All estimates refer to the total number of individuals in the population, including immature birds,
and are the most recent estimates available. In most cases, the estimates given in the tables are
taken directly from the source references. However, all estimates have been rounded to a maximum
of three significant figures, and rough estimates, particularly those given as a broad range,
have been rounded to two significant figures.
Most population estimates included in this edition have been derived from censuses made
towards the end of the non-breeding season or from estimations of breeding pairs.
Waterbird populations tend to be at their lowest and most stable at these times.
Individual numbers usually peak after the breeding season due to first year
recruitment and suffer high and variable mortality over the non-breeding
season making these times unsuitable for population estimation. To allow
for the element of immature or non-breeding birds in each population,
estimates given by original sources in the form of number of breeding
pairs (or males or females if only these were counted) have been multiplied
by three to give the total population size, as suggested by Meininger et al. (1995).
Estimates given in the form of breeding adults or mature individuals (i.e. twice
the number of breeding pairs) have been multiplied by a factor of 1.5. There is
no intentional overlap between the populations of a species, and therefore all
estimates for the populations of a species can be summed to produce a species total.
Letter codes applied in earlier editions are replaced by numeric ranges. The range
1,000,000 – 1,000,001 individuals shall be interpreted as being equivalent to the
former letter code E, i.e. more than 1 million individuals. Similarly for instances
where no upper estimate is given, the lower estimate + 1 is provided e.g. >25,000 is
presented as 25,000-25,001. In cases where there is no lower estimate available, the
range given is 1 to upper estimate e.g. 1-10,000.
Start & End Year
All population size estimates require a start and end year that determines the period when the
estimate was calculated. The methods/means by which the estimate is calculated may vary,
so the following conventions were used:
• Same start and end year, if the whole population has been censused in the same year. If the source of information is a boreal winter census (Dec-Jan) and covers two years, add the first year as the start and the next year as end year.
• Start and end year different, if the population has been censused/estimated in different years during a longer period. The start year will be year in which the first part of the population was censused/estimated and the end year will be year in which the last part of the population was censused/estimated.
• If the estimate comes from a publication but it is not clear when the estimate was made, use the year of the publication and subtract 1 year for both start and end year.
• If the estimate comes from a number of publications or from a review of publications but it is not clear when the estimate was made, use the year of the earliest publication – 1 as start year and the year of the last publication – 1 as the end year
Minimum & Maximum
All populations are given number to represent a min and max estimate.
The system of use of Coded Ranges used in previous editions of the WPE
has been discontinued. For some populations the same value has been entered for min
and max, but this should be avoided where possible, unless the estimate quality
is accurate down to the individual bird. This may also be applied if a statistically
adequate census is undertaken, although even these include errors of estimate.
For extinct species, a zero is entered as min and max. A minimum and maximum of -1
indicates that the size of the population is unknown.
From CSR5 and WPE5 onwards, four categories of quality assessments are used:
1. No estimate: No population estimate is available;
2. Best guess: Population estimate is only possible with large or uncertain ranges;
3. Expert opinion: Population estimate is based on incomplete survey and monitoring data and population size has been developed employing some expert opinion for extrapolating from this data with more accuracy than a best guess;
4. Census based: Population estimate is based on almost complete census or statistically adequate sampling.
References for population estimates
In cases where numerous sources have been used to calculate the estimate, all sources are given, and explanation is provided in the Notes.
More information concerning how the population size estimate was derived from the available sources is provided here if necessary.