Package index • tidysdm

Data preparation

Functions for sampling background/pseudo-absences points and thinning the data.

sample_background(): Sample background points for SDM analysis

sample_background_time(): Sample background points for SDM analysis for points with a time point.

sample_pseudoabs(): Sample pseudo-absence points for SDM analysis

sample_pseudoabs_time(): Sample pseudo-absence points for SDM analysis for points with a time point.

thin_by_cell(): Thin point dataset to have 1 observation per raster cell

thin_by_cell_time(): Thin point dataset to have 1 observation per raster cell per time slice

thin_by_dist(): Thin points dataset based on geographic distance

thin_by_dist_time(): Thin points dataset based on geographic and temporal distance

make_mask_from_presence(): Make a mask from presence data

Choice of predictor variables

Functions for removing collinearity and visualising the distribution of predictors.

filter_collinear(): Filter to retain only variables that have low collinearity

plot_pres_vs_bg(): Plot presences vs background

geom_split_violin(): Split violin geometry for ggplots

dist_pres_vs_bg(): Distance between the distribution of climate values for presences vs background

Recipes

Functions for recipes with spatial SDM data (additional steps can be added with standard recipes functions).

recipe(<sf>) spatial_recipe(): Recipe for sf objects

check_sdm_presence(): Check that the column with presences is correctly formatted

Models specification

Predefined model specifications (custom models can be added with standard parsnip model specifications).

sdm_spec_glm(): Model specification for a GLM for SDM

sdm_spec_gam(): Model specification for a GAM for SDM

gam_formula(): Create a formula for gam

sdm_spec_rand_forest() sdm_spec_rf(): Model specification for a Random Forest for SDM

sdm_spec_boost_tree(): Model specification for a Boosted Trees model for SDM

sdm_spec_maxent(): Model specification for a MaxEnt for SDM

maxent(): MaxEnt model

regularization_multiplier() feature_classes(): Parameters for maxent models

Data splitting

Functions for splitting the data into folds (in additiona to standard spatialsample functions.

spatial_initial_split(): Simple Training/Test Set Splitting for spatial data

blockcv2rsample(): Convert an object created with blockCV to an rsample object

check_splits_balance(): Check the balance of presences vs pseudoabsences among splits

autoplot(<spatial_initial_split>): Create a ggplot for a spatial initial rsplit.

grid_offset(): Get default grid cellsize for a given dataset

grid_cellsize(): Get default grid cellsize for a given dataset

Metrics

Specialised metrics for SDM, and methods of metrics from yardstick adapted to work on sf objects

sdm_metric_set(): Metric set for SDM

optim_thresh(): Find threshold that optimises a given metric

boyce_cont() boyce_cont_vec(): Boyce continuous index (BCI)

kap_max() kap_max_vec(): Maximum Cohen's Kappa

tss_max() tss_max_vec(): Maximum TSS - True Skill Statistics

tss(): TSS - True Skill Statistics

average_precision(<sf>) brier_class(<sf>) classification_cost(<sf>) gain_capture(<sf>) mn_log_loss(<sf>) pr_auc(<sf>) roc_auc(<sf>) roc_aunp(<sf>) roc_aunu(<sf>): Probability metrics for sf objects

Ensemble

Functions for creating an ensemble selecting the best set of parameters for each model.

simple_ensemble(): Simple ensemble

autoplot(<simple_ensemble>): Plot the results of a simple ensemble

repeat_ensemble(): Repeat ensemble

add_member(): Add best member of workflow to a simple ensemble

add_repeat(): Add repeat(s) to a repeated ensemble

collect_metrics(<simple_ensemble>) collect_metrics(<repeat_ensemble>): Obtain and format results produced by tuning functions for ensemble objects

control_ensemble_grid() control_ensemble_resamples() control_ensemble_bayes(): Control wrappers

explain_tidysdm(): Create explainer from your tidysdm ensembles.

Project SDM to present, past and future

Functions for making predictions.

calib_class_thresh(): Calibrate class thresholds

predict(<repeat_ensemble>): Predict for a repeat ensemble set

predict(<simple_ensemble>): Predict for a simple ensemble set

predict_raster(): Make predictions for a whole raster

Managing extrapolation

Functions for managing extrapolation.

clamp_predictors(): Clamp the predictors to match values in training set

extrapol_mess(): Multivariate environmental similarity surfaces (MESS)

Other

Additional helpful functions.

niche_overlap(): Compute overlap metrics of the two niches

km2m(): Convert a geographic distance from km to m

y2d(): Convert a time interval from years to days

filter_high_cor(): Deprecated: Filter to retain only variables below a given correlation threshold

pairs(<stars>): This is a wrapper around graphics::pairs() that accepts stars objects. It is adapted from a similar function in the terra package.

Example data

Example datasets and simple models used in the documentation

horses: Coordinates of radiocarbon dates for horses

lacerta: Coordinates of presences for Iberian emerald lizard

lacertidae_background: Coordinates of presences for lacertidae in the Iberian peninsula

lacerta_ensemble: A simple ensemble for the lacerta data

lacerta_rep_ens: A repeat ensemble for the lacerta data