numOSL (version 1.0)

sgcED: Fitting dose response curve and calculating equivalent dose using standardised growth curves (SGC) method

Description

Its a wrapped version of function calED(), fitting OSL dose response curve and calculating equivalent dose values using standardised growth curves (SGC) method after Roberts HM and Duller GAT (2004), the errors of equivalent doses are estimated after Duller GAT (2007). It is based on the fact that using information collected as part of standard SAR measurements, it is possible to construct reproducible patterns of growth from multiple-grain aliquots of a large number of samples (Roberts HM and Duller GAT, 2004).

Usage

sgcED(Curvedata, Ltx, 
      model = c("line","exp","line+exp"), origin = FALSE,
      nstart = 100, upb = 1, ErrorMethod = c("mc","sp"),
      nsim = 1000, plot = TRUE, samplename = NULL, outfile = NULL)

Arguments

Curvedata

data.frame(required): three columns, the same as that in function calED

Ltx

data.frame(required): two columns (standardized signals and its standard errors), from which equivalent doses can be estimated by interpolating

model

character(with default): a model("linear", "exponential" or "linear+exponential") used for fitting the dose response curve, the same as that in function calED

origin

logical(optional): whether force the fitting to pass the origin (x=0,y=0) or not

nstart

numeric(with default): maximum number of attempts that used to initialize parameters in curve fitting, the same as that in function calED

upb

numeric(with default): upper boundary of b value, initial b values will be generated uniformly from (0, upb), the same as that in function calED

ErrorMethod

character(with default): method ("sp" or "mc") for estimating the standard errors of equivalent doses, see calED for more information

nsim

numeric(with default): maximum simulative number allowed if using Monte Carlo method for error assessing

plot

logical(with default): whether drawing a plot or not

samplename

character(optional): name of the sample

outfile

character(optional): if specified, calculated equivalent doses will be written to a file of name outfile in .csv format and saved to the current work directory

Value

• Return a visible list that contains following elements:
• LMpars
  characteristic parameters and standard errors of the dose response curve obtained by Levenberg-Marquardt method
• residual
  sum of square of residual errors
• fit.value
  observations .VS. fitted values
• ED
  estimated equivalent doses (standard errors)

Details

see function calED() for details.

References

Roberts, H.M. and Duller, G.A.T., 2004. Standardised growth curves for optical dating of sediment using multiple-grain aliquots. Radiation Measurements 38(2), pp. 241-252.

Duller, G.A.T., 2007. Assessing the error on equivalent dose estimates derived from single aliquot regenerative dose measurements. Ancient TL 25(1), pp. 15-24.

Roberts, H.M., Durcan, J.A., Duller, G.A.T., 2009. Exploring procedures for the rapid assessment of optically stimulated luminescence range-finder ages. Radiation Measurements, 44(5-6), pp. 582-587.

Jorge More, Burton Garbow, Kenneth Hillstrom, User Guide for MINPACK-1, Technical Report ANL-80-74, Argonne National Laboratory, 1980.

Further reading

Long, H., Lai, Z.P., Fan, Q.S., Sun, Y.J., Liu, X.J., 2010. Applicability of a quartz OSL standardised growth curve for De determination up to 400 Gy for lacustrine sediments from the Qaidam Basin of the Qinghai-Tibetan Plateau. Quaternary Geochronology 5(2-3), pp. 212-217.

Yang, L.H., Lai, Z.P., Long, H., Zhang, J.R., 2011. Construction of a quartz OSL standardised growth curve (SGC) for aeolian samples from the Horqin dunefield in northeastern China. Geochronometria 38(4), pp. 391-396.

See Also

calED

Examples

ltx1<-c(0.032,1.61,2.55,3.21,3.87,0.031,1.55) # Lx/Tx for the first aliquot
 ltx2<-c(0.025,1.44,2.47,3.35,4.17,0.033,1.47) # Lx/Tx for the second aliquot
 ltx3<-c(0.027,1.51,2.68,3.52,4.41,0.021,1.39) # Lx/Tx for the third aliquot
 ltx4<-c(0.018,1.71,2.28,3.81,4.03,0.017,1.62) # Lx/Tx for the four aliquot
 ltx5<-c(0.026,1.49,1.99,3.43,4.17,0.015,2.01) # Lx/Tx for the five aliquot
 ltx<-cbind(ltx1,ltx2,ltx3,ltx4,ltx5)
 ltx<-cbind(apply(ltx,MARGIN=1,mean),
            apply(ltx,MARGIN=1,sd))  # means an standard deviations
 redose<-c(0,18,36,54,72,0,18)       # the same ReDose for the five aliquots
 Curvedata<-data.frame(redose,ltx)
 Ltx<-data.frame(c(0.5,1.0,1.8,2.3,2.8,3.1,3.6,4.0),
                 rep(0.1,8))         # Lx/Txs from which EDs are expected
 sgcED(Curvedata,Ltx,model="line+exp",origin=TRUE) # fitting y=a*(1-exp(-b*x))+c*x