BTO Research Report No. 321
Preliminary analyses of Breeding Bird Survey (BBS) mammal data
Newson, S.E. & Noble, D.G.
In 2001, a group of organizations interested in the surveillance and monitoring of mammal populations in the UK met to discuss priorities and possible approaches. This group, currently informally known as the UK Mammal Network, identified a number of schemes (currently in operation or in the planning) stage that could be integral components of a UK-wide mammal monitoring strategy. One of the schemes is the multi-species mammal monitoring carried out by many participants in the BTO/RSPB/JNCC Breeding Bird Survey (BBS). Although summaries of the mammal component are regularly reported in the annual BBS reports (see Noble et al. 2001), there have, to date, been relatively few analyses of these data or attempts to identify population trends. This report provides an overview of the BBS mammal data collected to date, presents some preliminary population trends for mammal species monitored in sufficient numbers, and outlines approaches for future analyses and reporting of these data. This is NOT intended to be an exhaustive report on UK mammal trends as revealed by the BBS. Clearly, there is considerable scope for more comprehensive examination of these data, but that would require much greater effort than was possible in this contract.


Executive Summary

1. In 1995 the scope of the BTO/RSPB/JNCC Breeding Bird Survey (BBS) was expanded to also record British mammals. This was the first multi-species, annual mammal survey to be carried out in the UK, with the focus on medium to large-sized easily identifiable species, although observers can record any mammal species seen or known to be present. Summaries of these data are regularly reported through the annual BBS reports, although there have been few analyses of these data or attempts to calculate population trends. In this report, we assess the effectiveness of the BBS for monitoring UK mammal populations and of detecting significant changes in abundance or distribution..

2. Preliminary analyses of BBS mammal data for 1995-2000 demonstrates that national populations of Brown Hare, Mountain/Irish Hare, Grey Squirrel, Red Fox, Red Deer, Fallow Deer, Roe Deer, Reeve's Muntjac and Rabbit can be monitored by this survey. Of these, the Roe Deer increased significantly over this period, whilst all remaining species (with the exception of Reeve's Muntjac), exhibited significant inter-year variation during this period. As expected, the BBS field methodology is not effective for monitoring bats, most mustelids, small mammals, and cetaceans and rare or highly localised species (e.g. Red Squirrel, Wildcat or Chinese Water Deer).
3. For a number of species, there are insufficient data to calculate reliable indices of abundance, but a large amount of indirect information on presence/absence from field signs or local knowledge of their presence in that year. For example, Badger, Mole, Hedgehog and Brown Rat. For these species, the change in presence on BBS was modelled to see whether there were sufficient data to produce reliable indices of abundance. Whilst the sample was adequate to do this, there were a number of problems with the resulting indices. Indices calculated for the first year or so may reflect awareness by the observer of the presence of the species and changes in the survey form may explain an apparent increase in these species in 1998. However, this is a method that may become more useful as data from further years are collected. This method may also be most appropriate for herding deer species, such as Red, Fallow and Sika Deer for which there was a large variance associated with indices based on counts and where modelling presence/absence may provide a more accurate means of monitoring change in their populations.
4. As with many analyses there is a conflict between maximising the sample size required to identify change with confidence and narrowing down the area to understand the exact pattern of change. In this study, we examine the use of count data to compare population trends for five species for which there was sufficient data (Brown Hare, Rabbit, Grey Squirrel, Red Fox and Roe Deer) within three broad regions. Of these, the Brown Hare showed a significantly larger decline in the southwest than in the north and southeast of Britain. Further analyses could examine pair-wise differences between regions for more species where the sample size in one region is limiting and additional analyses could change the boundaries of regions to better understand trends in different geographic areas.
5. Whilst data for a large proportion of species recorded by the BBS are insufficient to calculate indices of abundance or presence/absence, these data do provide important information on their distribution and relative abundance. Also, because observers record habitat information, we may be able to identify the habitat requirements of these species – essential for conservation initiatives and useful for planning of further targeted studies.
6. As mentioned above, the collection of habitat data on BBS squares is important for our understanding of species population trends. Trends may be unrelated to habitat, but could be directly related to a particular change in a specific habitat in which they occur. In this study, we calculate separate habitat-specific indices for four species (Brown Hare, Rabbit, Grey Squirrel and Red Fox) in two or more dominant habitats. These suggest that Brown Hare declined on farmland between 1995 and 2000, whilst Rabbit also declined on farmland, but increased on grassland over this period, although abundance fluctuated widely between years. Abundance of Grey Squirrel also fluctuated over this period, whilst the Red Fox declined in urban areas but increased on farmland.
Download in sections:
You will need the free Adobe Acrobat Reader to view this document.
Please cite as: Newson, S.E. & Noble, D.G., (2003), BTO Research Report No. 321