bury: Age-Depth Modeling

Description

Computes the age-depth curve from the output of the MCMC algorithm and the known depth of each dated samples.

Usage

bury(object, depth, ...)
# S4 method for EventsMCMC,numeric
bury(object, depth, span = 0.75, degree = 2)
# S4 method for AgeDepthModel
predict(object, newdata)
# S4 method for AgeDepthModel,missing
plot(
  x,
  level = 0.95,
  calendar = get_calendar(),
  main = NULL,
  sub = NULL,
  ann = graphics::par("ann"),
  axes = TRUE,
  frame.plot = axes,
  panel.first = NULL,
  panel.last = NULL,
  ...
)

Value

bury() returns an AgeDepthModel object.
predict() returns an EventsMCMC object.
plot() is called it for its side-effects: it results in a graphic being displayed (invisibly returns x).

Arguments

object: An EventsMCMC object.
depth: A numeric vector giving of the depths of the dated samples.
...: Other graphical parameters may also be passed as arguments to this function, particularly, border, col, lwd, lty or pch.
span: A length-one numeric vector giving the regression parameter (see stats::loess()).
degree: A length-one numeric vector giving the degree of the polynomials to be used (see stats::loess()).
newdata: A numeric vector giving the depths at which ages will be predicted. If missing, the original data points are used.
x: An AgeDepthModel object.
level: A length-one numeric vector giving the confidence level.
calendar: A aion::TimeScale object specifying the target calendar (see aion::calendar()).
main: A character string giving a main title for the plot.
sub: A character string giving a subtitle for the plot.
ann: A logical scalar: should the default annotation (title and x and y axis labels) appear on the plot?
axes: A logical scalar: should axes be drawn on the plot?
frame.plot: A logical scalar: should a box be drawn around the plot?
panel.first: An an expression to be evaluated after the plot axes are set up but before any plotting takes place. This can be useful for drawing background grids.
panel.last: An expression to be evaluated after plotting has taken place but before the axes, title and box are added.

Author

A. Philippe

Details

We assume it exists a function \(f\) relating the age and the depth \(age = f(depth)\). We estimate the function using local regression (also called local polynomial regression): \(f = loess(age ~ depth)\). This estimated function \(f\) depends on the unknown dates. However, from the posterior distribution of the age/date sequence, we can evaluate the posterior distribution of the age function for each desired depth.

References

Jha, D. K., Sanyal, P. & Philippe, A. (2020). Multi-Proxy Evidence of Late Quaternary Climate and Vegetational History of North-Central India: Implication for the Paleolithic to Neolithic Phases. Quaternary Science Reviews, 229: 106121. tools:::Rd_expr_doi("10.1016/j.quascirev.2019.106121").

Ghosh, S., Sanyal, P., Roy, S., Bhushan, R., Sati, S. P., Philippe, A. & Juyal, N. (2020). Early Holocene Indian Summer Monsoon and Its Impact on Vegetation in the Central Himalaya: Insight from dD and d13C Values of Leaf Wax Lipid. The Holocene, 30(7): 1063-1074. tools:::Rd_expr_doi("10.1177/0959683620908639").

Examples

Run this code

## Coerce to MCMC
eve <- matrix(rnorm(6000, (1:6)^2), ncol = 6, byrow = TRUE)
eve <- as_events(eve, calendar = CE())

## Compute an age-depth curve
age <- bury(eve, depth = 1:6)
plot(age)

## Predict new values
new <- predict(age, newdata = 1.5:5.5)
summary(new)

plot(eve)
plot(new)

Run the code above in your browser using DataLab