Movatterモバイル変換

CRAN Task View: Analysis of Ecological and Environmental Data

Introduction

This Task View contains information about using R to analyse ecological and environmental data.

The base version of R ships with a wide range of functions for use within the field of environmetrics. This functionality is complemented by a plethora of packages available via CRAN, which provide specialist methods such as ordination & cluster analysis techniques. A brief overview of the available packages is provided in this Task View, grouped by topic or type of analysis. As a testament to the popularity of R for the analysis of environmental and ecological data, aspecial volume of theJournal of Statistical Software was produced in 2007.

Those interested in environmetrics should consult theSpatial view. Complementary information is also available in theCluster, andSpatioTemporal task views.

If you have any comments or suggestions for additions or improvements, then please contact the maintainer or submit an issue or pull request in the GitHub repository linked above.

A list of available packages and functions is presented below, grouped by analysis type.

General packages

These packages are general, having wide applicability to the environmetrics field.

• PackageEnvStats is the successor to the S-PLUS moduleEnvironmentalStats, both by Steven Millard. Auser guide in the form of a book has been released.

Modelling species responses and other data

Analysing species response curves or modelling other data often involves the fitting of standard statistical models to ecological data and includes simple (multiple) regression, Generalized Linear Models (GLM), extended regression (e.g. Generalized Least Squares [GLS]), Generalized Additive Models (GAM), and mixed effects models, amongst others.

• The base installation of R provideslm() andglm() for fitting linear and generalized linear models, respectively.
• Generalized least squares and linear and non-linear mixed effects models extend the simple regression model to account for clustering, heterogeneity and correlations within the sample of observations. Packagenlme provides functions for fitting these models. The package is supported by Pinheiro & Bates (2000)Mixed-effects Models in S and S-PLUS, Springer, New York. An updated approach to mixed effects models, which also fits Generalized Linear Mixed Models (GLMM) and Generalized non-Linear Mixed Models (GNLMM) is provided by thelme4 package, though this is currently beta software and does not yet allow correlations within the error structure.
• Recommended packagemgcv fits GAMs and Generalized Additive Mixed Models (GAMM) with automatic smoothness selection via generalized cross-validation. The author ofmgcv has also written a companion monograph, Wood (2017)Generalized Additive Models; An Introduction with R Second Edition. Chapman Hall/CRC, which has an accompanying packagegamair.
• Alternatively, packagegam provides an implementation of the S-PLUS functiongam() that includes LOESS smooths.
• Proportional odds models for ordinal responses can be fitted usingpolr() in theMASS package, of Bill Venables and Brian Ripley.
• A negative binomial family for GLMs to model over-dispersion in count data is available inMASS.
• Models for overdispersed counts and proportions
  • Packagepscl also contains several functions for dealing with over-dispersed count data. Poisson or negative binomial distributions are provided for both zero-inflated and hurdle models.
  • aod provides a suite of functions to analyse overdispersed counts or proportions, plus utility functions to calculate e.g. AIC, AICc, Akaike weights.
• Detecting change points and structural changes in parametric models is well catered for in thesegmented package and thestrucchange package respectively.segmented is discussed in anR News article

Tree-based models

Tree-based models are being increasingly used in ecology, particularly for their ability to fit flexible models to complex data sets and the simple, intuitive output of the tree structure. Ensemble methods such as bagging, boosting and random forests are advocated for improving predictions from tree-based models and to provide information on uncertainty in regression models or classifiers.

Univariate trees

Tree-structured models for regression, classification and survival analysis, following the ideas in the CART book, are implemented in

• recommended packagerpart
• party provides an implementation of conditional inference trees which embed tree-structured regression models into a well-defined theory of conditional inference procedures

Multivariate trees

Multivariate trees are available in

• packageparty can also handle multivariate responses.

Ensembles of trees

Ensemble techniques for trees:

• The Random Forest method of Breiman and Cutler is implemented inrandomForest, providing classification and regression based on a forest of trees using random inputs
• Packageipred provides functions for improved predictive models for classification, regression and survival problems.

Graphical tools for the visualization of trees are available in packagemaptree.

Packagesmda andearth implement Multivariate Adaptive Regression Splines (MARS), a technique which provides a more flexible, tree-based approach to regression than the piecewise constant functions used in regression trees.

Ordination

R and add-on packages provide a wide range of ordination methods, many of which are specialized techniques particularly suited to the analysis of species data. The two main packages areade4 andvegan.ade4 derives from the traditions of the French school of “Analyse des Donnees” and is based on the use of the duality diagram.vegan follows the approach of Mark Hill, Cajo ter Braak and others, though the implementation owes more to that presented in Legendre & Legendre (1988)Numerical Ecology, 2^nd English Edition, Elsevier. Where the two packages provide duplicate functionality, the user should choose whichever framework that best suits their background.

• Principal Components (PCA) is available via theprcomp() function.rda() (in packagevegan),pca() (in packagelabdsv) anddudi.pca() (in packageade4), provide more ecologically-orientated implementations.
• Redundancy Analysis (RDA) is available viarda() invegan andpcaiv() inade4.
• Canonical Correspondence Analysis (CCA) is implemented incca() in bothvegan andade4.
• Detrended Correspondence Analysis (DCA) is implemented indecorana() invegan.
• Principal coordinates analysis (PCO) is implemented indudi.pco() inade4,pco() inlabdsv,pco() inecodist, andcmdscale() in packageMASS.
• Non-Metric multi-Dimensional Scaling (NMDS) is provided byisoMDS() in packageMASS andnmds() inecodist.nmds(), a wrapper function forisoMDS(), is also provided by packagelabdsv.vegan provides helper functionmetaMDS() forisoMDS(), implementing random starts of the algorithm and standardized scaling of the NMDS results. The approach adopted byvegan withmetaMDS() is the recommended approach for ecological data.
• Coinertia analysis is available viacoinertia() andmcoa(), both inade4.
• Co-correspondence analysis to relate two ecological species data matrices is available incocorresp.
• Canonical Correlation Analysis (CCoA - not to be confused with CCA, above) is available incancor() in standard package stats.
• Procrustes rotation is available inprocrustes() invegan andprocuste() inade4, with bothvegan andade4 providing functions to test the significance of the association between ordination configurations (as assessed by Procrustes rotation) using permutation/randomization and Monte Carlo methods.
• Constrained Analysis of Principal Coordinates (CAP), implemented incapscale() invegan, fits constrained ordination models similar to RDA and CCA but with any dissimilarity coefficient.
• Fuzzy set ordination (FSO), an alternative to CCA/RDA and CAP, is available in packagefso.fso complements a recent paper on fuzzy sets in the journalEcology by Dave Roberts (2008, Statistical analysis of multidimensional fuzzy set ordinations.Ecology89(5), 1246-1260).

Model-based multivariate analysis

Multivariate model-based methods follow typical statistical modeling principles, but for multivariate responses. Model-based ordination methods reduce dimensionality of a model component (usually predictor effects of a random-effect covariance matrix), so that they share features with both ordination methods (the ordination) and regression (e.g., information criteria and residual diagnostics). It thus requires specifying a response distribution, and link function, instead of a dissimilarity measure. Unlike “classical” ordination methods, it is usually required to specify the number of ordination axesa priori of fitting the model. The following packages have different features and functionalities, but most support creating ordinations.

• VGAM package implements ordination based on fixed effects. Ordination plots are constructed with thebiplot() orlvplot() functions. Implemented ordination methods include,
  • Unconstrained ordination with thercim() orgrc() functions.grc() implements Goodman’s RC association model, andrcim() generically fits row-column interaction models.
  • Constrained ordination with linear responses using therrvglm() function.
  • Constrained Quadratic Ordination (CQO; formerly known as Canonical Gaussian Ordination (CGO)), which is a maximum likelihood estimation alternative to Canonical Correspondence Analysis (CCA) fit by Quadratic Reduced Rank Vector GLMs, with thecqo() function.
  • Constrained Additive Ordination (CAO), an extension of CQO to flexible response curves, with thecao() function.
• mvabund does not perform ordination, but fits multivariate models following GLM principles, potentially with a residual correlation structure for species. This is implemented with themanylm(),manyglm(),traitglm(), andmanyany(), functions. Thecoefplot() function plots species responses to predictors with confidence intervals, and hypothesis testing based on resampling strategies is available via theanova() function.
• boral, which stands for “Bayesian Ordination and Regression AnaLysis” fits Joint Species Distribution Models (JSDMs) with a latent variable formulation for a variety of responses in a Bayesian framework where the model is fitted throughR2jags. The package supports residual and unconstrained ordinations, potentially with spatially structured latent variables, and stacked models (i.e., GLMs), with site random effects. It is possible to fit models with species observations in different types, e.g., to combine binary and count data in a single model. With functional traits, boral fits a fourth-corner model. Boral allows to perform automatic variable selection via stochastic search variable selection, i.e., by using a spike-and-slab prior. Since boral fits the models with only one MCMC chain, it can be difficult to assess convergence of the model, though the Geweke diagnostic statistic is available to help with that. The ordination can be visualized with thelvsplot() function.
• Hmsc fits Joint Species Distribution Models (JSDMs) with a latent variable formulation for normal responses, binary responses, and counts in a Bayesian framework. “HMSC” stands for “Hierarchical modeling of species communities”, and consequently the package allows to fit hierarchical models for multivariate responses. Its main function isHmsc(), latent variables can be visualized with thebiPlot() function. It has many different tools, including the option to separately formulate models for different sampling levels, to include spatial effects, to include additional random effects, to phylogenetically structure species responses to environmental predictors, or to hierarchically model species associations with predictors. HMSC determines the number of latent variables from the data, which thus do not need to be specified.
• gllvm fits latent variable models for ordination and JSDM purposes to a range of responses, with random-effects, in a relatively fast manner usingTMB. Its main functiongllvm() allows to fit unconstrained, constrained, and concurrent ordinations. Unlike in VGAM, unconstrained ordination is based on a random effects formulation. Constrained ordination is supported both as fixed and random effects formulation. Concurrent ordination by definition always includes random effects, but is also supported as a fully random effects specification with random slopes. Fourth-corner models with latent variables and random slopes are also supported. The ordination can be visualized with theordiplot() function, which also allows visualization of statistical uncertainty of the site scores.
• ecoCopula with main functionscgr() andcord() uses a Gaussian copula approach to fit multivariate models. Both functions first require fitting a secondary model, from which residuals are extracted to which the package fits its method.cgr() fits a graphical model, with the purpose of visualizing pairwise associations of species. The resulting network graph can be visualized using theplot() method. Thecord() function fits a model-based ordination with Gaussian copulas, which can be visualized usingplot(, biplot = TRUE). No estimates of statistical uncertainties are available.
• glmmTMB generically fits random-effects models usingTMB, and can thus fit model-based unconstrained ordination with additional random-effects. Its main function isglmmTMB(), and model-based ordination is fitted using therr() covariance structure in the model. There is no function to visualize the ordination at present.
• sjSDM fits JSDMs to gaussian, Bernoulli, or Poisson responses, not with a latent variable formulation but with an elastic net penalty approach. As such, it not a model-based ordination method. It allows to include spatial effects, and few other extensions. It allows computation with both CPU and GPU resources, which makes for a very fast fitting method that scales well for large datasets.
• gjam fits generalized joint attribute models in a Bayesian framework to a variety of response types. The package can post-hoc perform ordination with PCA or NMDS, and perform dimension reduction by setting parameters for the covariance matrix ingjam(), but does not have the ability to fit a model-based ordination. It is possible fit trait response models, and to include a measure of sampling effort in the models. It is possible to fit models with species observations in different types, and to include some additional random effect to account for clustering of observations. Output can be plotted withgjamPlot() or an ordination withgjamOrdination().
• spOccupancy fits occupancy models for single- and multi-species responses, and can be used to fit JSDMs that account for imperfect detection.
  • mvgam fits Bayesian State-Space Generalized Additive Models for many data types including counts, binary observations, proportions, non-negative real values and unbounded real values.mvgam can fit a wide range of hierarchical ecological models including N-mixture models, Latent Variable-based Joint Species Distribution Models, and multivariate time series models with imperfect detection.

Dissimilarity coefficients

Much ecological analysis proceeds from a matrix of dissimilarities between samples. A large amount of effort has been expended formulating a wide range of dissimilarity coefficients suitable for ecological data. A selection of the more useful coefficients are available in R and various contributed packages.

Standard functions that produce, square, symmetric matrices of pair-wise dissimilarities include:

• dist() in standard package stats
• daisy() in recommended packagecluster
• vegdist() invegan
• dsvdis() inlabdsv
• Dist() inamap
• distance() inecodist
• a suite of functions inade4

Functiondistance() in packageanalogue can be used to calculate dissimilarity between samples of one matrix and those of a second matrix. The same function can be used to produce pair-wise dissimilarity matrices, though the other functions listed above are faster.distance() can also be used to generate matrices based on Gower’s coefficient for mixed data (mixtures of binary, ordinal/nominal and continuous variables). Functiondaisy() in packagecluster provides a faster implementation of Gower’s coefficient for mixed-mode data thandistance() if a standard dissimilarity matrix is required. Functiongowdis() in packageFD also computes Gower’s coefficient and implements extensions to ordinal variables.

Cluster analysis

Cluster analysis aims to identify groups of samples within multivariate data sets. A large range of approaches to this problem have been suggested, but the main techniques are hierarchical cluster analysis, partitioning methods, such ask -means, and finite mixture models or model-based clustering. In the machine learning literature, cluster analysis is an unsupervised learning problem.

TheCluster task view provides a more detailed discussion of available cluster analysis methods and appropriate R functions and packages.

Hierarchical cluster analysis:

• hclust() in standard package stats
• Recommended packagecluster provides functions for cluster analysis following the methods described in Kaufman and Rousseeuw (1990)Finding Groups in data: an introduction to cluster analysis, Wiley, New York
• hcluster() inamap
• pvclust is a package for assessing the uncertainty in hierarchical cluster analysis. It provides approximately unbiasedp -values as well as bootstrapp -values.

Partitioning methods:

• kmeans() in stats providesk -means clustering
• cmeans() ine1071 implements a fuzzy version of thek -means algorithm
• Recommended packagecluster also provides functions for various partitioning methodologies.

Mixture models and model-based cluster analysis:

• mclust andflexmix provide implementations of model-based cluster analysis.
• prabclus clusters a species presence-absence matrix object by calculating an MDS from the distances, and applying maximum likelihood Gaussian mixtures clustering to the MDS points. The maintainer’s, Christian Hennig, website contains several publications in ecological contexts that useprabclus, especially Hausdorf & Hennig (2007;Oikos 116 (2007), 818-828).

Ecological theory

There is a growing number of packages and books that focus on the use of R for theoretical ecological models.

• vegan provides a wide range of functions related to ecological theory, such as diversity indices (including the “so-called” Hill’s numbers [e.g. Hill’s N² ] and rarefaction), ranked abundance diagrams, Fisher’s log series, Broken Stick model, Hubbell’s abundance model, amongst others.
• untb provides a collection of utilities for biodiversity data, including the simulation ecological drift under Hubbell’s Unified Neutral Theory of Biodiversity, and the calculation of various diagnostics such as Preston curves.
• PackageBiodiversityR provides a GUI for biodiversity and community ecology analysis.
• Functionbetadiver() invegan implements all the diversity indices reviewed in Koleff et al (2003;Journal of Animal Ecology 72(3), 367-382 ).betadiver() also provides aplot method to produce the co-occurrence frequency triangle plots of the type found in Koleff et al (2003).
• Functionbetadisper(), also invegan, implements Marti Anderson’s distance-based test for homogeneity of multivariate dispersions (PERMDISP, PERMDISP2), a multivariate analogue of Levene’s test (Anderson 2006;Biometrics 62, 245-253 ). Anderson et al (2006;Ecology Letters 9(6), 683-693 ) demonstrate the use of this approach for measuring beta diversity.
• TheFD package computes several measures of functional diversity indices from multiple traits.

Population dynamics

Estimating animal abundance and related parameters

This section concerns estimation of population parameters (population size, density, survival probability, site occupancy etc.) by methods that allow for incomplete detection. Many of these methods use data on marked animals, variously called ‘capture-recapture’, ‘mark-recapture’ or ‘capture-mark-recapture’ data.

• Rcapture fits loglinear models to estimate population size and survival rate from capture-recapture data as described byBaillargeon and Rivest (2007).
• secr estimates population density given spatially explicit capture-recapture data from traps, passive DNA sampling, automatic cameras, sound recorders etc. Models are fitted by maximum likelihood. The detection function may be half-normal, exponential, cumulative gamma etc. Density surfaces may be fitted. Covariates of density and detection parameters are specified via formulae.
• unmarked fits hierarchical models of occurrence and abundance to data collected on species subject to imperfect detection. Examples include single- and multi-season occupancy models, binomial mixture models, and hierarchical distance sampling models. The data can arise from survey methods such temporally replicated counts, removal sampling, double-observer sampling, and distance sampling. Parameters governing the state and observation processes can be modelled as functions of covariates.
• PackageRMark provides a formula-based R interface for the MARK package which fits a wide variety of capture-recapture models. See theRMark website and aNOAA report (PDF) for further details.
• Packagemarked provides a framework for handling data and analysis for mark-recapture.marked can fit Cormack-Jolly-Seber (CJS) and Jolly-Seber (JS) models via maximum likelihood and the CJS model via MCMC. Maximum likelihood estimates for the CJS model can be obtained using R or via a link to the Automatic Differentiation Model Builder software. Adescription of the package was published in Methods in Ecology and Evolution.
• mrds fits detection functions to point and line transect distance sampling survey data (for both single and double observer surveys). Abundance can be estimated using Horvitz-Thompson-type estimators.
• Distance is a simpler interface tomrds for single observer distance sampling surveys.
• dsm fits density surface models to spatially-referenced distance sampling data. Count data are corrected using detection function models fitted usingmrds orDistance. Spatial models are constructed as inmgcv.
• singleRcapture provides methods for estimating the population size of hard-to-reach populations using single-source capture-recapture methods. It implements zero-truncated, zero-one-truncated, zero-truncated one-inflated, and one-inflated zero truncated count regression models, as well as the Chao and Zelterman models. It provides fit assessment functions, diagnostic plots and four methods for estimating the variance of the population size.

Packagessecr can also be used to simulate data from the respective models.

See also theSpatioTemporal task view for analysis of animal tracking data underMoving objects, trajectories.

Modelling population growth rates:

• Packagepopbio can be used to construct and analyse age- or stage-specific matrix population models.

• Packagebiogrowth implements functions for modelling the growth of populations. It includes functions for model fitting and prediction under either constant or dynamic environmental conditions. It also includes functions for model comparison/selection.

Environmental time series

• Time series objects in R are created using thets() function, though seetseries orzoo below for alternatives.
• Classical time series functionality is provided by thear(), andarima() functions in standard package stats for autoregressive (AR), moving average (MA), autoregressive moving average (ARMA) and integrated ARMA (ARIMA) models.
• Theforecast package provides methods and tools for displaying and analysing univariate time series forecasts including exponential smoothing via state space models and automatic ARIMA modelling
• Thedse(archived) package provide a variety of more advanced estimation methods and multivariate time series analysis.
• Packagestseries andzoo provide general handling and analysis of time series data.
• Irregular time series can be handled using packagezoo as well as byirts() in packagetseries.
• pastecs provides functions specifically tailored for the analysis of space-time ecological series.
• strucchange allows for testing, dating and monitoring of structural change in linear regression relationships.
• Detecting change points in time series data — seesegmented above.
• Thesurveillance package implements statistical methods for the modelling of and change-point detection in time series of counts, proportions and categorical data. Focus is on outbreak detection in count data time series.
• Packagedynlm provides a convenient interface to fitting time series regressions via ordinary least squares
• Packagedyn provides a different approach to that ofdynlm, which allows time series data to be used with any regression function written in the style oflm such aslm(),glm(),loess(),rlm() andlqs() fromMASS,randomForest() (packagerandomForest),rq() (packagequantreg) amongst others, whilst preserving the time series information.
• Theopenair provides numerous tools to analyse, interpret and understand air pollution time series data
• ThebReeze package is a collection of widely used methods to analyse, visualize, and interpret wind data. Wind resource analyses can subsequently be combined with characteristics of wind turbines to estimate the potential energy production.
• TheRbeast package provides a Bayesian model averaging method to decompose time series into abrupt changes, trend, and seasonality and can be used for changepoint detection, time series decomposition, and nonlinear trend analysis.

Additionally, a fuller description of available packages for time series analysis can be found in theTimeSeries task view.

Spatial data analysis

See theSpatial CRAN Task View for an overview of spatial analysis in R.

Extreme values

ismev provides functions for models for extreme value statistics and is support software for Coles (2001)An Introduction to Statistical Modelling of Extreme Values , Springer, New York. Other packages for extreme value theory include

• evir
• evd
• evdbayes, which provides a Bayesian approach to extreme value theory
• extRemes

See also theExtremeValue task view for further information.

Phylogenetics and evolution

Packages specifically tailored for the analysis of phylogenetic and evolutionary data include:

• ape
• ouch

UseRs may also be interested in Paradis (2006)Analysis of Phylogenetics and Evolution with R, Springer, New York, a book in the“Use R!” book series from Springer.

Soil science

Several packages are now available that implement R functions for widely-used methods and approaches in pedology.

• soiltexture provides functions for soil texture plot, classification and transformation.
• aqp contains a collection of algorithms related to modelling of soil resources, soil classification, soil profile aggregation, and visualization.
• The Soil Water project on R-Forge.R-project.org provides packages providing soil water retention functions, soil hydraulic conductivity functions and pedotransfer functions to estimate their parameter from easily available soil properties. Two packages form the project:soilwaterfun andsoilwaterptf.

Hydrology and Oceanography

A growing number of packages are available that implement methods specifically related to the fields of hydrology and oceanography. Also see theExtreme Value and theClimatology sections for related packages.

• topmodel(archived) is a set of hydrological functions including an R implementation of the hydrological model TOPMODEL, which is based on the 1995 FORTRAN version by Keith Beven. New functionality is being developed as part of theRHydro package on R-Forge.
• Packageseacarb provides functions for calculating parameters of the seawater carbonate system.
• Stephen Sefick’sStreamMetabolism package contains function for calculating stream metabolism characteristics, such as GPP, NDM, and R, from single station diurnal Oxygen curves.
• Packageoce supports the analysis of Oceanographic data, including ADP measurements, CTD measurements, sectional data, sea-level time series, and coastline files.
• ThensRFA package provides collection of statistical tools for objective (non-supervised) applications of the Regional Frequency Analysis methods in hydrology.
• Theboussinesq package is a collection of functions implementing the one-dimensional Boussinesq Equation (ground-water).
• rtop is a package for geostatistical interpolation of data with irregular spatial support such as runoff related data or data from administrative units.
• A related package isqualV which provides quantitative and qualitative criteria to compare models with data and to measure similarity of patterns

Climatology

Several packages related to the field of climatology.

• seas implements a number of functions for analysis and graphics of seasonal data.
• RMAWGEN is set of S3 and S4 functions for spatial multi-site stochastic generation of daily time series of temperature and precipitation making use of Vector Autoregressive Models.

Palaeoecology and stratigraphic data

Several packages now provide specialist functionality for the import, analysis, and plotting of palaeoecological data.

• Transfer function models including weighted averaging (WA), modern analogue technique (MAT), Locally-weighted WA, & maximum likelihood (aka Gaussian logistic) regression (GLR) are provided by therioja andanalogue packages.
• Import of common, legacy, palaeo-data formats is provided by packagevegan (Cornell format).
• Stratigraphic data plots can be drawn usingStratiplot() function inanalogue and functionsstrat.plot() andstrat.plot.simple in therioja package. Also see thetidypaleo package, which provides tools to produce stratigraphic plots usingggplot(). Ablog post by the maintainer of thetidypaleo package, Dewey Dunnington, shows how to use the package to create stratigraphic plots.
• analogue provides extensive support for developing and interpreting MAT transfer function models, including ROC curve analysis. Summary of stratigraphic data is supported via principal curves in theprcurve() function.

Other packages

Several other relevant contributed packages for R are available that do not fit under nice headings.

• Andrew Robinson’sequivalence package provides some statistical tests and graphics for assessing tests of equivalence. Such tests have similarity as the alternative hypothesis instead of the null. The package contains functions to perform two one-sided t-tests (TOST) and paired t-tests of equivalence.
• Thomas Petzoldt’ssimecol package provides an object oriented framework and tools to simulate ecological (and other) dynamic systems within R. See thesimecol website and aR Newsarticle on the package for further information.
• Functions for circular statistics are found inCircStats andcircular.
• Packagee1071 provides functions for latent class analysis, short time Fourier transform, fuzzy clustering, support vector machines, shortest path computation, bagged clustering, naive Bayes classifier, and more…
• Packagepgirmess provides a suite of miscellaneous functions for data analysis in ecology.
• mefa provides functions for handling and reporting on multivariate count data in ecology and biogeography.
• Sensitivity analysis of models is provided by packagesensitivity.sensitivity contains a collection of functions for factor screening and global sensitivity analysis of model output.
• Functions to analyse coherence, boundary clumping, and turnover following the pattern-based metacommunity analysis ofLeibold and Mikkelson (2002) are provided in themetacom package.
• Growth curve estimation via noncrossing and nonparametric regression quantiles is implemented in packagequantregGrowth. A supporting paper isMuggeo et al. (2013).
• Thesiplab package provides an R platform for experimenting with spatially explicit individual-based vegetation models. A supporting paper isGarcía, O. (2014).
• PMCMRplus provides parametric and non-parametric many-to-one and all-pairs multiple comparison procedures for continuous or at least interval based variables. The package provides implementations of a wide range of tests involving pairwise multiple comparisons.

CRAN packages

Related links

• The vegan development site on GitHub
• Thomas Yee’s VGAM package for R
• The cocorresp development site on GitHub
• Brodgar
• More information on the ade4 package can be found on the ADE4 website
• Mike Palmer’s Ordination website
• Thomas Petzoldt’s page about ecological modelling with R
• The FLR project web page for Fisheries Library in R.
• The gllvm website with vignettes

Other resources

• CRAN Task View:Cluster
• CRAN Task View:ExtremeValue
• CRAN Task View:Spatial
• CRAN Task View:SpatioTemporal
• CRAN Task View:TimeSeries
• R-Forge Project:RHydro
• R-Forge Project:soilwaterfun
• R-Forge Project:soilwaterptf
• GitHub Project:tidypaleo