Package 'tidypaleo'

Title: Tidy Tools for Paleoenvironmental Archives
Description: Provides a set of functions with a common framework for age-depth model management, stratigraphic visualization, and common statistical transformations. The focus of the package is stratigraphic visualization, for which 'ggplot2' components are provided to reproduce the scales, geometries, facets, and theme elements commonly used in publication-quality stratigraphic diagrams. Helpers are also provided to reproduce the exploratory statistical summaries that are frequently included on stratigraphic diagrams. See Dunnington et al. (2021) <doi:10.18637/jss.v101.i07>.
Authors: Dewey Dunnington [aut, cre, cph]
Maintainer: Dewey Dunnington <[email protected]>
License: MIT + file LICENSE
Version: 0.1.3.9000
Built: 2024-11-06 03:11:41 UTC
Source: https://github.com/paleolimbot/tidypaleo

Help Index

Use an age depth model as a second ggplot axis

Description

Use an age depth model as a second ggplot axis

Usage

age_depth_as_sec_axis(x, primary = c("depth", "age"), ...)

Arguments

x

An age_depth_model
primary

Specify the primary axis as 'age' or 'depth'
...

Passed to sec_axis

Value

A ggplot2 sec_axis for use in scale_x_continuous, scale_y_continuous, or their reverse variants.

Examples

library(ggplot2)
alta_lake_adm <- age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm,
  age = age_year_ad
)

ggplot(alta_lake_210Pb_ages, aes(y = depth_cm, x = age_year_ad)) +
  geom_path() +
  geom_point() +
  scale_y_reverse(sec.axis = age_depth_as_sec_axis(alta_lake_adm))

Age-depth model interpolators/extrapolators

Description

Age-depth model interpolators/extrapolators

Usage

age_depth_interpolate(x, y)

age_depth_extrapolate(x, y, x0 = last, y0 = last, slope = NULL)

age_depth_exact(x, y)

age_depth_na(x, y)

Arguments

x

A paired vector of x values
y

A paired vector of y values
x0

The x value to anchor the transform
y0

The y value to anchor the transform
slope

The slope (in units of y/x) to use for the transform

Value

A list with component functions trans and inverse

Examples

age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm, age = age_year_ad,
  age_max = age_year_ad + age_error_yr,
  age_min = age_year_ad - age_error_yr,
  extrapolate_age_below = ~age_depth_extrapolate(
    tail(.x, 3), tail(.y, 3), x0 = dplyr::last, y0 = dplyr::last
  ),
  extrapolate_age_above = ~age_depth_extrapolate(
    head(.x, 3), head(.y, 3), x0 = dplyr::first, y0 = dplyr::first
  )
)

Create age depth models

Description

Create age depth models

Usage

age_depth_model(
  .data = NULL,
  depth,
  age,
  age_min = NA_real_,
  age_max = NA_real_,
  interpolate_age = age_depth_interpolate,
  extrapolate_age_below = ~age_depth_extrapolate(.x, .y, x0 = last, y0 = last),
  extrapolate_age_above = ~age_depth_extrapolate(.x, .y, x0 = first, y0 = first),
  interpolate_age_limits = age_depth_exact,
  extrapolate_age_limits_below = age_depth_na,
  extrapolate_age_limits_above = age_depth_na
)

Arguments

.data

A data frame
depth, age, age_min, age_max

Expressions evaluated in .data that provide the known depths, known ages, and error information if available. These expressions are evaluated like they are within mutate if .data is present.
interpolate_age, extrapolate_age_below, extrapolate_age_above

These arguments provide the rules for interpolating and extrapolating ages based on depths.
interpolate_age_limits, extrapolate_age_limits_below, extrapolate_age_limits_above

These arguments provide the rules for interpolating and extrapolating age min and max values based on depths.

Value

An age depth model object.

Examples

age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm, age = age_year_ad,
  age_max = age_year_ad + age_error_yr,
  age_min = age_year_ad - age_error_yr
)

Alta Lake Lead-210 and Carbon-14 Ages

Description

The alta_lake_210Pb_ages object contains raw depths and ages for Alta Lake (Whistler, British Columbia, Canada) core AL-GC2 (Dunnington et al. 2016; Dunnington 2015). For these values, ages were calculated from Lead-210 activities using the constant rate of supply (CRS) model (Appleby and Oldfield 1983). The alta_lake_14C_ages object contains one uncalibrated Carbon-14 measurement from the same core. The alta_lake_bacon_ages object contains the combined result of the Lead-210 and the Carbon-14 ages as modelled by the rbacon package (Blaauw and Christen 2011).

Usage

alta_lake_210Pb_ages

alta_lake_14C_ages

alta_lake_bacon_ages

alta_lake_geochem

Format

An object of class tbl_df (inherits from tbl, data.frame) with 9 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 1 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 58 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 192 rows and 9 columns.

References

Appleby, P. G., and F. Oldfield. "The Assessment of 210Pb Data from Sites with Varying Sediment Accumulation Rates." Hydrobiologia 103, no. 1 (July 1, 1983): 29–35. doi:10.1007/BF00028424.

Blaauw, Maarten, and J. Andrés Christen. "Flexible Paleoclimate Age-Depth Models Using an Autoregressive Gamma Process." Bayesian Analysis 6, no. 3 (September 2011): 457–74. doi:10.1214/ba/1339616472.

Dunnington, Dewey W., Ian S. Spooner, Chris E. White, R. Jack Cornett, Dave Williamson, and Mike Nelson. "A Geochemical Perspective on the Impact of Development at Alta Lake, British Columbia, Canada." Journal of Paleolimnology 56, no. 4 (November 2016): 315–330. doi:10.1007/s10933-016-9919-x.

Dunnington, Dewey W. "A 500-Year Applied Paleolimnological Assessment of Environmental Change at Alta Lake, Whistler, British Columbia, Canada." M.Sc. Thesis, Acadia University, 2015. https://scholar.acadiau.ca/islandora/object/theses:411.

Examples

alta_lake_210Pb_ages
alta_lake_14C_ages
alta_lake_bacon_ages

Coerce and validate transforms and functions that produce them

Description

Coerce and validate transforms and functions that produce them

Usage

as_trans_factory(factory, env = parent.frame())

validate_trans_factory(factory, x = 1:3, y = 1:3)

validate_trans(trans, x = 1:3, y = 1:3)

Arguments

factory

A function that produces a transform object
env

The calling environment, for transform factories that are calls or rlang lambda-style functions.
x

The test x data
y

The test y data
trans

A transform object

Value

The input, invisibly.

Examples

as_trans_factory(age_depth_interpolate)

Facet for relative abundance data

Description

Provides a number of modifications to the plot that are necessary for relative abundance plots of a number of species. See scale_x_abundance, facet_grid, facet_grid, label_species, label_geochem, and rotated_facet_labels rotated_axis_labels for examples of how to customize the default behaviour.

Usage

facet_abundanceh(
  taxon,
  grouping = NULL,
  rotate_facet_labels = 45,
  labeller = label_species,
  scales = "free_x",
  space = "free_x",
  dont_italicize = c("\\(.*?\\)", "spp?\\.", "-complex", "[Oo]ther"),
  ...
)

facet_abundance(
  taxon,
  grouping = NULL,
  rotate_facet_labels = 0,
  labeller = label_species,
  scales = "free_y",
  space = "free_y",
  dont_italicize = c("\\(.*?\\)", "spp?\\.", "-complex", "[Oo]ther"),
  ...
)

facet_geochem_wraph(
  param,
  grouping = NULL,
  rotate_axis_labels = 90,
  scales = "free_x",
  labeller = label_geochem,
  renamers = c(`^d([0-9]+)([HCNOS])$` = "paste(delta ^ \\1, \\2)", `^210Pb$` =
    "paste({}^210, Pb)", `^Pb210$` = "paste({}^210, Pb)"),
  units = character(0),
  default_units = NA_character_,
  ...
)

facet_geochem_wrap(
  param,
  grouping = NULL,
  scales = "free_y",
  labeller = label_geochem,
  renamers = c(`^d([0-9]+)([HCNOS])$` = "paste(delta ^ \\1, \\2)", `^210Pb$` =
    "paste({}^210, Pb)", `^Pb210$` = "paste({}^210, Pb)"),
  units = character(0),
  default_units = NA_character_,
  ...
)

facet_geochem_grid(
  param,
  grouping = NULL,
  rotate_axis_labels = 0,
  scales = "free_y",
  space = "fixed",
  labeller = label_geochem,
  renamers = c(`^d([0-9]+)([HCNOS])$` = "paste(delta ^ \\1, \\2)", `^210Pb$` =
    "paste({}^210, Pb)", `^Pb210$` = "paste({}^210, Pb)"),
  units = character(0),
  default_units = NA_character_,
  ...
)

facet_geochem_gridh(
  param,
  grouping = NULL,
  rotate_axis_labels = 90,
  scales = "free_x",
  space = "fixed",
  labeller = label_geochem,
  renamers = c(`^d([0-9]+)([HCNOS])$` = "paste(delta ^ \\1, \\2)", `^210Pb$` =
    "paste({}^210, Pb)", `^Pb210$` = "paste({}^210, Pb)"),
  units = character(0),
  default_units = NA_character_,
  ...
)

Arguments

taxon, param

A call to vars, defining the column that identifies the taxon (parameter).
grouping

A call to vars, identifying additional grouping columns
rotate_facet_labels, rotate_axis_labels

Facet (axis) label rotation (degrees)
labeller

Labeller to process facet names. Use label_species to italicize species names, label_geochem to perform common formatting and units, or label_value to suppress.
space, scales

Modify default scale freedom behaviour
dont_italicize

Regular expressions that should not be italicized
...

Passed to facet_grid (abundance) or facet_wrap (geochem).
renamers

Search and replace operations to perform in the form search = replace. Replace text can (should) contain backreferences, and will be parsed as an expression (see plotmath). Use NULL to suppress renaming.
units

A named list of values = unit
default_units

The default units to apply

Value

A subclass of ggplot2::facet_grid() or ggplot2::facet_wrap().

Examples

library(ggplot2)

ggplot(keji_lakes_plottable, aes(x = rel_abund, y = depth)) +
  geom_col_segsh() +
  scale_y_reverse() +
  facet_abundanceh(vars(taxon), grouping = vars(location)) +
  labs(y = "Depth (cm)")

ggplot(keji_lakes_plottable, aes(y = rel_abund, x = depth)) +
  geom_col_segs() +
  scale_x_reverse() +
  facet_abundance(vars(taxon), grouping = vars(location)) +
  labs(x = "Depth (cm)")

ggplot(alta_lake_geochem, aes(x = value, y = depth)) +
  geom_lineh() +
  geom_point() +
  scale_y_reverse() +
  facet_geochem_wrap(vars(param), units = c(C = "%", Cu = "ppm", Ti = "ppm"), nrow = 1) +
  labs(x = NULL, y = "Depth (cm)")

ggplot(alta_lake_geochem, aes(x = value, y = depth)) +
  geom_lineh() +
  geom_point() +
  scale_y_reverse() +
  facet_geochem_gridh(vars(param), units = c(C = "%", Cu = "ppm", Ti = "ppm")) +
  labs(x = NULL, y = "Depth (cm)")

ggplot(alta_lake_geochem, aes(y = value, x = depth)) +
  geom_line() +
  geom_point() +
  scale_x_reverse() +
  facet_geochem_grid(vars(param), units = c(C = "%", Cu = "ppm", Ti = "ppm")) +
  labs(y = NULL, x = "Depth (cm)")

Useful geometries for strat diagrams

Description

Useful geometries for strat diagrams

Usage

geom_col_segsh(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  xend = 0,
  arrow = NULL,
  arrow.fill = NULL,
  lineend = "butt",
  linejoin = "round",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

geom_col_segs(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  yend = 0,
  arrow = NULL,
  arrow.fill = NULL,
  lineend = "butt",
  linejoin = "round",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

GeomColSegsh

GeomColSegs

GeomLineh

Arguments

mapping, data, stat, position, arrow, arrow.fill, lineend, linejoin, na.rm, show.legend, inherit.aes, ...

See geom_segment.
xend, yend

The end of the horizontal or vertical segment bars, respectively.

Format

An object of class GeomColSegsh (inherits from GeomSegment, Geom, ggproto, gg) of length 4.

An object of class GeomColSegs (inherits from GeomSegment, Geom, ggproto, gg) of length 4.

An object of class GeomLineh (inherits from GeomPath, Geom, ggproto, gg) of length 2.

Value

A ggplot2 layer

Connect observations in the vertical direction

Description

Connect observations in the vertical direction

Usage

geom_lineh(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE,
  ...
)

Arguments

mapping, data, stat, position, na.rm, show.legend, inherit.aes, ...

See geom_line.

Value

A ggplot2 layer.

Exaggerated geometries that do not train scales

Description

Exaggerated geometries that do not train scales

Usage

geom_point_exaggerate(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  exaggerate_x = 1,
  exaggerate_y = 1,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

geom_line_exaggerate(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  exaggerate_x = 1,
  exaggerate_y = 1,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

geom_lineh_exaggerate(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  exaggerate_x = 1,
  exaggerate_y = 1,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

geom_area_exaggerate(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  exaggerate_x = 1,
  exaggerate_y = 1,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

geom_areah_exaggerate(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  exaggerate_x = 1,
  exaggerate_y = 1,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

mapping, data, stat, position, na.rm, show.legend, inherit.aes, ...

See parent geometries
exaggerate_x, exaggerate_y

The factor by which to exaggerate x or y values

Value

A subclass of ggplot2::Geom.

Examples

library(ggplot2)

ggplot(keji_lakes_plottable, aes(x = rel_abund, y = depth)) +
  geom_lineh_exaggerate(exaggerate_x = 2, lty = 2) +
  geom_col_segsh() +
  scale_y_reverse() +
  facet_abundanceh(vars(taxon), grouping = vars(location)) +
  labs(y = "Depth (cm)")

Vertical ribbons and area plots

Description

Vertical ribbons and area plots

Usage

geom_ribbonh(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

GeomRibbonh

geom_areah(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "stackv",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE,
  ...
)

GeomAreah

Arguments

mapping, data, stat, position, na.rm, show.legend, inherit.aes, ...

See geom_ribbon.

Format

An object of class GeomRibbonh (inherits from Geom, ggproto, gg) of length 6.

An object of class GeomAreah (inherits from GeomRibbonh, Geom, ggproto, gg) of length 4.

Examples

library(ggplot2)

# Generate data
huron <- data.frame(year = 1875:1972, level = as.vector(LakeHuron))
h <- ggplot(huron, aes(y = year))

h + geom_ribbonh(aes(xmin=0, xmax=level))
h + geom_areah(aes(x = level))

# Add aesthetic mappings
h +
  geom_ribbonh(aes(xmin = level - 1, xmax = level + 1), fill = "grey70") +
  geom_lineh(aes(x = level))

Halifax lakes water chemistry and top/bottom diatom counts

Description

A subset of well-labeled, clean diatom count data for 44 Halifax-area (Nova Scotia) lakes, an analysis of which has been published by Ginn et al. (2015).

Usage

halifax_lakes_plottable

Format

An object of class tbl_df (inherits from tbl, data.frame) with 114 rows and 5 columns.

Source

Neotoma paleoecology database (https://www.neotomadb.org)

References

Ginn, Brian K., Thiyake Rajaratnam, Brian F. Cumming, and John P. Smol. "Establishing Realistic Management Objectives for Urban Lakes Using Paleolimnological Techniques: An Example from Halifax Region (Nova Scotia, Canada)." Lake and Reservoir Management 31, no. 2 (April 3, 2015): 92-108. doi:10.1080/10402381.2015.1013648.

Examples

halifax_lakes_plottable

Keji lakes core diatom counts

Description

A subset of well-labeled, clean diatom count data for 3 Keji-area (Nova Scotia) lakes, which form part of the analysis in Ginn et al. (2007).

Usage

keji_lakes_plottable

Format

An object of class tbl_df (inherits from tbl, data.frame) with 202 rows and 5 columns.

Source

Neotoma paleoecology database (https://www.neotomadb.org)

References

Ginn, Brian K., Brian F. Cumming, and John P. Smol. "Long-Term Lake Acidification Trends in High- and Low-Sulphate Deposition Regions from Nova Scotia, Canada." Hydrobiologia 586, no. 1 (July 1, 2007): 261-75. doi:10.1007/s10750-007-0644-3.

Examples

keji_lakes_plottable

Kellys Lake Data

Description

Geochemistry measurements and Cladocera counts from Kellys Lake, Cape Breton Island, Nova Scotia, Canada.

Usage

kellys_lake_cladocera

kellys_lake_geochem

kellys_lake_ages

Format

An object of class tbl_df (inherits from tbl, data.frame) with 300 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 305 rows and 9 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 14 rows and 5 columns.

References

Joshua Kurek, Ian Spooner, and Dewey Dunnington (unpublished data).

Geochem facet labelers

Description

Geochem facet labelers

Usage

label_geochem(
  labels,
  units = character(0),
  default_units = NA_character_,
  geochem_facet = 1,
  renamers = c(`^d([0-9]+)([HCNOS])$` = "paste(delta ^ \\1, \\2)", `^210Pb$` =
    "paste({}^210, Pb)", `^Pb210$` = "paste({}^210, Pb)"),
  multi_line = TRUE
)

Arguments

labels

A data.frame of facet label values
units

A named list of values = unit
default_units

The default units to apply
geochem_facet

Which facet to apply formatting
renamers

Search and replace operations to perform in the form search = replace. Replace text can (should) contain backreferences, and will be parsed as an expression (see plotmath). Use NULL to suppress renaming.
multi_line

See label_parsed

Value

A ggplot2::labeller()

Examples

library(ggplot2)

ggplot(alta_lake_geochem, aes(x = value, y = depth)) +
  geom_lineh() +
  geom_point() +
  scale_y_reverse() +
  facet_wrap(
    vars(param),
    labeller = purrr::partial(label_geochem, geochem_facet = "param"),
    nrow = 1,
    scales = "free_x"
  ) +
  labs(x = NULL, y = "Depth (cm)")

Species facet labellers

Description

Use these to label species with partial italic formatting. See label_parsed.

Usage

label_species(
  labels,
  dont_italicize = c("\\(.*?\\)", "spp?\\.", "-complex", "[Oo]ther"),
  species_facet = 1,
  multi_line = TRUE
)

Arguments

labels

A data.frame of facet label values
dont_italicize

Regular expressions that should not be italicized
species_facet

Which facet(s) contain species values
multi_line

See label_parsed

Value

A ggplot2::labeller()

Examples

library(ggplot2)

ggplot(keji_lakes_plottable, aes(x = rel_abund, y = depth)) +
  geom_col_segsh() +
  scale_y_reverse() +
  facet_grid(
    cols = vars(taxon),
    rows = vars(location),
    scales = "free_x",
    space = "free_x",
    labeller = purrr::partial(label_species, species_facet = "taxon")
  ) +
  labs(y = "Depth (cm)")

Add a dendrogram as a layer or facet

Description

Add a dendrogram as a layer or facet

Usage

layer_dendrogram(
  object,
  mapping,
  ...,
  colour = "black",
  size = 0.5,
  linetype = 1,
  alpha = NA,
  sequential_facets = TRUE
)

plot_layer_dendrogram(object, mapping, ..., panel_label = "CONISS")

layer_zone_boundaries(
  object,
  mapping,
  ...,
  linetype = 2,
  alpha = 0.7,
  colour = "black",
  size = 0.5
)

Arguments

object

A nested_hclust object.
mapping

Map at least one axis (x or y) to a qualifier, like aes(x = depth) or similar.
...

Use facet_var = "CONISS" or similar to name the panel
linetype, alpha, colour, size

Customize the apperance of boundary/dendrogram segment lines
sequential_facets

TRUE will result in the panel containing the dendrogram added to the right of the plot.
panel_label

Use to label a pane on a stanalone dendrogram plot

Value

A ggplot2::layer()

Examples

library(ggplot2)
library(dplyr, warn.conflicts = FALSE)

alta_coniss <- nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
) %>%
  nested_chclust_coniss()

ggplot() +
  layer_dendrogram(alta_coniss, aes(y = depth)) +
  scale_y_reverse()

Add scores to a plot

Description

Add scores to a plot

Usage

layer_scores(
  object,
  mapping = NULL,
  which = "PC1",
  key = "param",
  value = "value",
  scores_geom = list(ggplot2::geom_path(), ggplot2::geom_point()),
  sequential_facets = TRUE
)

plot_layer_scores(
  object,
  mapping,
  which = "PC1",
  key = "param",
  value = "value",
  ...
)

Arguments

object

A nested_prcomp or similar object
mapping

A mapping created with aes
which

Which principal components to plot
key

The column name to use for the principal component names
value

The column name to use for the principal component score values
scores_geom

One or more geometries to which scores should be applied.
sequential_facets

TRUE will result in the panel containing the dendrogram added to the right of the plot.
...

Passed to layer_scores()

Value

A list() that can be addeed to a ggplot2::ggplot()

Examples

library(ggplot2)
library(dplyr, warn.conflicts = FALSE)

alta_pca <- nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
) %>%
  nested_prcomp()

ggplot() +
  layer_scores(alta_pca, aes(value, depth), which = "PC1") +
  scale_y_reverse()

plot_layer_scores(alta_pca, aes(y = depth), which = c("PC1", "PC2")) +
  scale_y_reverse()

Long Lake Carbon-14 Ages

Description

This object contains several uncalibrated Carbon-14 measurements from Long Lake (Nova Scotia-New Brunswick Border Region, Canada) core LL-PC2 (Dunnington et al. 2017; White 2012). The long_lake_bacon_ages object contains the result of the Carbon-14 ages as modelled by the rbacon package (Blaauw and Christen 2011).

Usage

long_lake_14C_ages

long_lake_bacon_ages

long_lake_plottable

Format

An object of class tbl_df (inherits from tbl, data.frame) with 5 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 86 rows and 5 columns.

An object of class tbl_df (inherits from tbl, data.frame) with 51 rows and 10 columns.

References

Blaauw, Maarten, and J. Andrés Christen. "Flexible Paleoclimate Age-Depth Models Using an Autoregressive Gamma Process." Bayesian Analysis 6, no. 3 (September 2011): 457–74. doi:10.1214/ba/1339616472.

Dunnington, Dewey W., Hilary White, Ian S. Spooner, Mark L. Mallory, Chris White, Nelson J. O’Driscoll, and Nic R. McLellan. "A Paleolimnological Archive of Metal Sequestration and Release in the Cumberland Basin Marshes, Atlantic Canada." FACETS 2, no. 1 (May 23, 2017): 440–60. doi:10.1139/facets-2017-0004.

White, Hilary E. "Paleolimnological Records of Post-Glacial Lake and Wetland Evolution from the Isthmus of Chignecto Region, Eastern Canada." M.Sc. Thesis, Acadia University, 2012. https://scholar.acadiau.ca/islandora/object/theses:247.

Examples

long_lake_14C_ages
long_lake_bacon_ages
long_lake_plottable

Perform an analysis on a nested data matrix

Description

Perform an analysis on a nested data matrix

Usage

nested_analysis(
  .data,
  .fun,
  ...,
  .output_column = "model",
  .reserved_names = NULL,
  .env = parent.frame()
)

Arguments

.data

A data frame with a list column of data frames, possibly created using nested_data.
.fun

A model function
...

Passed to fun
.output_column

A column name in which the output of .fun should be stored.
.reserved_names

Names that should not be allowed as columns in any data frame within this object
.env

Passed to as_function

Value

.data with an additional list column of fun output

Examples

nd <- nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
)

na <- nested_analysis(nd, vegan::rda, X = data)
plot(na)

Prepare a parameter-long data frame for statistical analysis

Description

Prepare a parameter-long data frame for statistical analysis

Usage

nested_data(
  .data,
  qualifiers = NULL,
  key = NULL,
  value,
  fill = NA,
  select_if = ~TRUE,
  filter_all = any_vars(TRUE),
  trans = identity,
  groups = NULL
)

unnested_data(.data, ...)

Arguments

.data

Data in parameter-long form
qualifiers

Columns that add context to observations (e.g., depth, zone, core)
key

The column name that contains the column names of the data matrix
value

The column name that contains the values
fill

If a key/value combination doesn't exist in the input, this value will be assigned in the data matrix. Generally, using NA for geochemical data and 0 for relative abundance data is advised.
select_if

Use ~TRUE to keep all columns; use ~all(is.finite(.)) to keep columns with all finite values. See select_if.
filter_all

Use any_vars(TRUE) to keep all observations; use all_vars(is.finite(.)) to keep only observations with finite (non-missing) values. See filter_all.
trans

A function that will be applied to all columns, column-wise. Use identity to perform no transformation, use scale to scale each column to a mean of zero and variance of 1. See mutate_all.
groups

Use group_by or this argument to group by one or more columns (e.g., core or lake)
...

One or more columns to unnest.

Value

A nested data matrix, which is composed of a tibble::tibble() with tibble list-columns data, discarded_rows, discarded_columns, and qualifiers.

Examples

nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
)

Nested (Constrained) hierarchical clustering

Description

Powered by chclust and hclust; broken stick using bstick.

Usage

nested_hclust(
  .data,
  data_column = "data",
  qualifiers_column = "qualifiers",
  distance_fun = stats::dist,
  n_groups = NULL,
  ...,
  .fun = stats::hclust,
  .reserved_names = character(0)
)

nested_chclust_conslink(
  .data,
  data_column = "data",
  qualifiers_column = "qualifiers",
  distance_fun = stats::dist,
  n_groups = NULL,
  ...
)

nested_chclust_coniss(
  .data,
  data_column = "data",
  qualifiers_column = "qualifiers",
  distance_fun = stats::dist,
  n_groups = NULL,
  ...
)

Arguments

.data

A data frame with a list column of data frames, possibly created using nested_data.
data_column

An expression that evalulates to the data object within each row of .data
qualifiers_column

The column that contains the qualifiers
distance_fun

A distance function like dist or vegdist.
n_groups

The number of groups to use (can be a vector or expression using vars in .data)
...

Passed to chclust or hclust.
.fun

Function powering the clustering. Must return an hclust object of some kind.
.reserved_names

Names that should not be allowed as columns in any data frame within this object

Value

.data with additional columns

References

Bennett, K. (1996) Determination of the number of zones in a biostratigraphic sequence. New Phytologist, 132, 155-170. doi:10.1111/j.1469-8137.1996.tb04521.x (Broken stick)

Grimm, E.C. (1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers & Geosciences, 13, 13-35. doi:10.1016/0098-3004(87)90022-7

Juggins, S. (2017) rioja: Analysis of Quaternary Science Data, R package version (0.9-15.1). (https://cran.r-project.org/package=rioja).

See hclust for hierarchical clustering references

Examples

library(tidyr)
library(dplyr, warn.conflicts = FALSE)

nested_coniss <- keji_lakes_plottable %>%
  group_by(location) %>%
  nested_data(depth, taxon, rel_abund, fill = 0) %>%
  nested_chclust_coniss()

# plot the dendrograms using base graphics
plot(nested_coniss, main = location, ncol = 1)

# plot broken stick dispersion to verify number of plausible groups
library(ggplot2)

nested_coniss %>%
  select(location, broken_stick) %>%
  unnest(broken_stick) %>%
  tidyr::gather(type, value, broken_stick_dispersion, dispersion) %>%
  ggplot(aes(x = n_groups, y = value, col = type)) +
  geom_line() +
  geom_point() +
  facet_wrap(vars(location))

Nested Principal Components Analysis (PCA)

Description

Powered by prcomp. When creating the nested_data, the data should be scaled (i.e, trans = scale) if all variables are not in the same unit.

Usage

nested_prcomp(.data, data_column = .data$data, ...)

Arguments

.data

A data frame with a list column of data frames, possibly created using nested_data.
data_column

An expression that evalulates to the data object within each row of .data
...

Passed to prcomp.

Value

.data with additional columns 'model', 'loadings', 'variance' and 'scores'

Examples

library(dplyr, warn.conflicts = FALSE)

nested_pca <- alta_lake_geochem %>%
  nested_data(
    qualifiers = c(depth, zone),
    key = param,
    value = value,
    trans = scale
  ) %>%
  nested_prcomp()

# get variance info
nested_pca %>% unnested_data(variance)

# get loadings info
nested_pca %>% unnested_data(loadings)

# scores, requalified
nested_pca %>% unnested_data(c(qualifiers, scores))

Plot an age depth model using base graphics

Description

Plot an age depth model using base graphics

Usage

## S3 method for class 'age_depth_model'
plot(
  x,
  xlab = "depth",
  ylab = "age",
  xlim = NULL,
  ylim = NULL,
  add = FALSE,
  ...
)

Arguments

x

An age_depth_model
xlab, ylab

Axis labels
xlim, ylim

Axis limits
add

Pass TRUE to skip creating a new plot
...

Passed to points to customize points display

Value

The input, invisibly

Examples

adm <- age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm, age = age_year_ad,
  age_max = age_year_ad + age_error_yr,
  age_min = age_year_ad - age_error_yr
)

plot(adm)

Plot a nested analysis

Description

Calls plot or another (base) plotting function on all models, arranging the output in subplots.

Usage

## S3 method for class 'nested_analysis'
plot(x, ..., main = "", nrow = NULL, ncol = NULL)

plot_nested_analysis(
  .x,
  .fun,
  ...,
  nrow = NULL,
  ncol = NULL,
  .model_column = .data$model,
  .output_column = NULL
)

## S3 method for class 'nested_prcomp'
biplot(x, ..., nrow = NULL, ncol = NULL)

Arguments

x, .x

A nested_analysis object (or subclass)
...

Passed to the plot function. Tidy evaluation is supported, and arguments are evaluated within a transposed version of x for each row.
main

The plot title
nrow, ncol

Force a number of rows or columns in the output
.fun

A function that produces graphical output
.model_column

The column containing the model
.output_column

The column in which the output of the plot function should be placed

Value

the input, invisibly

Examples

nd <- nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
)

na <- nested_analysis(nd, vegan::rda, X = data)
plot(na)

Predict age and depth values

Description

Predict age and depth values

Usage

## S3 method for class 'age_depth_model'
predict(object, .data = NULL, depth = NULL, age = NULL, ...)

predict_depth(object, age)

predict_age(object, depth)

Arguments

object

An age_depth_model object
.data

Optional input data frame
depth, age

Specify exactly one of these to predict the other.
...

Unused

Value

A data frame with the same number of observations as the input age or depth vector.

Examples

adm <- age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm, age = age_year_ad,
  age_max = age_year_ad + age_error_yr,
  age_min = age_year_ad - age_error_yr
)

predict(adm, depth = 1:5)

Common plot modifications for stratigraphic plots

Description

Common plot modifications for stratigraphic plots

Usage

rotated_facet_labels(
  angle = 45,
  direction = "x",
  remove_label_background = TRUE
)

rotated_axis_labels(angle = 90, direction = "x")

Arguments

angle

The angle at which labels should be rotated
direction

The axes along which the operations should be performed
remove_label_background

Whether or not label backgrounds should be removed along rotated label axes

Value

An object or list of objects that can be added to a ggplot

Examples

library(ggplot2)

ggplot(mpg, aes(cty, hwy)) +
  geom_point() +
  facet_wrap(vars(class)) +
  rotated_facet_labels(45, "x")

Scales for relative abundance values

Description

Continuous scales that (1) always start at 0, (2) always have the same breaks, and (3) expand using a constant rather than a percentage. These scales assume that data are in percentages (i.e., range 0 to 100 rather than 0 to 1).

Usage

scale_x_abundance(
  ...,
  limits = c(0, NA),
  breaks = seq(10, 90, 30),
  minor_breaks = seq(0, 100, 10),
  expand = c(0, 1)
)

scale_y_abundance(
  ...,
  limits = c(0, NA),
  breaks = seq(10, 90, 30),
  minor_breaks = seq(0, 100, 10),
  expand = c(0, 1)
)

Arguments

...

Passed to scale_y_continuous or scale_x_continuous
limits

Limits for the scale
breaks

Where to place labels on the scale
minor_breaks

Where to place minor breaks
expand

A vector of expantion constants

Value

A scale_y_continuous or scale_x_continuous

Examples

library(dplyr, warn.conflicts = FALSE)
library(ggplot2)

keji_lakes_plottable %>%
  filter(taxon == "Other", location == "Beaverskin Lake") %>%
  ggplot(aes(rel_abund, depth)) +
  geom_col_segsh() +
  scale_x_abundance() +
  scale_y_reverse()

Age-depth scales

Description

Age-depth scales

Usage

scale_y_depth_age(
  model = NULL,
  age_name = "age",
  age_breaks = waiver(),
  age_labels = waiver(),
  ...
)

scale_y_age_depth(
  model = NULL,
  reversed = FALSE,
  depth_name = "depth",
  depth_breaks = waiver(),
  depth_labels = waiver(),
  ...
)

scale_x_depth_age(
  model = NULL,
  age_name = "age",
  age_breaks = waiver(),
  age_labels = waiver(),
  ...
)

scale_x_age_depth(
  model = NULL,
  reversed = FALSE,
  depth_name = "depth",
  depth_breaks = waiver(),
  depth_labels = waiver(),
  ...
)

Arguments

model

An age-depth model, or NULL to suppress the second axis
age_name, depth_name

Label for the second axis
age_breaks, depth_breaks

Breaks for the second axis
age_labels, depth_labels

Labels for each break on the second axis
...

Passed to scale_y_continuous or scale_x_continuous
reversed

Reverse the primary age axis (for years BP or similar)

Value

A scale_y_continuous or scale_x_continuous

Examples

library(ggplot2)
library(dplyr, warn.conflicts = FALSE)

adm <- age_depth_model(
  alta_lake_210Pb_ages,
  depth = depth_cm, age = age_year_ad
)

alta_lake_geochem %>%
  filter(param == "Cu") %>%
  ggplot(aes(value, depth)) +
  geom_point() +
  scale_y_depth_age(adm)

Change facet ordering behaviour

Description

Normally, facets are ordered using as.factor on all values that occur within layer data, which means that when adding additional layers, any ordering is not preserved unless the factor levels are identical on all factors. This function changes this behaviour such that facet levels are combined in layer order. This is useful when adding standalone layers to a plot without disturbing the existing order.

Usage

sequential_layer_facets(reverse = FALSE)

Arguments

reverse

Use TRUE to process layers in reverse order

Value

An object that can be added to a ggplot2::ggplot()

Examples

library(ggplot2)

p <- ggplot(mapping = aes(x, y)) +
  geom_point(data = data.frame(x = 1:5, y = 1:5, facet = "b")) +
  geom_point(data = data.frame(x = 1:5, y = 1:5, facet = "a")) +
  facet_wrap(vars(facet))

p
p + sequential_layer_facets()

Display a dendrogram as a ggplot2 layer

Description

Display a dendrogram as a ggplot2 layer

Usage

stat_nested_hclust(
  mapping = NULL,
  data = NULL,
  geom = "segment",
  position = "identity",
  ...,
  inherit.aes = TRUE,
  show.legend = NA
)

StatNestedHclust

Arguments

mapping

A mapping created using aes. Must map x OR y to a qualifier.
data

A nested_hclust object
geom

Any geom that takes x, xend, y, and yend. Probably geom_segment is the only one that makes sense.
position

Position adjustment
...

Passed to the the stat/geom (see geom_segment)
inherit.aes

Inherit aesthetics from ggplot()?
show.legend

Show mapped aesthetics in the legend?

Format

An object of class StatNestedHclust (inherits from Stat, ggproto, gg) of length 4.

Value

A ggplot2::Stat

Examples

library(ggplot2)
library(dplyr, warn.conflicts = FALSE)

alta_coniss <- nested_data(
  alta_lake_geochem,
  qualifiers = c(age, depth, zone),
  key = param,
  value = value,
  trans = scale
) %>%
  nested_chclust_coniss()

ggplot(alta_coniss) +
  stat_nested_hclust(aes(model = model, y = depth)) +
  scale_y_reverse()

A Paleo-friendly ggplot2 theme

Description

Essentially, this is ggplot2::theme_bw() with a few modifications

Usage

theme_paleo(...)

Arguments

...

Passed to ggplot2::theme_bw()

Value

A complete ggplot2::theme()

Examples

library(ggplot2)

ggplot(mpg, aes(cty, hwy)) +
  geom_point() +
  theme_paleo()