral. Can be a list of integers, if object is of type list. If the input is vector (e.g., c(1,2)) the 2nd value will be interpreted as the maximum signal integral for the Tx curve. Can be set to NA, in this case no integrals are taken into account.

background.integral.min

integer (required): lower bound of the background integral. Can be a list of integers, if object is of type list. If the input is vector (e.g., c(1,2)) the 2nd value will be interpreted as the minimum background integral for the Tx curve. Can be set to NA, in this case no integrals are taken into account.

background.integral.max

integer (required): upper bound of the background integral. Can be a list of integers, if object is of type list. If the input is vector (e.g., c(1,2)) the 2nd value will be interpreted as the maximum background integral for the Tx curve. Can be set to NA, in this case no integrals are taken into account.

OSL.component

character or integer (optional): s single index or a character defining the signal component to be evaluated. It requires that the object was processed by [OSLdecomposition::RLum.OSL_decomposition]. This argument can either be the name of the OSL component assigned by [OSLdecomposition::RLum.OSL_global_fitting] or the index in the descending order of decay rates. Then "1" selects the fastest decaying component, "2" the second fastest and so on. Can be a list of integers or strings (or mixed) If object is a list and this parameter is provided as list it alternates over the elements (aliquots) of the object list, e.g., list(1,2) processes the first aliquot with component 1 and the second aliquot with component 2. NULL does not process any component.

rejection.criteria

list (with default): provide a named list and set rejection criteria in percentage for further calculation. Can be a list in a list, if object is of type list. Note: If an unnamed list is provided the new settings are ignored!

Allowed arguments are recycling.ratio, recuperation.rate, palaeodose.error, testdose.error, exceed.max.regpoint = TRUE/FALSE, recuperation_reference = "Natural" (or any other dose point, e.g., "R1"). Example: rejection.criteria = list(recycling.ratio = 10). Per default all numerical values are set to 10, exceed.max.regpoint = TRUE. Every criterion can be set to NA. In this value are calculated, but not considered, i.e. the RC.Status becomes always 'OK'

dose.points

numeric (optional): a numeric vector containing the dose points values. Using this argument overwrites dose point values extracted from other data. Can be a list of numeric vectors, if object is of type list

trim_channels

logical (with default): trim channels per record category to the lowest number of channels in the category by using trim_RLum.Data. Applies only to OSL and IRSL curves. For a more granular control use trim_RLum.Data before passing the input object.

mtext.outer

character (optional): option to provide an outer margin mtext. Can be a list of characters, if object is of type list

plot

logical (with default): enables or disables plot output.

plot_onePage

logical (with default): enables or disables on page plot output

plot.single

logical (with default) or numeric (optional): single plot output (TRUE/FALSE) to allow for plotting the results in single plot windows. If a numeric vector is provided the plots can be selected individually, i.e. plot.single = c(1,2,3,4) will plot the TL and Lx, Tx curves but not the legend (5) or the growth curve (6), (7) and (8) belong to rejection criteria plots. Requires plot = TRUE.

onlyLxTxTable

logical (with default): If TRUE the dose response curve fitting and plotting is skipped. This allows to get hands on the Lx/Tx table for large datasets without the need for a curve fitting.

...

further arguments that will be passed to the function plot_GrowthCurve or calc_OSLLxTxRatio (supported: background.count.distribution, sigmab, sig0). Please note that if you consider to use the early light subtraction method you should provide your own sigmab value!

Value

A plot (optional) and an RLum.Results object is returned containing the following elements:

data

data.frame containing De-values, De-error and further parameters

LnLxTnTx.values

data.frame of all calculated Lx/Tx values including signal, background counts and the dose points

rejection.criteria

data.frame with values that might by used as rejection criteria. NA is produced if no R0 dose point exists.

Formula

formula formula that have been used for the growth curve fitting

The output should be accessed using the function get_RLum.

Details

The function performs an analysis for a standard SAR protocol measurements introduced by Murray and Wintle (2000) with CW-OSL curves. For the calculation of the Lx/Tx value the function calc_OSLLxTxRatio is used. For changing the way the Lx/Tx error is calculated use the argument background.count.distribution and sigmab, which will be passed to the function calc_OSLLxTxRatio.

What is part of a SAR sequence?

The function is rather picky when it comes down to accepted curve input (OSL,IRSL,...) and structure. A SAR sequence is basically a set of \(L_{x}/T_{x}\) curves. Hence, every 2nd curve is considered a shine-down curve related to the test dose. It also means that the number of curves for \(L_{x}\) has to be equal to the number of \(T_{x}\) curves, and that hot-bleach curves do not belong into a SAR sequence; at least not for the analysis. Other curves allowed and processed are preheat curves, or preheat curves measured as TL, and irradiation curves. The later one indicates the duration of the irradiation, the dose and test dose points, e.g., as part of XSYG files.

Argument object is of type list

If the argument object is of type list containing only RLum.Analysis objects, the function re-calls itself as often as elements are in the list. This is useful if an entire measurement wanted to be analysed without writing separate for-loops. To gain in full control of the parameters (e.g., dose.points) for every aliquot (corresponding to one RLum.Analysis object in the list), in this case the arguments can be provided as list. This list should be of similar length as the list provided with the argument object, otherwise the function will create an own list of the requested length. Function output will be just one single RLum.Results object.

Please be careful when using this option. It may allow a fast an efficient data analysis, but the function may also break with an unclear error message, due to wrong input data.

Working with IRSL data

The function was originally designed to work just for 'OSL' curves, following the principles of the SAR protocol. An IRSL measurement protocol may follow this procedure, e.g., post-IR IRSL protocol (Thomsen et al., 2008). Therefore this functions has been enhanced to work with IRSL data, however, the function is only capable of analysing curves that follow the SAR protocol structure, i.e., to analyse a post-IR IRSL protocol, curve data have to be pre-selected by the user to fit the standards of the SAR protocol, i.e., Lx,Tx,Lx,Tx and so on.

Example: Imagine the measurement contains pIRIR50 and pIRIR225 IRSL curves. Only one curve type can be analysed at the same time: The pIRIR50 curves or the pIRIR225 curves.

Supported rejection criteria

[recycling.ratio]: calculated for every repeated regeneration dose point.

[recuperation.rate]: recuperation rate calculated by comparing the Lx/Tx values of the zero regeneration point with the Ln/Tn value (the Lx/Tx ratio of the natural signal). For methodological background see Aitken and Smith (1988). As a variant with the argument recuperation_reference another dose point can be selected as reference instead of Ln/Tn.

[testdose.error]: set the allowed error for the test dose, which per default should not exceed 10%. The test dose error is calculated as Tx_net.error/Tx_net. The calculation of the \(T_{n}\) error is detailed in calc_OSLLxTxRatio.

[palaeodose.error]: set the allowed error for the De value, which per default should not exceed 10%.

Irradiation times

The function makes two attempts to extra irradiation data (dose points) automatically from the input object, if the argument dose.points was not set (aka set to NULL).

  1. It searches in every curve for an info object called IRR_TIME. If this was set, any value set here is taken as dose point.

  2. If the object contains curves of type irradiation, the function tries to use this information to assign these values to the curves. However, the function does not overwrite values preset in IRR_TIME.

Note

This function must not be mixed up with the function Analyse_SAR.OSLdata, which works with Risoe.BINfileData objects.

The function currently does support only 'OSL', 'IRSL' and 'POSL' data!

Function version

0.10.3

How to cite

Kreutzer, S., 2024. analyse_SAR.CWOSL(): Analyse SAR CW-OSL measurements. Function version 0.10.3. In: Kreutzer, S., Burow, C., Dietze, M., Fuchs, M.C., Schmidt, C., Fischer, M., Friedrich, J., Mercier, N., Philippe, A., Riedesel, S., Autzen, M., Mittelstrass, D., Gray, H.J., Galharret, J., Colombo, M., 2024. Luminescence: Comprehensive Luminescence Dating Data Analysis. R package version 0.9.26. https://r-lum.github.io/Luminescence/

References

Aitken, M.J. and Smith, B.W., 1988. Optical dating: recuperation after bleaching. Quaternary Science Reviews 7, 387-393.

Duller, G., 2003. Distinguishing quartz and feldspar in single grain luminescence measurements. Radiation Measurements, 37 (2), 161-165.

Murray, A.S. and Wintle, A.G., 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32, 57-73.

Thomsen, K.J., Murray, A.S., Jain, M., Boetter-Jensen, L., 2008. Laboratory fading rates of various luminescence signals from feldspar-rich sediment extracts. Radiation Measurements 43, 1474-1486. doi:10.1016/j.radmeas.2008.06.002

See also

calc_OSLLxTxRatio, plot_GrowthCurve, RLum.Analysis, RLum.Results, get_RLum

Author

Sebastian Kreutzer, Institute of Geography, Heidelberg University (Germany) , RLum Developer Team

Examples


##load data
##ExampleData.BINfileData contains two BINfileData objects
##CWOSL.SAR.Data and TL.SAR.Data
data(ExampleData.BINfileData, envir = environment())

##transform the values from the first position in a RLum.Analysis object
object <- Risoe.BINfileData2RLum.Analysis(CWOSL.SAR.Data, pos=1)

##perform SAR analysis and set rejection criteria
results <- analyse_SAR.CWOSL(
object = object,
signal.integral.min = 1,
signal.integral.max = 2,
background.integral.min = 900,
background.integral.max = 1000,
log = "x",
fit.method = "EXP",
rejection.criteria = list(
  recycling.ratio = 10,
  recuperation.rate = 10,
  testdose.error = 10,
  palaeodose.error = 10,
  recuperation_reference = "Natural",
  exceed.max.regpoint = TRUE)
)

#> [plot_GrowthCurve()] Fit: EXP (interpolation) | De = 1668.25 | D01 = 1982.76



##show De results
get_RLum(results)
#>         De De.Error      D01 D01.ERROR D02 D02.ERROR Dc       n_N De.MC Fit
#> 1 1668.251 48.06899 1982.757  100.0438  NA        NA NA 0.4854696  1663 EXP
#>   HPDI68_L HPDI68_U HPDI95_L HPDI95_U RC.Status signal.range background.range
#> 1 1602.275 1709.013 1569.894 1761.183        OK        1 : 2       900 : 1000
#>   signal.range.Tx background.range.Tx ALQ POS
#> 1         NA : NA             NA : NA   1   1
#>                                  UID
#> 1 2024-11-18-01:33.0.878804486710578

##show LnTnLxTx table
get_RLum(results, data.object = "LnLxTnTx.table")
#>      Name Repeated Dose  LnLx   LnLx.BG TnTx   TnTx.BG    Net_LnLx
#> 1 Natural    FALSE    0 20391  60.15842 4771  73.50495 20330.84158
#> 2      R1    FALSE  450  7591  65.70297 4977  81.56436  7525.29703
#> 3      R2    FALSE 1050 15150  96.17822 4960 101.92079 15053.82178
#> 4      R3    FALSE 2000 23700 118.49505 5064 121.06931 23581.50495
#> 5      R4    FALSE 2550 31094 155.92079 5715 145.84158 30938.07921
#> 6      R5     TRUE  450  9376 122.37624 5860 127.60396  9253.62376
#> 7      R0    FALSE    0   192  94.39604 6107 117.64356    97.60396
#>   Net_LnLx.Error Net_TnTx Net_TnTx.Error       LxTx  LxTx.Error
#> 1      143.03415 4697.495       69.24882 4.32801767 0.070695494
#> 2       87.39498 4895.436       71.06916 1.53720681 0.028578369
#> 3      123.18402 4858.079       71.66840 3.09871888 0.052275097
#> 4      154.46828 4942.931       72.07168 4.77075371 0.076258410
#> 5      176.39564 5569.158       75.88226 5.55525214 0.082052531
#> 6       97.54912 5732.396       77.37976 1.61426805 0.027647946
#> 7       14.86509 5989.356       78.87190 0.01629624 0.002491179
#>                                  UID
#> 1 2024-11-18-01:33.0.878804486710578
#> 2 2024-11-18-01:33.0.878804486710578
#> 3 2024-11-18-01:33.0.878804486710578
#> 4 2024-11-18-01:33.0.878804486710578
#> 5 2024-11-18-01:33.0.878804486710578
#> 6 2024-11-18-01:33.0.878804486710578
#> 7 2024-11-18-01:33.0.878804486710578