Tips for bird surveys and censuses in countries
without existing monitoring schemes
State Museum of Natural History Stuttgart
Rosenstein 1, 70191 Stuttgart, Germany
Email: [email protected]
Swen C. Renner
Institute of Experimental Ecology, University of Ulm
Albert-Einstein Allee 11, 89069 Ulm, Germany
Email: [email protected]
Zoological Museum, University of Copenhagen
Universitetsparken 15, DK-2100 Copenhagen, Denmark
Email: [email protected]
Birds are important environmental indicators and, for a long time, have been
included when surveying biodiversity. This has led to a whole array of
publications, some of which are available over the Internet, making them easily
accessible worldwide. Here we provide practical guidance, with relevant source
references, for how to plan and conduct bird surveys and censuses, especially
in tropical environments.
Keywords: biodiversity, bird counts, assessments, monitoring techniques
When surveying biodiversity, birds are usually included because they have been
more completely charted (in terms of taxonomy and distributions) than any other
taxonomic group, and because good field guides, and even bioacoustics data,
are available for identification for most parts of the world. Furthermore, a large
number of skilled birdwatchers are often keen to volunteer in bird monitoring
projects. In order to develop bird monitoring as an effective tool in conservation
biology, a whole array of literature about bird census and monitoring techniques
has been published, the standard book by the late Bibby and his colleagues
(2000) covering most of them.
In 1998, Bibby et al. presented a guide especially designed for expeditions.
Gilbert and a team of specialists published a comprehensive book on monitoring
techniques for all sorts of UK bird species ranging from songbirds to raptors to
waterfowl (Gilbert et al., 1998). A “Best Practice Guide for Wild Bird Monitoring”
was published in 2008 by Voříšek and his colleagues focusing mainly on
Europe and giving an overview of existing monitoring schemes. It can be
downloaded, making it easily accessible to birdwatchers and ornithologists
worldwide (http://www.ebcc.info/index.php?ID=365). More articles with
information on specialized count procedures can be found in Gibbons &
Gregory (2006). So why include birds in this manual? The goal of this article is
to give practical advice on how to plan bird surveys and censuses in countries
where monitoring schemes are lacking, and to provide useful Internet links. This
is by no means a complete treatment of methods, study design, data
management and analysis as this would be far beyond the scope of this
chapter. All of these, sometimes rather complicated topics, are covered well in
the publications mentioned above (also see Gregory et al., 2004).
This chapter is written from a European perspective. Whereas the general
biology and life cycle of European and North American birds is rather well
known, we know far less about birds from many other parts of the world. It is
useful and important that scientists from developed countries contribute to the
study of biodiversity outside Europe, especially in the tropics. To be sustainable,
the long-term monitoring in developing countries should be locally based
(Danielsen et al., 2006, 2007). Although hard to achieve, this goal should
always be kept in mind, and is indeed often feasible once local communities
experience how simple monitoring systems can be used proactively to manage
their own resources.
Preparation for the survey
To start with, we advise contacting local ornithologists, to tell them about the
plans and to ask them if they support the idea of a survey and if they would
become a partner. It is important to find the right people to work with, people
that are accepted locally but ideally biologists by training. In most countries
birdwatchers and ornithologists are associated with the Birdlife International
Partner and can be tracked down by visiting the Birdlife International homepage,
a network of birdwatchers and ornithologists worldwide :
http://www.birdlife.org/worldwide/national/index.html. A local partner can help to
answer the following questions: (1) Have surveys already been carried out in
the target area or are some being planned? (2) Do other monitoring schemes
already exist? (3) When is the best time for a survey (season and time of day)?
(4) Are permits needed and how are they to be obtained? If a certain survey or
monitoring scheme is already in place one should consider choosing a similar
method to make data comparable between sites. A good example for a largescale bird survey is the second South African Bird Atlas Project SABAP2
(http://sabap2.adu.org.za/index.php). On the homepage there are good
descriptions of survey methods and databasing procedures including various
For your partnership to work, consider that volunteer schemes, as developed in
Europe and South Africa (breeding bird surveys, international waterfowl counts
etc.) rarely exist in developing countries. People have to work to survive and
can rarely afford a hobby like birding. Therefore, find out what the “normal” fees
and salaries are. Discuss this issue with your local partner beforehand. Out in
the villages, it may be useful to make agreements about donations and salaries
with a village chairman, or council, rather than with the individual helpers. Such
discussions may be cumbersome (for you) and you may feel awkward, but
nothing is worse than having to sort out conflicts afterwards. Sometimes it can
be useful to have a small contract telling the nature of your cooperation who is
responsible for which aspect, signed by all parties involved and a copy resting
with each party.
When planning a survey in an unfamiliar region, prepare yourself beforehand,
as this will save a lot of time in the field. This includes surveying existing
ornithological literature about birds in the target area, and to identify species of
particular interest. For most parts of the world there are field guides for birds
and CDs with bird calls and songs. The quality of these guides greatly varies
and they rarely include juvenile birds. Some are heavy to take into the field. A
simple although somewhat drastic trick is to ask a book binder to split your book
into two – one with the plates (to bring into the field) and one with the more
extensive text that you may decide to leave at home or at base camp. Some
publishers have already caught up with this idea, i.e. for West Africa (Borrow &
Demey, 2004), New Zealand (Robertson & Heather, 2004) or South Asia
(Rasmussen & Anderton, 2005). A good source to check what species occur in
an area is http://www.birdtours.co.uk/ which is a collection of trip reports by
travelling birders all around the world, including up to date maps, tips on where
to stay and who the useful local contacts are.
Calls and Songs
We very much recommend using a MP3 player with headphones and
microphone that can easily be taken into the field. With a special amplifier, a
directional microphone can be used, increasing the range and quality of the
recordings. The calls from a CD can be transferred onto the player. Most
modern recordings will offer files in mpg-format. If not, the sound files on audio
CDs can be transcribed into mpg-files easily, using freeware available (i.e.
http://www.freerip.com/). Do observe copyright laws and make sure your
download is really for free. Free resources are online sound libraries:
If you are after species from an area where sound recordings are not yet
available, you can check with the Wildlife Section of the British Library National
Sound Archive (NSA, http://www.bl.uk/soundarchive) or the Library of Natural
Sounds (LNS) at Cornell Laboratory of Ornithology (www.birds.cornell.edu).
Fees may apply. Once you have all songs and calls you need, arranging the
files in folders is useful, so that they can be easily found when in the field (i.e. in
alphabetical or systematic order, whatever the preference). Time permitting,
calls of species one expects in the target area can be put into a separate folder.
The recording function of most players is usually good enough to make a (low
quality) recording of a bird call or song that cannot be identified at once. These
can later be sent to a specialist to aid identification. Take a player with regular
AA or AAA batteries as those usually can be bought in most countries or, when
sunny enough, use small solar battery rechargers. Avoid complicated
recharging systems that you need electrical power and adaptors for, as
electrical power may not be available at base camp. In the headphone slot one
can usually plug small active speakers (working with batteries) that can be used
for playback. We do not recommend the use of playback but for certain species
it may be necessary, especially to detect cryptic or understorey species.
Playback can disturb birds, especially in the breeding grounds. Therefore it
should be only used if absolutely necessary and then only very briefly, i.e. for a
maximum of five minutes. As soon as there is a reaction, stop. Keep in mind
that a bird may not visibly respond, yet may still be disturbed.
Bird collections in museums
Bird collections are good places to brush up bird identification skills before going
into the field. If a survey is for scientific and conservation purposes, most
museums will allow such studies. See for example:
To find out whether a museum holds the required specimens, the bird curator
should be contacted well in advance and arrangements made to see
specimens. Bird curators or collection staff should explain the best way how to
handle bird skins. Always handle them with great care because they are meant
to be used by generations to come! A large number of web pages are useful to
study birds. Many of them are accessible through the GBIF platform
(http://www.gbif.org/). A useful searchable database is http://avibase.bsceoc.org giving links to selected Google images, distribution maps, taxonomy,
ITIS, Birdlife and Wikipedia. ORNIS, the Ornithological Information System, is
linked to GBIF and allows searching 42 mostly American bird databases,
including museum specimens (http://olla.berkeley.edu/ornisnet/). The site offers
a list of the respective curators including email addresses. For the Neotropics,
another good source is http://neotropical.birds.cornell.edu/portal/home. The
Zoological Museum, University of Copenhagen offers online access to birds
collected in Tanzania
When to count
The timing of a bird survey will depend on the life cycle of a particular bird
species if monitoring one species. Of course, many surveys would be for more
than one species. To get an inventory for a given area, typically the major
breeding season is best suitable for monitoring the community. Some birds
migrate and will be absent from the chosen study area for part of the year. This
can include a smaller scale i.e. altitudinal movement, but can range to longdistance migration. For most surveys, the time when males are singing on their
territory and the birds constructing their nests, are the most suitable, since birds
are most active then. Once sitting on the nest, birds often become very silent
and cryptic, making it hard for anybody to detect them. In many tropical areas
breeding seasons are not as synchronized as in more temperate regions. This
means that often only some bird species are breeding whereas other, often
closely related species may breed much later. Furthermore, only some
individuals within a population may be breeding. The timing of breeding greatly
depends on the altitude of your study site and the weather. Even rainforests can
be rather dry in certain years. Humidity greatly affects food availability and
triggers the onset of the breeding season. Low temperatures in mountain
regions can defer the onset of the breeding season. It is often very hard to
predict these patterns, even for experienced local people. If logistically possible,
we recommend to count at least two times a year.
You should aim to find all species possible including breeding birds, wintering
birds, cryptic species, nocturnal species, understorey species, rare species, bird
colonies and mixed species flocks. Special methods are available for most or
can be adapted from closely related species (see Gilbert et al., 1998). The more
time you spend in an area, the more species you are likely to find. These
species discovery curves (Fig. 1) are quite useful as they help you to identify the
point in time when the number of new species discovered in an area becomes
rather low. For economic reasons and depending on the question being
investigated, one may decide to stop data collection at that point (see below for
Fig. 1. Species discovery curve, with the total cumulative number of species discovered
each day (the time unit could also be hours). At first, the number of species increases
rapidly until, after a certain number of days, this number levels off. After that it takes
many days to discover a few more species, some of which may only be occasional
visitors of the area (after Bibby et al., 1998).
Birds live in almost any habitat you can think of. Clearly, you need to adapt your
survey method to the habitat. As an example of two extremes, consider a desert
with very low vegetation on one hand, and a dense rainforest with trees ranging
up to 40 m into the sky on the other. It is therefore useful to split your survey
area into different habitats (i.e. forest, scrub, desert, alpine etc.) and to adjust
your survey method accordingly (i.e. spacing of transects, distance of sample
points etc.). If you want to compare different habitats you should, however, use
the same protocol. Bibby et al. (1998, 2000), Gilbert et al. (1998), Gregory et al.
(2004) and Gibbons & Gregory (2006) offer good advice on this topic.
High-resolution maps are essential for a good survey and we recommend
obtaining maps prior to surveys. In the capitals of most countries you can
contact ministries for geography/geology or local cartographic services. Google
Earth images can give you a good idea of your study area, and for many areas
high-resolution images are now available. Sometimes they are a few years old
and therefore of limited value in areas under rapid “development”.
Birds have very different life styles. Some spend almost all their time up in the
air (e.g. swifts), some are flightless (e.g. Kakapo, Kiwi) or virtually so and skulk
around in the understorey vegetation. Birds that live up in the canopy are often
almost impossible to detect, as are birds that hide in dense foliage. A few birds
are rather curious and easy to detect, while others are extremely shy. Species
detectability depends very much on bird behaviour but also on weather
conditions and the skill of the observer. Observers who recognise all of the bird
calls and songs of an area will naturally discover more bird species than
observers without these skills.
Birds also sing at very different times of day, some start very early in the
morning, or advertise their territory around daybreak by giving a single call,
some sing at night (e.g. owls). Whereas for some species it may be useful to
count breeding territories identified by their territorial song, for other species it
may work better to count them during foraging or when they are flying to a roost
(i.e. gulls or terns). Many tropical forest birds move around in multi-species
feeding parties, or become active only when such parties pass through their
territory and provide effective antipredator vigilance. Thus, it is of great
advantage to pay attention to these bird parties, which often follow the same
route day after day (Poulsen, 1996). In the neotropics, mixed flocks are known
to follow ant swarms (Vallely, 2001; Roberts et al., 2000).
Although the books by Bibby et al. (1998, 2000) and Gilbert et al. (1998) offer
more detailed guidance, the annual cycles of birds outside Europe and North
America are, in comparison, poorly known. Indigenous people are often the only
ones that can give you some ideas about certain species (Ng’weno, 2008).
Sometimes their stories may make Europeans sneer because they contain a lot
of mystery. Don’t sneer at the stories but try to interpret them. A bird spending
winter in a tree cavity and coming out of its hole when the thunder arrives could
simply mean that it is a migratory species returning with the rains. Documenting
all bird behaviour and observations during surveys is a very worthwhile
Local knowledge and training of locals
Once in the field, it is very useful to ask a local guide, hunter or project partner
to accompany you and to talk to local communities. Going through a bird book
with indigenous people can give you priceless information. Depending on where
in the world you are, birds are often part of the day-to-day diet of people, so
they may have extensive knowledge about them. Talking to the elders of a
village one may also find out which species used to occur in an area but may
now have become scarce or have disappeared (Ng’weno, 2008). However,
when it comes to smaller, similar-looking species, locals often cannot
distinguish them as they lack binoculars. It is useful to bring extra binoculars
that you can give to local guides or other project participants. To get hold of
them ask friends and colleagues back home if they have a pair they don’t need
anymore. The most important achievement of your trip may not be your species
list, but instead the training of locals in bird identification, to make them
interested and to possibly teach them how to carry out a monitoring scheme
http://www.springerlink.com/content/100125/) (Danielsen et al. 2005, 2006).
Like other monitoring approaches, locally-based methods may be less precise
and biased, but may on the other hand be very effective tools for locally based
resource management, once locals realize how data can be used for
empowerment, e.g. for rapid management decisions to counter habitat loss by
interventions from foreigners and from corrupt administrations. Typically, data
collected by locals may lead to prompt and local decisions, while data collected
by scientists feed into long-term government regulation (Danielsen et al., 2007).
Thus, a combination of both is needed.
2.10. Personal safety
In remote areas always take somebody with you. Local people often know an
area very well and have a fabulous sense of orientation. Put an emergency
mechanism in place in case you do not return in time. Let other people know
where you are going and how long for. When discovering a rare animal the
temptation to leave known ground becomes very high and suddenly you don’t
know how to get back. If one person remains on a path, the second person can
go off in search for the animal, remaining in shouting distance you will always
find your way back. A handheld GPS and a compass are very useful, but you
need to know how to use them. Familiarize yourself with these on known
ground. It is good to note the direction of larger roads, or rivers, mountain
ranges, steep valleys as they can lead you back in case you get lost. Always
think about the basics: enough water, emergency food, sun and mosquito
protection, small headlamp (LED’s), raingear if needed, small first aid kit,
waterproof matches etc. A very powerful yet lightweight torch is the Supernova
run with LED’s which can function as a signal light (or to be used as a spotlight
to see owls…).
Before your trip make sure you have all necessary vaccinations. When mistnetting this includes one against rabies, as bats that are sometimes caught are
known to have transmitted this disease. Make sure you have enough
medication for all likely diseases with you. When leaving medication behind give
it to a local doctor or hospital. Being in remote places it is always useful to know
first aid and to be able to diagnose diseases, not only concerning yourself but
also the people that work with you (Werner, 1979; Merry, 1994; AAOS, 2007). If
one does become sick, one should always go to a doctor. Usually, local doctors
have a very good knowledge of local diseases.
Short overview of methods available
For anybody planning a survey we strongly recommend to thoroughly study
Bibby et al. (1998) as it offers in depth advice on many relevant topics
(http://biology.kenyon.edu/courses/biol229/fieldmanual%20birds.pdf). The best
way to learn is to join a professional team for a few days, to get some first field
experience and training. The Tropical Biology Organisation offers a wide range
of training courses (http://www.tropicalbiology.org/).
There are methods that will give you an idea of the species present in an area
(qualitative data) but not how many of them (quantitative data). When
introducing the factor “time” or “space” into a simple species survey, you can
very quickly improve the quality of your data (species discovery curve,
encounter rates, MacKinnon index or timed species counts, see Table 1).
Quantitative methods are, in general, more time consuming and require more
skill. There are point counts, line counts and even the mapping of territories. In
quantitative methods often the distance between observer and bird has to be
estimated (see below).
The key decisions are (from Gregory et al., 2004):
Do we want to estimate population size accurately or will an index meet our
needs? In other words, are we interested in absolute or relative abundance
Where will we undertake the survey?
Should we cover the whole area of interest, or only sample part of it?
If we plan to sample, how should we select the study sites?
What geographical sampling units will we use? Mapped grid squares, forest
blocks, or other parcels of land?
What field method will we use?
What are the recording units: individuals, singing males, breeding pairs,
nests or territories?
How will the subsequent data analysis be carried out?
How will the results be reported and used?
You need to adapt your method to: a) the question you are asking, b) your skill,
c) the time available, and d) the habitat. In open habitats, distance sampling
may be easy and therefore the method of choice, but this will be hard in
rainforest. Notice, though, that comparisons between habitats require that
similar methods are being used.
When trying to survey a dense lowland rainforest you will quickly notice your
limitations. It will be hard to see birds and to estimate distances to vocalizing
birds. This makes standardised quantitative sampling difficult or even
impossible (Bibby et al., 2000). With the many logistic constraints during
fieldwork in such environments, it is therefore important to consider how to best
spend the time available. Rather than working hard to obtain perfect quantitative
data from a single site, it may be better to use the time to get semi-quantitative
data for several sites. This approximates random sampling of the
metacommunity and, in addition, gives some information about variation across
different habitats. Small samples mean that some rare species are unrecorded,
and this truncation of the community (Preston, 1948) reduces the possibility to
discriminate between different abundance models. However, even incomplete
samples will suffice to identify dense (viable) populations of species of
conservation concern, and will allow estimates of species richness.
Unless you have been there, you will not know what your survey area is really
like. Plan a pilot survey of at least two weeks to a) get to know your species and
the habitat, b) to try out methods and c) to practise them.
Qualitative methods, relative abundance
Simple species list. All species are noted, regardless of time of day or
season. The presence or absence of rare or threatened species (Birdlife
International 2000, http://www.iucnredlist.org/) is the key for conservation
and management of a site. The problem with simple species lists is that
there is no control for observer effort. Chance observations will obviously
increase with the time you spend in the field and some cryptic resident
species don’t show up immediately.
Species discovery curves. Species discovery curves that record survey
effort can be obtained by recording the time spent in the field for each
observer. It is important that observers work at different areas or at different
times. By also noting the date and time each species was discovered, some
simple analysis becomes possible. Having separate lists for different areas
may enable you to come up with further detail, e.g. if you split your area into
degraded and natural forest, you can make a simple comparison
Encounter rates. Encounter rates are calculated for each species by
dividing the number of birds recorded by the number of hours spent
searching, giving a figure of birds per hour for each species. When doing
this separately for different habitats, more detailed information can be
obtained. However, beware that encounter rates will vary with the structure
of the vegetation. In dense vegetation encounter rates may be lower than in
more open habitat. Encounter rates are not a substitute for true density
estimates but they allow a comparison of relative abundance. Abundance
categories can be scored.
As an example (from Lowen et al., 1996), for each species assume the
number of individuals/100 field hours to be your value, then you could use
the following abundance categories: rare (< 0.1), uncommon (0.1-2.0),
frequent (2.1-10.0), common (10.1-40.0) or abundant (>40.0). If these
categories do not work for your data you can of course adapt them
accordingly. To avoid counting birds several times, it is important that
different observers move to a starting point at some distance from base
camp and plan their routes so that they are not overlapping. Birds that call
loudly will be recorded more often than more quiet ones and the likelihood
to record a species will depend on its state in the annual cycle.
MacKinnon lists (MacKinnon & Phillips, 1993). They are often used in
“Rapid Assessments” (Herzog et al., 2002). Make a list by recording each
new species seen until you reach e.g. 20 species; then start again with a
new list. Any one species will only be recorded once in your first list of 20,
but may be recorded again in subsequent lists. Analysis of ten or more lists
for a given area will give a good picture of its avifauna. Plotting the
cumulative total number of species recorded against the number of lists
made, this produces a species discovery curve whose steepness reflects
species richness and indicates how many more species are likely to be
found in an area (Fig. 2). If you are in a habitat that is species poor, you
may decide to use a lower number than 20, maybe 15 or even 10 (Poulsen
et al., 1997). You need to try this out. Your speed of walking will greatly
affect the kind of species you will encounter. Whilst walking slowly through
the forest one will get a good number of scrub or canopy dwelling birds, but
for some ground-dwelling birds such as pheasants, pittas or thrushes,
moving quickly but silently through the forest will yield better results, as birds
have less time to react to your approach. It is recommended to discover
every bird that is active within 50 m from the transect line (Schieck, 1997)
and thus it is not practically possible to achieve a constant walking speed.
If this method is used to describe community structure, there are some
fundamental flaws, as some records will be single birds and others will be
flocks with many individuals of the same species; fortunately this problem is
not so serious in tropical forests where most species appear in pairs and
family groups (2-4 individuals). The data can be much improved by writing
down the number of individuals and use this raw data for the final analysis
of community composition and species richness (Herzog et al., 2002).
Fig. 2. Species curves derived from MacKinnon lists, simplified and altered from
MacKinnon & Phillips (1993).
Timed species counts. Timed species counts (TSCs) are especially useful
for open habitats (Pomeroy & Tengecho, 1986), but as mentioned above
this approach may not be useful off-trail in thick vegetation. Data for TSCs
are recorded in six columns, corresponding to six 10-min intervals during an
hour-long survey. The observer walks at a slow pace (about 1-2 km/h). For
the first 10 min, every species seen or heard is noted down in the first
column, regardless of the number of individuals. For the second 10 min-
period, any species not already recorded is noted in the second column and
so on. For one observation hour each species is only noted once. A
minimum of 15 surveys should be carried out for a site, corresponding to 15
observation hours. Pomeroy & Tengecho (1986) suggest to physically cover
an area of 1 km² for each count. Depending on the habitat this may of
course be modified.
Positioning of sampling points. There are various possibilities to place
sampling points or beginnings of transects (Fig. 3), each one with
advantages and disadvantages. For statistical reasons it is important to
place all sampling points at random or at least to place your first point in a
line of points at random. When counting along roads or existing paths (Fig.
3A) it is likely that not all species are discovered because the path does not
run through all the different habitats. More importantly, the presence of the
road or path may influence the species (or numbers) present. Placing the
points randomly may give better coverage (Fig. 3B), however, choosing a
completely random approach may leave some areas unsampled. A stratified
random sample (Fig. 3C) using a grid (at least 500 m apart) is the best
choice. In each resulting square one point is chosen at random. If points
from two neighbouring squares are too close to each other (so that possibly
the same birds are counted twice) then it may be useful to omit that point
and choose a new one.
Fig. 3. Three possibilities for positioning point counts or beginnings of transects. a. along
existing roads or paths; b. entirely random or c. stratified random (after Bibby et al.,
Easy, no data
No control of
between areas or
Estimate of the
total number of
Different sites and
counts can be
Plotting of data
analysis, but simple
individuals of a
species per unit
Crude comparison of
species within a site
and within species
Differences in species
accounted for. To
count all individuals of
all species present
can be a practical
based on the
block of effort.
Crude comparison of
species within a site
and within species
between sites. Data
collection is simple,
allowing the observer
freedom to roam.
by observer skill and
Differences in species
based on the
of time. Plotting
Crude comparison of
species within a site
and within species
between sites. Data
collection is fairly
simple, allowing the
observer freedom to
when use of
net length and
birds, get to know the
Proper training and
required, timeconsuming, mostly
limited to understorey
species, not costeffective, capture
Table 1. Bird survey and count methods (adapted from Bibby et al., 1998).
There are some practical considerations for choosing a certain sampling
pattern and distance between sampling units (Robinson et al., 2000). In a
mountainous rainforest with dense understorey using existing paths or
roads may be the method of choice because: a) observers could get lost
otherwise, b) finding random points would be very time consuming and c) to
get to these points a lot of vegetation needs to be cut down which is not
only time consuming but may also be quite destructive. Furthermore, some
canopy species are impossible to detect when obscured by foliage, thus
using a road may enable you to see them (MacKinnon & Phillips, 1993).
When cutting transects during the breeding season there is a danger to
destroy nests and you will open pristine forest to people and animals that
may follow your tracks to exploit the forest. The disadvantage is obvious –
you will not cover your study area evenly, thereby not encountering some
bird species that you may have found using a random approach. In forest,
two sampling points should be at least 150 to 200 m apart, in open habitat
even further. Doing your survey in open farmland savannah it will be more
easy to set up transects or to find random points.
Distance sampling (from Bibby et al., 1998). Quantitative methods often
require the estimation of the distance between the observer and the bird
(Fig. 4). Errors can be minimized by practising beforehand, and it is
important that different team members synchronize their estimates. Optical
range finders can be useful when you see a bird (but not when you hear it).
If you sample from points you can mark certain distances in advance.
However, in dense vegetation neither of these methods will work. Estimation
of the distance to a calling bird can be practised by placing a tape recorder
at various distances. To make things easier one can use distance bands,
i.e. within 5, 10, 15, 20, 30, 40, 60, 80, 100 m of the observer.
Fig. 4. d1 (perpendicular distance) can be calculated using d2 and the sighting angle θ
For the analysis of these data the software “Distance” can be used (Laake
et al., 1994). Please check Bibby et al. (1998) for further detail. Distance
sampling can be used both in line counts and point counts (see Table 2 for
advantages of each method).
Point Counts. Points are usually laid out on a random transect, i.e. every
50 m. One walks from one point to the other, stops at the point for a
predetermined amount of time (i.e. 5-10 min) to count all birds present
(individuals and species) and then walks to the next point to repeat this and
so forth. Distance between points and amount of time spent counting need
to be adapted to the habitat. For more details see Bibby et al. (1998).
Line Counts. You walk continuously along a certain line and record all bird
contacts either side of the track. Walking speed should be constant, a goal
very hard to achieve, especially in dense forest. Avoid counts along streams
and rivers as your splashing about will flush birds along the river often long
before you have even had a glimpse on them. For more details see Bibby et
+++ extensive, open and uniform
+++ dense forest or scrub
+++ mobile, large or conspicuous
species and those that easily flush
+++ cryptic, shy, skulking species
+++ low population densities or species
+++ high population densities or
Cover the ground quickly and efficiently
recording many birds
Time is lost whilst walking
between points, but at point more
time for observation and
Double counting minor as observer is
on the move
Double counting potential
Birds are not so much attracted to
Curious birds may be attracted to
+++ when easy access
+++ when difficult access
Can be used for bird-habitat studies
Better suited to bird habitat
Errors in distance estimation have a
smaller influence on density estimates
(because the area sampled increases
linearly from the transect line)
Errors in distance estimation can
have a larger influence on density
estimates (because the area
sampled increases geometrically
from the sampling point)
Table 2. Differences between point and line counts (modified from Gregory et al., 2004),
+++ = good for.
Territory mapping and other methods. For more advanced methods like
territory mapping we refer to the literature (i.e. Bibby et al., 2000). A method
especially designed for tropical environments is “Multi Time-Window
Transect-Mapping” (see Jahn, in press).
Mist-netting is useful to learn the birds in an area and to discover skulking
understorey species but it is not time effective and the data obtained are not
useful for a quantitative description of the local bird community. According to
Remsen & Good (1996), the record of birds through mist-netting measures the
activity of birds within 3-4 m above the ground rather than the community
composition in the habitat. Mist-netting is very time-consuming, extremely
weather dependant and should only be carried out by well-trained people. A
good description of how to responsibly use mist nets to catch birds is given in
Redfern & Clark (2001). Training is offered in many countries by institutions that
organize the national bird ringing schemes.
Fig. 5. A mist-net of 6 m height using
bamboo poles to catch birds in
Madagascar. (Photo: F. Woog).
Also see Hofmann et al. (this manual) for a detailed description of the use of
mist-nets for catching bats. For most passerines, 16 mm Nylon mesh is used,
nets are 6 or 12 m long and have 4-5 shelves (giving them a height of about 2.5
m). Mistnets can be placed on ridges, in thickets or at forest edges, where birds
often pass close to the ground. If nets are set in wrong places a lot of damage
can be done to the birds (bird colonies, roosting places and nests have special
rules). Generally, avoid putting the nets in the sun, where they are easily visible,
and where captured birds can rapidly dehydrate. Taking birds out of the nets,
and handling them, has to be done properly and enough people need to be
present to monitor the nets that have been put up at close intervals (every hour,
or more often under warm conditions or when it drizzles; but close the nets
when it rains!). After extraction from the net birds are usually placed in doublesowed light cotton bags for further processing.
Birds can transmit diseases. To reduce this risk, make sure that bird bags stay
dry and clean (turn them inside out and shake them after each capture, and
wash them often). Make sure to thoroughly disinfect nets, especially when
moving between countries or continents. It is unacceptable to use mist-nets first
in an European country and then use them, unwashed, in a pristine tropical
forest (the same is of course true for all your camping and outdoor equipment).
Also observe principles of hygiene: wash your hands, preferably with
disinfective soap. Do not take soiled bird bags into the tent you sleep in.
When you have the bird in your hand, make careful records (photos, weight, fat
score, brood patch, moult stage, wing length and other measurements)
(Svensson, 1992; Baker, 1993) and, depending on additional research
questions, take a blood sample for DNA-analysis or feather for the analyses of
stable isotopes (see paragraph on collecting birds). Bird ectoparasites are often
poorly known and collecting them may yield new species (preserved in 2 ml
vials in 70% alcohol, use very fine tweezers).
Canopy nets can increase the number of species you catch. They are not easy
to handle. The best way to learn about these is to ask somebody that uses
To quantify birds, mist-netting is not really a good method and will only be useful
to compare relative abundances of selected understorey species (see Table 1).
If this is planned, the birds need to be marked to avoid double counting
recaptures as new captures. The most useful are bird rings, but if these are not
available tail feathers can be clipped systematically to identify birds which have
already been caught. Clippings should be as small as possible. As birds moult
their feathers at least annually, these markings remain temporary. Markrecapture methods are useful for a wide variety of studies and purposes and
can help to estimate abundance of selected species quite precisely or establish
local movement pattern (for information about data analysis of capturerecapture data see http://warnercnr.colostate.edu/~gwhite/mark/mark.htm).
A note of caution – locals often ask for mist-nets (as a means of catching birds
for food, plumage or for the pet-trade). Never leave mist-nets unattended. When
storing them make sure they are as safe as your money and passport. On the
other hand, some locals may be afraid of the nets and will not pass by. Make
sure you talk to the village people that may encounter the nets, ask them for
their permission and explain what you are doing. In some areas with large
game, goats, cows or monkeys it may not be advisable to use mist-nets as
these animals can easily destroy your expensive nets and may severely injure
themselves during the process of entanglement.
For a bird survey and monitoring scheme the collection of birds is usually not
necessary, but sometimes new species remain undiscovered because birds
were not collected. There are all sorts of opinions about collecting birds (see
Remsen, 1995; Collar, 2000). Habitat loss, agricultural practice and world-wide
climate warming are the real threats to birds, and in comparison the “sacrificing”
of a few birds for science means nothing for most species except when they are
very rare. But all collecting should be done legally and justified by some clear
purpose, such as needs for documenting new-discovered populations and
potential new taxa. In many developing countries there exists a sort of split
moral – one for indigenous people, one for industrial enterprises and one for
(foreign) biologists. Whereas indigenous people hunt birds for a living, and
international companies destroy vast pristine natural areas (i.e. for mining, dam
projects etc.) often, especially foreign, biologists are not being granted collection
and export permits for birds. Even the collection of feathers and birds found
dead along roadsides may not be permitted and the export from the country of
origin and import into your home country is not easy (i.e. concerning species
listed under CITES or under health regulations).
However, it is always useful to prepare birds found dead or for a specific
scientific question, and local partners can advise on what to do concerning
permits. The easiest is usually a cooperation with a museum or university where
specimens can be kept before permits have been worked out. It is fair enough
that these institutions often want a share of the collected material for their own
collections. For preparation techniques see Wagstaffe & Fidler (1968), Harrison
& Cowle (1970), Piechocki (1998), Winker (2000) and Hofmann et al. (this
volume). To see a video on bird preparation, paste
mms://18.104.22.168/bts/birdprep.wmv into your browser.
In order to preserve bird skins under field conditions it is best to prepare the skin
right away, and to quickly dry it properly (i.e. using a kerosene lamp for heating,
when it rains). Remove the brain and as much tissue as possible. Salt can be
used as a cheap and easily available preservative agent. Thymol can help to
prevent bacterial growth and moulding, i.e. put some crystals into your air-tight
storage containers, and if needed silica gel. If one has no time to make skins in
the field or is working in climates with a high humidity, birds can be put into 70
% alcohol (1/3 animal, 2/3 alcohol). When preserving a complete animal without
skinning it you need to inject alcohol with a syringe into the internal cavity of the
animal and the brain (through the nose). Because the alcohol gets diluted by the
fluids of the animal, it is advisable to change it after a few days. Alcohol can
wash out certain colours, which is a disadvantage. For transport, the alcohol
can be drained and specimens be put into double Ziplock bags. It is also useful
to collect a tissue sample (i.e. muscle or liver) in the field. This is to be stored in
pure 90-95% alcohol or EDTA-buffer. For good practices in tissue conservation:
For many research purposes it may suffice to take a blood or feather sample
and then releasing the bird (note however, that official permits may still be
needed!). Blood samples are useful for genetic studies on various levels not
only for speciation but also for population differences (Gaunt, 1999; Dawson,
2005). Stable isotopes found in feathers can give you an indication where a
migratory bird grew a feather (Bearhop et al., 2000; Wassenaar & Hobson,
2001). A small drop of blood is taken from the wing vein or in species with soft
legs (like swans, geese and ducks) or young birds from the leg vein and placed
in small vials containing buffer (200-300 μl blood in storage buffer containing
10% EDTA, 1% SDS, 0.5% NaF, 0.5% thymol and 100 mM Tris, pH 7.4 (Wink,
2006) or alternatively DMSO-buffer (SSDE) consisting of 20% DMSO, 0.25M
EDTA pH 8.0 saturated with NaCl). These samples can be stored at ambient
temperatures, but longer-term storage at -20°C (or lower) is recommended.
Care should be taken, that syringes or buffer do not contain heparin, as this will
inhibit the PCR reactions. The procedure should be learned from another
ornithologist that has used the method before. Veterinarians often do not have
experience extracting blood from birds.
When collecting samples make sure they are properly labelled, a collection
number is not enough. A proper label should at minimum contain date (write out
the month i.e. 11 Dec 2009, not 11.12.2009 as this may be read 11/12/2009
which could be interpreted as 12 Nov 2009, always write the complete year as
09 could mean 1909, 1809 or 2009), exact location including country, species,
collector, collection number. Never trust that you will remember to do this later.
http://olla.berkeley.edu/ornisnet/?q=node/5 gives detailed tools and guidelines
for geo-referencing. If possible, note longitude and latitude (i.e. read from your
GPS or map).
Field notes should always be detailed, with date, time of day, weather
observation (that may influence your survey results, e.g. heavy rainfall) and, if
possible, number of individuals encountered, and, if discernible, their sex and
age. If you see a group of peacocks, for example, note number of males,
females, immatures and juveniles. If you encounter birds that you do not know,
try to take a photograph, take a sound recording or immediately make a small
sketch of what it looked like. If you don’t have enough time in the field or you
can’t write things down whilst you are observing birds, MP3 players can serve
as dictaphone. However, always think about the time it will take to transcribe the
information from your recordings. One can only guess how many recordings
have been made in ornithological research without ever having been analyzed.
Modern digital cameras offer a unique possibility for improving field
identification. For instance, when a mixed feeding party of birds passes through
the vegetation, take as many photographs as you can, and by zooming in
afterwards you can identify birds that you did not immediately have time to
identify (or later blow up the images on your computer screen). In this way you
may sometimes be able to reliably identify every bird in the party.
It is often useful to enter data in forms prepared in advance, as this may
facilitate later data entry into your database. The forms should mirror the
structure of the database you intend to use. Enter your data as quickly as
When working in wet climates, working with normal paper is a challenge as at
the end of your expedition you may end up with a heap of ‘papier mâché’.
Fortunately there are solutions, i.e. “Write in the Rain” notebooks and copying
paper (Darling Corporation, http://www.riteintherain.com). These items are not
cheap but are well worth the investment. If you use pencil you can drop them in
a river and will still be able to read what you have written.
Database and data analysis
This should be in place before you start the survey, as they are very much
linked. Good guidance can be found in Bibby et al. (1998, 2000) and Voříšek et
al. (2008). The statistical approaches for estimating bird abundance from bird
counts and taking detectability into account (i.e. Kéry, 2008) are well beyond the
scope of this chapter. Whenever in doubt, consult a professional ornithologist or
statistician before you start fieldwork.
Case study 1 from a tropical cloud forest (The Chelemhá,
Combined line-point counts are often used in the tropics. Almost any method is
biased to sample the entire bird community (Terborgh et al., 1990; Poulsen,
1994; Remsen, 1994; Remsen & Good, 1996). Therefore, a combination of
several methods is sometimes essential to get a complete species list and
estimate relative abundance. In Guatemala, it proved essential to combine point
counts with transects, since a large part of the bird community would have been
missed if using only one method (Renner, 2003). To circumvent losing some
essential species, point counts were combined with transects counts: point
count sites were established each 25 m along transects. At each 25 m mark, all
birds sighted or heard within a nominal distance of 100 m were recorded for five
minutes. After the five minutes, the distance to the next point count locality was
slowly followed in the shortest possible way covering the distance in about the
same time. Transects were 150 m apart totalling 3,300 m. The local cloud
forests, the major habitat in Chelemhá, fortunately do not have a very dense
understorey, hence only minor efforts were needed to establish the pointtransects. The bordering secondary vegetation, however, was very dense, and
establishing trails to count birds was impossible (the dense secondary
vegetation was a wall consisting of 2 cm thick stems of plants only 20 cm apart
at the time of monitoring). The Guatemalan authorities and the land owner were
interested in the results of the bird survey but because the area was a non-use
forest reserve, all cutting of vegetation was banned. The establishment of point
sites and transects was therefore a trade-off between scientific desire (random)
and conservation (using existing trails as much as possible and minimizing the
impact on the area). To diminish effects of detectability (Hines, 2006;
MacKenzie et al., 2002, 2003), all point counts/transects were visited three
times a year. The data were used to establish relative abundance of all bird
species and to determine presence of species (Magurran, 1988; Rosenzweig,
1995). Results showed that while more species were present in secondary
forest, all species of conservation concern were only present in natural forest
(Renner, 2003, 2005).
Case study 2 from cloud forests in the mountains of the tropical
Because of the steepness of the terrain, impenetrable vegetation and lack of
trails it was difficult to standardise the sampling and to obtain reliable bird
density data (Bibby et al., 2000). Rather than trying to get absolute quantitative
data from one or two study plots, the study aimed to obtain semi-quantitative
data for comparing the community composition of samples over several sites
Avian community data were obtained during transect walks. All visual and
acoustical records of birds within 50 m (Schieck, 1997) were noted while
walking very slowly and quietly through the terrain and as "randomly" (with
frequent changes in direction) as topography and vegetation permitted (Fjeldså,
1999; Herzog et al., 2002). Species accumulation curves level out rapidly (much
more so than with point sampling!) and high correlations between relative
species abundance data obtained this way and by point-counts in the same
area suggests that observations made during "random" walks are not
significantly more biased than those obtained by more standardised point
counts (Fjeldså, 1999). The main advantages of this "random-walk" approach
are the broad sampling of the study area, time-efficiency (all bird observations
being used, unlike in point counts; see below) and relative observer
independence (Sauer et al., 1994) compared to timed-species-count methods.
At each study site, data were collected within 1-1½ km² and over 2-4 days.
Variation in the extent of study plots is not of a magnitude that requires
adjustment for area differences. Study sites of this size will represent habitat
mosaics (of different associations of forest trees, tree-fall gaps, landslides and
glades) but the study plot was large enough to find most birds on the move,
singly or in mixed feeding parties. Walking speed varied (as the vegetation is
sometimes nearly impenetrable) but was usually ca. 500 m per hour. On
average, 0.4 birds were identified per minute. This rate could be raised by
walking faster, but the data will then be more biased towards easily detectable
The observed species richness is constrained by sample size, and for
comparison it is necessary to estimate species richness by extrapolation. Such
estimators reach their own asymptote much sooner than sample-based
rarefaction curves, they level off and approximate empirical asymptotes well.
The simplest approach (which can be applied currently, during field work) is to
use the Chao 1 formula (Colwell & Coddington, 1994):
S1 = Sobs + a2/2b (Sobs being the number of species recorded, a being the
number of singletons = number of species recorded only once, b the number of
A more sophisticated estimation can be done later using Colwell’s software
EstimateS (http://viceroy.eeb.uconn.edu/estimates). Ranked abundance curves
can be constructed from the total list of observed birds along the route,
assuming that the attentive observer is able to detect all birds (at least those
which are active) within 50 m from a transect. In most cases, about 500 bird
identifications will provide a good sample for describing the bird community
within a study plot.
John MacKinnon and Karen Phillips on birdwatching in forest:…” Watching birds
in tall forests is not easy. You may walk for an hour without seeing anything
then suddenly be surrounded by so many twittering birds that you cannot focus
on any. A bird may be so high up and so obscured by foliage that you cannot
get a good view. In the rain, water on your lenses may blur your vision…
leeches are an accepted irritation”…
We do sincerely thank Michel Louette (Royal Museum for Central Africa,
Tervuren, Belgium), Dieter Oschadleus (Animal Demography Unit, University of
Cape Town, South Africa), Olaf Jahn (Aves & Conservación, Corporación
Ornitológica del Ecuador), Richard Gregory (RSPB, England, EBCC), Andy
Musgrove and David Noble (BTO, Thetford, England) and Hans-Günther Bauer
(Max-Planck Institute for Ornithology and EBCC, Radolfzell, Germany) for
helpful discussions, ideas and comments on earlier drafts of this manuscript.
AMERICAN ACADEMY OF ORTHOPAEDIC SURGEONS (AAOS) 2008. Wilderness First
Aid. Jones and Bartlett Publishers. Boston, Toronto, London, Singapore: 369
BAKER, K. 1993. Identification Guide to European Non-Passerines. BTO Guide
24. Thetford: 332 pp.
BEARHOP, S., PHILLIPS, R.W., THOMPSON, D.R., WALDRON, S. & FURNESS, R.W.
2000. Variability in mercury concentrations of Great Skuas Catharacta skua: the
influence of colony, diet and trophic status inferred from stable isotope
signatures. Marine Ecology Progress Series 195: 261-268.
BIBBY, C., JONES, M. & MARSDEN, S. 1998. Expedition Field techniques: Bird
Surveys. Royal Geographical Society, London. [Online], Available:
http://biology.kenyon.edu/courses/biol229/fieldmanual%20birds.pdf [2009, Dec].
BIBBY, C.J., BURGESS, N.D., HILL, D.A. & MUSTOE, S.H. 2000. Bird Census
Techniques, 2nd edition, Academic Press, London: 277pp.
BIRDLIFE INTERNATIONAL 2000. Threatened Birds of the world. Lynx Edicions and
Birdlife International, Barcelona and Cambridge 852 pp.
BORROW, N. & DEMEY, R. 2004. Birds of Western Africa. Helm Field Guides,
London: 551 pp.
COLLAR, N. 2000. Opinion. Collecting and conservation: cause and effect. Bird
Conservation International 10(1): 1-15.
COLWELL, R.K. & CODDINGTON, J.A. 1994. Estimating terrestrial biodiversity
through extrapolation. Philosophical Transactions of the Royal Society (Series
B) 345: 101-118.
DANIELSEN, F., BURGESS, N.D. & BALMFORD, A. 2005. Monitoring matters:
examining the potential of locally-based approaches. Biodiversity and
Conservation 14: 2507-2542.
DANIELSEN, F., BURGESS, N.D., BALMFORD, A., FJELDSÅ, J., ANDRIANANDRASANA,
H.T., BECKER, C.D., BENNUM, L., BRASHARES, J.S., CHRISTIANSEN, S., DONALD,
P.G., EGUINO, S., ENGHOFF, M., FUNDER, M., GRAY, M., HÜBERTZ, H., JONES,
J.P.G., OETTING, I., POULSEN, M.K., RIJSOORT, J., STUART-HILL, G., TOPPJØRGENSEN, E., TOWNSEND, W.R., UYCHIAOCO, A.J., WHITTEN, T. & YONTEN, D.
2006. Monitoring matters: evaluating locally-based biodiversity monitoring in
developing countries. Oryx 40: 14-15.
DANIELSEN, F., MENDOZA, M.M., TAGTAG, A., ALVIOLA, P.A., BALETE, D.S., JENSEN,
A.E., ENGHOFF, M. & POULSEN, M.K. 2007. Increasing conservation management
action by involving local people in natural resource monitoring. Ambio 36: 566570.
DAWSON, A. 2005. Techniques in physiology and genetics. In : Sutherland, W.J.,
Newton, I. & Green R.E. (Eds). Bird Ecology and Conservation. A Handbook of
Techniques. Oxford University Press, Oxford: 211-231.
FJELDSÅ, J. 1999. The impact of human forest disturbance on the endemic
avifauna of the Udzungwa mountains, Tanzania. Bird Conservation International
GAUNT, A.S., ORING, L.W., ABLE, K.P., ANDERSON, D.W., BAPTISTA, L.F., BARLOW,
J.C. & WINGFIELD, J.C. 1999. Guidelines to the use of wild birds in research. The
Ornithological Council, Washington DC: 56 pp.
GIBBONS, D.W. & GREGORY, R.D. 2006. Birds. In : Sutherland W.J. (Ed.).
Ecological Census Techniques, 2nd Edition. Cambridge University Press,
GILBERT, G., GIBBONS, D.W. & EVANS, J. 1998. Bird Monitoring Methods: a
manual of techniques for key UK species. RSPB, Sandy: 464 pp.
GREGORY R.D, GIBBONS, D.W & DONALD, P.F. 2004. ‘Bird census and survey
techniques’. In : Sutherland, W.J., Newton, I. & Green R.E. (Eds). Bird Ecology
and Conservation. A Handbook of Techniques. Oxford University Press, Oxford:
HARRISON, C.J. & COWLES, G.S. 1970. Birds: Instructions for Collectors. 1st
edition. British Museum, London: 48 pp.
HERZOG, S.K, KESSLER, M. & CAHILL, T.M. 2002. Estimating species richness of
tropical bird communities from rapid assessment data. The Auk 119: 749-769.
HINES, J.E. 2006. PRESENCE2- Software to estimate patch occupancy and
related parameters. United States Geological Survey–Patuxent Wildlife
Research Center (USGS-PWRC). [Online], Available: http://www.mbrpwrc.usgs.gov/software/presence.html [2009, Dec].
JAHN, O. in press. Bird communities of the Ecuadorian Chocó: a case study in
conservation. Bonner Zoologische Monographien 56.
KÉRY, M. 2008. Estimating abundance from bird counts: Binomial Mixture
Models uncover complex covariate relationships. The Auk 125(2): 336-345.
LAAKE, J.L., BUCKLAND, S.T., ANDERSON, D.R. & BURNHAM, K.P. 1994. Distance
User’s guide Version 2.1. Colorado Cooperative Fish and Wildlife Research
Unit, Colorado State University, Fort Collins: 84 pp.
LOWEN, J.C., BARTRINA, L., CLAY, R. & TOBIAS, J. 1996. Biological surveys and
conservation priorities in eastern Paraguay. CSB Conservation Publications,
Cambridge: 186 pp.
MACKENZIE, D.I., NICHOLS, J.D., LACHMAN, G.B., DROEGE, S., ROYLE, J.A. &
LANGTIMM, C.A. 2002. Estimating site occupancy rates when detection
probabilities are less than one. Ecology 83: 2248-2255.
MACKENZIE, D.I., NICHOLS, J.D., HINES, J.E., KNUTSON, M.G. & FRANKLIN, A.B.
2003. Estimating site occupancy, colonization and local extinction when a
species is detected imperfectly. Ecology 84: 2200-2207.
MACKINNON, J. & PHILLIPS, K. 1993. A Field Guide to the Birds of Sumatra, Java
and Bali. Oxford University Press, Oxford: 491 pp.
MAGURRAN, A.E. 1988. Ecological diversity and its measurement. Princeton
University Press, Princeton: 179 pp.
MERRY, W. 1994. The Official Wilderness First Aid Guide. McClelland & Stewart,
Toronto: 408 pp.
NG’WENO, F. 2008. Pitfalls in ethno-ornithological research. In : Harebottle D.M.,
Craig A.J.F.K., Anderson M.D., Rakotomanana H. & Muchai M. (Eds).
Proceedings of the 12th Pan-African Ornithological Congress. Animal
Demography Unit, Cape Town: 44-45.
PIECHOCKI, R. & ALTNER, H.-J. 1998. Makroskopische Präparationstechnik. Teil
I, Wirbeltiere. Verlag Gustav Fischer, Jena, Stuttgart, Lübeck, Ulm: 461 pp.
POMEROY, D. & TENGECHO, B. 1986. Studies of birds in a semi-arid area of
Kenya. III – The use of “timed species-counts” for studying regional avifaunas.
Journal of Tropical Ecology 2: 231-247.
POULSEN, B.O. 1994. Mist-netting as a census method for determining species
richness and abundances in an Andean cloud forest bird community. Le Gerfaut
POULSEN, B.O. 1996. Structure, dynamics and home-range of two mixed
species flocks and species composition in a montane Alder dominated
secondary forest in Ecuador. Journal of Avian Biology 12: 333-343.
POULSEN, B.O., KRABBE, N., FRØLANDER, A., HINOJOSA, M.B., & QUIROGA, C.I.
1997. A rapid assessment of Bolivian and Ecuadorian montane avifaunas using
20-species lists: efficiency, biases and data gathered. Bird Conservation
International 7: 53-67.
PRESTON, F.W. 1948. The commoness and rarity of species. Ecology 29(3):
RASMUSSEN, P.C. & ANDERTON, J.C. 2005. Birds of South Asia. The Ripley
Guide. Vols. 1 and 2. Smithsonian Institution and Lynx Edicions, Washington,
D.C. and Barcelona: 378 & 685 pp.
REDFERN, P.F. & CLARK, J.A. 2007. Ringer’s manual. BTO, Thetford: 270 pp.
REMSEN, J.V. 1994. Use and misuse of bird lists in community ecology and
conservation. The Auk 111: 225-227.
REMSEN, J.V. 1995. The importance of continued collecting of bird specimens to
ornithology and conservation. Bird Conservation International 5: 145-180.
REMSEN, J.V. & GOOD, D.A. 1996. Misuse of data from mist-net captures to
assess relative abundance in bird populations. The Auk 113: 381-398.
RENNER, S.C. 2005. The Resplendent Quetzal (Pharomachrus mocinno) in the
Sierra Yalijux, Alta Verapaz, Guatemala. Journal of Ornithology 146:79-84.
RENNER, S.C. 2003. Structure and diversity of cloud forest bird communities in
Alta Verapaz, Guatemala, and implications for conservation. Niedersächsische
http://webdoc.sub.gwdg.de/diss/2003/renner/renner.pdf [2009, Dec].
ROBERTS, D.L., COOPER, R.J. & PETIT, L.J. 2000. Flock characteristics of antfollowing birds in premontane moist forest and coffee agroecosystems.
Ecological Applications 10(5): 1414-1425.
ROBERTSON, H. & HEATHER, B. 2004. The Hand Guide to the Birds of New
Zealand. Oxford University Press, London: 168 pp.
ROBINSON, W.D., BRAWN, J.D. & ROBINSON, S.K. 2000. Forest bird community
structure in central Panama: influence of spatial scale and biogeography.
Ecological Monographs 70: 209-235.
ROSENZWEIG, M.L. 1995. Species diversity in space and time. Cambridge
University Press, Cambridge: 436 pp.
SAUER, J.R., PETERJOHN, B.G. & LINK, W.A. 1994. Observer differences in the
North American Breeding Bird Survey. The Auk 111: 50-62.
SCHIECK, J. 1997. Biased detection of bird vocalizations affects comparisons of
bird abundance among forested habitats. Condor 99: 179-190.
SVENSSON, L. 1992. Identification Guide to European Passerines. 4th edition,
revised and enlarged, self-published, Stockholm: 367 pp.
TERBORGH, J., ROBINSON, S.K., PARKER, T.A., MUNN, C. & PIERPONT, N. 1990.
Structure and organization of an Amazon forest bird community. Ecological
Monographs 60: 213-238.
VALLELY, A.C. 2001. Foraging at army ant swarms by fifty bird species in the
highlands of Costa Rica. Ornitologia Neotropical 12: 271–275.
VOŘÍŠEK, P., KLVAŇOVÁ, A., WOTTON, S. & GREGORY, R.D. 2008. A Best Practice
Guide For Wild Bird Monitoring Schemes. Royal Society for the protection of
Birds, Sandy: 150 pp.
WAGSTAFFE, R. & FIDLER, J.H. 1968. Preservation of Natural History Specimens.
Littlehampton Book Services Ltd., vol.2, part 2: Zoology: Vertebrates. Witherby,
Philosophical Library of London: 115 pp.
WASSENAAR, L.I. & HOBSON, K.A. 2001. A stable-isotope approach to delineate
catchment areas of avian migration monitoring stations in North America.
Environmental Science and Technology 35: 1845-1850.
WERNER, D. 1979. Where there is no doctor. A village health care handbook for
Africa. McMillan, Oxford: 440 pp.
WINK, M. 2006. Use of DNA markers to study bird migration. Journal of
Ornithology 147: 234-44.
WINKER, K. 2000. Obtaining, preserving, and preparing birds. Journal of Field
Ornithology 71: 250-297.
Useful Internet links
http://olla.berkeley.edu/ornisnet/ (American Bird collections, ORNIS)
Online access to bird collection data.
South-African Bird Atlas.
Guidelines to the use of wild birds in research.
European Bird census council.
http://www.riteintherain.com (Darling Corporation)
Data analysis of capture-recapture data.
Voříšek et al., 2008. A Best Practice Guide for Wild Bird Monitoring Schemes.
Sutherland et al., 2004. Ecological Census Techniques, see section on “Gratis