Many species occur in more than one habitat and their trends can be different in those different habitats. To produce indicators for a particular habitat type, PECBMS includes the species that were classified as characteristic for that habitat and then uses all count data for those species. Wouldn´t it be better to use only the data from the habitat type under concern?
In an ideal world the answer is probably yes, but there are practical barriers to achieving this at a European scale in the near future, and you could argue that using all the count data for an individual species is preferable because it provides a more robust estimate of trend. The assumption behind our procedure to use all count data is that population trend of species selected as characteristic for a particular habitat indeed reflect changes in that habitat because the vast majority of its population uses that habitat for breeding or feeding. The alternative approach of only taking trend data from farmland plots for farmland birds, or forest plots for forest birds, is attractive because there is a direct link between the birds and the habitat, but producing habitat-specific trends requires high quality computerised data on habitat and birds, and this is not currently possible in a number of countries. So far, experience at the national level, e.g. from UK, where habitat-specific indicators have been produced too, shows that both approaches can give very similar results. This is a finding now echoed in other European countries too. So the UK has decided not to use habitat-specific indicators in part because the trends were so similar to those using all count data, and in part because it wishes to maintain continuity in the pre-established indicator series (and not keep ´moving the goalposts´). However, it would be worthwhile exploring this issue further and the PECBMS plans to do so in the future. This is certainly possible as information on habitat type at sample plots is routinely collected by nearly all national monitoring schemes. The challenge is to combine the bird and habitat data in each country and produce habitat-specific trends appropriate for each species.