Farmland birds have a well documented record of recent population declines and range contractions. Defra has adopted a Public Service Agreement target to reverse the long-term decline in farmland birds by 2020, as measured by the population trends of 20 farmland bird species.
For widespread but declining species, agri-environment schemes provide a mechanism for simultaneously reversing declines of a range of species over a large area. In 1998, MAFF initiated the Pilot Arable Stewardship Scheme, which was designed to test the efficacy of a range of management prescriptions in delivering farmland wildlife. MAFF ran a monitoring scheme, from 1998 to 2001, to test the delivery of these options for a range of taxa. While plants and invertebrates showed many positive responses to the scheme, there were few effects on birds, possibly because the monitoring occurred too soon for birds to have responded. A resurvey, 5 years after the Scheme started, is needed to give a more robust picture of the effects of the scheme on birds and so to provide a basis for management and policy advice. This will have two main elements: Module 1 will repeat the breeding bird surveys of the original monitoring project, allowing us to determine whether any effects of AS on breeding birds can be detected at the farm-scale. Module 2 will repeat the winter bird surveys of the original monitoring project, allowing us to determine whether there are more farm-scale effects on winter birds.
The findings will provide a rigorous scientific basis for recommendations relating to the further development and implementation of arable agri-environment measures. The research may also provide a critical advance towards quantifying the scale at which resource provision is required to achieve recovery of farmland bird populations.
The project will benefit from additional funding from EN and `in kind` support from RSPB.
1. To carry out a program of monitoring of winter bird populations on pilot Arable Stewardship farms and control farms in winter 2002-3, to determine whether winter bird populations have responded positively to AS, relative to a baseline on non-AS farms. This will be achieved by comparison between AS and control sites in 2002-3 and between the change in numbers between 1998-9 and 2002-3 on AS and control sites.
2. To carry out a program of monitoring of breeding bird populations on pilot Arable Stewardship farms and control farms in summer 2003, to determine whether breeding bird populations have responded positively to AS, relative to a baseline on non-AS farms. This will be achieved by comparison between AS and control sites in 2003 and between the change in numbers between 1999 and 2003 on AS and control sites.
As with the original monitoring scheme, assessment of the effect of Arable Stewardship on birds will be made at the whole-farm scale and will be achieved by comparison of responses on farms in the scheme with those on ‘control’ farms which had not entered the scheme.
WINTER BIRDS: ESTIMATION OF POPULATION DENSITIES ON AGREEMENT AND CONTROL FARMS
In each pilot area, 20 scheme and 20 control farms will be censused twice in winter 2002-3, once between 1st October and 31st December and once between 1st January and 31st March. Methods will be as for the original monitoring project. Each site-census will be completed in a single day. On each site visit, counts will be made of all birds feeding in (or ‘hunting over’, in the case of raptors) each field. The observer will walk a predetermined number of transects across each field (walked along the longest axis), based on 50m separation, in order to flush all birds in the field (Wilson et al. 1996; Perkins et al. 2000). This method achieves complete counts of birds except where the vegetation within a field is particularly tall and dense. In these cases, closer transects will be used. Some species tend to use only the field edge, so the two outer transects will be within 10m of the field boundary (Perkins et al. 2000). Double-counting of birds will be minimised by observers counting birds flushed to other fields or to other parts of the field being censused, and ignoring them on subsequent encounter (Wilson et al. 1996; Perkins et al. 2000). Censuses will start at least one hour after sunrise and be completed at least one hour before sunset, to avoid missing birds because they are yet to arrive from, or have already left for, roost sites. Censuses will not be conducted in periods of heavy rain, strong wind or poor visibility and census routes will be reversed between visits to minimise any effects of time of day on the presence and detectability of birds.
The primary analysis will be at the farm-scale, to determine whether farm-scale effects of stewardship can be detected. Data for species/guilds will be analysed using the count of birds on the whole site as the dependent variable in a log-linear regression analysis, controlling for the effect of site area on number of birds detected. Analysis will be made of the difference in change in numbers between years 1-5 between AS and control farms. If the data allow, it may also be possible to analyse data by field (in a general linear mixed model, with ‘site’ as a random factor), allowing us to determine whether birds show association with particular AS features.
BREEDING BIRDS: ESTIMATION OF POPULATION DENSITIES ON AGREEMENT AND CONTROL FARMS
We will monitor birds at 25 AS and 25 control farms in each pilot area (East Anglia and West Midlands). Methods will be as for the original monitoring project. Survey will be based on the methods of the Breeding Bird Survey (BBS: Noble et al. 2000). At each site, a 2000m survey route will be established along field boundary features. Where possible (and according to the existence of maps of survey routes from the original monitoring project) the transect route will be the same as that used in the original monitoring project. Although desirable to follow the same transect route, this is not essential, as we are looking for effects which are consistent across the farm (N. Aebischer, pers. Comm.). To avoid double counting of birds, the route will not return to within 200m of a previous part of its course. If a 2000m long transect cannot be achieved because of small size of the site, the longest possible length will be used (in the original project, only 10 out of 100 sites had transects below 1.8km, the minimum length being 1.4km).
Two bird recording visits will be made to each site in summer 2003, once in the early part (early April to mid- May) and once in the late part (mid-May to late June) of the breeding season, as with BBS. As with BBS, a preliminary visit will be made (in March) to finalise the transect route. Counts will begin between 0600 and 0700 and last about 90 minutes. Only one site will be surveyed per day. All birds seen or heard within 100m of the transect route, including flying birds that are actively hunting (raptors, hirundines, swifts), will be recorded. Surveys will not be conducted in periods of heavy rain, strong wind or poor visibility and survey routes will be reversed between visits to minimise any effects of time of day on the presence and detectability of birds.
Data for each species will be analysed using the count of birds in the 100m either side of the transect as the dependent variable in a log-linear regression analysis, controlling for the effect of length of the transect on number of birds detected. Analysis will be made of the difference in change in numbers between years 1-5 between stewardship and control farms.
Findings will be disseminated to Defra through interim reports in February and December 2003 and a final report in February 2004. Research results will be communicated to a wide scientific audience. We envisage that several papers will be submitted to refereed scientific journals. Articles will also be written for the popular scientific and farming press. Opportunities will also be sought for publication of results in conference proceedings. In any case we envisage that presentations will be made and national and international scientific conferences such as The British Ecological Society Winter Meeting.