Package 'rshift'

Title: Paleoecology Functions for Regime Shift Analysis
Description: Contains a variety of functions, based around regime shift analysis of paleoecological data. Citations: Rodionov() from Rodionov (2004) <doi:10.1029/2004GL019448> Lanzante() from Lanzante (1996) <doi:10.1002/(SICI)1097-0088(199611)16:11%3C1197::AID-JOC89%3E3.0.CO;2-L> Hellinger_trans from Numerical Ecology, Legendre & Legendre (ISBN 9780444538680) rolling_autoc from Liu, Gao & Wang (2018) <doi:10.1016/j.scitotenv.2018.06.276> Sample data sets lake_data & lake_RSI processed from Bush, Silman & Urrego (2004) <doi:10.1126/science.1090795> Sample data set January_PDO from NOAA: <https://www.ncei.noaa.gov/access/monitoring/pdo/>.
Authors: Alex H. Room [aut, cre, cph] , Felipe Franco-Gaviria [ctb, fnd] , Dunia H. Urrego [ctb, fnd] , The authors of the dependency Rust crates [ctb] (see inst/AUTHORS file for details)
Maintainer: Alex H. Room <[email protected]>
License: MIT + file LICENSE
Version: 3.1.1
Built: 2024-12-24 06:58:49 UTC
Source: CRAN

Help Index

Converts absolute abundance data to a percentage of total abundance for each site

Description

Converts absolute abundance data to a percentage of total abundance for each site

Usage

absolute_to_percentage(data, col, site)

Arguments

data

The dataframe to be used.
col

The column that change is being measured on.
site

The column containing the site of each sample.

Value

The ‘data' dataframe with an added ’percentage' column.

Hellinger transform

Description

Hellinger transforms data (Legendre and Legendre, Numerical Ecology)

Usage

Hellinger_trans(data, col, site)

Arguments

data

The dataframe to be used.
col

The column that change is being measured on.
site

The column containing the site of each sample.

Value

The ‘data' dataframe with an added ’hellinger_trans_vals' column.

Pacific Decadal Oscillation in January

Description

A dataset containing January PDO values. Subset of the data from NOAA: <https://www.ncei.noaa.gov/access/monitoring/pdo/>.

Usage

data(January_PDO)

Format

A data frame with 104 rows and 2 variables

Details

  • PDO - Pacific Decadal Oscillation in January for the given year.

  • Age - the year for which the PDO was measured.

DCA-ordinated pollen data from Lake Consuelo

Description

A dataset containing pre-processed DCA-ordinated data from Bush, Silman & Urrego (2004) <doi:10.1126/science.1090795>.

Usage

data(lake_data)

Format

A data frame with 39 rows and 2 variables

Details

  • DCA1 - DCA values for each timepoint from the raw dataset.

  • Age - timepoint of each sample that has been DCA-ordinated.

DCA-ordinated pollen data from Lake Consuelo with RSI values

Description

A dataset containing pre-processed DCA-ordinated data from Bush, Silman & Urrego (2004) <doi:10.1126/science.1090795>. This data has been processed using Rodionov(lake_data, "DCA1", "Age", l=5, merge=TRUE)

Usage

data(lake_RSI)

Format

A data frame with 39 rows and 3 variables

Details

  • DCA1 - DCA values for each timepoint from the raw dataset.

  • Age - timepoint of each sample that has been DCA-ordinated.

  • RSI - Regime Shift Index (see docs for Rodionov()) for each timepoint.

Lanzante L-test

Description

performs the L-method for detection of regime shifts (Lanzante, 1996)

Usage

Lanzante(data, col, time, p = 0.05, merge = FALSE)

Arguments

data

The dataframe to be used.
col

The column we are measuring change on.
time

The column containing time units (e.g. age of a subsample)
p

The largest p-value you want to check regime shifts for. Defaults to p = 0.05.
merge

Sets the result to be either a regime-shift only table (if FALSE), or an addition to the original table (if TRUE)

Value

If merge = FALSE (default), produces a 2-column table of time (the time value for each regime shift) and p (the p-value for each regime shift). If merge = TRUE, returns the original dataset with an extra p-value column, giving the p-value for each time unit - 0 for non-shift years.

Examples

Lanzante(lake_data, "DCA1", "Age")
Lanzante(lake_data, "DCA1", "Age", p=0.10, merge=TRUE)

Calculate means for each regime

Description

calculates the mean for each regime in a regime shift analysis.

Usage

regime_means(data, col, rsi)

Arguments

data

The dataframe that will be used.
col

The column we are measuring change on.
rsi

The column containing RSI values.

Value

A vector of the mean value for each regime.

Examples

regime_means(lake_RSI, "DCA1", "RSI")

Rodionov (2004)'s STARS algorithm

Description

performs STARS analysis (Rodionov, 2004) on a dataset

Usage

Rodionov(data, col, time, l, prob = 0.05, startrow = 1, merge = FALSE)

Arguments

data

The dataframe to be used.
col

The column we are measuring change on.
time

The column containing time units (e.g. age of a subsample)
l

The cut-off length of a regime; affects sensitivity (see Rodionov, 2004)
prob

The p-value for significance of a regime shift. Defaults to p = 0.05.
startrow

What row the analysis starts at. Defaults to 1.
merge

Sets the result to be either a regime-shift only table (if FALSE), or an addition to the original table (if TRUE)

Value

If merge = FALSE (default), produces a 2-column table of time (the time value for each regime shift) and RSI (the regime shift index for each regime shift). If merge = TRUE, returns the original dataset with an extra RSI column, giving the regime shift index for each time unit - 0 for non-shift years.

Examples

Rodionov(lake_data, "DCA1", "Age", l=5)
Rodionov(lake_data, "DCA1", "Age", l=5, prob=0.01, startrow=2, merge=TRUE)

Rolling autocorrelation

Description

finds lag-1 autocorrelation in a rolling window; can be used to predict resilience (Liu, Gao, & Wang, 2018)

Usage

rolling_autoc(data, col, l)

Arguments

data

The dataframe that will be used.
col

The column we are measuring change on.
l

The time interval (no. of columns) used in the autocorrelation.

Value

A table of rolling lag-1 autocorrelation values.

Regime Shift Index graph

Description

creates two graphs, one of data and one of the RSI, as seen in Rodionov (2004)

Usage

RSI_graph(data, col, time, rsi, mean_lines = FALSE)

Arguments

data

The dataframe that will be used.
col

The column we are measuring change on.
time

The column containing time units (e.g. age of a subsample)
rsi

The column containing RSI values (for best visualisation use Rodionov() with merge=TRUE)
mean_lines

If true, add lines over the data indicating the mean of each regime.

Value

Two graphs, one on top of the other; one of col against time and one of RSI against time.

Examples

RSI_graph(lake_RSI, "DCA1", "Age", "RSI")

Calculates the mean for each regime in a regime shift analysis.

Description

Calculates the mean for each regime in a regime shift analysis.

Usage

rust_regime_means(col, rsi)

Arguments

col

The column we are measuring change on.
rsi

The column containing RSI values.

Calculate STARS RSI points and return to R as a vector

Description

Calculate STARS RSI points and return to R as a vector

Usage

rust_rodionov(vals, t_crit, l)

Arguments

vals

The column we are measuring change on
t_crit

The critical value of a t-distribution at the desired p-value
l

The cut-off length of a regime; affects sensitivity