Taxonomic, phylogenetic and functional diversity on one scale, in one tidy workflow.
Reviewers increasingly expect taxonomic (TD), phylogenetic (PD) and functional (FD) diversity to be reported together and to be comparable. tidydiversity expresses all three as Hill numbers of the same order q, in units of effective numbers of species, and standardises them by sample size or sample coverage. Data go in as a long data frame and results come back as tibbles.
In memory of Filip Damen (about 1966-2026). tidydiversity is dedicated to the Wanang leader who refused the loggers, kept his forest for science, and built a research station, a school and a health post with his community. The
wanang_butterfliesdata come from the forest he protected. Read his story.
Species counts only. Observed and asymptotic Hill numbers, sample coverage, rarefaction and extrapolation, matching iNEXT.
Add a phylogeny with type = "PD", tree = tree. Effective number of
lineages at a reference time, matching iNEXT.3D.
Add a trait distance matrix with type = "FD", distance = d. Effective
number of functionally distinct species at a threshold tau, or the area
under the tau profile, matching iNEXT.3D.
Community ecology in R is spread over many packages: iNEXT and iNEXT.3D for Hill numbers, vegan for ordination and permutation tests, betapart for beta diversity, FD, TPD, ade4 and phytools for traits and phylogenies. Each has its own input format (wide matrices, lists, row names) and its own output (lists, S4 objects, printed tables). Most of an analysis ends up as glue code: reshaping data frames by hand, matching species names across matrices, pulling numbers out of nested lists, and re-drawing every plot. tidydiversity replaces that glue with one consistent, tested interface.
| You usually hand-roll | With tidydiversity |
|---|---|
| Pivot long field data to a site x species matrix, align row names with the site table | Long data in, one row per site and species: div_hill(data, site, species, abundance)
|
| Match species names between the community matrix, the trait table and the tree | Matched and checked for you, with a clear error naming the missing species |
| Loop over seasons or treatments and bind the results |
.by = season, or group_by()
|
| Dig estimates out of iNEXT, vegan or FD list objects | Every result is a tibble; tidy() and glance() for models |
| Rebuild CANOCO-style biplots, rarefaction curves and multi-panel figures |
autoplot(), plot_panel(), plot_save() at journal sizes |
| Write permutation tests for CWM, null models, forward selection |
trait_cwm_test() (max test), trait_ses(), ord_select()
|
| Feed diversity into tidymodels with leakage-free resampling |
model_group_folds(), model_case_weights(), step_hill()
|
beta_hill(), Chao et al. 2023), so regions sampled with unequal
effort can be compared.beta_welch(), Hamidi et al. 2019), site-blocked
cross-validation, and precision weights from estimation error.div_ and beta_ verb.Hot loops are rewritten as compiled C++ (Rcpp) kernels or as vectorised matrix algebra. Every rewrite is tested against the reference package and gives identical results. Timings on one CPU core, with the Brazil rainforest data of iNEXT.3D (3868 individuals, 425 species, 40 sample sizes) and the DaPhnE tree (1641 species):
| Task | Before | tidydiversity | How |
|---|---|---|---|
| PD rarefaction and extrapolation | 3.9 s | 0.4 s | C++ hypergeometric kernel |
| FD rarefaction and extrapolation | 3.6 s (iNEXT.3D) | 0.2 s | C++ kernel |
| Blomberg’s K with 999 randomisations | 4.4 s per K (phytools) | 1.5 s in total | contrasts with precomputed tree terms |
| Community TPDs and TPD indices | 2.5 s (TPD) | < 0.1 s | matrix algebra |
Optional packages (vegan, FD, ade4, TPD, phytools, tidymodels) load only
when a function needs them, so library(tidydiversity) stays light.
site, species and
abundance, as you would in dplyr..by and respects group_by().tidy() and glance() methods, a recipes
step, and rsample folds.autoplot() returns ordinary ggplot objects you can
extend with +; patchwork composes them.div_ (alpha
diversity), beta_, ord_, trait_, phylo_, model_, stems_
and plot_.| Topic | tidydiversity | Usually needs |
|---|---|---|
| Hill numbers, coverage, rarefaction/extrapolation (TD, PD, FD) | div_*() |
iNEXT, iNEXT.3D, hillR |
| Beta diversity partitioning, distance decay and decay models |
beta_pair(), beta_multi(), beta_decay(), beta_decay_model()
|
betapart |
| Pairwise Hill-number dissimilarity (Sorensen, Jaccard, Horn, Morisita-Horn) | beta_hill_pair() |
vegan |
| Coverage-standardised beta diversity (TD, PD, FD) | beta_hill() |
iNEXT.beta3D |
| Ordination (PCA to db-RDA) with biplots |
ord_fit(), autoplot()
|
vegan, ggvegan, ggord |
| PERMANOVA (classic and Welch-type), dispersion, variation partitioning, forward selection |
beta_permanova(), beta_welch(), beta_dispersion(), ord_varpart(), ord_select()
|
vegan, adespatial |
| Indicator species, corrected for multiple testing | beta_indicator() |
indicspecies |
| Traits: CWM, FD indices, null models, ITV, fourth corner | trait_*() |
FD, gawdis, picante, cati, ade4, weimea |
| Trait probability densities | trait_tpd() |
TPD |
| Phylogenetic signal and distances | phylo_*() |
phytools, picante, ape |
| Modelling with tidymodels |
model_*(), step_hill()
|
hand-written glue |
| ForestGEO census tables and demography | stems_*() |
fgeo, CTFS R package |
| Partial-effect and surface plots; overdispersed counts |
plot_partial(), plot_surface(), model_count()
|
visreg, plotmo, aods3 |
| Publication figures and tables |
autoplot(), plot_*(), div_table()
|
ggplot2, patchwork, gt |
Install a ready-built binary from r-universe (Windows, macOS and Linux; no compiler needed):
install.packages("tidydiversity",
repos = c("https://jmozez.r-universe.dev", "https://cloud.r-project.org"))
Or build the latest development version from GitHub source (needs Rtools on Windows or the Xcode command-line tools on macOS):
# install.packages("pak")
pak::pak("jmozez/tidydiversity")
The source code lives on GitHub, https://github.com/jmozez/tidydiversity, with a mirror on GitLab, https://gitlab.com/jimmy.moses/tidydiversity.
New to tidydiversity or to diversity analysis? The User guide explains, chapter by chapter, how to install and run tidydiversity in RStudio and what every analysis means: the concept and theory, runnable code on real field data, figures and tables, how to interpret them and how to report them.
tidydiversity does no geospatial work. Its companion package tidymapr draws study-site and diversity maps, reads and writes vector and raster files, downloads climate, elevation and boundary layers, computes spatial statistics and fits species distribution models. It takes the tidy tables tidydiversity uses and returns tibbles and ggplots, so the two packages share one workflow, the way dplyr and ggplot2 do:
pak::pak("jmozez/tidymapr")
library(tidydiversity)
library(tidymapr)
div_estimate(png_ants, q = 0, .by = season) |>
dplyr::filter(season == "dry") |>
dplyr::left_join(geo_example_sites(), by = "site") |>
plot_geo(estimate, lon, lat, label = site)
png_ants records ground-foraging ants at nutrient baits along the Mt
Wilhelm elevational gradient in Papua New Guinea, in the dry and the wet
season (Moses et al. 2023). Each row is one species at one bait, so
diversity is estimated from the number of baits at which each species
was found.
library(tidydiversity)
library(dplyr)
div_hill(png_ants, q = c(0, 1, 2), .by = season)
#> # A tibble: 30 x 5
#> season site q n diversity
#> <chr> <chr> <dbl> <dbl> <dbl>
#> 1 dry MW0200 0 47 16
#> 2 dry MW0200 1 47 12.3
#> 3 dry MW0200 2 47 9.91
#> 4 wet MW0200 0 76 21
#> 5 wet MW0200 1 76 10.9
#> 6 wet MW0200 2 76 6.99
#> 7 dry MW0700 0 23 15
#> 8 dry MW0700 1 23 11.8
#> 9 dry MW0700 2 23 8.67
#> 10 wet MW0700 0 58 28
#> # i 20 more rows
Compare sites at equal sample completeness. By default each season is standardised to the lowest coverage that all its sites reach at twice their sample size, so no site is extrapolated too far:
set.seed(1)
div_standardise(png_ants, .by = season, nboot = 50) |>
filter(q == 0)
#> # A tibble: 10 x 10
#> season site q m method coverage diversity se lower upper
#> <chr> <chr> <dbl> <dbl> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 dry MW0200 0 11.6 rarefaction 0.565 7.96 0.458 7.07 8.86
#> 2 wet MW0200 0 45.1 rarefaction 0.796 15.7 1.42 12.9 18.4
#> 3 dry MW0700 0 46.0 extrapolation 0.565 26.0 5.31 15.6 36.4
#> 4 wet MW0700 0 116 extrapolation 0.796 42.2 6.36 29.8 54.7
#> 5 dry MW1200 0 11.0 rarefaction 0.565 7.17 0.669 5.86 8.48
#> 6 wet MW1200 0 40.6 rarefaction 0.796 17.1 1.82 13.5 20.6
#> 7 dry MW1700 0 3.07 rarefaction 0.565 2.32 0.226 1.88 2.77
#> 8 wet MW1700 0 25.4 rarefaction 0.796 8.36 1.58 5.26 11.5
#> 9 dry MW2200 0 4.76 rarefaction 0.565 3.48 0.363 2.77 4.20
#> 10 wet MW2200 0 8.97 rarefaction 0.796 4.01 0.585 2.86 5.15
Every example uses published field data, rebuilt from the original
deposits by scripts in data-raw/.
| Dataset | What it holds | Source |
|---|---|---|
png_ants |
Ants at nutrient baits, Mt Wilhelm, 200-3700 m, dry and wet season | Moses et al. (2023) Global Ecology and Biogeography, doi:10.1111/geb.13757 |
png_tree_ants |
Every ant nest in 1254 trees at 200, 900 and 1800 m | Plowman et al. (2020) Ecography, doi:10.1111/ecog.04730 |
png_frogs |
3390 frogs of 55 species at eight elevations, Mt Wilhelm | Dahl et al. (2024) Biotropica, doi:10.1111/btp.13283 |
wanang_butterflies |
Butterflies on 36 days over three years, Wanang forest plot | Luk et al. (2019) Biotropica, doi:10.1111/btp.12671 |
wanang_caterpillars |
9363 reared caterpillars of 849 morphospecies, primary and secondary forest at Wanang | Butterill and Redmond (2023) GBIF, doi:10.15468/7e96ht (CC BY 4.0) |
The same verbs give all three dimensions. Here are rarefaction and extrapolation curves for two Brazilian rainforest tree assemblages, using the abundance data, phylogeny and functional distances shipped with iNEXT.3D.
library(ggplot2)
data(list = c(
"Brazil_rainforest_abun_data", "Brazil_rainforest_phylo_tree",
"Brazil_rainforest_distance_matrix"
), package = "iNEXT.3D")
wide <- Brazil_rainforest_abun_data
brazil <- tibble(
site = rep(colnames(wide), each = nrow(wide)),
species = rep(rownames(wide), ncol(wide)),
n = unlist(wide, use.names = FALSE)
) |>
filter(n > 0)
curves <- bind_rows(
div_rarefy(brazil, abundance = n, type = "TD") |> mutate(type = "TD"),
div_rarefy(brazil,
abundance = n, type = "PD",
tree = Brazil_rainforest_phylo_tree
),
div_rarefy(brazil,
abundance = n, type = "FD",
distance = as.matrix(Brazil_rainforest_distance_matrix)
)
)
curves$type <- factor(curves$type, levels = c("TD", "PD", "FD"))
ggplot(curves, aes(m, diversity, colour = site)) +
geom_line(aes(linetype = method == "extrapolation")) +
geom_point(data = filter(curves, method == "observed"), size = 2) +
facet_wrap(vars(q, type),
ncol = 3, scales = "free_y",
labeller = labeller(q = function(x) paste("q =", x), .multi_line = FALSE)
) +
scale_linetype_manual(values = c("solid", "dashed"), guide = "none") +
labs(
x = "Number of individuals",
y = "Effective number of species", colour = NULL
) +
theme_tidydiversity()

Hill numbers of order q share one unit, the effective number of equally abundant (and, for PD and FD, equally distinct) species, so TD, PD and FD of the same order can be compared directly, and their ratios read as measures of phylogenetic or functional redundancy (Chao, Chiu and Jost 2014, doi:10.1146/annurev-ecolsys-120213-091540; Chao et al. 2019, doi:10.1002/ecm.1343). Standardising every dimension by the same sample size or coverage removes the effect of unequal sampling effort before you compare sites.
| Family | Functions |
|---|---|
| Community data |
as_community(), community_wide(), community_long(), add_site_data()
|
Alpha diversity (div_) |
div_hill(), div_estimate(), div_coverage(), div_rarefy(), div_standardise(), div_completeness(), div_evenness(), div_boot()
|
Beta diversity (beta_) |
beta_pair(), beta_multi(), beta_hill(), beta_hill_pair(), beta_decay(), beta_decay_model(), beta_dist(), beta_permanova(), beta_welch(), beta_dispersion(), beta_indicator()
|
Ordination (ord_) |
ord_fit(), ord_test(), ord_select(), ord_varpart()
|
Traits (trait_) |
trait_dist(), trait_cwm(), trait_cwm_test(), trait_fd_indices(), trait_rao(), trait_ses(), trait_itv(), trait_fourth_corner(), trait_tpd(), trait_tpd_dissim()
|
Phylogeny (phylo_) |
phylo_signal(), phylo_dist(), phylo_combine_dist()
|
Modelling (model_) |
model_group_folds(), model_case_weights(), step_hill(), tidy(), glance()
|
ForestGEO (stems_) |
stems_as_trees(), stems_abundance(), stems_basal_area(), stems_size_class(), stems_demography(), stems_as_community()
|
| Count models and partial plots |
model_count(), plot_partial(), plot_surface()
|
Plots and tables (plot_) |
autoplot(), plot_panel(), plot_save(), plot_palette(), plot_partial(), plot_surface(), plot_phylo(), plot_trait_space(), plot_venn(), theme_tidydiversity(), div_table()
|
| Example data |
png_ants, png_tree_ants, png_frogs, wanang_butterflies, wanang_caterpillars, data_aravo()
|
Every plot that autoplot() draws, using png_ants.
Solid lines are rarefied, dashed lines extrapolated to twice the sample size; points are the observed samples (wet season).

The same curves plotted against estimated sample coverage, the basis for coverage-standardised comparison.

Observed Hill numbers across orders q from 0 (richness) to 3 (dominant species), one line per site.

Observed q = 0, 1 and 2 diversity at each elevation, one line per season.

Open circles are observed values, filled circles the asymptotic estimates, with 95% bootstrap intervals.

Estimated sample coverage of each site against its number of bait occurrences.

Total Sorensen dissimilarity between every pair of sites within each season.

Multi-site Sorensen dissimilarity of each season split into turnover and nestedness.

Gamma, alpha and beta diversity of each season at 90% sample coverage, with 95% bootstrap intervals.

Pairwise total, turnover and nestedness dissimilarity against elevation difference, with quasibinomial (logit) trends that stay within 0-1.

Example output of every main function on png_ants. Long tables show
their first rows.
div_hill()Observed Hill numbers: n occurrences, diversity of order q per
site.
| season | site | q | n | diversity |
|---|---|---|---|---|
| dry | MW0200 | 0 | 47 | 16.000 |
| dry | MW0200 | 1 | 47 | 12.278 |
| dry | MW0200 | 2 | 47 | 9.906 |
| wet | MW0200 | 0 | 76 | 21.000 |
| wet | MW0200 | 1 | 76 | 10.876 |
| wet | MW0200 | 2 | 76 | 6.993 |
| dry | MW0700 | 0 | 23 | 15.000 |
| dry | MW0700 | 1 | 23 | 11.758 |
div_estimate()Asymptotic estimates: observed and estimate Hill numbers with
bootstrap se, lower and upper 95% limits.
| season | site | q | observed | estimate | se | lower | upper |
|---|---|---|---|---|---|---|---|
| dry | MW0200 | 0 | 16.000 | 18.447 | 5.913 | 16.162 | 53.059 |
| dry | MW0200 | 1 | 12.278 | 15.183 | 1.700 | 12.278 | 18.515 |
| dry | MW0200 | 2 | 9.906 | 12.284 | 1.859 | 9.906 | 15.928 |
| wet | MW0200 | 0 | 21.000 | 32.941 | 11.434 | 23.457 | 79.036 |
| wet | MW0200 | 1 | 10.876 | 13.793 | 2.024 | 10.876 | 17.761 |
| wet | MW0200 | 2 | 6.993 | 7.600 | 1.099 | 6.993 | 9.754 |
| dry | MW0700 | 0 | 15.000 | 83.870 | 39.982 | 38.949 | 213.049 |
| dry | MW0700 | 1 | 11.758 | 38.453 | 30.595 | 11.758 | 98.417 |
div_coverage()Sample coverage: n occurrences, s_obs observed species, f1
singletons, f2 doubletons and estimated coverage.
| season | site | n | s_obs | f1 | f2 | coverage |
|---|---|---|---|---|---|---|
| dry | MW0200 | 47 | 16 | 5 | 5 | 0.898 |
| wet | MW0200 | 76 | 21 | 11 | 5 | 0.857 |
| dry | MW0700 | 23 | 15 | 12 | 1 | 0.482 |
| wet | MW0700 | 58 | 28 | 17 | 3 | 0.709 |
| dry | MW1200 | 29 | 13 | 7 | 3 | 0.766 |
| wet | MW1200 | 57 | 20 | 9 | 3 | 0.844 |
| dry | MW1700 | 17 | 5 | 1 | 2 | 0.953 |
| wet | MW1700 | 34 | 10 | 6 | 2 | 0.827 |
div_rarefy()Rarefaction and extrapolation: diversity of order q at sample size
m, whether method is rarefaction, observed or extrapolation, the
expected coverage, and lower and upper limits (NA when
nboot = 0).
| site | q | m | method | diversity | coverage | lower | upper |
|---|---|---|---|---|---|---|---|
| MW0200 | 0 | 1 | rarefaction | 1.000 | 0.132 | NA | NA |
| MW0200 | 0 | 38 | rarefaction | 14.140 | 0.780 | NA | NA |
| MW0200 | 0 | 75 | rarefaction | 20.855 | 0.855 | NA | NA |
| MW0200 | 0 | 76 | observed | 21.000 | 0.857 | NA | NA |
| MW0200 | 0 | 77 | extrapolation | 21.143 | 0.859 | NA | NA |
| MW0200 | 0 | 102 | extrapolation | 24.211 | 0.895 | NA | NA |
| MW0200 | 0 | 127 | extrapolation | 26.482 | 0.923 | NA | NA |
| MW0200 | 0 | 152 | extrapolation | 28.162 | 0.943 | NA | NA |
div_standardise()Coverage standardisation: diversity of order q at a common
coverage for every site, with the matching sample size m, the
method used, and bootstrap se, lower and upper limits.
| season | site | q | m | method | coverage | diversity | se | lower | upper |
|---|---|---|---|---|---|---|---|---|---|
| dry | MW0200 | 0 | 11.646 | rarefaction | 0.565 | 7.963 | 0.458 | 7.067 | 8.860 |
| dry | MW0200 | 1 | 11.646 | rarefaction | 0.565 | 7.099 | 0.513 | 6.093 | 8.105 |
| dry | MW0200 | 2 | 11.646 | rarefaction | 0.565 | 6.236 | 0.534 | 5.189 | 7.283 |
| wet | MW0200 | 0 | 45.094 | rarefaction | 0.796 | 15.651 | 1.415 | 12.877 | 18.425 |
| wet | MW0200 | 1 | 45.094 | rarefaction | 0.796 | 9.570 | 1.071 | 7.471 | 11.668 |
| wet | MW0200 | 2 | 45.094 | rarefaction | 0.796 | 6.630 | 0.816 | 5.031 | 8.228 |
| dry | MW0700 | 0 | 46.000 | extrapolation | 0.565 | 25.975 | 5.306 | 15.575 | 36.374 |
| dry | MW0700 | 1 | 46.000 | extrapolation | 0.565 | 17.858 | 6.141 | 5.821 | 29.895 |
beta_pair()Pairwise beta diversity: Sorensen turnover (Simpson) and nestedness
components, which sum to total dissimilarity.
| season | site_1 | site_2 | turnover | nestedness | total |
|---|---|---|---|---|---|
| dry | MW0200 | MW0700 | 0.667 | 0.011 | 0.677 |
| dry | MW0200 | MW1200 | 0.769 | 0.024 | 0.793 |
| dry | MW0200 | MW1700 | 0.800 | 0.105 | 0.905 |
| dry | MW0200 | MW2200 | 1.000 | 0.000 | 1.000 |
| dry | MW0700 | MW1200 | 0.615 | 0.027 | 0.643 |
| dry | MW0700 | MW1700 | 0.800 | 0.100 | 0.900 |
| dry | MW0700 | MW2200 | 0.833 | 0.071 | 0.905 |
| dry | MW1200 | MW1700 | 1.000 | 0.000 | 1.000 |
beta_multi()Multi-site beta diversity: n_sites in each group with the multi-site
turnover, nestedness and total.
| season | n_sites | turnover | nestedness | total |
|---|---|---|---|---|
| dry | 5 | 0.827 | 0.051 | 0.877 |
| wet | 5 | 0.790 | 0.074 | 0.864 |
beta_decay()Distance decay: the pairwise components plus each site’s gradient_1
and gradient_2 values and their distance.
| season | site_1 | site_2 | turnover | nestedness | total | gradient_1 | gradient_2 | distance |
|---|---|---|---|---|---|---|---|---|
| dry | MW0200 | MW0700 | 0.667 | 0.011 | 0.677 | 200 | 700 | 500 |
| dry | MW0200 | MW1200 | 0.769 | 0.024 | 0.793 | 200 | 1200 | 1000 |
| dry | MW0200 | MW1700 | 0.800 | 0.105 | 0.905 | 200 | 1700 | 1500 |
| dry | MW0200 | MW2200 | 1.000 | 0.000 | 1.000 | 200 | 2200 | 2000 |
| dry | MW0700 | MW1200 | 0.615 | 0.027 | 0.643 | 700 | 1200 | 500 |
| dry | MW0700 | MW1700 | 0.800 | 0.100 | 0.900 | 700 | 1700 | 1000 |
| dry | MW0700 | MW2200 | 0.833 | 0.071 | 0.905 | 700 | 2200 | 1500 |
| dry | MW1200 | MW1700 | 1.000 | 0.000 | 1.000 | 1200 | 1700 | 500 |
beta_decay_model()Distance-decay model of similarity: the fitted a and rate b, the
half_distance over which similarity halves, the pseudo_r2, aic and
a site-permutation p.value.
| season | model | component | n_pairs | a | b | half_distance | pseudo_r2 | aic | p.value |
|---|---|---|---|---|---|---|---|---|---|
| dry | exponential | total | 10 | 0.257 | 0.001 | 924.159 | 0.143 | -9.214 | 0.349 |
| wet | exponential | total | 10 | 0.897 | 0.002 | 293.331 | 0.694 | -17.577 | 0.017 |
beta_hill_pair()Pairwise Hill-number beta diversity of order q with the four
normalised dissimilarities (first rows shown).
| season | site_1 | site_2 | q | gamma | alpha | beta | dis_sorensen | dis_jaccard | dis_turnover_sorensen | dis_turnover_jaccard |
|---|---|---|---|---|---|---|---|---|---|---|
| dry | MW0200 | MW0700 | 0 | 26 | 15.5 | 1.677 | 0.677 | 0.808 | 0.677 | 0.808 |
| dry | MW0200 | MW1200 | 0 | 26 | 14.5 | 1.793 | 0.793 | 0.885 | 0.793 | 0.885 |
| dry | MW0200 | MW1700 | 0 | 20 | 10.5 | 1.905 | 0.905 | 0.950 | 0.905 | 0.950 |
| dry | MW0200 | MW2200 | 0 | 22 | 11.0 | 2.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| dry | MW0700 | MW1200 | 0 | 23 | 14.0 | 1.643 | 0.643 | 0.783 | 0.643 | 0.783 |
| dry | MW0700 | MW1700 | 0 | 19 | 10.0 | 1.900 | 0.900 | 0.947 | 0.900 | 0.947 |
beta_hill()Coverage-standardised beta diversity: gamma, alpha and beta of
order q at a common coverage, the sample sizes used (m_gamma,
m_alpha) and four normalised dissimilarities.
| season | q | coverage | m_gamma | m_alpha | gamma | alpha | beta | n_sites | dis_sorensen | dis_jaccard | dis_turnover_sorensen | dis_turnover_jaccard |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| dry | 0 | 0.9 | 150.161 | 244.117 | 44.361 | 14.476 | 3.064 | 5 | 0.516 | 0.842 | 0.516 | 0.842 |
| dry | 1 | 0.9 | 150.161 | 244.117 | 30.995 | 9.929 | 3.122 | 5 | 0.707 | 0.707 | 0.530 | 0.850 |
| dry | 2 | 0.9 | 150.161 | 244.117 | 23.324 | 6.968 | 3.347 | 5 | 0.877 | 0.587 | 0.587 | 0.877 |
| wet | 0 | 0.9 | 259.764 | 480.767 | 64.058 | 23.502 | 2.726 | 5 | 0.431 | 0.791 | 0.431 | 0.791 |
| wet | 1 | 0.9 | 259.764 | 480.767 | 35.366 | 12.345 | 2.865 | 5 | 0.654 | 0.654 | 0.466 | 0.814 |
| wet | 2 | 0.9 | 259.764 | 480.767 | 21.377 | 7.081 | 3.019 | 5 | 0.836 | 0.505 | 0.505 | 0.836 |
beta_welch()Welch-type PERMANOVA (W*d), robust to unequal dispersion: the
statistic, the numbers of groups and sites, and the permutation
p.value.
| statistic | n_groups | n_sites | p.value |
|---|---|---|---|
| 1.29 | 2 | 20 | 0.179 |
If you use tidydiversity in published work, please cite it. In R, citation("tidydiversity")
prints the reference:
Moses, J. (2026). tidydiversity: Taxonomic, Phylogenetic and Functional Diversity with Tidy Hill Numbers. R package version 0.2.0. Zenodo. https://doi.org/10.5281/zenodo.24182679
@Manual{tidydiversity,
title = {tidydiversity: Taxonomic, Phylogenetic and Functional Diversity with Tidy Hill Numbers},
author = {Jimmy Moses},
year = {2026},
note = {R package version 0.1.0},
doi = {10.5281/zenodo.24182679},
url = {https://github.com/jmozez/tidydiversity},
}
Please also cite the methods you use: each help page and User guide chapter lists the original references.
MIT © Jimmy Moses