choosePositions {Bchron}

Compute positions to date next which result in maximal decrease of chronological uncertainty

Description

This function finds, for a given current chronology, created via Bchronology, which positions (depths) to date next If N = 1 it just finds the position with the biggest uncertainty If N>1 it puts a date at the N = 1 position and re-runs Bchronology with the extra psuedo date. It uses the unCalibrate function with the un-calibrated age estimated at the median of the chronology and the sd as specified via the newSds argument. Other arguments specify the new thicknesses, calibration curves, and outlier probabilities for newly inserted psuedo-dates.

Usage

choosePositions(
  bchrRun,
  N = 1,
  newSds = 30,
  newThicknesses = 0,
  positions = bchrRun$predictPositions,
  newCalCurve = "intcal20",
  newOutlierProb = 0.05,
  level = 0.5,
  plot = TRUE,
  count = 1,
  linesAt = NULL
)

Arguments

bchrRun

A run of the current chronology as output from Bchronology

N

The number of new positions required

newSds

The new standard deviations of the psuedo-added dates

newThicknesses

The new thicknesses of the psuedo-added dates

positions

The positions allowed to estimate the new positions to date. Defaults to the value of predictPositions from the Bchronology run

newCalCurve

The new calibration curve of the psuedo-added dates

newOutlierProb

The new outlier probabilities of the psuedo-added dates

level

The confidence level required for minimising the uncertainty. Defaults to 50%. (Note: this will be estimated more robustly than the 95% level)

plot

Whether to plot the chronologies as they are produced

count

Counter function (not for use other than by the function itself)

linesAt

Horizontal line positions (not for use other than by the function itself)

Value

Some plots and the positions to date next

See Also

Bchronology for the main function to create chronologies, unCalibrate for the ability to invert calendar dates for a given calibration curve.

Examples

data(Glendalough)
GlenOut <- Bchronology(
  ages = Glendalough$ages,
  ageSds = Glendalough$ageSds,
  calCurves = Glendalough$calCurves,
  positions = Glendalough$position,
  positionThicknesses = Glendalough$thickness,
  ids = Glendalough$id,
  predictPositions = seq(0, 1500, by = 10)
)

# Find out which two positions (depths) to date if we have room for two more dates
# Here going to choose 3 new positions to date
newPositions <- choosePositions(GlenOut, N = 3)
print(newPositions)

# Suppose you are only interested in dating the new depths at 500, 600, or 700 cm
newPositions2 <- choosePositions(GlenOut,
  N = 2,
  positions = seq(500, 700, by = 10)
)
print(newPositions2)