WeBS Alerts methods - Interpretation of Accounts

The Alerts report can be navigated in two principal ways, either by selecting from a list of protected sites or by selecting from a list of species.

Selection by site

Selecting a given site returns a 'Table of Alerts' for all species listed as features on the site in question.  A given site may be considered important for a species for the local breeding population, numbers that visit the site during passage periods or over-wintering numbers. Breeding numbers are not monitored by WeBS and the way in which numbers recorded by WeBS Core counts relate to numbers using a site during passage is unresearched. Consequently, only over-wintering numbers are assessed for Alert status. In the future it may be possible to assess the Alert status of passage numbers once turnover of individuals at sites is better understood.

Summaries of the trends in waterbird numbers are provided through a drop-down species list at the foot of the 'Table of Alerts'. The aim here is not to suggest what factors explain the numeric changes in bird numbers that have caused Alerts to be triggered but to highlight cases that warrant further investigation.

Plots of trends are available for all evaluated species. These plots show the annual index values and fitted trend over time upon which Alert status is assessed. The index for the base (most recent) winter is scaled to 100 and the reference winter for alert assessment is the penultimate winter. In addition to trend plots for the site in question, regional and / or national trend is also given to provide context. In this regard, for sites in Northern Ireland both the regional and wider context is provided by the Northern Ireland trend. For sites in Scotland Wales and England, the wider perspective is provided by the trend for Great Britain while the regional context is provided by trends based on the old Environment Agency (EA) regions and Scottish Environmental Protection Agency (SEPA) regions in which the site is found. For sites that straddle these regions (Humber Flats, Marshes and Coast; Severn Estuary; Thames Estuary and marshes; The Dee Estuary; Upper Solway Flats and Marshes) a trend based on all monitored sites within the two adjoining regions are used. In practise, and with the exception of the Upper Solway Flats and Marshes, where the appropriate SEPA region is used in conjunction with the North West England region, it has been necessary to use all of Scotland for region context for Scottish sites because of the relative low number of sites monitored by WeBS in many areas within Scotland.

The total count for a given site will often be derived from amalgamating counts from smaller parts of that site. On any one occasion counts for one or more of these parts may be unavailable. It is standard WeBS practise to assess the impact of missing coverage separately for each site / species combination. Where those parts of a site not counted or for which the count has been flagged as an undercount on a given occasion would typically hold 25% or more of the overall site total the amalgamated count is flagged as incomplete. When calculating indices, such counts are replaced by imputed values where they are lower than expected (Underhill & Pr?s-Jones 1994). Sometimes the calculation of an Alert may involve a comparison between winters, at least one of which has a value based on a high proportion of imputed data. Where data for adjacent winters are relatively complete this is not a serious concern because of the smoothing technique used. However, where data for a number of consecutive winters rely heavily on imputed data then the resulting Alert could be considered less reliable. To aid the identification of these potentially less reliable Alerts, the index values associated with the smooth trend (on which the Alert status is calculated) are identified by by red symbols rather than green symbols where 50% or more of the index value results from imputed bird numbers.

Comparisons of site trends with those of the regional or national trends help to focus attention on where to seek an explanation. When the site trend reflects trends occurring throughout the country then the decline is less likely to be due to local factors that can be addressed by sympathetic site management, but more likely to be due to wide scale processes such as climate change. Similarly, when the site trend reflects the regional trend then broad-scale explanations may be sought, for example general improvements in water quality throughout the region, redistribution of the species within the United Kingdom due to climate change or buffer effects as numbers nationally increase or decline. However, where observed declines at a site are sufficient to trigger an Alert but the numbers in the region or country are stable or increasing then it is appropriate to seek local explanations such as habitat loss or degradation, local improvements to water quality or increased disturbance.

These comparisons are provided by a plot derived from a Generalised Linear Model in which the number of birds wintering at a site as a proportion of the overall numbers wintering in the region or country-wide is modelled as a function of winter. In this way, the proportion of the region’s birds occurring at a given site is calculated across years. The 95% confidence limits obtained encapsulate several facets of uncertainty. Firstly, they encapsulate confidence in the calculation of the proportion as it varies with the total number of birds in the region / country (e.g. we can be statistically more confident that a site truly holds 10% of the regional total based on a site count of 100 birds and a regional count of 1000 birds than based on a count of 10 birds on a site and 100 in the region). Secondly, for the majority of species (those that are indexed using data from more than one month) the confidence limits also encapsulate between-month variation in the average proportion a site holds in a given winter. Monthly values for a given winter are treated as repeat observations on the premise that those months chosen for indexing / alerts are those when the numbers are relatively stable. (for full details with examples see Banks and Austin 2004).

Selection by species

Selecting a given species returns a 'Table of Alerts' which reports the Alerts assessment for Great Britain and the constituent countries of the UK. At the national level, trends are considered for each of Northern Ireland, Wales, England and Scotland. A wider perspective is provided by consideration of trends within Great Britain as a whole. Trends for the United Kingdom as a whole are not considered because generally the time series of WeBS Core Counts for sites in Northern Ireland is shorter than for sites in Great Britain and so the degree of impution would be higher than would be desireable.  However, this will be less of an issue by the time the next full release of WeBS Alerts and the assessment of UK trends will be reconsidered then.  There is also an assessment of the overall trend across the UK SPA Suite for the species in question.  This has been split into sites within Northern Ireland and sites within Great Britain for the reasons just given.  Alerts for all individual sites in the SPA Suite for the species in question are also given. Accounts for the SPA Suite for each species follows the same pattern as used for site accounts showing the trend across the SPA Suite for each species in question, the country-wide trend and the comparison of the SPA suite as a proportion of the country-wide total.


WeBS Alerts citation

This report should be cited as: Cook, A.S.C.P., Barimore, C., Holt, C.A., Read, W.J. & Austin, G.E. (2013). Wetland Bird Survey Alerts 2009/2010: Changes in numbers of wintering waterbirds in the Constituent Countries of the United Kingdom, Special Protection Areas (SPAs) and Sites of Special Scientific Interest (SSSIs). BTO Research Report 641. BTO, Thetford. http://www.bto.org/webs-alerts

WeBS is a partnership between the British Trust for Ornithology, the Royal Society for the Protection of Birds and the Joint Nature Conservation Committee (the last on behalf of the statutory nature conservation bodies: Natural England, Natural Resources Wales and Scottish Natural Heritage and the Department of Agriculture, Environment and Rural Affairs, Northern Ireland) in association with the Wildfowl and Wetlands Trust. Email: webs [at] bto.org.