SNI: Calculate signal-to-noise index (SNI) for paleoecological...

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

Calculate signal-to-noise index (SNI) for paleoecological timeseries records.

Description

This function computes SNI as described in Kelly et al. 2011. Note that your data must be interpolated to constant sample resolution (yr/sample) before input to the function. The function makes no assumption about prior analysis on the input series, i.e. any background and threshold methods may be used. However, input data should still be consistent with the interpretation that a value (column 2) greater than the corresponding threshold value (column 3) is a signal sample, whereas a value below the threshold is noise. Refer to Kelly et al. 2010 for details and discussion.

Usage

SNI(ProxyData, BandWidth)

Arguments

ProxyData	Matrix of input data with one row per sample, containing: Column 1 age associated with the sample (yr) Column 2 proxy accumulation rate (e.g. CHAR) of the sample (pieces/cm^2/yr) Column 3 threshold value (pieces/cm^2/yr)
BandWidth	Width of moving window for computing SNI.

Value

The function was modified by Walter Finsinger to return only a subset of the original SNI_output list:

$SNI_sm

Lowess-smoothed SNI values.

$SNI_raw

Raw SNI values.

In the original script, the returned data list contained the computed SNI and related data, with one row for each row in the input variable (ProxyData):

$SNI_sm

The smoothed SNI computed for each sample.

$winInd

Indexes of the first and last samples included in each moving window. E.g. SNI_output$winInd(X) == [A, B] indicates that the moving window used to calculate SNI for the Xth sample contained all samples between A and B, inclusive.

$popN

The CHAR values of all samples in the noise (N) population (samples in the moving window with CHAR below threshold)

$popS

The CHAR values of all samples in the signal (S) population (samples in the moving window with CHAR at or above threshold)

$meanN

Mean CHAR of the samples in popN

$stdN

standard deviation of the samples in popN

$CF

The correction factor used in computing SNI. Equal to (v - 2)/v, where v is the number of samples in popN

Author(s)

Ryan Kelly (University of Illinois, USA)

References

Supplementary materials to the publication: Kelly, R., Higuera, P., Barrett, C., & Hu, F. (2011). A signal-to-noise index to quantify the potential for peak detection in sediment–charcoal records. Quaternary Research, 75(1), 11-17. https://doi.org/10.1016/j.yqres.2010.07.011

See Also

Supplementary materials: https://static.cambridge.org/content/id/urn:cambridge.org:id:article:S0033589400006785/resource/name/S0033589400006785sup001.txt

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