R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Relative sea level rate (RSL) estimation

BchronRSL

R Documentation

Relative sea level rate (RSL) estimation

Description

This function takes output from a Bchronology run, some RSL mean and standard deviations, and fits an errors-in-variables polynomial regression model which takes account of the age uncertainty. It then esimates rates and accelerations in RSL with appropriate uncertainty quantification

Usage

BchronRSL(BchronologyRun, RSLmean, RSLsd, degree = 1, iterations = 10000, 
burn = 2000, thin = 8)

Arguments

`BchronologyRun`	Output from a run of `Bchronology`
`RSLmean`	A vector of RSL mean estimates of the same length as the number of predictPositions given to the `Bchronology` function
`RSLsd`	A vector RSL standard deviations of the same length as the number of predictPositions given to the `Bchronology` function
`degree`	(optional) The degree of the polynomial regression: linear=1 (default), quadratic=2, etc. Supports up to degree 5, though this will depend on the data given
`iterations`	(optional) The number of MCMC iterations to run
`burn`	(optional) The number of starting iterations to discard
`thin`	(optional) The step size of iterations to discard

Details

This function fits an errors-in-variables regression model to relative sea level (RSL) data. An errors-in-variables regression model allows for uncertainty in the explanatory variable, here the age of sea level data point. The algorithm is more fully defined in the reference below

Value

An object of class BchronRSLRun with elements

`BchronologyRun`	The output from the run of `Bchronology`
`samples`	The posterior samples of the regression parameters
`degree`	The degree of the polynomial regression
`RSLmean`	The RSL mean values given to the function
`RSLsd`	The RSL standard deviations as given to the function
`const`	The mean of the predicted age values. Used to standardise the design matrix and avoid computational issues

Author(s)

Andrew Parnell <andrew.parnell@ucd.ie>

References

Andrew C. Parnell and W. Roland Gehrels (2013) 'Using chronological models in late holocene sea level reconstructions from salt marsh sediments' In: I. Shennan, B.P. Horton, and A.J. Long (eds). Handbook of Sea Level Research. Chichester: Wiley

Examples

## Not run: # Load in data
data(TestChronData)
data(TestRSLData)

# Run through Bchronology
RSLrun = Bchronology(ages=TestChronData$ages,ageSds=TestChronData$ageSds,
positions=TestChronData$position,positionThicknesses=TestChronData$thickness,
ids=TestChronData$id,calCurves=TestChronData$calCurves,
predictPositions=TestRSLData$Depth)

# Now run through BchronRSL
RSLrun2 = BchronRSL(RSLrun,RSLmean=TestRSLData$RSL,RSLsd=TestRSLData$Sigma,degree=3)

# Summarise it
summary(RSLrun2)

# Plot it
plot(RSLrun2)
## End(Not run)