w o r k i n g p a p e r n o .

41 APRI L 2 0 1 2

A Decision Tree for Monitoring Wildlife to Assess

the Effectiveness of Conservation Interventions
Samantha Strindberg & Tim O’Brien
WORKING PAPER NO. 41
APRIL 2012

A Decision Tree for Monitoring

Wildlife to Assess the Effectiveness
of Conservation Interventions

By Samantha Strindberg & Tim O'Brien

Wildlife Conservation Society

Global Conservation Program
2300 Southern Boulevard
Bronx, NY 10460

This projected was funded in part by USAID, the Laikipia Wildlife Forum, the Mpala Research Centre, and the Wildlife
Conservation Society. We thank the WCS Latin America and Caribbean Program staff who attended the August 2009 work-
shop in Peru for interesting discussions on monitoring and for insights into their wildlife survey work. We thank Margaret
Kinnaird for many insights on monitoring, and Alfred DeGemmis for assistance in layout and editing of this document.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions i
 WCS Working Papers: ISSN 1530-4426
Online posting: ISSN 1534-7389

Cover images: Front Cover: Babirusa at salt lick, North Sulawesi © M. Kinnaird. Back
Cover: Leopard photographed during carnivore survey, Mpala Ranch, Kenya © M.
Kinnaird/T. O'Brien.

Copyright:
The contents of this paper are the sole property of the authors and cannot be
reproduced without permission of the authors.

The Wildlife Conservation Society saves wildlife and wild places worldwide. We
do so through science, global conservation, education, and the management
of the world’s largest system of urban wildlife parks, led by the flagship Bronx
Zoo. Together these activities change attitudes towards nature and help people
imagine wildlife and humans living in harmony. WCS is committed to this mission
because it is essential to the integrity of life on Earth.

Over the past century, WCS has grown and diversified to include four zoos, an
aquarium, over 100 field conservation projects, local and international educa-
tion programs, and a wildlife health program. The WCS Working Paper Series is
designed to share with the conservation and development communities in a timely
fashion information from the various settings where WCS works. These Papers
address issues that are of immediate importance to helping conserve wildlife and
wild lands either through offering new data or analyses relevant to specific conser-
vation settings, or through offering new methods, approaches, or perspectives on
rapidly evolving conservation issues. The findings, interpretations, and conclusions
expressed in the Papers are those of the author(s) and do not necessarily reflect
the views of the Wildlife Conservation Society. For a complete list of WCS Working
Papers, please see the end of this publication.

ii Wildlife Conservation Society | WORKING PAPER NO. 41

TABLE OF CONTENTS
Basic Monitoring Considerations …………………………………………………….. 1

The General Sampling Framework ...………………………………………………….3

Designing Monitoring Programs ................………………………………………… 3

Power Analysis ….............................................…………………………………….. 5

Decision Tree for Monitoring Wildlife …...…………………………………………..6

Occupancy ..............................................…………………………………………….. 8

Species Richness …………………………….....................................……………… 10

Standardization of Monitoring Methods …………………………………………… 11

Conclusions ……..............................................……………………………………… 12

Table 1: Applications of Monitoring Methods at Field Sites…......……….. 13

Appendix 1: Frequently Used Survey Techniques………….……………………. 14

Appendix 2: Literature and Internet Resources..............................……….. 18
Appendix 3: Monitoring Applications at WCS Sites..........................…...... 25

WCS Working Paper Series …………………………………………………………….. 27

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions iii
A Decision Tree for Monitoring
Wildlife to Assess the
Effectiveness of Conservation
Interventions

Basic Monitoring Considerations

Worldwide, biodiversity is being lost at a rate comparable in magnitude only to

a handful of cataclysmic mass extinction events in the Earth’s geological history.
Loss of biodiversity has major implications for ecosystem health and function, pro-
vision of goods and services, and the impoverishment of quality of life. Biodiversity
loss can be thought of as the sum of decline and loss of many individual species.
Stemming biodiversity loss, therefore, requires that we reduce the decline and loss
of individual species and communities through effective interventions and man-
agement. This can be accomplished through better conservation management of
species, habitats and ecosystems.
Monitoring is a crucial component of good conservation management (Salafsky
et al., 2001; Open Standards for the Practice of Conservation, 2007). It allows us
to assess whether or not threats are decreasing, and/or wildlife populations are
increasing or remaining stable. It requires that we identify the most important
threats, where they occur within the landscape or seascape of interest, and how
they change over time. Through monitoring we can test our assumptions as to
whether our interventions actually lead to what we want to achieve, or whether
they are wasted efforts.
Monitoring tracks progress over time towards a clearly defined objective. We
can only monitor if we have a clear idea of what we hope to achieve, thus setting
explicit objectives lies at the core of effective monitoring. Monitoring assumes suf-
ficient knowledge of the system of interest to allow us to set explicit objectives in
contrast to research that gathers information about the unknown.
Ideally we would want to monitor the conservation strategies (also referred to
as interventions or activities), the threats and the conservation targets themselves
to get the most information about the effectiveness of our actions. We would mon-
itor our strategies to make sure that they are being implemented as we planned
(e.g., Are trained guards getting out on patrol?). Since our strategies are chosen
to reduce levels of threat to wildlife and their habitat, we monitor our success in
reducing threats to assess whether or not our interventions were worthwhile (e.g.,
Is there a reduction in the number of arms and cartridge shells in the area being
patrolled?). Lastly, we look at the status of the wildlife species or habitat that form
our conservation targets to see whether it improves when our interventions are

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 1
 implemented successfully, and threats are reduced (e.g., Are elephant populations
doing better due to the reduction of poaching with firearms?).
The improved state of our conservation targets is the ultimate indicator of suc-
cess. Knowing that state gives us the greatest level of confidence in our interven-
tions. Yet, this level of monitoring is often the most difficult to implement, costs the
most, and may have longer lag-times (see Figure 1). If we monitor the strategies
and threat reductions as proxies for our progress there are definite tradeoffs. The
time frame for seeing results and the costs of monitoring generally declines as we
move from directly monitoring changes in wildlife and their habitats, to monitor-
ing reductions in threats, to monitoring whether or not our strategies were imple-
mented as planned. However, using these proxies that change within a shorter
time frame also lowers our level of confidence in our actual conservation success
(Wilkie et al., 2002 and 2006).
As we will see in a subsequent section, even if we decide to monitor the conser-
vation target directly, different types of indicators can be chosen for this and can
give different results with varying associated levels of confidence in those results.
For the remainder of this document we focus on monitoring our conservation
targets, specifically wildlife, rather than monitoring the threats or interventions,
although the techniques that will be covered can readily be applied to some forms
of threat monitoring.

Figure 1: The relationship between confidence in monitoring results, monitoring cost and time to
see impact of management activities for the different components that could be monitored over time.
Monitoring strategies, threats or conservation targets are frequently referred to as measuring our
outputs, outcomes and impacts, respectively.

Monitoring tracks changes over time and/or space and this distinguishes it from
a sample survey, which estimates conditions at a single point in time or space. Thus
monitoring uses survey results at many instances in time/space. The next section
considers a general sampling framework upon which the monitoring results are
built.

2 Wildlife Conservation Society | WORKING PAPER NO. 41

The General Sampling Framework

Usually, the geographic areas of interest (study sites, landscapes) for monitor-
ing wildlife are large and difficult to access. Thus when designing a survey we will
seldom be able to cover the entire area of interest, but instead select a manage-
able sub-region. Within that sub-region referred to as the survey area, we usually
attempt to cover the entire area or we select sampling units.
If E(C) is the expected value of the count statistic C (number of animals counted
or number of occupied sampling units observed) and p is the detection probability,
then the relationship between the count statistic and the true population size or
occupancy N is given by:

(1)

When detection is 100% (p =1), the count statistic provides an accurate esti-
mate of N. However, when p < 1 the count statistic provides a biased estimate of
N. For example, if 10 animals were observed and in fact p = 1/2 then half of the 20
animals in the survey area were missed. Once the detection probability has been
estimated, then the estimate of abundance or occupancy can be obtained from
count statistics as follows:

(2)

Note that the hats (^) indicate estimated parameters. The equation is general-
ized as follows to incorporate the proportion of the survey area covered :

(3)

This formulation is known as the canonical estimator. The various methods used
to estimate abundance, density, occupancy, and species richness can be expressed
in terms of the canonical estimator (Williams et al., 2002).

Designing Monitoring Programs

Yoccoz et al. (2001) emphasize the need to pay attention to three basic ques-
tions when developing monitoring programs: (1) Why monitor? (2) What should
be monitored? and (3) How should monitoring be carried out? With respect to
'why monitor,' programs to monitor species arise for a number of reasons and at a
number of spatial scales. Species conservation can occur at the site (population),
landscape/seascape (metapopulation) or global range. Once the key threats to
the species have been identified and the conservation activities planned, then the
monitoring program to assess the effectiveness of the interventions can be put in
place. The important part of planning a species monitoring program is to have a

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 3
 clear idea a priori of the objectives of the monitoring program. Before formulating
a monitoring program, or in parallel to improve its formulation over time, research
may be needed to gain better scientific understanding of the ecological and
human-influence factors that affect state variables (density, occupancy), vital rates,
or some combination. When direct interventions are applied to address threats to
a species’ persistence, it is possible to gain insights from monitoring data when a
priori hypotheses are used to make comparisons among alternatives. Combining
monitoring with management interventions may yield information about the cur-
rent population status and the impact of management activities.
‘What to monitor’ follows from the monitoring program objectives. Objectives
should focus on state variables and rate parameters that characterize the system
dynamics (Williams et al., 2002). In species monitoring, the state variable may
include abundance, density or occupancy. In biodiversity monitoring, the state
variable can be a measure of species richness, or some combination of ‘abundance
and diversity’ (Magurran, 2004). The rate parameters may be birth, death, immigra-
tion, emigration, extinction and colonization. Abundance can be measured directly
(an estimate of numbers of animals or the biomass of the species), or indirectly (a
measure of occupancy for a species). In addition, it can be measured by means of
information collected on the animals themselves or on their sign. For communities,
often it is desirable to include some measure of abundance/biomass/occupancy in
the monitoring metric. This increases the complexity of the monitoring program
but provides better information on the tradeoffs between species richness, species
abundance and species evenness, and a better understanding of system function.
In general, a monitoring program’s design and field implementation details will
depend on the choice of conservation target and the selected monitoring metric.
‘How to monitor’ should follow best practices for sampling. There is a large
literature on species and community-level monitoring. Much of this literature
is devoted to the ‘How’ question and the merits of indices requiring calibration
versus estimators of absolute abundance. The ideal monitoring program would
account for variation in detectability across individuals, over time, and across space
(Pollock et al., 2002; Moore and Kendall, 2004; Buckland et al., 2005). It would also
account for spatial variation and survey error. Accounting for variation in detection
is normally done by estimating the detection probability (may also be referred to
as a sighting or capture probability) for a population of individuals at a time and
at a site, and correcting the count C (number of observed individuals, number of
observed occupied sites, number of observed species) by the estimate of detection
probability, p, as described above.
The ease with which counts can be obtained and p estimated varies widely for
state variables of abundance, biomass, occupancy, and species richness. Usually,
it will be easier to collect data on occupancy and species richness than on abun-
dance and biomass when working with mammals, birds, herptiles and fish. There is
a temptation to use the counts directly as indices of the variable of interest under
the assumption that detection probabilities are either equal or are constant over
space and time (Conroy, 1996). This is usually not a good idea. For example, when
monitoring abundance over time, let measure the rate of change in population
size between time (or space) i and time j. is calculated as the ratio of abundance,
Nj /Ni . The counts Ci and Cj , at times i and j, are used as indices of abundance
and is estimated as:

4 Wildlife Conservation Society | WORKING PAPER NO. 41

 (4)

The expected value of is estimated as:

(5)

where the expected value of the counts is equal to the product of abundance and
detection probability. If detection probabilities remained constant across space
and time then the use of a count statistic is justifiable as a proxy for changes in
the parameter being monitored, because the count would be expected to track
changes in that parameter. For example if abundance increases, then the count
also increases and similarly a decline in abundance is reflected by a decline in the
count. Unfortunately, detection probabilities are seldom constant in space and
time and thus need to be estimated to enable reliable trend estimation from the
raw counts. Without an estimate of the detection probability, it is usually impos-
sible to interpret due to the unpredictable and unknown fluctuations in the
relationship between C and N. An index based on counts only may have a smaller
variance than the corresponding unbiased abundance estimate incorporating
detectability, which is desirable as this makes it easier to detect a trend (see next
section on Power Analysis for other factors that impact one’s ability to detect a
trend). However, the gain in precision is offset by the unpredictable loss of accu-
racy. It is best to avoid precise metrics with unknown bias. Thus when designing a
monitoring program we recommend first selecting unbiased metrics facilitating a
reliable interpretation of trends and then focusing on improving precision.

Power Analysis

The ability of a monitoring program to detect a real effect (or a response to

the management strategy) when it exists is called the power of the sampling pro-
gram and analysis. Power increases with increasing sample size, and increasing
size of the effect or response. Power decreases as the variance and standard error
increases. Power analysis is most useful when planning a study or monitoring pro-
gram. Power analysis can be used to explore the relationship between the range
of possible sample sizes, response sizes that are important, levels of variance that
are expected to occur (usually from literature or pilot data), and the desired level of
statistical power. The goal is to be able to design a monitoring program (the sam-
pling) that will detect the effect or response with sufficient sensitivity to be used
as a basis for management decisions. Low power in a monitoring program means
high uncertainty in interpreting the data. Unfortunately, power analyses are rarely
conducted prior to setting up a monitoring design.
There are ongoing debates regarding appropriate methods for analysis of
trends. In addition, there are discussions about whether or not conservation man-
agement questions should be posed in a hypothesis testing framework, which

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 5
 most power analyses assume. Many argue that decision making in the face of
uncertainty should at least rely on multiple hypotheses and that the associated
models be used to help make these decisions (Kendall, 2001; Williams et al., 2002;
Nichols and Williams, 2006; Gerrodette, 2011). These methods work best in data-
rich environments, but are increasingly being used in situations of data paucity and
limited technical capacity (Yoccoz et al., 2001). Although power analyses placed
in a hypothesis testing framework are perhaps not ideal, they do promote more
careful thought about the data requirements for a monitoring program and are
very informative in terms of illustrating how difficult it may be to show that our
conservation actions are effective.
Making decisions for conservation management within a Bayesian framework is
a different increasingly popular approach (Wade, 2001; Hoyle and Maunder, 2004;
Wade et al., 2007). Bayesian methods are well-suited to problems involving the
interpretation of monitoring data. Proponents of the methods argue that they pro-
vide a much more intuitive approach to decision making in the face of uncertainty.
Bayesian analysis permits the integration of information and data from a variety of
sources in a single framework and explicitly considers uncertainty in the decision–
making process. Just as decision making can be done in a Bayesian framework,
similarly the monitoring techniques themselves can be used or the trend analysis
can be done taking either a frequentist or Bayesian analysis approach (Williams et
al., 2002; McCarthy, 2007; Royle and Dorazio, 2008; Barker and Link, 2010).

Decision Tree for Wildlife Monitoring

A decision tree for selecting a method for wildlife

monitoring can quickly become very complex. At
the highest level, we need to determine whether we
want to monitor wildlife or habitat. Within each of
these, we can monitor at different spatial scales: a
site, landscape/seascape level or range level. We can
monitor at a number of levels of organization and
increasing complexity. At the species level, we can
monitor a single population (site), a group of popula-
tions (metapopulation) or the entire range (single
population to meta-metapopulation depending on
the species distribution). At the community level
we have a population of species occurring at a site
(community) or at a group of sites (metacommunity).
Elephant monitoring, Mpala Ranch, Kenya. © M. Kinnaird In addition, we need to estimate density/abundance,
occupancy, demographic rates or a combination of these. As a demonstration of
concept, we restrict the decision tree to density/ abundance estimation for a single
population of a single wildlife species.
We view the decision tree as composed of several nodes where branching deci-
sions occur. Having decided to restrict the example to estimate density/abundance
for a population of a wildlife species we are already four decisions into the decision
tree (Decision 1: wildlife; Decision 2: species; Decision 3: population; Decision 4:
density/abundance).

6 Wildlife Conservation Society | WORKING PAPER NO. 41

Decisions

Decision 1: Wildlife vs. Habitat

This is a basic division in conservation management. In WCS, many programs
focus on wildlife recovery and the prevention of future reductions to the popula-
tion at a site, and the management interventions are focused on that species (e.g.
reducing poaching). Alternatively, a wildlife program may focus on species habitat
management (e.g. reducing deforestation). The decision here usually is based on
the nature of the threat. Although some programs may be single species programs
and focus mainly on wildlife monitoring, and some may be community level pro-
grams, it is also the case that programs focus on both wildlife and habitats.

Decision 2: Species vs. Community

Species programs are most widespread in WCS (Landscape Species, Global
Priority Species, and endangered species). However, increasingly, we are manag-
ing and monitoring at the community level (e.g., bushmeat trade, ornamental fish
trade, carnivore community conservation).

Decision 3: Population vs. Metapopulation

Most WCS programs are site-based and deal with single populations. Global
Priority Species programs deal with species recovery or prevention of future
declines over a set of sites, usually semi-isolated and characterized as a metapo-
pulation.

Decision 4: State/rate variable - Abundance vs. Occupancy vs. Demographic rates

The choice of state or rate variable(s) to monitor depends on the nature of
the monitoring program, spatial scale, and funding. Demographic monitoring
usually occurs at a very small spatial scale (10’s of km2), and at one or a few sites.
Abundance/density monitoring usually occurs at a moderate spatial scale (100’s of
km2) at one or few sites. Occupancy monitoring often occurs at large scales (1000’s
of km2) and at many sites (note that certain abundance/density estimation tech-
niques also provide estimates of some demographic rates). Generally the cost per
unit effort is highest for demographic data, followed by abundance/density and
then occupancy.

Decision 5: Detection Probability = 1 vs. <1

When we assume that detection probability is one, we are obligated to verify
this assumption. It is often approximately true when surveying large animals in
open habitats in relatively small areas, and when conducting demographic moni- Camera trap photo of chimpanzee with tool.
© P. Boundja
toring at local scales (i.e. cohort-based primate demography). The assumption can
be verified through pilot studies that correct for imperfect detection, use of 2-stage
sampling, and use of double observer methods. If we assume that detection prob-
ability is less than one, we confront three alternatives: (1) detection probability
is known and fixed. In this situation we would apply a correction factor to get an
unbiased estimate of the state or rate variable; (2) detection probability is unknown
but fixed. In this case, we can monitor changes in the state or rate variable using
biased indices under the assumption that bias is constant; and (3) detection prob-

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 7
 ability is unknown and not fixed. This is the most likely case and requires correction
for detection bias.

Decision 6: Complete coverage vs. sample survey

We can either have complete spatial coverage of the area of interest or we con-
duct a sample survey over a portion of the area. A population closure assumption
is required to bind the study in space and time. For large spatial scale monitoring,
we usually conduct sample surveys, but there are examples of complete coverage
surveys.

Decision 7: Recognize individuals vs. sub-population vs. no discrimination

The ability to identify individual animals or distinguish between sub-groups in
the population will dictate the options for monitoring methods. If it is not possible
to identify individuals, distance sampling, occupancy or sign survey methods may
be appropriate. If sub-groups within the population are recognizable (e.g. marked
vs. unmarked, males vs. females, adults vs .juveniles), then mark-resight or change-
in-ratio methods may work. If individuals can be recognized, then Spatially Explicit
Capture-Recapture (SECR), capture-recapture or mark-resight methods may be
applied.

Decision 8: Animal vs. sign

A basic question for designing monitoring programs is whether the target spe-
cies is directly or indirectly observable. If the animal is directly observable, we can
use observation-based counting techniques. If the animal is indirectly observable
(cryptic, nocturnal) we may use passive detectors (camera traps, hair snares) or
active trapping. Alternatively, we may decide to use sign (dung, nests, tracks, feed-
ing, hair, acoustic cues) to indicate presence and/or to estimate abundance. Sign is
usually directly observable, but often requires ancillary information on deposition
and decay rates and age in order to interpret the meaning of estimates based on
sign. It may make sense to choose the survey target based on the comparative
detection probabilities and costs.

Note that not all decisions need to be made in all cases. Sometimes making
certain choices eliminates the need for other decisions (see figures 2a and b for
details). In the next two sections we briefly describe some of the more widely used
methods other than those relevant for density/abundance estimation of a wildlife
population, namely occupancy and species richness.

Occupancy

Often, estimation of abundance or density is logistically or financially pro-

hibitive. In these cases, an alternative state variable to consider is the proportion
of area occupied (PAO). PAO is an estimate of the species distribution or the area of
use, based on three possible states: the site is occupied and a species is detected,
the site is unoccupied and the species is not detected, and the site is occupied but
the species is not detected. The concept is similar to abundance estimation, but
instead of estimating number of animals, we estimate number of occupied sam-

8 Wildlife Conservation Society | WORKING PAPER NO. 41

pling units (MacKenzie et al., 2002, 2006). Because we need to account for animals
that are present but not detected, estimation of a detection probability is a key fea-
ture of occupancy analysis. Estimating PAO has a number of practical advantages.
The data are relatively easy to collect, we can use multiple sampling methods, and
the interpretation is straightforward. PAO may be the most reliable metric for large
landscapes because it is likely to be more robust to local and stochastic effects than
estimates of abundance or density. PAO can also be easily related to covariates of
interest such as habitat and exploitation.
Occupancy analysis offers great flexibility in design and analysis. The basic
feature of the method is the need for replicate visits to the sampling unit. The
replicates however, may be spatial or temporal, and may be carried out by repeat
visits by the same investigator or simultaneous visits by several independant
investigators. We can consider multiple occupancy states (classes of relative abun-
dance for example), habitat suitability, and other covariates that affect detection
and occupancy. We can also use open and closed models that allow us to monitor
occupancy over time in a single analytical framework. Open models also allow for
estimation of extinction and colonization rates and may be used to track commu-
nity and meta-population dynamics (MacKenzie et al., 2003, 2006).

Figure 2a: If it is assumed that detection of the target is certain (p = 1), then this requires verification. If spatial coverage is complete
and all individuals in a population are counted, then this is referred to as a census. A census is rarely possible, but if everything is observ-
able and counted in an area of interest, then no statistical analysis is required, as the result is simply a single number with no asso-
ciated variance. The target may be individual animals (scenario 1) or their sign (scenario 2). An example of the former is the 2008 Ewaso
Nyiro elephant survey that attempted to count all elephants in the Ewaso Nyiro watershed (30,000 km2) of northern Kenya. Examples of
the latter might include bird call cue counts or fixed width elephant dung counts in a small area. In this situation, ancillary information is
required to interpret the sign (estimates of calling/deposition rates and also decay rates for dung).
If complete spatial coverage is not possible, then a sample survey is conducted in a set of sampling units and the results extrapolated to the
entire area of interest to obtain the population size. It is assumed that all animals (scenario 3) or sign (scenario 4) within a sampling unit
are detected and counted without error and again requires deposition and decay rate estimates for the latter. Thus the variance associated
with the density or abundance estimates is solely due to spatial distribution of individual animals or sign. A sample survey requires careful
definition of the study area, and may benefit from stratification to improve precision, needs decisions to be made about sampling effort
to obtain an acceptable balance between precision and costs, and finally the sampling units should be defined and ideally be located by
means of a random design or systematic design with a random start. Aerial surveys in open habitats that follow strip transects or surveys
of dung in sampling plots are some examples.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 9
Figure 2b: When detection is less than one and varies over time and space, we must know or estimate the detection probability to obtain
unbiased estimates. The potential methods available in this case are independent of whether or not spatial coverage is complete. In the
former case the variance of the estimate will be solely due to variance in detection and in the latter case the variance of the estimate will be
composed of spatial variance and variance in detection (and decisions will need to be made about the definition of the study area, potential
stratification, amount of sampling effort, and definition and placement of the sampling units).
If the target has characteristics that can be used for identification (either due to trapping and marking of individuals or natural markings),
then it may be possible to identify individuals (spotted cats, ringed birds). When dealing with animals (scenario 5), capture-recapture (spa-
tially explicit or not), mark-resight or band return techniques might be used, for example. When dealing with sign (scenario 6), DNA analysis
and identification of individuals permits the use of capture-recapture techniques, for example. If it is only possible to identify an individual as
belonging to a sub-population (scenario 7), then once the sub-population has been defined it is possible to apply change-in-ratio or removal
techniques, for example. When identifying characteristics are not available or not made use of, then for both animals (scenario 8) and sign
(scenario 9) methods such as distance sampling, double observer or temporal removal may be used (again for sign deposition and decay
rates are ideally required).

Species Richness

Species richness, defined as the number of species occupying a delineated area,

is an increasingly important state variable for conservation of communities and
biodiversity. There are many approaches to estimating species richness including
the extrapolation of species-area or species-effort curves, the use of parametric
models of species abundance based on count statistics, use of taxon ratios, and
estimation of species richness based on sampling. For reviews of these methods
see Magurran (1988, 2003). Burnham and Overton (1979) suggested the use of
population estimation models to incorporate heterogeneity in species-specific
detection probabilities into estimates of species richness. They recommend a
model similar to the Mh estimator in closed population estimation permitting het-
erogeneous capture probabilities among individuals (Otis et al., 1978; Williams et
al., 2002). An extension of this approach to multi-year studies allows estimation of

10 Wildlife Conservation Society | WORKING PAPER NO. 41

the rate parameters of local extinction and colonization and can be implemented
using Pollock’s (1982) robust design and the software package COMDYN.
Cam et al. (2002) present a probabilistic, non-parametric estimator of spe-
cies richness for use with species accumulation data. They make the connection
between species richness estimation and abundance estimation using capture-
removal models in which the detection probability changes after the first detec-
tion (Model Mb: Otis et al., 1978). Removal models are appropriate for species
accumulation data because a species is removed from the population after its first
detection so the only statistics are the number of new species detected at each
sample period.
Species richness also can be estimated from detection/non-detection data
(MacKenzie et al., 2006). Occupancy models are especially useful for estimation
involving site level species richness when a list of potential species occurring at the
site or in the region is available. Species richness at a particular site will be deter-
mined by local environmental conditions (i.e., habitat) and by the regional species
pool that contains all possible species for the area. When the regional species pool
is known, each species may serve as a “site” in the context of occupancy sampling.
The estimate of occupancy is interpreted as the proportion of species from the
regional pool that occur at the site. The “number of sites occupied” is the estimate
of species richness. Occupancy modeling allows tracking of changes in species
richness over time and the modeling of covariates that might affect detectability,
extinction or colonization (MacKenzie et al., 2006). The use of covariates is not
possible in the capture-recapture models of Cam et al. (2000) because undetected
species are not used in the estimation, providing no ability to use covariate infor-
mation of such species.

Standardization of Monitoring Methods

Although, progress has been made on standardizing methods across land-

scapes or for a particular species within WCS (e.g. tiger and prey monitoring, forest
elephants and apes) a great deal of work remains in this regard. In addition, a key
element is tying this standardization to strategic planning and the development
of monitoring frameworks that clearly detail our goals and objectives in terms of
wildlife or habitat status and the desired reduction of threats. Without explicit goals
and objectives it is impossible to know whether or not we are successful in our
work and which activities tend to be successful.
As an example, a 2009 workshop on “Strategic Planning for Conservation
Management Across Landscapes” that included managers from a number of WCS
Latin America and Caribbean Program landscapes, as well as species specialists
focused on monitoring in different landscapes, provided some instructive ideas
and insights. After a review of monitoring techniques, many discussions, and
consideration of past experience, workshop participants thought it made sense
to standardize sampling designs and monitoring protocols for species, communi-
ties (human and wildlife) and potentially also indicators that are monitored on a
regional level (multiple country programs, multiple sites within countries). This
may include technical aspects of sample design, analytical methods, and develop-
ment of aggregate or headline indicators.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 11
 To form a picture of the most commonly used wildlife monitoring methods
and the associated target species, we conducted a survey of the workshop partici-
pants. The results lend weight to the idea that the set of methods being used (or
available) is reasonably small (see Table 1), which should make standardization a
reasonable possibility. The results of this survey also gives some indication where
resources might initially be invested to have the largest impact.
A few key techniques, such as capture-recapture (with camera trapping or DNA-
based), distance sampling, catch per unit effort (preferably with associated model-
based analysis to account for imperfect detectability that is unknown and not fixed
across time or space), questionnaire surveys, could be the focus of this standard-
ization. A first step would be the collection of existing monitoring protocols from
the field sites or other sources and then to standardize and improve these where
necessary. Collation and development of protocols and implementation manuals
to guide development of sampling designs and analysis of species and communi-
ties of interest would be made available more broadly to WCS staff and others via a
website, which could include links to already existing, good protocols available on
other websites, as well as other resource materials (list servers, papers).
Aside from the further development of these protocols and implementation
manuals, workshop participants thought it would be useful to put together an
overview paper describing the various techniques and their applications (e.g., the
variety of applications of presence surveys) with a synthesis of best practices across
WCS that could be used as an overview working paper for reference. For all proto-
cols workshop participants asked that we consider options for pooling data across
studies in order to improve accuracy and precision.

Conclusions

It certainly seems to be the case that

methods available for monitoring wildlife
are a fairly specialized and small set. In addi-
tion, in most cases the options in terms of
choosing a method will be further limited
by the characteristics of the conservation
target and the habitat. Finding a balance
between the costs of implementing the
method, the available technical capacity
and the required monitoring information
that can appropriately inform management
will also be part of the decision making
process.
Market scene in Sulawesi. This relatively small set of methods avail-
© M. Kinnaird able for monitoring wildlife and the desire
to monitor the effectiveness of our conser-
vation actions argues for an appropriate and comparable application of methods
across WCS. Although methods will always be implemented in a manner that takes
into account the characteristics of a particular species, there are still measures that
can be taken to ensure that methods are correctly implemented and standardized

12 Wildlife Conservation Society | WORKING PAPER NO. 41

where appropriate.
Some of the more commonly used methods shown in the nodes of figures 2a
and b and mentioned previously are briefly detailed in Appendix 1. A literature
review with some of the key references and internet resources is listed in Appendix
2. A small set of examples of monitoring applications at WCS sites that make use of
these methods is given in Appendix 3.

Table 1. Application of monitoring methods at 10 landscape sites in Latin

America based on an analysis of 120 combinations of taxonomic group,
species, project goals and indicators. Numbers indicate frequency of use.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 13
 Appendix 1. Overview of Frequently Used Wildlife
Monitoring Survey Techniques for Estimating
Abundance, Occupancy, or Demographic Rates

Occupancy Methods

Occupancy methods estimate the proportion of a habitat or number of patches

occupied when detection is incomplete (Mackenzie et al., 2002, 2003, 2006). The
analysis recognizes three states: occupied and detected, occupied and not detect-
ed, and not occupied. It provides estimates of the probability that a sampling unit
is occupied and the probability that an individual animal (or sign, if sign surveys are
used) is detected. It requires replicated observations on each sampling unit and it
allows for covariates that might affect occupancy or detection to be incorporated
into the analysis. The basic method assumes demographic and spatial closure dur-
ing a sampling period (referred to as a season) such that the occupancy status does
not change and that sampling units states are independent. Additional assump-
tions include no errors in identifying species and that observations are indepen-
dent. There are analysis options that relax most of these assumptions should this
be needed. The methods are continually evolving and some of the analysis options
currently available include:

• Single Season – estimates the proportion of occupied sampling units, detection prob-
ability, and estimates of covariate effects.
• Multiple Seasons – estimates include above plus estimates of colonization and extinc-
tion rates of sampling units, and estimates of covariate effects on rates.
• Species Interactions – allows the estimation of co-occurrence of species.
• Spatial Autocorrelation – relaxes the assumption that sampling units are spatially
independent.
• Multi-Method – allows detection probabilities to be different for different methods of
observation.
• Multi-State – allows the estimation of the probability that animals are in a given state,
given that they are present, which is especially useful for relative abundance data.
Multi-state models allow a species to occupy a site at different levels of abundance
and to evaluate factors affecting the occurrence and abundance of a species on the
landscape.
• Point Count – estimates population size from point-count data.
• Habitat Suitability – estimates occupancy as a function of site suitability.
• Simultaneous Modeling of Habitat Suitability, Occupancy and Relative Abundance
– allows for estimation of transition probabilities between habitats and abundance.

The Presence software facilitates analysis of occupancy data and can be used for
single species studies, community level studies and estimation of species richness.
It is available as a free download from Patuxent Software Archive (http://www.
mbr.pwrc.usgs.gov/software.html). Occupancy analysis can also be carried out in R
using the Unmarked package (http://github.com/rbchan/unmarked).

14 Wildlife Conservation Society | WORKING PAPER NO. 41

Distance Sampling

Distance sampling is one of a number of survey methods that can be used to

estimate animal density or abundance (Buckland et al., 2001). The key to distance
sampling is recording perpendicular distance to each observation (or radial dis-
tances for points) and fitting a detection function to these data that can be used to
estimate both the proportion of animals detected and counted and the proportion
of the survey area covered. Thus, the canonical estimator (Eq. 3) can be applied
to the raw counts to obtain an unbiased estimate of abundance. Ideally transect
lines or points are located randomly with respect to the distribution of the animals,
which helps ensure valid statistical inference. Additional assumptions when using
the standard method include that objects of interest on the line or point are detect-
ed with certainty, animals are detected at their initial location, measurements are
exact, and that detections are independent events.
For distance sampling to be successfully applied it is essential that detectability
decreases as distance from the transect line or point increases and that the dis-
tance between the observer and each target can be obtained accurately. Distance
sampling works well for populations in well-defined groups or detected through a
flushing response and can be very efficient and cost-effective for large populations,
populations at low or medium individual or group density, and populations
sparsely distributed over large geographic regions. In particular, point transects
might be most appropriate for populations at high density, for multi-species
surveys (e.g. songbirds), or when habitat is patchy or terrain is difficult, making it
problematic to walk along predetermined lines. Advantages are that the detection
function is robust to unmodeled heterogeneity and that repeated surveys are not
required unlike occupancy or capture-recapture surveys. Distance sampling meth-
ods are continually evolving with innovations in spatial modeling using distance
sampling data, incorporating covariates into the detection function, combining
distance sampling with mark-recapture methods, automated survey design, for
example (Buckland et al., 2004)
Fortunately, the freely available Distance software exists to help with distance
sampling design and analysis (Thomas et al., 2010). It comes with a comprehensive
online users' guide and can be downloaded from the Distance website (www.
ruwpa.st-and.ac.uk).

Capture-Recapture

Capture-recapture techniques comprise a continually evolving set of methods

to estimate state and rate parameters. The methods require recaptures (active or
passive) of animals that can be individually identified or sub-populations that can
be recognized either through tags or natural marking (or through DNA). A key
assumption is that marked animals are representative of the entire population
of interest and that marks are not lost (or do not change in the case of natural
markings). Unmodeled heterogeneity in capture probabilities create biases in
the estimates and every attempt must be made to account for this heterogeneity
that may be due, for example, to reactions to physical trapping, differences in the
natural behavior of individuals or changes in behavior over time. Some of the most

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 15
 well known capture-recapture methods include known fate models, Cormack-
Jolly-Seber models, closed models, band recovery/exploitation models, multi-state
models or combinations of these.
The Mark software that offers an astonishing list of analysis options is the state
of the art software for the analysis of capture-recapture data (www.cnr.colostate.
edu/~gwhite/ software.html). Mark's online help is comprehensive and in addi-
tion the e-book compiled by Evan Cooch, Program Mark: A Gentle Introduction,
provides a wealth of a information(http://www.phidot.org/software/mark/docs/
book/).

Spatially Explicit Capture-Recapture

Spatially explicit capture-recapture (SECR) uses the locations where each animal
is detected to fit a spatial model of the detection process, and hence to obtain esti-
mates of population density unbiased by edge effects and incomplete detection.
Previously, the conventional approach to the analysis of animal density from trap
surveys was to apply closed capture-recapture model analyses, and, then convert
resulting estimates of abundances to densities using a wide range of essentially
ad hoc methods. While these approaches appear to work adequately in practice,
little had been known about the range of conditions under which they work well.
This is because most real world study situations involve study areas of odd shapes
and sizes and difficult terrain that makes setting traps challenging and conditions
assumed by ad hoc approaches may not apply. Detections may take place by
means of live-capture traps, with animals uniquely marked; they also may be sticky
traps or snags that passively sample hair, from which individuals are distinguished
by their DNA microsatellites, or cameras that take photographs from which indi-
viduals are recognized by their natural marks.
The Density software uses maximum likelihood to estimate the density of ani-
mal populations from spatially explicit capture-recapture data (www.otago.ac.nz/
density). The SECR library developed for the R statistical software implements an
even wider range of spatially explicit capture-recapture analysis options using
maximum likelihood methods. The SPACECAP library for R implements a set of
Bayesian spatially explicit capture-recapture models. It was developed specifically
for tiger camera trap data.

Mark-Resight

Mark-resight methods rely on resightings rather than recaptures of individuals

or recognition of marked sub-populations. They are a variation on the mark-re-
capture theme in that they account for imperfect detection during the estimation
process and in addition utilize information on the sightings of unmarked individu-
als. Previously the main focus of mark-resight methods was on abundance estima-
tion (Neal et al., 1993; Bowden and Kufeld 1995); however, recent developments in
mark-resight models now permit the use of the robust design and an integrated
approach to estimate survival and transition rates between observable and unob-
servable states, as well as allowing for individual heterogeneity in sightability
(McClintock et al., 2006; McClintock and White 2009). Fortunately, these recently
developed analysis options are available in the Mark software package.

16 Wildlife Conservation Society | WORKING PAPER NO. 41

Demography

Some studies are interested in monitoring demographic performance as a func-

tion of individual performance. In these studies identified individuals are followed
as a cohort over time. Either a census can be conducted where all members of the
cohort are identified and their presence in the population verified. New individuals
are added (births and immigrations) and disappearances are noted (death, emigra-
tion). The assumption is that all individuals in the population are accounted for
in each census. Examples include primate and elephant demography monitoring
which typically assumes the cohort is defined by the area occupied, or an area
visited.
Demographic analysis can be implemented in mathematical software pack-
ages such as MATHEMATICA or MATLAB. The Demography library for R implements
functions for demographic analysis including lifetable calculations, Lee-Carter
modeling, functional data analysis of mortality rates, fertility rates, net migration
numbers, and stochastic population forecasting. Survival analysis is featured in
many statistical software packages such as SPSS and SAS. University of Vermont
Cooperative Fish and Wildlife Research Unit Spreadsheet Project offers Excel
spreadsheets for age- and stage-structured life table analysis as well as instructions
in the use of life tables. The specialized software Mayfield provides simple analysis
options for nest survival.
An alternative to life table analysis is to use open population capture-recapture
analysis in conjunction with Pollock’s robust design to estimate demographic
parameters (survival, mortality, immigration, emigration) for populations that
include a marked sub-population. Mark is the most complete software for estimat-
ing demographic parameters from data that include marked individuals. In addition
to standard models, Mark includes the ability to incorporate covariates that might
affect parameters, e.g. analysis options for nest survival data. Analysis options pre-
viously available in other specialized software have been mainly incorporated into
Mark. For example, Jolly-Seber-type models for open population mark-recapture
data available in POPAN (POPulation Analysis) can now be accessed in Mark.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 17
 Appendix 2. Literature and Internet Resources
Amstrup, S.C., McDonald, T.L., and Manly, B.F.J. (eds.). 2005. Handbook of
Capture-Recapture Analysis. Princeton University Press, Princeton.
Anderson, D. R. 1975. Optimal exploitation strategies for an animal popu-
lation in a Markovian environment: A theory and an example. Ecology
56: 1281-1297.
Anderson, D.R. 2001. The need to get the basics right in wildlife field stud-
ies. Wildl. Soc. Bull. 29: 1294-1297.
Bailey, L.L, Hines, J.E., Nichols, J.D., and Mackenzie, D.I. 2007. Sampling
design trade-offs in occupancy studies with imperfect detection: exam
ples and software. Ecological Applications 17: 281–290.
Barbraud, C., Nichols, J. D., Hines, J. E., and Hafner, H. 2003. Estimating rates
of extinction and colonization in colonial species. Oikos 101: 113–126.
Barker, R.J and Link, W.A. Bayesian Inference: with ecological applications.
Academic Press.
Boitani, L., Fuller, T. (eds.). 2000. Research Techniques in Animal Ecology,
2nd ed. Columbia University Press.
Borchers, D. L., Buckland, S.T. and Zucchini, W. 2002. Estimating Animal
Abundance: Closed Populations. Springer Verlag.
Borchers, D.L., and Efford, M.G. 2008. Spatially explicit maximum likelihood
methods for capture-recapture studies. Biometrics 64: 377-385.
Boulinier, T., Nichols, J.D., Sauer, J.R., Hines, J.E., and Pollock, K.H. 1998.
Estimating species richness: the importance of heterogeneity in species
detectability. Ecology 79: 1018-1028.
Bowden, D. C., and R. C. Kufeld. 1995. Generalized mark-resight popu-
lation size estimation applied to Colorado moose. Journal of Wildlife
Management 59: 840-851.
Brownie, C., Anderson, D. R., Burnham, K. P., and Robson, D. R. 1985.
Statistical inference from band recovery data - a handbook, 2nd ed. U.
S. Fish and Wildl. Serv. Resour. Publ. 156: 1-305.
Buckland, S. T., Anderson, D. R., Burnham, K. P., Laake, J. L., Borchers,
D.L., and Thomas, L. 2001. Distance Sampling: Estimating Abundance of
Biological Populations. Oxford Univ. Press, Oxford.
Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L, Borchers, D.L., and
Thomas, L. (eds.). 2004. Advanced Distance Sampling. Oxford University
Press.
Buckland, S.T. 2006. Point-transect surveys for songbirds: robust method-
ologies. The Auk 123: 345-357.
Buckland, S.T., Marsden, S.J. and Green, R.E. 2008. Estimating bird abun-
dance: making methods work. Bird Conservation International 18: S91-
S108.
Buckland, S.T., Plumptre, A.J., Thomas, L. and Rexstad, E.A. 2010. Design
and analysis of line transect surveys for primates. International Journal
of Primatology 31: 833-847.

18 Wildlife Conservation Society | WORKING PAPER NO. 41

Burnham, K. P., and Overton, W. S. 1979. Robust estimation of population
size when capture probabilities vary among animals. Ecology 62: 927
936.
Burnham, K. P., Anderson, D. R., White, G. C., Brownie, C., and Pollock, K. P.
1987. Design and analysis of methods for fish survival experiments
based on release-recapture. Am. Fish. Soc. Monogr. 5: 1-437.
Burnham, K.P. and D.R. Anderson. 2002. Model Selection and Multimodal
Inference: A Practical Information-Theoretic Approach. Second Edition.
Springer, New York.
Caughley, G. 1977. Analysis of Vertebrate Populations. Wiley, New York.
Caswell, H. 2001. Matrix population models: Construction, analysis and
interpretation. 2nd ed. Sinauer, Sunderland, MA.
Chatfield, C. 1981. Introduction to Multivariate Analysis. Chapman and
Hall/CRC.
Cochran, W.G. 1997. Sampling Techniques. Wiley, New York.
Collen, B., Pettorelli, N., Durant, S., Baillie, J., and Krueger, L. (eds.)
Biodiversity monitoring and conservation: bridging the gaps between
global commitment and local action, Blackwell Publishing. In press.
Conroy, M.J. 1996. Abundance indices. In: Measuring and Monitoring
Biological Diversity: Standard Methods for Mammals (Wilson, D.E., Cole,
F.R., Nichols, J.D., Rudran, R., and Foster, M.S. (eds.), pp 179-192,
Smithsonian Institution Press, Washington, D.C.
Cooch, E. and White G. (eds.). 2006. Program MARK: A Gentle Introduction.
5th Edition. http://www.phidot.org/software/mark/docs/book/.
Efford, M.G. 2004. Density estimation in live-trapping studies. Oikos 106:
598-610.
Efford, M.G. 2009. DENSITY4.4: Software for spatially explicit capture-
recapture. Dept. Zoology, University of Otago, Dunnedin, New Zealand.
http://www.otago.ac.nz/density.
Efford, M.G., D.K. Dawson, and C.S. Robbins. 2004. DENSITY: software for
analyzing capture-recapture data from passive detector arrays. Anim.
Biodivers. and Conserv. 27: 217-228.
Elzinga, C.L., Salzer, D.W., Willoughby, J.W., Gibbs, J.P. (eds.). 2001.
Monitoring Plant and Animal Populations. Blackwell Science.
Farnsworth, G.L., Pollock, K.H., Nichols, J.D., Simons, T.R., Hines, J.E. and
Sauer, J.R. 2002. A removal model for estimating detection probabilities
from point count surveys. Auk 119: 414-425.
Gerrodette, T. 2011. Inference without significance: measuring support for
hypotheses rather than rejecting them. Marine Ecology 32: 404-418.
Hedges, S. (ed.) Monitoring elephants and assessing threats: a manual for
researchers, managers and conservationists . Centre for Wildlife Studies,
Bangalore, India. In press.
Hilborn, R., and Mangel, M. 1997. The Ecological Detective. Confronting
Models with Data. Princeton Univ. Press, Princeton, New Jersey.
Hoyle, S. D. and Maunder, M. N., 2004. A Bayesian integrated population
dynamics model to analyze data for protected species. Animal
Biodiversity and Conservation 27: 247–266.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 19
 Johnson, F. A., Moore, C. T., Kendall, W. L., Dubosky, J. A., Caithamer, D. F.,
Kelley, J. R., Jr., and Williams, B. K. 1997. Uncertainty and the manage
ment of mallard harvests. J. Wildl. Manage. 61: 202-216.
Karanth, K.U. and J.D. Nichols, (eds.). 2002. Monitoring Tigers and Their
Prey: A Manual for Researchers, Managers, and Conservationists in
Tropical Asia. Centre for Wildlife Studies, Bangalore, India. 193 pp.
Kendall, W.L. 2001. Using models to facilitate complex decisions, p.147-
170. In Modeling in Natural Resource Management (T.M. Shenk and A.B.
Franklin, eds.) Island Press, Washington.
Krebs, C.J. 1998. Ecological Methodology (2nd Edition). Benjamin
Cummings.
Kremen, C., A. M. Merenlender, and D. D. Murphy. 1994. Ecological moni-
toring: a vital need for integrated conservation and development pro
grams in the tropics. Conservation Biology 8: 1-10.
Krzanowski, W.J. 1998. An Introduction to Statistical Modelling. Oxford
University Press.
Lancia, R. A., Nichols, J. D., and Pollock, K. H. 1994. Estimating the num
ber of animals in wildlife populations. In Research and Management
Techniques for Wildlife and Habitats (T. Bookhout, ed.), pp. 215-253. The
Wildlife Society, Bethesda, Maryland.
Lebreton, J. D., Burnham, K. P., Clobert, J., and Anderson, D. R. 1992.
Modelling survival and testing biological hypotheses using marked ani
mals: a unified approach with case studies. Ecol. Monogr. 62: 67-118.
Link, W.A. and Sauer, J.R. 1997. Estimation of population trajectories from
count data. Biometrics 53: 63-72.
Link, W.A. and Sauer, J.R. 1998. Estimation of population change from
count data: application to the North American Breeding Bird Survey.
Ecol. Appl. 8: 258-268.
Link, W.A. and Suaer, J.R. 2002. A hierarchical analysis of population
change with application to Cerulean warblers. Ecology 83: 2832-2840.
Long, R. A., MacKay, P., Ray, J., and Zielinski, W. (eds.) 2008. Noninvasive
Survey Methods for Carnivores. Island Press, Washington, D.C.
McGarigal, K., Cushman, S., and Stafford, S. 2002. Multivariate Statistics for
Wildlife and Ecology Research. Springer.
MacKenzie, D.I. and Kendall W.L. 2002. How should detection probability
be incorporated into estimates of relative abundance? Ecology
83: 2387-2393.
Mackenzie, D.I., Nichols, J.D., Lachman, G.B., Droege, S., Royle, J.A., and.
Langtimm. C.A. 2002. Estimating site occupancy when detection prob
abilities are less than one. Ecology 83: 2248-2255.
Mackenzie, D.I., Nichols, J.D., Hines, J.E., Knutson, M.G., and Franklin, A.B.
2003. Estimating site occupancy, colonization and local extinction
probabilities when a species is not detected with certainty. Ecology 84:
2200–2207.
MacKenzie, D.I., Nichols, J.D., Royle, J.A., Pollock, K.P., Bailey, L.L. and Hines,
J.E. 2006. Occupancy Estimation and Modeling: Inferring Patterns and
Dynamics of Species Occurrence. Academic Press, New York.

20 Wildlife Conservation Society | WORKING PAPER NO. 41

Manly, B.F.J. 1991. Randomization and Monte Carlo methods in biology.
Chapman and Hall, New York.
Manly, B.F.J. 2004. Multivariate Statistical Methods: A Primer (3rd ed.).
Chapman and Hall, New York.
Magurran, A.E. 1988. Ecological diversity and its measurement. Princeton
University Press, New Jersey.
Magurran, A.E. 2003. Measuring biological diversity. Blackwell
Publishing, Oxford.
Manly, B.F., McDonald, L, Thomas, D.L., McDonald, T.L., and Erickson, W.P.
2002. Resource Selection by Animals: Statistical Design and Analysis for
Field Studies (2nd ed.). Springer.
McCarthy, M. A. 2007. Bayesian Methods for Ecology. Cambridge
University Press.
McClintock, B.T., G.C. White and K. P. Burnham. 2006. A robust design
mark-resight abundance estimator allowing heterogeneity in resight
ing probabilities. Journal of Agricultural, Biological, and Environmental
Statistics 11: 231-248.
McClintock, B.T., and G.C. White. 2009. A less field-intensive robust design
for estimating demographic parameters with mark–resight data.
Ecology 90: 313–320.
McClintock, B.T., G.C. White, M.F. Antolin and D.W. Tripp. 2009. Estimating
abundance using mark-resight when sampling is with replace-
ment or the number of marked individuals is unknown. Biometrics
65: 237-246.
McComb, B., Zuckerberg, B., Vesely, D. and Jordan, C. 2010. Monitoring
Animal Populations and Their Habitats: A Practitioner's Guide CRC
Press, Boca Raton, Florida.
Mood, A. M., Graybill, F. A., and Boes, D. C. 1974. Introduction to the Theory
of Statistics, 3rd ed. McGraw-Hill, New York.
Moore, C. T. and Kendall, W. L., 2004. Costs of detection bias in index–
based population monitoring. Animal Biodiversity and Conservation 27:
287–296.
Morgan, B.J.T. 2001. Applied Stochastic Modelling. Arnold Publishers.
Morrison, M.L., Block, W.M., Strickland, M.D. and Kendall, W.L. 2001. Wildlife
Study Design. Springer, New York.
Neal, A. K., G. C. White, R. B. Gill, D. F. Reed and J. H. Olterman. 1993.
Evaluation of mark-resight model assumptions for estimating mountain
sheep numbers. Journal of Wildlife Management 57: 436-450.
Nichols,J.D., Boulinier, T., Hines, J.E., Pollock, K.H., and Sauer, J.R. 1998.
Estimating rates of local extinction, colonization and turnover in animal
communities. Ecol. Appl. 8: 1213-1225.
Nichols, J.D. and Williams, B.K. 2006. Monitoring for conservation. Trends
in Ecology and Evolution 21: 668-673.
O’Connell, A. F., Nichols, J. D., and Karanth, K. U. (Eds.) 2010. Camera Traps
in Animal Ecology: Methods and Analyses. Springer Verlag, New York.
Otis, D.L., Burnham, K.P., White, G.C. and Anderson. D.R. 1978. Statistical
inference from capture data on closed animal populations. Wildl.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 21
 Monogr. 62: 1-135.
Pollock, K.P. 1982. A capture-recapture design robust to unequality of cap
ture. J. Wildl. Manage. 46: 757-760.
Pollock, K.H., J.D. Nichols, C. Brownie, and J.E. Hines. 1990. Statistical infer-
ence for capture-recapture experiments. Wildl. Monogr. 107: 1-97.
Pollock, K.P., Nichols, J.D., Simons, T.R., Farnsworth, G.L., Bailey, L.L., and
Sauer, J.R. 2002. Large scale wildlife monitoring studies: Statistical
methods for design and analysis. Environmetrics. 13: 105-119.
Rolf, F.J and Sokal, R. R. 1994. Biometry (3rd ed.). W. H. Freeman.
Royle, J.A. 2004. N-mixture models for estimating population size from
spatially replicated counts. Biometrics 60: 108-115.
Royle, J.A. and Nichols, J.D. 2003. Estimating abundance from repeated
presence absence data or point counts. Ecology 84: 777-790.
Royle, J.A., Karanth, K.U., Gopalaswamy, A.M., and Kumar, N.S. 2009.
Bayesian inference in camera trapping studies for a class of spatial cap-
ture-recapture models. Ecology 90: 3233-3244.
Royle, J.A., Nichols, J.D., Karanth, K.U., and Gopalaswamy, A.M. 2009. A hier-
archical model for estimating density in camera trap studies. J. Appl.
Ecol. 46: 118-127.
Royle, J.A., and Young, K.V. 2008. A hierarchical model for spatial capture-
recapture data. Ecology 89: 2281-2289.
Royle, J.A., and Dorazio, R.M. 2008. Hierarchical Modeling and Inference
in Ecology: The Analysis of Data from Populations, Metapopulations
and Communities. Academic Press.
Salafsky, N., Margoluis, R.and Redford, K.. 2001. Adaptive Management:
A tool for conservation practitioners. Biodiversity Support Program.
Washington DC.
Scheiner, S.M. and Gurevitch, J. (eds.). 2001. Design and Analysis of
Ecological Experiments (2nd ed.). Oxford University Press.
Scott et al. (eds.) 2002. Predicting Species Occurrences: Issues of Accuracy
and Scale. Island Press. Washington DC.
Seber, G.A.F. 1965. A note on the multiple-recapture census. Biometrika
52: 249-259.
Seber, G.A.F. 1982 The Estimation of Animal Abundance and related
Parameters, 2nd Ed. Macmillan, New York.
Skalski and Robson. 1992. Techniques for Wildlife Investigations. Academic
Press, San Diego.
Thompson, S.K. 1992. Sampling. Wiley, New York.
Thompson, D.L., Cooch, E.G., and Conroy, M.J. (eds.). 2009. Modeling
demographic processes in marked populations. New York: Springer.
Thompson, W.L., White, G.C., and Gowan, C. 1998. Monitoring Vertebrate
Populations, Academic Press.
Thompson, W.L. (ed.). 2004. Sampling Rare or Elusive Species. Island Press,
Washington D.C.
Wade, P. R. 2001. The conservation of exploited species in an uncer-
tain world: novel methods and the failure of traditional techniques.
In Reynolds, J., Mace, G. M. Redford, K. H. and Robinson, J. G. (editors).

22 Wildlife Conservation Society | WORKING PAPER NO. 41

 Conservation of exploited species. Cambridge University Press.
Wade, P. R., Watters G.M., Gerrodette, T. and Reilly S.B. 2007. Depletion of
spotted and spinner dolphins in the eastern tropical Pacific: modeling
hypotheses for their lack of recovery. Marine Ecology-Progress Series
343: 1-14.
Walters, C. J. 1986. Adaptive Management of Renewable Resources.
MacMillan, New York.
White, G. C., Anderson, D. R., Burnham, K. P., and Otis, D. L. 1982. Capture-
recapture removal methods for sampling closed populations. Los
Alamos Nat. Lab. Publ. LA-8787-NERP. Los Alamos, NM.
White, G.C. and Burnham, K.P. 1999. Program MARK: Survival rate esti-
mation from both live and dead encounters. Bird Study 46(Suppl.):
120-139.
White, G. C., and Garrott, R. A. 1990. Analysis of Wildlife Radio-Tracking
Data. Academic Press, San Diego, CA.
Wilkie, D. and the Living Landscapes Program. 2002. Monitoring conserva
tion project effectiveness. Bulletin 6, Wildlife Conservation Society,
Living Landscapes Program, Bronx, NY.
Wilkie, D. and the Living Landscapes Program. 2006. Measuring our
effectiveness—a framework for monitoring. Technical Manual 3, Wildlife
Conservation Society, Living Landscapes Program, Bronx, NY.
Williams, B.K., J.D. Nichols, and M.J. Conroy. 2002. Analysis and manage-
ment of animal populations. Academic Press, San Diego, CA.
Yoccoz, N.G., Nichols, J.D. and Bouliniar, T. 2001. Monitoring of biological
diversity in space and time. Trends in Evolution and Ecology 16: 446-
453.
Zuur, A.F., Ieno, E.N., Walker, N., Saveliev, A.A., Smith, G.M. 2009. Mixed
Effects Models and Extensions in Ecology with R. New York: Springer.
574 p.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 23
 Internet Resources
Patuxent Wildlife Research Center
http://www.mbr-pwrc.usgs.gov/software
CAPTURE
PRESENCE
MAYFIELD
COMDYN
Many others
Colorado State University Department of Fishery and Wildlife Biology
http://www.cnr.colostate.edu/~gwhite/software.html
MARK
University of Otago
http://www.otago.ac.nz/density
DENSITY
Evan Cooch's software page
http://www.phidot.org/software
Links to other population analysis software
Research Unit for Wildlife Population Assessment
http://www.ruwpa.st-and.ac.uk/
DISTANCE
University of Vermont, Vermont Cooperative Fish and Wildlife
Research Unit Spreadsheet Project
http://www.uvm.edu/rsenr/vtcfwru/spreadsheets/
Spreadsheet exercises for population analysis

24 Wildlife Conservation Society | WORKING PAPER NO. 41

Appendix 3. Examples of WCS Wildlife Monitoring
Applications

Multiple Landscapes Monitoring Using Standardized Methods

Tigers Forever
Tiger prey - line transect density estimation (DISTANCE), occupancy and
point abundance estimation (PRESENCE)
Tigers - density (camera trapping but different analytical methods at
different sites)
Tigers - occupancy (PRESENCE)

Humpback Whales
Population estimation using DNA or fluke identification - capture
recapture (CAPTURE, MARK)

Tropical Ecology Assessment and Monitoring

Terrestrial wildlife - camera-trapping and the Wildlife Picture Index
(PRESENCE)
Tree/Liana - plot-based cohorts
Climate

Albertine Rift
Terrestrial wildlife camera traps
Birds - point transects (DISTANCE)
Primates - line transects (DISTANCE)
Climate

Landscape-Scale Monitoring Single Sites

Congo Africa
Great apes - line transect sign surveys (DISTANCE)
Forest Elephants - line transect sign surveys (DISTANCE)

Sudan
Large mammals - line transect aerial surveys (assume detectability is
certain)

Zambia
Large mammals - line transect aerial surveys (assume detectability is
certain)

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 25
 Kenya
Large mammals, terrestrial birds - line-transect surveys (DISTANCE),
camera traps
Elephants - cohort-based, line transect surveys (DISTANCE), aerial sur-
veys
Livestock-wildlife interactions - camera traps
Savanna and forest birds - point count surveys

Site-Scale Monitoring (Small Spatial Scale)

Indonesia
Siamang/gibbon demography - cohort
Vegetation dynamics - plot-based cohort
Primates, hornbills, ungulates, birds - line/point transect (DISTANCE),
camera trapping (PRESENCE)

Bangladesh
Bottle-nosed dolphins - photo-identification mark-resight (MARK)

Belize
Turtles - line transects (DISTANCE), capture-based mark-resight (MARK)
Atoll fished species - strip transects, plot-based on patch reefs

26 Wildlife Conservation Society | WORKING PAPER NO. 41

WCS WORKING PAPER SERIES
WCS Working Paper No. 1
Bleisch, William V. (1993) Management Recommendations for Fanjing
Mountain Nature Reserve and Conservation at Guizhou Golden Monkey &
Biodiversity. (38 pp.)
WCS Working Paper No. 2
Hart, John A. & Claude Sikubwabo. (1994) Exploration of the Maiko
National Park of Zaire, 1989-1994, History, Environment and the
Distribution and Status of Large Mammals. (88 pp.)
WCS Working Paper No. 3
Rumiz, Damian & Andrew Taber. (1994) Un Relevamiento de Mamíferos y
Algunas Aves Grandes de la Reserva de Vida Silvestre Ríos Blanco y Negro,
Bolívia: Situación Actual y Recomendaciones. (40 pp.)
WCS Working Paper No. 4
Komar, Oliver & Nestor Herrera. (1995) Avian Density at El Imposible
National Park and San Marcelino Wildlife Refuge, El Salvador. (76 pp.)
(English and Spanish)
WCS Working Paper No. 5
Jenkins, Jerry. (1995) Notes on the Adirondack Blowdown of July 15th, 1995:
Scientific Background, Observations, and Policy Issues. (93 pp.)
WCS Working Paper No. 6
Ferraro, Paul, Richard Tshombe, Robert Mwinyihali, and John Hart. (1996)
Projets Integres de Conservation et de Developpement; un Cadre pour
Promouvoir la Conservation et la Gestion des Ressources Naturalles. (105 pp.)
WCS Working Paper No. 7
Harrison, Daniel J. & Theodore G. Chapin. (1997) An Assessment of
Potential Habitat for Eastern Timber Wolves in the Northeastern United States
and Connectivity with Occupied Habitat on Southeastern Canada. (12 pp.)
WCS Working Paper No. 8
Hodgson, Angie. (1997) Wolf Restoration in the Adirondacks? The Question
of Local Residents. (85 pp.)
WCS Working Paper No. 9
Jenkins, Jerry. (1997) Hardwood Regeneration Failure in the Adirondacks:
Preliminary Studies of Incidence and Severity. (59 pp.)
WCS Working Paper No. 10
García Víques, Randall. (1996) Propuesta Técnica de Ordenamiento
Territorial con Fines de Conservación de Biodiversidad en Costa Rica:
Proyecto GRUAS. (114 pp.)
WCS Working Paper No. 11
Thorbjarnarson, John & Alvaro Velasco. (1998) Venezuela’s Caiman Harvest
Program: A historical perspective and analysis of its conservation benefits. (67
pp.) (English with Spanish Abstract)
WCS Working Paper No. 12
Bolze, Dorene, Cheryl Chetkiewicz, Qui Mingjiang, and Douglas Krakower.
(1998) The Availability of Tiger-Based Traditional Chinese Medicine Products
and Public Awareness about the Threats to the Tiger in New York City’s
Chinese Communities: A Pilot Study. (28 pp.)

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 27
 WCS Working Paper No. 13
O’Brien, Timothy, Margaret F. Kinnaird, Sunarto, Asri A. Dwiyahreni,
William M. Rombang, and Kiki Anggraini. (1998) Effects of the 1997 Fires
on the Forest and Wildlife of the Bukit Barisan Selatan National Park,
Sumatra. (16 pp.) (English with Bahasa Indonesia Summary)
WCS Working Paper No. 14
McNeilage, Alistair, Andrew J. Plumptre, Andy Brock-Doyle, and Amy
Vedder. (1998) Bwindi Impenetrable National Park, Uganda. Gorilla and large
mammal census, 1997. (52 pp.) (English with French Summary)
WCS Working Paper No. 15
Ray, Justina C. (2000) Mesocarnivores of Northeastern North America: Status
and Conservation Issues. (84 pp.)
WCS Working Paper No. 16
Kretser, Heidi. (2001) Adirondack Communities and Conservation Program:
Linking Communities and Conservation Inside the Blue Line. (62 pp.)
WCS Working Paper No. 17
Gompper, Matthew. (2002) The Ecology of Coyotes in Northeastern North
America: Current Knowledge and Priorities for Future Research.
WCS Working Paper No. 18
Weaver, John L. (2001) The Transboundary Flathead: A Critical Landscape
for Carnivores in the Rocky Mountains. (64 pp.)
WCS Working Paper No. 19
Plumptre, Andrew J., Michel Masozera, Peter J. Fashing, Alastair McNeilage,
Corneille Ewango, Beth A. Kaplin, and Innocent Liengola. (2002) Biodiversity
Surveys of the Nyungwe Forest Reserve In S.W. Rwanda. (95 pp.)
WCS Working Paper No. 20
Schoch, N. (2003) The Common Loon in the Adirondack Park: An Overview
of Loon Natural History and Current Research. (64 pp.)
WCS Working Paper No. 21
Karasin, L. (2003) All-Terrain Vehicles in the Adirondacks: Issues and
Options. (72 pp.)
WCS Working Paper No. 22
Clarke, Shelly. (2002) Trade in Asian Dry Seafood, Characterization,
Estimation & Implications for Conservation. (92 pp.)
WCS Working Paper No. 23
Mockin, Miranda H., E.L. Bennett, and D.T. LaBruna. (2005) Wildlife
Farming: A Viable Alternative to Hunting in Tropical Forests? (32 pp.)
WCS Working Paper No. 24
Ray, Justina C., Luke Hunter, and Joanna Zigouris. (2005) Setting
Conservation and Research Priorities for Larger African Carnivores. (211 pp.)
WCS Working Paper No. 25
Redford, Kent H., and Michael Painter. (2006) Natural Alliances Between
Conservationists and Indigenous Peoples. (24 pp.)
WCS Working Paper No. 26
Agrawal, Arun and Kent Redford. (2006) Poverty, Development, and
Biodiversity Conservation: Shooting in the Dark? (50 pp.)
WCS Working Paper No. 27
Sickler, Jessica, John Fraser, Sarah Gruber, Paul Boyle, Tom Webler, and Diana
Reiss. (2006) Thinking About Dolphins Thinking. (64 pp.)

28 Wildlife Conservation Society | WORKING PAPER NO. 41

WCS Working Paper No. 28
Castillo, Oscar, Connie Clark, Peter Coppolillo, Heidi Kretser, Roan McNab,
Andrew Noss, Helder Quieroz, Yemeserach Tessema, Amy Vedder, Robert
Wallace, Joseph Walston, and David Wilkie. (2006) Casting for Conservation
Actors: People, Partnerships and Wildlife. (85 pp.)
WCS Working Paper No. 29
Redford, Kent H., and Eva Fearn, eds. (2007) Protected Areas and Human
Displacement: A Conservation Perspective. (148 pp.)
WCS Working Paper No. 30
Redford, Kent H., and Eva Fearn, eds. (2007) Ecological Future of Bison in
North America: A Report from a Multi-stakeholder, Transboundary Meeting.
(64 pp.)
WCS Working Paper No. 31
Smith, Brian D., Robert G. Shore, and Alvin Lopez. (2007) Status and
Conservation of Freshwater Populations of Irrawaddy Dolphins. (115 pp.)
WCS Working Paper No. 32
Redford, Kent H. and Eva Fearn, eds. (2007) Protected Areas and Human
Livelihoods. (198 pp.)
WCS Working Paper No. 33
Beckmann, J. P., L. Karasin, C. Costello, S. Matthews, and Z. Smith. (2008)
Coexisting with Black Bears: Perspectives from Four Case Studies Across
North America. (73 pp.)
WCS Working Paper No. 34
Painter, M., A. R. Alves, C. Bertsch, R. Bodmer, O. Castillo, A. Chicchón,
F. Daza, F. Marques, A. Noss, L. Painter, C. Pereira de Deus, P. Puertas,
H. L. de Queiroz, E. Suárez, M. Varese, E. M. Venticinque, R.Wallace (2008)
Landscape Conservation in the Amazon Region: Progress and Lessons. (72
pp.)
WCS Working Paper No. 35
Brodie, Jedediah F. (2008) A review of American bison (Bos bison)
demography and population dynamics. (50 pp.)
WCS Working Paper No. 36
Redford, Kent H., and Catherine Grippo, eds. (2008) Protected Areas,
Governance, and Scale.
WCS Working Paper No. 37
Estes, Richard D. and Rod East. (2009) Status of the Wildebeest
(Connochaetes Taurinus) in the Wild 1967-2005
WCS Working Paper No. 38
Olupot, William, Alastair J. McNeilage and Andrew J. Plumptre (2009) An
Analysis of Socioeconomics of Bushmeat Hunting at Major Hunting Sites in
Uganda.
WCS Working Paper No. 39
O'Brien, Tim. (2010) Wildlife Picture Index: Implementation Manual Version 1.0.
WCS Working Paper No. 40
Weaver, John L. (2011) Conservation Value of Roadless Areas for Vulnerable Fish
and Wildlife Species in the Crown of the Continent Ecosystem, Montana.

A Decision Tree for Monitoring Wildlife to Assess the Effectiveness of Conservation Interventions 29
Wildlife Conservation Society
2300 Southern Boulevard
Bronx, NY 10460
Tel: 718-220-5100
www.w.cs.org
www.conservationsupport.org