arco: Screen peaks when input data is charcoal areas

In wfinsinger/tapas: Trend And PeAkS analysis in paleo-ecological time series

Screen peaks when input data is charcoal areas

Description

Screens peaks when input data is charcoal-area records. Modified from the ARCO v1 function, available at urlhttps://github.com/wfinsinger/ARCO.

Usage

arco(
  Seedle.file,
  Smpl.file,
  FireA.file,
  FireC.file,
  n.boot = 10000,
  thresh.prob = 0.95,
  win.width = 1000,
  plotit = FALSE,
  breakage = FALSE,
  storedat = FALSE
)

Arguments

Seedle.file	A data frame with charcoal-particle areas. Should have as many rows as the number of charcoal particles and two columns in the following order: Column 1: Top depth of samples; Column 2: Charcoal-particle areas.
Smpl.file	A data frame with as many rows as the number of samples and seven columns in the following order: CmTop, CmBot, AgeTop, AgeBot, volume, charcoal counts, charcoal areas.
FireA.file	Output from peak-detection analysis based on charcoal areas. Generate it using first the tapas::peak_detection function (set argument min_CountP = NULL), and thereafter the tapas::tapas_export() function.
FireC.file	Output from peak-detection analysis based on charcoal counts. Generate it using first the tapas::peak_detection function (set argument 0 < min_CountP < 1), and thereafter the tapas::tapas_export() function.
n.boot	Number of bootstrap samples generated by the function to obtain a distribution of simulated charcoal-areas. By default, n.boot = 10000.
thresh.prob	The pth percentile threshold used to separate significant charcoal-area peaks. By default thresh.prob = 0.95.
win.width	The temporal span of the window from which bootstrap samples are generated. For each peak to be screened, particles are randomly drawn (with replacement) from all samples within a focal window, which is centered on the peak and has a full span of win.width. By default win.width = 1000.
plotit	Logical. If plotit = FALSE (default), plots are not sent to the device.
breakage	Logical. If breakage = FALSE, plots also C#/CA-ratios in one of the diagnostic plots. By default breakage = FALSE.
storedat	Logical. If storedat = TRUE, the function returns also other output data related to charcoal-particle areas.

Details

This screening procedure is specific for data sets comprising charcoal numbers (counts) and areas. It screens the charcoal-area estimates with respect to the number and size of charcoal particles.

The method begins with a charcoal-area data set analysed by existing methods to identify peaks representing fire episodes, e.g. the peak_detection() function with the argument min_CountP = NULL). To screen these peaks, the method uses bootstrap resampling of charcoal-particle areas observed in a user-defined subsection of the data set around each peak (by default win.width = 1000 to obtain the range of likely charcoal areas for different counts. Peaks with total area within the likely range of bootstrapped samples (e.g. p > 0.05) are flagged as potentially unreliable, whereas peak samples with total area significantly greater than expected by chance are deemed robust indicators of past fire episodes.

Value

By default (as with storedat = FALSE), the function returns a list with the same structure as the input list (FireA.file).

Author(s)

Ryan Kelly

Walter Finsinger

References

Finsinger, W., R. Kelly, J. Fevre, and E.K. Magyari. 2014. A guide to screening charcoal peaks in macrocharcoal-area records for fire episode reconstructions. The Holocene, 24:1002–1008. doi: 10.1177/0959683614534737

Higuera, P.E., L.B. Brubaker, P.M. Anderson, F.S. Hu, and T.A. Brown. 2009. Vegetation mediated the impacts of postglacial climate change on fire regimes in the south-central Brooks Range, Alaska. Ecological Monographs, 79:201–219.

wfinsinger/tapas documentation built on Aug. 22, 2024, 4:28 a.m.