Accredited official statistics

Wild bird populations in the UK: Frequently asked questions

Updated 12 November 2024

Frequently asked questions

Why monitor bird populations?

Bird populations have long been considered to provide a good indication of the broad state of wildlife in the UK and England. This is because they occupy a wide range of habitats and respond to environmental pressures that also operate on other groups of wildlife. In addition, there are considerable long-term data on trends in bird populations, allowing for comparisons between the short term and long term. Because they are a well-studied taxonomic group, drivers of change for birds are better understood than for other species groups, which enables better interpretation of any observed changes.

Which species of birds are included in the indices?

There are 6 main habitat/species groups representing farmland, woodland, water and wetland, seabird and upland habitats and wintering waterbirds. Species within each habitat type are included in the indices if they have a population of at least 500 breeding pairs (300 in England) and are common birds that are native to, and breed in England; the trends included refer to breeding populations (except for those used in the winter waterbird indicator). There is also an all-species indicator where all of the species included in the farmland, woodland, wetland and seabird indicators are included here, as well as species associated with urban habitats (for example, collared dove, house martin), with heathlands (for example, Dartford warbler, hobby) and species that occupy a range of habitats (for example, peregrine, black-headed gull). There are a few species in the upland indicator that have not yet been added to the all-species indicator (for example, golden eagle, merlin, black grouse) and these will be considered for addition following a consultation.

What are ‘unsmoothed’ and ‘smoothed’ indices and why are they used?

Two trends are referred to in the text: the unsmoothed indices show year-to-year fluctuation in populations, reflecting the observed changes in the survey results, and smoothed trends, which are used to formally assess the statistical significance of change over time. Smoothed trends are used for both long and short-term assessments as they reduce the short-term peaks and troughs resulting from, for example, year-to-year weather and sampling variations.

What is the difference between a ‘long-term’ and ‘short-term’ trend?

There are no differences between the way the trends are calculated as they both use the smoothed indices, it is only that in most cases the long-term trend goes back to the earliest data point, usually 1970, whereas the short-term trend looks at the most recent 5 years.

Why should users not place too much reliance on short-term changes from one year to another?

Looking at the data from year to year may include weather effect and sampling variations which would not provide reliable data as the trend is not over a longer time frame.

Why use indices rather than absolute numbers?

It is not possible to determine changes in the actual numbers of birds for each species in the UK and England each year, however it is possible to estimate the relative change, from counts on sample plots surveyed as part of a range of national monitoring schemes.

How are the individual species indices combined into a single indicator and why is it done?

The creation of the all-species wild bird indicator involves two steps: (1) the production of annual population indices for the individual species for which there is trend data, and (2) the amalgamation of these individual indices into a single aggregate index.

1. Indices for individual species: These are generated by a statistical analysis of representative sites resurveyed year after year (for example, in the Breeding Bird Survey) or based on annual or periodic estimates of total populations (for example, the Heronries Survey). The population trends for each species are made comparable by expressing them as indices relative to ‘100’ in the start year. Thus, each annual index shows relative changes in population size from the start year: a rise to 200 in the index reflects a doubling in numbers, a decline to 50 a halving.

2. Amalgamating into a single index: The all-species index is calculated as the geometric mean of all the individual indices, with no weightings - so each species has the same relative effect on the indicator. The geometric mean is used to ensure that a doubling in the population index of one species (for example, 100 to 200) is balanced by a halving (for example, 100 to 50). The geometric mean of 200 and 50 is 100.

The composite all species indicator shows the year-to-year fluctuations in population trends across all species that can be included, reflecting the observed changes in the annual survey results. Alongside this is the smoothed version of the trend, which is used to formally assess the statistical significance of change over time. The smoothed trend is derived using a published statistical methodology and is used for assessments as it reduces the short-term peaks and troughs resulting from, for example, year-to-year impacts of weather and sampling variations. The index is considered to give reliable medium to long-term trends but strong reliance should not be attached to short term changes from one year to the next.

What is the ‘geometric mean’, how is it calculated and why is it used?

The geometric mean is an average. It is calculated by multiplying a set of index values and taking the nth root, where n is the number of index values. The geometric mean is used to ensure that a doubling in the population index of one species (for example, 100 to 200) is balanced by a halving (for example, 100 to 50). It allows for each species to have the same relative effect on the indicator.

What does ‘modelling count data’ mean?

For those species covered by the Common Bird Census (CBC) and Breeding Bird Survey (BBS), all data from survey sites are used in the generation of trends, regardless of the habitat at these sites (for example, survey data from woodland sites is used in the trends for farmland species and vice versa). Trends are generated from the two data sources using the joint-model methods described by Noble et al. (2003a). Generalized linear models (GLMs) are used, with the application of a post-hoc smoothing spline to produce smoothed indices for each species, thereby removing short-term fluctuations that may be caused by sampling error, or minor fluctuations due to weather effects, for example. Such smoothing does however mean that the estimates for the final year of a trend must be treated with caution as they lack the smoothing effect of data in subsequent years. The nature of smoothed trends, in that data from any given year has an impact on trend values for earlier (and later) years means that existing species indices (and hence indicator) values will be different in subsequent annual revisions.

Data from the CBC and BBS are combined and analysed statistically in a single GLM with site and year effects, as described at BTO’s BirdTrends 2013. Equal weight is given to CBC and BBS sites by assigning each CBC site the mean of the BBS site weighting. Confidence limits on these species trends are generated by bootstrapping; repeated resampling (with replacement) to generate a sample of estimated trend values, with the 2.5% and 97.5% percentiles giving the 95% confidence limits around the trend value for each year.

What are ‘confidence intervals’ and why are they used?

The trends in this publication are based on estimates from surveys. Smoothed trends are presented with 95% confidence intervals (CI), which are a measure of the precision of these survey estimates. While the exact value for an indicator in a particular year cannot be known, a 95% CI means users can be 95% confident that the true value of the indicator falls within the confidence interval around it.

Why don’t wintering waterbirds and seabirds have confidence intervals?

Data from wintering waterbirds monitoring schemes are based largely on full counts at colonies or at wetland and coastal sites of markedly varying size. This means that bootstrapping methods cannot be applied reliably and hence trends for these groups are currently presented without confidence intervals.

There are no confidence intervals available for seabird indicators for the 2024 publication, pending consultation on a recent thorough review of the analytical methods used in the Seabird Monitoring Programme (SMP) species population indices, including confidence estimation. New analytical approaches will be developed and tested, and when available, the resulting trends will be incorporated into the aggregate seabird indicators. Trends should be interpreted with care due to the relatively small number of species included in some indices and the scarcity of data for some constituent species.

Why are the confidence intervals for some indices such as ‘breeding generalist farmland birds’ so much wider than they are for other indices such as ‘breeding generalist woodland birds’ and what is it indicating?

The size of confidence intervals (CIs) varies among habitat indicators because their width is influenced by the number of species in each indicator and the precision of the individual species trends that make up the indicator. The precision of these trends varies due to differences in sample size. More specifically, the CIs for generalist farmland birds indicator is wider than the CIs for generalist woodland birds indicator because there are only 7 individual species in the former indicator, whereas there are 12 species in the latter. Wider confidence intervals imply that the range of values within which users can be confident the true indicator value falls is greater than it would be for a narrower confidence interval. Therefore, users can be 95% confident that the true value of the breeding generalist woodland birds indicator falls within a narrower range of values than the true value of the breeding generalist farmland birds indicator.

Why are the confidence intervals for the England all species indicator wider than they are for the UK version?

The statistical approach to generating confidence intervals used in indicators with a large and diverse suite of species can be sensitive to individual species with sparse data in periods of the time series. This can be seen in the widening of confidence intervals in the England indicator. The species composition of this indicator will be reviewed, and a sensitivity analysis performed to determine whether these species should be retained and to carefully consider what the impact of their removal would be.

Why is the percentage change used to define a ‘weak increase’ different to that used to define a ‘weak decline’?

Asymmetric percentage change thresholds are used to define these classes as they refer to proportional change, where a doubling of a species index (an increase of 100%) is counterbalanced by a halving (a decrease of 50%).

What is ‘bootstrapping’ and why is it used?

Bootstrapping, a statistical method that estimates the uncertainty in a trend through repeated re-sampling and trend estimation. Confidence intervals for the estimated trends are calculated from percentiles (such as 2.5% and 97.5%) of the sample of estimated trend lines. The procedures currently used provide a measure, and level of confidence, of the indicator value in any particular year (for example, 2014) relative to the start year of the time series. It is feasible to calculate the change, and confidence in that change, for any other time period (for example, over five years from 2009 to 2014) but this requires all of the constituent population trends and their bootstraps to be recalculated using the start year (for example, 2009) as a baseline.

Why does the overall index for breeding wild birds in England not appear to reflect the large reductions in the population of farmland birds during the 1980s and why did the population of farmland birds suffer so much during that time?

The overall ‘all-species’ index is comprised of 130 species of birds in the UK and 118 in England, all widespread species with populations of at least 500 breeding pairs (300 in England) for which there are sufficient data. It represents 6 habitat types (including farmland birds) and a further 26 generalist species with no particular habitat preference. Species trends within this index vary widely, from species increasing several-fold (for example, buzzard, Cetti’s warbler, great spotted woodpecker, and collared dove) to those having declined to less than a tenth of their 1970 numbers (turtle dove, tree sparrow, willow tit and grey partridge). Within the UK index, for example, 28% of the 130 species increased, 42% showed little change and 30% declined between 1970 and 2023. The apparent declines in the farmland birds indicator (comprising of 19 species, so just 15% of the total species in the all-species indicator) are being mitigated by increases in some of the other species in the indicator.

The large declines in the abundance of many farmland birds have many known and potential causes. For a large part, declines have been caused by the changes in farming practices that have taken place since the 1950s and 60s, such as the loss of mixed farming, a move from spring to autumn sowing of arable crops, change in grassland management (for example, a switch from hay to silage production), increased pesticide and fertiliser use, and the removal of non-cropped features such as hedgerows. The rate of these changes, which resulted in the loss of suitable nesting and suitable feeding habitats, and a reduction in available food, was greatest during the late 1970s and early 1980s, the period during which many farmland bird populations declined most rapidly.

What are the main differences between ‘generalist’ and ‘specialist’ bird populations and why have they fared so differently since the 1970s?

Specialist bird species are considered to be largely or wholly dependent on one particular habitat whereas generalist bird species utilise a wider range of habitats. Specialist bird species may find it difficult to adapt to changes in their habitats while generalist bird species may find it easier to adapt because they are not as reliant on one specific habitat type. For example, the woodland bird index is comprised of trends for 37 species. Between 1970 and 2017, the index for woodland specialists which are highly dependent on woodland habitats declined by 46% while the index for woodland generalists, many of which have adapted to using gardens and wooded areas in farmland including woodland, showed a 14% increase.

Are species of birds included in the trends if they are introduced in subsequent years?

Yes. Species of birds which are introduced into the indices in later years will be included and trends calculated.