JSS Journal of Statistical Software

August 2011, Volume 43, Issue 10. [Link]

unmarked: An R Package for Fitting Hierarchical

Models of Wildlife Occurrence and Abundance

Ian J. Fiske Richard B. Chandler

North Carolina State University USGS Patuxent Wildlife Research Center

Abstract
Ecological research uses data collection techniques that are prone to substantial and
unique types of measurement error to address scientific questions about species abundance
and distribution. These data collection schemes include a number of survey methods
in which unmarked individuals are counted, or determined to be present, at spatially-
referenced sites. Examples include site occupancy sampling, repeated counts, distance
sampling, removal sampling, and double observer sampling. To appropriately analyze
these data, hierarchical models have been developed to separately model explanatory
variables of both a latent abundance or occurrence process and a conditional detection
process. Because these models have a straightforward interpretation paralleling mecha-
nisms under which the data arose, they have recently gained immense popularity. The
common hierarchical structure of these models is well-suited for a unified modeling in-
terface. The R package unmarked provides such a unified modeling framework, including
tools for data exploration, model fitting, model criticism, post-hoc analysis, and model
comparison.

Keywords: ecological, wildlife, hierarchical, occupancy, occurrence, distance, point count.

1. Introduction

1.1. Imperfect detection and data collection in ecological research

A fundamental goal of ecological research is to understand how environmental variables in-
fluence spatial or temporal variation in species abundance or occurrence. Addressing these
research questions is complicated by imperfect detection of individuals or species. Individuals
may go undetected when present for a variety of reasons including their proximity to the ob-
server, cryptic behavior, or camouflage. Thus, imperfect detection can introduce substantial
measurement error and obscure underlying ecological relationships if ignored.
2 unmarked: Analyze Wildlife Data in R

Season 1 Season 2
Visit 1 Visit 2 Visit 3 Visit 1 Visit 2 Visit 3
Site 1 3 3 2 0 0 0
Site 2 0 0 0 0 0 0
Site 3 4 6 5 5 5 3
Site 4 0 0 0 0 0 2

Table 1: An example of the data structure required by unmarked’s fitting functions. Counts
of organisms were made at M = 4 sites during T = 2 seasons with J = 3 visits (“observations”)
per season.

To accommodate imperfect detection of individuals or species, ecologists have developed spe-

cialized methods to survey wildlife populations such as site occupancy sampling, repeated
counts, distance sampling, removal sampling, and double observer sampling (see Section 2
and Williams et al. 2002 for definitions). This myriad of sampling methods are all unified by
a common repeated-measures type of sampling design in which J “observations” are made at
M spatial sample units during each of T seasons. As an example, Table 1 shows a typical
repeated count dataset.
In addition to the repeated-measures structure, there are two important features to note
about the data shown in Table 1. First, this design is often referred to as a ‘metapopulation
design’ because the population may be regarded as an aggregation of subpopulations and
because the spatial sampling is an explicit component of the problem. This is important
because it provides a basis for modeling variation among sites as a function of site-specific
covariates, and because metapopulation parameters such as local colonization and extinction
can be directly estimated. Second, the counts are of individuals that may not be uniquely
recognized, so although it may be possible to determine if three individuals were detected on
two different visits (as was the case at Site 1 in Table 1), it is not always possible to determine
if they were the same three individuals.
Both the metapopulation design and absence of individual recognition distinguish these
data from data collected using another large suite of ecological sampling methods known
as ‘capture-recapture’ methods. A vast number of models have been developed for capture-
recapture data, and many of them are implemented in the software program MARK (White
and Burnham 1999). The origins of unmarked (the package and its name) arose to fill the void
of models for data from studies of unmarked individuals involving explicit spatial sampling.

1.2. Hierarchical models for metapopulation designs

While many ecological studies produce data consistent with a metapopulation design, the
development and implementation of models for inference under individual sampling protocols
has proceeded in a piecemeal fashion, without a coherent analysis platform. Recently, however,
a broad class of hierarchical models (see Royle and Dorazio 2008 for a general treatment)
has been developed that offer a unified framework for analysis by formally recognizing that
observations are generated by a combination of (1) a state process determining abundance or
species occurrence at each site and (2) a detection process that yields observations conditional
on the state process. The model for the state process describes abundance or occurrence at
each site, but due to imperfect detection, these quantities cannot be observed directly and
are regarded latent variables.
 Journal of Statistical Software 3

This paper introduces unmarked, an R (R Development Core Team 2011) package that pro-
vides a unified approach for fitting this broad class of hierarchical models developed for
sampling biological populations. It is available from the Comprehensive R Archive Network
at [Link] Inference is based on the integrated
likelihood wherein the latent state variable is marginalized out of the conditional likelihood.

1.3. Scope and features of unmarked

unmarked provides tools to assist researchers with every step of the analysis process, includ-
ing data manipulation and exploration, model fitting, post-hoc analysis, model criticism, and
model selection. unmarked provides a growing list of model-fitting functions designed for
specific sampling methods. The fitting functions each find the maximum likelihood estimates
of parameters from a particular model (Section 3.3) and return an object that can be easily
manipulated. Methods exist for performing numerous post-hoc analyses such as requesting
linear combinations of parameters, back-transforming parameters to constrained scales, de-
termining confidence intervals, and evaluating goodness of fit. The model specification syntax
of the fitting functions was designed to resemble the syntax of R’s common fitting functions
such as lm for fitting linear models.
Although there is existing software for fitting some of these models (e.g., Hines and MacKenzie
2002), there are a number of advantages to a unified framework within R. Many researchers
are already familiar with R and use its powerful data manipulation and plotting capabilities.
Sometimes many species are analyzed in tandem, so that a common method of aggregating
and post-processing of results is needed, a task easily accomplished in R. Unlike other available
software, unmarked makes it possible to map habitat-specific abundance and species distribu-
tions when combined with R’s GIS capabilities. Another important advantage of unmarked’s
approach is that researchers can simulate and analyze data within the same computational
environment. This work flow permits simulation studies for power analysis calculations or the
effectiveness of future sampling designs. All of this is made much simpler by analyzing the
data within R, and using a single environment to complete all phases of the analysis is much
less error-prone than switching between applications.
In this paper, Section 2 gives a brief summary of many of the models unmarked is capable of
fitting. Section 3 describes general unmarked usage aided by a running data example.

2. Models implemented in unmarked

The list of models implemented in unmarked continues to grow as new models are developed.
Table 2 shows the models available as of version 0.9-0. Rather than describe each fitting
function in detail, this section provides a summary of several of the most common sampling
techniques and how unmarked can be used to model the resulting data.

2.1. Site occupancy models

Single season site occupancy model

An important state variable in ecological research is species occurrence (or ‘site occupancy’),
say Zi , a binary state variable such that Zi = 1 if site i is occupied by a species and Zi = 0
4 unmarked: Analyze Wildlife Data in R

Model Fitting Data Citation

function
Occupancy from detec- occu unmarkedFrameOccu MacKenzie et al. (2002)
tion/ non-detection data
Abundance from detec- occuRN unmarkedFrameOccu Royle and Nichols (2003)
tion/ non-detection data
Abundance from repeated pcount unmarkedFramePCount Royle (2004b)
counts
Abundance from distance distsamp unmarkedFrameDS Royle et al. (2004)
sampling
Abundance from multino- multinomPois unmarkedFrameMPois Royle (2004a)
mial counts
Abundance from repeated gmultmix unmarkedFrameGMM Chandler et al. (2011)
multinomial counts
Multi-season occupancy colext unmarkedMultFrame MacKenzie et al. (2003)
Multi-season abundance pcountOpen unmarkedFramePCO Dail and Madsen (2011)

Table 2: Models currently handled by unmarked along with their associated fitting functions
(Section 2) and data type (Section 3.1).

otherwise. Much interest is focused on estimating functions of species occurrence (e.g., number
of occupied sites) or to identify factors that are associated with changes in the probability of a
site being occupied, i.e., ψ = Pr(Zi = 1). To estimate these parameters, researchers employ a
sampling design, whereby surveyors visit a sample of M sites and record the binary response
Yij of species detection (Y = 1) or non-detection (Y = 0) during j = 1, . . . , Ji visits to the ith
site during a ‘season’ (MacKenzie et al. 2002). A key feature of species occurrence surveys
is that false absences are possible, so that a species might go undetected (Yij = 0) even if
it is present (Zi = 1). The detection probability parameter p accounts for this observation
process, and is defined as the probability of detecting a species that is present.
The key assumptions made when modeling these data are that the occupancy state at a site
remains constant throughout the season and repeated visits at a site are independent. In
order to meet the first assumption, a season will generally be a very short time frame, such
as a few months during a breeding season, or even a few minutes if repeated visits are made
in quick succession. Replicate samples (i.e., J > 1) provide information about the detection
rate separate from the occupancy rate.
The following hierarchical model describes the joint distribution of the observations condi-
tional on the latent occupancy state, and the marginal distribution of the latent occupancy
state variable:

Zi ∼ Bernoulli(ψ) for i = 1, 2, . . . , M
Yij |Zi ∼ Bernoulli(Zi p) for j = 1, 2, . . . , Ji

Removing the latent Z variables by marginalization yields the likelihood:

 
M Y
Y Ji 
pYij (1 − p)1−Yij ψ + I(Yi. = 0)(1 − ψ)


L(ψ, p| {Yij }) = , (1)
 
i=1 j=1

where I(.) is the indicator function taking the value 1 if its argument is true, and 0 otherwise.
 Journal of Statistical Software 5

The model and likelihood are easily generalizable to accommodate covariates on detection
probability or occupancy. Variables that are related to the occupancy state are modeled as

logit(ψi ) = x>
i β,

where xi is a vector of site-level covariates and β is a vector of their corresponding effect

parameters. Similarly, the probability of detection can be modeled with
>
logit(pij ) = vij α,
where vij is a vector of observation-level covariates and α is a vector of their corresponding
effect parameters. Examples of site-level covariates include habitat characteristics such as
vegetation height. Observation-level covariates could include time of day, date, wind speed,
or other factors that might affect detection probability. The function occu implements this
model.

Multi-season site occupancy model

Sometimes the study objective is to understand the dynamics of the occupancy state over time.
To obtain such information researchers conduct repeated occupancy studies (see above) at the
same sample of sites over consecutive seasons (MacKenzie et al. 2003) and seek to estimate
probabilities of colonization (γ) and extinction (), where colonization is the change of an
unoccupied site to occupied and extinction if the change of an occupied site to unoccupied. If
the occupancy status is assumed to evolve according to a Markov process, then a 2-state finite
hidden Markov model describes these data. Let Yijt denote the observed species occurrence
status at visit j during season t to site i. Then
Zi1 ∼ Bernoulli(ψ)
(
Bernoulli(γt−1 ) if Zi(t−1) = 0
Zit ∼ ,
Bernoulli(1 − t−1 ) if Zi(t−1) = 1
for t = 2, 3, . . . , T
Yijt |Zit ∼ Bernoulli(Zit p)

This model generalizes the single season site occupancy model by relaxing the population
closure assumption. To define the likelihood of this model, let φ0 = (ψ, 1 − ψ) and
QJ !
pI(Yijt = 0)(1 − p)
j=1 P
θYit =
I( Jj=1 Yijt = 0).

Furthermore, let the matrix of one-step Markov transition probabilities be

 
1− 
Φt =
1−γ γ
The likelihood is then given by

M
( (T −1 ) )
Y Y
L(ψ, , γ, p| {Yijt }) = φ0 Φt θYit θYiT , (2)
i=1 t=1
6 unmarked: Analyze Wildlife Data in R

which can be maximized by the colext function in unmarked. Again, each of the four model
parameters ψ, γ, , and p can be modeled as functions of covariates on the logit scale.

2.2. Abundance models

Repeated count data

Occupancy is a crude summary of population structure and dynamics and in practice consid-
erable effort is focused on obtaining information about abundance (Dorazio 2007). A common
sampling design for obtaining abundance information, analogous to the basic occupancy de-
sign described above, is to repeatedly visit a sample of M sites J times and record the number
of unique individuals observed at each site. Similar assumptions are made as with occurrence
data: (1) abundance at a site remains constant during a season and (2) counts at a site are
independent. Royle (2004b) presented the following hierarchical model for repeated count
data. Let Ni be the unobserved total number of individuals using a site and define Yij as the
number of individuals observed during the jth visit. Then,
Ni ∼ f (λ, θ) for i = 1, 2, . . . , M
Yij |Ni ∼ Binomial(Ni , p) for j = 1, 2, . . . , Ji ,
where λi is the abundance rate at site i and p is the detection probability. f is a discrete
distribution with support restricted to Ni ≥ 0 and θ are extra parameters of f other than the
location parameter, λ. unmarked currently supports f as Poisson or negative binomial. In
the negative binomial case, θ is a dispersion parameter, which is useful when overdispersion
is suspected. In the Poisson case, the overdispersion parameter θ is constrained to equal zero.
As with the occupancy model, covariates may be included at either the state (here, abundance)
or detection levels, but abundance is modeled through a log link to enforce its positivity
constraint.
log(λi ) = x>
i β,

where xi is a vector of site-level covariates and β is a vector of their corresponding effect

parameters. Similarly, the probability of detection can be modeled with
>
logit(pij ) = vij α,
where vij is a vector of observation-level covariates and α is a vector of their corresponding
effect parameters.
The Ni variables are latent and so analysis is carried out based on the integrated likelihood
obtained by marginalizing each Ni from the conditional likelihood (Royle et al. 2004):
   
M  ∞ J
Y X Y Ni! e−λ λNi 
L(λ, p| {Yij }) =  pYij (1 − p)Ni −Yij  . (3)
 (Ni − Yij )! Ni ! 
i=1 Ni =max(Yi ) j=1

This likelihood can be maximized using the pcount function.

The function pcountOpen implements a generalized form of this model designed for ‘open’
populations with temporal dynamics governed by apparent survival and recruitment param-
eters (Dail and Madsen 2011). Another related model was described by Royle and Nichols
 Journal of Statistical Software 7

(2003) and can be fit with the occuRN function. This model estimates abundance from site
occupancy data by exploiting the link between abundance and detection probability.

General multinomial-Poisson mixture model

Here we discuss a more general class of models that can be customized by the user for a variety
of sampling methods, the multinomial-Poisson model (Royle 2004a). The general form of this
model is

Ni ∼ Poisson(λ) for i = 1, 2, . . . , M
Yi |Ni ∼ Multinomial (Ni , π)

where Ni is the latent abundance at site i as with the repeated count model, and π =
(π1 , π2 , . . . , πJ )> is the vector of cell probabilities corresponding to the vector of counts Yi .
Examples of sampling methods that produce data of this form include removal sampling,
double observer sampling, and distance sampling. In the next sections, we discuss P these
specific cases. In general, π is determined byP the specific sampling method and j πj ≤ 1
because detection is imperfect and πJ+1 = 1 − j πj is the probability of the species escaping
detection.
Note that the replication here is of a different nature than in previously described models –
not over time necessarily – but still effectively replication as far as the design goes. That is,
there are repeated measurements at each site, but with a multinomial protocol, the replicate
counts are dependent instead of independent.
To illustrate the likelihood under a multinomial observation model, suppose that a sampling
method produces a multinomial observation with 3 observable frequencies at each site, ie
J = 3. As before, Ni are latent variables, and so inference is based on the integrated likelihood
of Yi which is:
 
N  ∞   −λ Ni 
Y X Ni ! Ni −Yi. e λ
L(λ, p| {Yi }) = π1Yi1 π2Yi2 π3Yi3 πJ+1 . (4)
 P Yi1 !Yi2 !Yi3 !Yi0 ! Ni ! 
i=1 Ni = j (Yij )

For this specific case of a multinomial-Poisson mixture, the likelihood simplifies analytically
(to a product-Poisson likelihood). The function multinomPois can be used to maximize this
likelihood.
For multinomial-Poisson sampling methods, the actual observations are an underlying cate-
gorical detection variable with M ≤ J levels so that the J-dimensional Yi is derived from the
M -dimensional raw counts in some sampling method-specific manner. Thus, it is necessary
to model the detection at the raw observation level, denoted pik for k = 1, 2, . . . , M at site i.
Then we derive the multinomial cell probabilities π i through the sampling technique-specific
relationship πij = g(pik ) where pik is the underlying probability of detection and g is some
sampling method-specific function.
Thus, the only two requirements to adapt unmarked’s general multinomial Poisson modeling
to a new sampling method is to specify g and a binary 0/1 matrix that describes the mapping
of elements of pi = (pi1 , . . . , piR )> to elements of π i . This mapping matrix, referred to in
unmarked as obsToY, is necessary to consistently clean missing values from the data and relate
observation-level covariates with the responses. The (j,k)th element of obsToY is 1 only if pik
is involved in the computation of πij . The detection function g is called piFun in unmarked.
8 unmarked: Analyze Wildlife Data in R

Covariates may be included in either the state (here, abundance) or detection models, through
pi (not π i ).
log(λi ) = x>
i β,

where xi is a vector of site-level covariates and β is a vector of their corresponding effect

parameters. Similarly, the probability of detection can be modeled with
>
logit(pij ) = vij α,
where vij is a vector of observation-level covariates and α is a vector of their corresponding
effect parameters.
We now describe two common sampling methods that can be modeled with the multinomial-
Poisson model: removal sampling and double observer sampling. These two methods are
included in unmarked, but additional methods, such as mark-recapture samples, may easily
be specified by defining a user-specified piFun function and obsToY matrix.

Removal sampling Popular in fisheries, removal sampling is commonly implemented by

visiting a sample of M sites J times each and trapping or otherwise capturing and then
removing individuals at each visit. Thus, Yij is the number of individuals captured at the jth
visit for j = 1, 2, . . . , J. We can specify g for removal sampling as follows. The probability of
an individual at site i being captured on the first visit is πi1 = pi1 . The probability of capture
on the jth visit is
j−1
Y
πij = (1 − pik )pij ,
k=1
for j = 2, . . . , J and the probability of not being sampled is
J
X
πi,J+1 = 1 − (πij ).
j=1

Thus, the mapping matrix is an J × J matrix with ones in the upper triangle,
 
1 1 ... 1
0 1 . . . 1
..  .
 
 .. . .
. . .
0 0 ... 1

Double observer sampling Double observer sampling involves collecting data by a team
of two surveyors simultaneously visiting a site. Each observer independently records a list
of detected organisms, and at the end of the survey the two observers attempt to reconcile
their counts. If individuals are not uniquely marked, this may be a difficult task in practice;
however, assuming that individuals can be distinguished, the data at each site are a vector
of length three (Yi ), corresponding to the numbers of individuals seen by only observer one,
only observer two, and both observers. Thus, for double observer sampling, g is defined as
follows  
pi1 (1 − pi2 )
 (1 − pi1 )pi2 
πi =  .
 pi1 pi2 
(1 − pi1 )(1 − pi2 )
 Journal of Statistical Software 9

The obsToY mapping matrix for double observer sampling is the following 3 × 2 matrix
 
1 0
0 1  .
1 1

Distance sampling One of the most widely used sampling methods in animal ecology
is distance sampling (Buckland et al. 2001), which involves recording the distance to each
individual detected at M sites (often referred to as ‘transects’) on a single occasion. Detection
probability is modeled as a function of distance (d) to the observer, for example using the half-
normal detection function p = exp(−d2 /(2σ 2 )) where σ is the half-normal shape parameter.
In practice, the distance measurements are often binned into J distance intervals, which allows
them to modeled as multinomial outcomes with cell probabilities πi computed as the product
of the probability of detection and the probability of occurrence in each distance interval
(Royle et al. 2004).
As currently implemented, the general form of the distance sampling model is identical to the
multinomial-Poisson mixture described above, with the sole difference being that instead of
>
logit(pij ) = vij α, (5)

we have
log(σi ) = vi> α, (6)

Here, the log link is required because σ is a positive shape parameter of the detection func-
tion. In addition, distance sampling data are associated with many unique attributes such as
distance interval cut-points and survey method (line-transect vs point count); therefore, we
created the specialized function distsamp to fit the multinomial-Poisson model to distance
sampling data.
Generalization The flexibility of the multinomPois function is extended even further in
the gmultmix function, which allows for a negative binomial mixture (Dorazio et al. 2005)
and relaxes the population closure assumption (Chandler et al. 2011).

3. unmarked usage
unmarked provides data structures, fitting syntax, and post-processing that form a cohesive
framework for the analysis of ecological data collected using a metapopulation design. In order
to achieve these goals, unmarked uses the S4 class system (Chambers 2008). As R’s most
modern system of class-based programming, S4 allows customization of functions, referred to
as methods, to specific object classes and superclasses. For example, when the generic predict
method is called with any unmarked model fit object as an argument, the actual predict
implementation depends on the specific model that was fit. Use of class-based programming
can provide more reliable and maintainable software while also making the program more
user-friendly (Chambers 2008).

3.1. Preparing data

unmarked uses a custom S4 data structure called the unmarkedFrame to store all data and
metadata related to a sampling design. Although this at first appears to add an extra layer of
10 unmarked: Analyze Wildlife Data in R

work for the user, there are several reasons for this design choice. The multilevel structure of
the models means that standard rectangular data structures such as [Link] or matrices
are not suitable for storing the data. For example, covariates might have been measured
separately at the site level and at the visit level. Furthermore, the length of the response vector
Yi at site i might differ from the number of observations at the site as in the multinomial
Poisson model. In some cases, metadata of arbitrary dimensions may need to be associated
with the data. For example, in distance sampling it is necessary to store the units of measure
and the survey design type. Aside from these technical reasons, Gentleman (2009) pointed
out that the use of such portable custom data objects can simplify future reference to previous
analyses, an often neglected aspect of research. Repeated fitting calls using the same set of
data require less code repetition if all data are contained in a single object. Finally, calls to
fitting functions have a cleaner appearance with a more obvious purpose when the call is not
buried in data arguments.
The parent S4 data class is called an unmarkedFrame and each unmarked fitting function
has its own data type that extends the unmarkedFrame. To ease data importing and con-
version, unmarked includes several helper functions to automatically convert data into an
unmarkedFrame: csvToUMF which imports data directly from a comma-separated value text
file, formatWide and formatLong which convert data from data frames, and the family of
unmarkedFrame constructor functions.
An unmarkedFrame object contains components, referred to as slots, which hold the data
and metadata. All unmarkedFrame objects contain a slot for the observation matrix y, a
[Link] of site-level covariates siteCovs, and a [Link] of observation-level covari-
ates obsCovs. The y matrix is the only required slot. Each row of y contains either the ob-
served counts or detection/non-detection data at each of the M sites. siteCovs is an M -row
[Link] with a column for each site-level covariate. obsCovs is an M J-row [Link]
with a column for each observation-level covariate. Thus each row of obsCovs corresponds to
a particular observation, with the order corresponding to site varying slower and observation
within site varying faster. Both siteCovs and obsCovs can contain NA values corresponding
to unbalanced or missing data. If a site-level covariate is missing, unmarked automatically
removes all data for that site prior to fitting the model. Missing values in the obsCovs are
handled by removing the corresponding occurrence or count observations such that the miss-
ing values make no contribution to the likelihood. unmarked provides constructor functions
to make creating unmarkedFrames straightforward. For each specific data type, specific types
of unmarkedFrames extend the basic unmarkedFrame to handle model-specific nuances.

Importing repeated count data

Here is an example of creating an unmarkedFrame for repeated count data (Section 2.2). First,
load Mallard (Anas platyrhynchos) point count dataset described in Kéry et al. (2005).

R> library("unmarked")
R> data("mallard")

Loading the mallard data makes three objects available within the R workspace. The matrix
mallard.y contains the number of mallards counted at each of M = 239 sites (rows) on J = 3
visits (columns). Counts from the first five sites are shown below:

R> mallard.y[1:5, ]
 Journal of Statistical Software 11

y.1 y.2 y.3

[1,] 0 0 0
[2,] 0 0 0
[3,] 3 2 1
[4,] 0 0 0
[5,] 3 0 3

The site-level covariates are columns of the [Link] [Link], which also has M =
239 rows.

R> [Link][1:5, ]

elev length forest

1 -1.173 0.801 -1.156
2 -1.127 0.115 -0.501
3 -0.198 -0.479 -0.101
4 -0.105 0.315 0.008
5 -1.034 -1.102 -1.193

The site-level covariates are elevation (elev), transect length (length), and the proportion
of forest covering the site (forest).
The observation-level covariates are a list named [Link] with separate M ×J matrices
for each observation-level covariate. Here, the two observation-level covariates are a measure
of survey effort (ivel) and the date of the survey (date). Both have been standardized to a
mean of zero and unit variance.

R> [Link]$ivel[1:5, ]

ivel.1 ivel.2 ivel.3

[1,] -0.506 -0.506 -0.506
[2,] -0.934 -0.991 -1.162
[3,] -1.136 -1.339 -1.610
[4,] -0.819 -0.927 -1.197
[5,] 0.638 0.880 1.042

R> [Link]$date[1:5, ]

date.1 date.2 date.3

[1,] -1.761 0.310 1.381
[2,] -2.904 -1.047 0.596
[3,] -1.690 -0.476 1.453
[4,] -2.190 -0.690 1.239
[5,] -1.833 0.167 1.381

The unmarkedFrame constructors can accept obsCovs in this list format or as a [Link]
in the format described in Section 3.1.
The following call to unmarkedFramePCount organizes the observations and covariates into an
object that can be passed to the data argument of the fitting function pcount.
12 unmarked: Analyze Wildlife Data in R

R> mallardUMF <- unmarkedFramePCount(y = mallard.y, siteCovs = [Link],

+ obsCovs = [Link])
R> summary(mallardUMF)

unmarkedFrame Object

239 sites
Maximum number of observations per site: 3
Mean number of observations per site: 2.76
Sites with at least one detection: 40

Tabulation of y observations:
0 1 2 3 4 7 10 12 <NA>
576 54 11 9 6 1 1 1 58

Site-level covariates:
elev length forest
Min. :-1.436e+00 Min. :-4.945e+00 Min. :-1.265e+00
1st Qu.:-9.565e-01 1st Qu.:-5.630e-01 1st Qu.:-9.560e-01
Median :-1.980e-01 Median : 4.500e-02 Median :-6.500e-02
Mean :-4.603e-05 Mean :-2.929e-05 Mean : 6.695e-05
3rd Qu.: 9.940e-01 3rd Qu.: 6.260e-01 3rd Qu.: 7.900e-01
Max. : 2.434e+00 Max. : 2.255e+00 Max. : 2.299e+00

Observation-level covariates:
ivel date
Min. :-1.753e+00 Min. :-2.904e+00
1st Qu.:-6.660e-01 1st Qu.:-1.119e+00
Median :-1.390e-01 Median :-1.190e-01
Mean : 1.504e-05 Mean : 7.259e-05
3rd Qu.: 5.490e-01 3rd Qu.: 1.310e+00
Max. : 5.980e+00 Max. : 3.810e+00
NA's : 5.200e+01 NA's : 4.200e+01

The summary reveals that only 40 sites have at least one detection. Note that the “number
of observations” in this case refers to number of visits. The term “observation” is used rather
than “visit” because the definition of the J replicate surveys depends upon the sampling
method. The tabulation of y observations provides additional evidence of sparse counts, with
no mallards being detected during 576 of the surveys. The 58 NA values correspond to missing
data.

Importing removal sampling data

To illustrate the slightly different syntax for removal sampling data example, we will import
data from a removal survey of Ovenbirds (Seiurus aurocapillus) described by Royle (2004a).
The data consist of a list named [Link] with a matrix named data containing the re-
moval counts for 4 visits and a [Link] called covariates containing site-level covariates
information.
 Journal of Statistical Software 13

R> data("ovendata")

This loads a list named [Link] with components for the observed count data and
covariates. The only additional specification required when creating an unmarkedFrame for
the multinomial-Poisson model is to specify the particular type of data as either removal,
double, or userDefined.

R> ovenFrame <- unmarkedFrameMPois([Link]$data,

+ siteCovs = [Link]$covariates, type = "removal")
R> ovenFrame[1:5, ]

Data frame representation of unmarkedFrame object.

y.1 y.2 y.3 y.4 site ufp trba
1 0 0 0 0 CAT003 23.43 62.5
2 1 0 0 0 CAT004 15.62 70.0
3 0 0 0 0 CAT011 35.54 90.0
4 0 0 0 0 CAT013 25.78 60.0
5 0 0 0 0 CAT017 25.00 122.5

Notice the bracket method of subsetting used to display data from the first five sites.

Preparing multi-season occupancy data

The data structures are more complex when surveys were conducted over more than one
season. The following example demonstrates how to prepare a toy dataset for the colext
fitting function. This detection/non-detection dataset has dimensions M = 4 sites, T = 2
seasons, and J = 3 visits per season.

R> y

season1v1 season1v2 season1v3 season2v1 season2v2 season2v3

site1 0 0 0 0 0 0
site2 1 1 1 1 1 1
site3 0 0 0 0 0 0
site4 1 1 1 1 1 1

R> umf <- unmarkedMultFrame(y = y, numPrimary = 2)

The function unmarkedMultFrame accepts siteCovs and obsCovs like all other unmarked-
Frame classes, and it also has an argument yearlySiteCovs that can accept a list of M × T
[Link] containing season-level covariates. The numPrimary argument specifies the num-
ber of seasons.

3.2. Manipulating unmarkedFrames

The various components of unmarkedFrames can be extracted and subsetted in a manner sim-
ilar to the methods used to manipulate standard R objects. Subsetting can be accomplished
using the ‘bracket’ notation. For example, the first five rows of data and all three removal
occasions can be extracted from the Ovenbird removal study using
14 unmarked: Analyze Wildlife Data in R

R> ovenFrame[1:5, 1:3]

Data frame representation of unmarkedFrame object.

y.1 y.2 y.3 site ufp trba
1 0 0 0 CAT003 23.43 62.5
2 1 0 0 CAT004 15.62 70.0
3 0 0 0 CAT011 35.54 90.0
4 0 0 0 CAT013 25.78 60.0
5 0 0 0 CAT017 25.00 122.5

In some cases, the covariate data may need to be manipulated after the unmarkedFrame
has been created. The following code demonstrates how to extract the site-level covariates,
standardize those that are continuous (columns 2 and 3), and reinsert them back into the
unmarkedFrame.

R> sc <- siteCovs(ovenFrame)

R> sc[, 2:3] <- scale(sc[, 2:3])
R> siteCovs(ovenFrame) <- sc

3.3. Fitting models

As introduced in Section 2, each type of survey design has a corresponding fitting function.
For example, to fit a repeated count model, we call pcount. Table 2 provides a summary
of all models that unmarked currently fits. With the exception of the ‘open’ population
models (colext, gmultmix, and pcountOpen), all fitting functions use a double right-hand
sided formula syntax representing the hierarchical model structure. Specifically, covariates
affecting the detection process are specified following the first tilde, and covariates of the
state process follow the second tilde. No left-hand side of the formula is specified because the
unmarkedFrame defines the response variable uniquely as the y slot.

Fitting a repeated count model

Continuing the Mallard example, the following call to pcount fits a binomial-Poisson mixture
model (Section 2.2). The following code specifies that detection probability p should be
modeled by day of year, including a quadratic term. We also wish to model abundance using
elevation and proportion of area forested. As described in Section 2.2, covariates of detection
are on the logit-scale and covariates of abundance are on the log-scale for the repeated count
model.

R> [Link].1 <- pcount(~ date + I(date^2) ~ elev + forest,

+ data = mallardUMF, K = 50)
R> [Link].1

This initial fit suggests that Mallard abundance decreases with increasing elevation and forest.
It also looks like a linear model might suffice for the detection model, so we subsequently fit
the linear detection model as follows:
 Journal of Statistical Software 15

R> [Link].2 <- pcount(~ date ~ elev + forest, data = mallardUMF,

+ K = 50)
R> [Link].2

This seems to be a better model according to both the Wald p value and AIC. The result
suggests that detection probability decreases during the course of a year.

Fitting a multinomial-Poisson model

Here we demonstrate fitting a multinomial-Poisson mixture model to removal sampling data.
The Ovenbird data has no observation-level covariates, so detection probability is assumed
constant across visits. It is not necessary to specify that removal sampling was used when
fitting the model because this information is already stored in the ovenFrame data. We
model abundance as a function of understory forest coverage (ufp) and average basal tree
area (trba).

R> [Link].1 <- multinomPois(~ 1 ~ ufp + trba, ovenFrame)

R> [Link].1

3.4. Examining fitted models

Objects returned by unmarked’s fitting functions also make use of the S4 class system. All
fitted model objects belong to the unmarkedFit parent class. Thus, common operations such
as extracting coefficient estimates, covariance matrices for estimates, and confidence intervals
have been adapted to behave similar to R’s base functions.
For example, we can extract estimated coefficients either from the entire model, or from the
state or detection levels by specifying the type argument.

R> coef([Link].2)

lam(Int) lam(elev) lam(forest) p(Int) p(date)

-1.8565048 -1.2355497 -0.7555178 0.2977558 -0.4006066

R> coef([Link].2, type = "state")

lam(Int) lam(elev) lam(forest)

-1.8565048 -1.2355497 -0.7555178

To check which types are available for a model, use the names method.

R> names([Link].2)

[1] "state" "det"

Similarly, the vcov function extracts the covariance matrix of the estimates, using the ob-
served Fisher information by default. vcov also accepts a type argument, as does the conve-
nience method SE, which returns standard errors from the square root of the diagonal of the
covariance matrix.
16 unmarked: Analyze Wildlife Data in R

R> vcov([Link].2, type = "det")

p(Int) p(date)
p(Int) 0.03549024 0.00344853
p(date) 0.00344853 0.01302865

Extracting confidence intervals proceeds in a similar fashion. By default, the asymptotic

normal approximation is used.

R> confint([Link].1, type = "state", level = 0.95)

0.025 0.975
lambda(Int) -0.1299722 0.33470834
lambda(ufp) -0.1471741 0.34776195
lambda(trba) -0.4361311 0.09440937

Profile confidence intervals are also available upon request. This can take some time, however,
because for each parameter, a nested optimization within a root-finding algorithm is being
used to find the profile limit.

R> ci <- confint([Link].1, type = "state", level = 0.95, method = "profile")

R> ci

0.025 0.975
lambda(Int) -0.1390786 0.32676614
lambda(ufp) -0.1477724 0.34811770
lambda(trba) -0.4368444 0.09469605

The profile confidence intervals and normal approximations are quite similar here.

The nonparametric bootstrap

Nonparametric bootstrapping can also be used to estimate the covariance matrix. unmarked
implements a two-stage bootstrap in which the sites are first drawn with replacement, and then
within each site, the observations are drawn with replacement. First, bootstrap draws must be
taken using the nonparboot function, which returns a new version of the unmarkedFit object
with additional bootstrap sampling information. Thus, this new fit must be stored, either in
a new fit object or the same one, and then subsequently queried for bootstrap summaries.
To illustrate, we use the removal sampling data instead of the repeated count data because
computations are much faster; however, bootstrapping is available for any of the models in
unmarked.

R> [Link](1234)
R> [Link].1 <- nonparboot([Link].1, B = 100)
R> SE([Link].1, type = "state")

lambda(Int) lambda(ufp) lambda(trba)

0.1185431 0.1262615 0.1353444
 Journal of Statistical Software 17

R> SE([Link].1, type = "state", method = "nonparboot")

lambda(Int) lambda(ufp) lambda(trba)

0.1256949 0.1156153 0.1197142

The bootstrapping and asymptotic standard errors are similar. Additional bootstrap samples
can be drawn by calling nonparboot again.

R> [Link].1 <- nonparboot([Link].1, B = 100)

Linear combinations of estimates

Often, meaningful hypotheses can be addressed by estimating linear combinations of coeffi-
cient estimates. Linear combinations of coefficient estimates can be requested with linearComb.
Continuing the Ovenbird example, the following code estimates the log-abundance rate for a
site with ufp = 0.5 and trba= 0.

R> (lc <- linearComb([Link].1, type = "state", coefficients = c(1, 0.5, 0)))

Linear combination(s) of Abundance estimate(s)

Estimate SE (Intercept) ufp trba

0.153 0.13 1 0.5 0

Multiple sets of coefficients may be supplied as a design matrix. The following code requests
the estimated log-abundance for sites with ufp = 0.5 and trba = 1.

R> (lc <- linearComb([Link].1, type = "state",

+ coefficients = matrix(c(1, 0.5, 0, 1, 1, 0), 2, 3, byrow = TRUE)))

Linear combination(s) of Abundance estimate(s)

Estimate SE (Intercept) ufp trba

1 0.153 0.130 1 0.5 0
2 0.203 0.166 1 1.0 0

Standard errors and confidence intervals are also available for linear combinations of parame-
ters. By requesting nonparametric bootstrapped standard errors, unmarked uses the samples
that were drawn earlier.

R> SE(lc, method = "nonparboot")

[1] 0.1354092 0.1659877

Back-transforming linear combinations of coefficients

Estimates of linear combinations back-transformed to the native scale are likely to be more
interesting than the direct linear combinations. For example, the logistic transformation is
18 unmarked: Analyze Wildlife Data in R

applied to estimates of detection rates, resulting in a probability bound between 0 and 1. This
is accomplished with the backTransform. Standard errors of back-transformed estimates are
estimated using the delta method. Confidence intervals are estimated by back-transforming
the confidence interval of the original linear combination.

R> (btlc <- backTransform(lc))

Backtransformed linear combination(s) of Abundance estimate(s)

Estimate SE LinComb (Intercept) ufp trba

1 1.16 0.151 0.153 1 0.5 0
2 1.22 0.203 0.203 1 1.0 0

Transformation: exp

R> SE(btlc)

[1] 0.1510141 0.2032272

R> confint(btlc)

0.025 0.975
1 0.9033915 1.501747
2 0.8846296 1.695386

3.5. Model selection

unmarked performs AIC-based model selection for structured lists of unmarkedFit objects. To
demonstrate, we fit a few more models to the Ovenbird removal data, including an interaction
model, two models with single predictors, and a model with no predictors.

R> [Link].2 <- update([Link].1, formula = ~ 1 ~ ufp * trba)

R> [Link].3 <- update([Link].1, formula = ~ 1 ~ ufp)
R> [Link].4 <- update([Link].1, formula = ~ 1 ~ trba)
R> [Link].5 <- update([Link].1, formula = ~ 1 ~ 1)
R> preddata <- predict([Link].4, type = "state", appendData = TRUE)
R> library("ggplot2")
R> qplot(trba, Predicted, data = preddata, geom = "line",
+ xlab = "Scaled Basal Tree Area", ylab = "Estimated Abundance") +
+ geom_ribbon(aes(x = trba, ymin = lower, ymax = upper), alpha = 0.1) +
+ theme_bw()

It looks like the best model includes only tree basal area as a predictor of abundance. We
can examine this relationship using the predict method and the ggplot2 package (Wickham
2009), see Figure 1.
Next, we organize the fitted models with the fitList function and the use the modSel method
to rank the models by AIC.
 Journal of Statistical Software 19

2.5

2.0
Estimated Abundance

1.5

1.0

0.5

−2 −1 0 1 2
Scaled Basal Tree Area

Figure 1: Examine estimated abundance for Ovenbird removal data. Band is 95% confidence
interval.

R> fmList <- fitList(`lam(ufp+trba)p(.)` = [Link].1,

+ `lam(ufp*trba)p(.)` = [Link].2, `lam(ufp)p(.)` = [Link].3,
+ `lam(trba)p(.)` = [Link].4, `lam(.)p(.)` = [Link].5)
R> modSel(fmList)

nPars AIC delta AICwt cumltvWt

lam(trba)p(.) 3 324.77 0.00 0.35 0.35
lam(ufp)p(.) 3 325.73 0.96 0.21 0.56
lam(ufp+trba)p(.) 4 326.14 1.37 0.17 0.73
lam(.)p(.) 2 326.28 1.51 0.16 0.90
lam(ufp*trba)p(.) 5 327.17 2.40 0.10 1.00

predict functions much like linearComb except that new data can be passed to it as a
[Link] rather than a design matrix. When the first argument given to predict is a list
of models created by fitList, predict computes model-averaged predictions, which may be
useful in the presence of high model selection uncertainty.

3.6. Goodness of fit and the parametric bootstrap

To conduct goodness of fit tests, unmarked provides a generic parametric bootstrapping
function parboot. It simulates data from the fitted model and applies a user-defined function
that returns a fit-statistic such as the Pearson’s χ2 .
20 unmarked: Analyze Wildlife Data in R

60
50
40
Frequency

30
20
10
0

200 250 300 350 400

χ2

Figure 2: Graphical assessment of model fit by parametric bootstrapping. The dashed line
is the observed chi-squared statistic. The histogram approximates the expected sampling
distribution.

R> [Link](1234)
R> chisq <- function(fm) {
+ observed <- getY(fm@data)
+ expected <- fitted(fm)
+ sum((observed - expected)^2/expected)
+ }
R> pb <- parboot([Link].1, statistic = chisq, nsim = 200)
R> plot(pb, main = "")

The above call to plot with a parametric bootstrap object as the argument produces a useful
graphic for assessing goodness of fit (Figure 2). The plot suggests that the model adequately
explains these data.
Beyond serving as a tool to evaluate goodness of fit, parboot can be used to characterize
uncertainty in any derived quantity of interest.

4. Future directions for unmarked development

unmarked has become a stable and useful platform for the analysis of ecological data, but
several areas of development could improve its utility. First, new models need to be added
to cover the range of sampling techniques and population dynamics commonly encountered.
 Journal of Statistical Software 21

Sampling method Population dynamics

Closed Open to Open to
temporary demographic
emigration processess
Occurrence sampling occu — colext
Repeated counts pcount — pcountOpen
Removal sampling, multinomPois gmultmix —
double observer sampling,
and other multinomial designs
Distance sampling distsamp — —

Table 3: Model fitting functions classified by sampling method and population dynamics.
Missing cells indicate models that have not been developed but are likely to be investigated
in the future.

Table 3 illustrates some of the gaps that need to be filled. In most cases, models to fill these
gaps have not been developed so more research is needed.
Second, each of the models in unmarked assumes independence among sites. However, ecol-
ogists often use sampling methods such as cluster sampling that induce spatial dependence.
Typically, this is done for logistical convenience, but because few methods are available to
account for spatial correlation and imperfect detection probability, the spatial dependence is
often ignored. Rather than this being a weakness of the sampling design, we envision that
this dependence can be used as information regarding the spatial distribution of individuals.
Third, many of the likelihoods are written in pure R, which can be slow for large problems.
We are currently translating many of these functions into C++ with the help of the R package
Rcpp (Eddelbuettel and François 2011).
Finally, Markov chain Monte Carlo methods could be implemented for all of these models
allowing for Bayesian inference. An important advantage of Bayesian analysis over classical
methods is that the latent abundance or occurrence variables can be treated as formal pa-
rameters. Thus posterior distributions could easily be calculated for derived parameters such
as the proportion of sites occupied. Bayesian analysis also would provide a natural frame-
work for incorporating additional sources of random variation. For example, one could model
heterogeneity among sites not accounted for by covariates alone.

Acknowledgments
The authors thank Andy Royle of the United States Geological Survey’s Patuxent Wildlife
Research Center, who provided initial funding and inspiration for this work. Andy Royle and
Robert Dorazio made valuable suggestions that greatly improved an earlier version of the
manuscript.

References

Buckland ST, Anderson DR, Burnham KP, Laake JL, Borchers DL, Thomas L (2001). Intro-
duction to Distance Sampling. Oxford University Press, New York.
22 unmarked: Analyze Wildlife Data in R

Chambers JM (2008). Software for Data Analysis: Programming with R. Springer-Verlag,

New York.

Chandler RB, Royle JA, King DI (2011). “Inference about Density and Temporary Emigration
in Unmarked Populations.” Ecology, 92(7).

Dail D, Madsen L (2011). “Models for Estimating Abundance from Repeated Counts of an
Open Metapopulation.” Biometrics, 67(2), 577–587.

Dorazio RM (2007). “On the Choice of Statistical Models for Estimating Occurrence and
Extinction from Animal Surveys.” Ecology, 88(11), 2773–2782.

Dorazio RM, Jelks HL, Jordan F (2005). “Improving Removal-Based Estimates of Abundance
by Sampling a Population of Spatially Distinct Subpopulations.” Biometrics, 61(4), 1093–
1101.

Eddelbuettel D, François R (2011). “Rcpp: Seamless R and C++ Integration.” Journal of

Statistical Software, 40(8), 1–18. URL [Link]

Gentleman R (2009). R Programming for Bioinformatics. Chapman & Hall/CRC, New York.

Hines JE, MacKenzie DI (2002). PRESENCE. USGS-PWRC, Laurel. URL [Link]

[Link]/software/[Link].

Kéry M, Royle JA, Schmid H (2005). “Modeling Avian Abundance from Replicated Counts
Using Binomial Mixture Models.” Ecological Applications, 15(4), 1450–1461.

MacKenzie DI, Nichols JD, Hines JE, Knutson MG, Franklin AB (2003). “Estimating Site
Occupancy, Colonization, and Local Extinction when a Species Is Detected Imperfectly.”
Ecology, 84(8), 2200–2207.

MacKenzie DI, Nichols JD, Lachman GB, Droege S, Royle JA, Langtimm CA (2002). “Es-
timating Site Occupancy Rates when Detection Probabilities Are less than One.” Ecology,
83(8), 2248–2255.

R Development Core Team (2011). R: A Language and Environment for Statistical Computing.
R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http:
//[Link]/.

Royle JA (2004a). “Generalized Estimators of Avian Abundance from Count Survey Data.”
Animal Biodiversity and Conservation, 27(1), 375–386.

Royle JA (2004b). “N -Mixture Models for Estimating Population Size from Spatially Repli-
cated Counts.” Biometrics, 60(1), 108–115.

Royle JA, Dawson DK, Bates S (2004). “Modeling Abundance Effects in Distance Sampling.”
Ecology, 85(6), 1591–1597.

Royle JA, Dorazio RM (2008). Hierarchical Modeling and Inference in Ecology. Academic
Press, London.

Royle JA, Nichols JD (2003). “Estimating Abundance From Repeated Presence-Absence Data
or Point Counts.” Ecology, 84(3), 777–790.
 Journal of Statistical Software 23

White GC, Burnham KP (1999). “Program MARK: Survival Estimation from Populations of
Marked Animals.” Bird Study, 46(001), 120–138.

Wickham H (2009). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag, New
York.

Williams BK, Nichols JD, Conroy MJ (2002). Analysis and Management of Animal Popula-
tions. Academic Press, San Diego.

Affiliation:
Ian Fiske
Department of Statistics
North Carolina State University
2311 Stinson Drive
Campus Box 8203
Raleigh, NC 27695-8203, United States of America
E-mail: ijfiske@[Link]

Richard Chandler
USGS Patuxent Wildlife Research Center
Gabrielson Lab, Room 226
12100 Beech Forest Rd.
Laurel, MD 20708, United States of America
E-mail: rchandler@[Link]

Journal of Statistical Software [Link]

published by the American Statistical Association [Link]
Volume 43, Issue 10 Submitted: 2010-05-06
August 2011 Accepted: 2011-07-29