J. Clarke. Ivy Cottage, Kemerton, Tewkesbury, Glos. GL20 7HY

This was the third year of the project. Publicised in local Church magazines and by 'co-ordinators' in each village, the aim was to achieve similar coverage to 2002. During the winter JC and supporters made 50 'nest-boxes' - basically a half coconut shell mounted on a back-plate and sheltered by a flat roof that overhangs the shell. These were sited on house, garage or garden walls at or near known or traditional Spotted Flycatcher nest sites.

The British Trust for Ornithology provided a grant of £100 towards the costs of the project. The BTO approached JC to see if this project would fit in with national trials of their 'Constant Nest Monitoring Scheme'. JC agreed to include the Bredon Hill project as the eighth trial. By coincidence the Bredon Hill project is very similar to the Constant Nest Monitoring Scheme in that a 'survey plot' is selected, a species of bird chosen, and then all pairs using the plot are followed throughout the season - each nesting attempt being monitored. The major difference is that a 'plot' is supposed to be a few acres - perhaps a small wood, a few fields or a short stretch of river. The Bredon Hill project comprises 9 'plots' - each an entire village (or much of it). This of course will involve a considerable increase in paperwork.

Data from the survey is supplied to the BTO Nest Record Scheme. In 2002 the Bredon Hill project supplied the 11,000th Spotted Flycatcher nest record since the scheme started in 1939. This large database provides important insights into population changes and breeding behaviour. This year JC will attempt to input nest records via a new computer programme known as 'Integrated Population Monitoring Reporter' (IPMR).

The villages covered by the survey were Ashton under Hill, Beckford, Bredon, Bredons Norton, Conderton, Kemerton, Kinsham, Overbury and Westmancote (Lower and Upper).

As in previous years the survey relies heavily on observations by the local community, backed up by co-ordinators and/or JC walking the villages searching for Spotted Flycatchers. This year, visits were made to check every nest box site for signs of nesting. As usual, 'cold searches' were conducted, usually from the road to check all suitable Spotted Flycatcher habitats within the villages. Sightings and reports were followed up with visits by JC to locate or confirm the pair and then to locate and monitor the nests. Requests were received from two other villages to check individual nests and this was done, but the data is not included here.

Nests were located and checked using a 15cm x 6cm mirror attached to an extending handle and wherever possible the use of ladders was again avoided.

Information collected was recorded first in a ledger and later transferred to a computer spreadsheet. That data includes information on nest building, clutch size, brood size and fledging. Miscellaneous notes were taken on observations by JC and local observers.

All nine villages were surveyed at least once and the more likely areas therein checked several times. Coverage was not evenly spread as some properties cannot be fully viewed from the road but this 'variable' remains constant for each year. If a pair of Spotted Flycatchers (or a single bird) was reported but not confirmed by a visit then further checks were made to find the birds or evidence of breeding. It is unlikely that more than a few pairs were missed in the areas covered by this and previous year's surveys.

Cover achieved was similar to 2002 with very little restriction of access. It is unlikely that a significant number of pairs were missed.

Bad weather during migration and a cold spell on their arrival in Britain appears to have seriously affected Spotted Flycatcher numbers within the survey area - and indeed elsewhere. We may never know if the birds died on migration or were forced to move on to other areas - or even remained in France and Iberia. Many arrived very late at their breeding sites. A few arrived on time (around 8-10th May) but then either interrupted or delayed nest-building until late May/early June. Most birds were not reported until the last week in May or into June. 3 pairs fledged second broods between 16th and 18th August,

The number of pairs located in Kemerton fell to 6 (from 11 in 2001); boosted by late arrivals Overbury reached seven pairs (same as 2002 but down from 10 in 2001); and Conderton dropped from 5 to 3 pairs.

The nests of two pairs were not located - due to time available. Some were not found until shortly before and in one case, shortly after fledging. fiveout of 41 nests monitored were 'predated' or failed before or at a very early stage of egg laying. These nests are not normally included in Nest Record data. Nevertheless, they are significant when looking at the number of nesting attempts made by each pair. One pair nested in an inaccessible spot -on the end of a branch of a Scots Pine. Where a nest such as this could not be fully monitored the pairs are not included in laying/fledging data.

Around 100 householders and property owners co-operated in the survey, receiving visits from JC and local co-ordinators.

Spotted Flycatchers were again not found in the eastern end of Bredon or in Kinsham. Only one pair turned up in Westmancote but two were found in Bredons Norton.

Full details of the data are not provided here but to summarise:

1. 30 pairs were located in the villages covered (40 in 2002) - of which 28 could be monitored satisfactorily (31 in 2002).

2. 43 nests were located (50 in 2002) of which 39 (44 in 2002) were successfully monitored.

3. 18 of the 43 nests failed (10/50 in 2002). 11 pairs failed at the first attempt. Seven of the 28 monitored pairs are thought to have failed to breed (2/31 in 2002). Of those, 4 gave up after a single attempt and three after a second attempt. 2 pairs succeeded at the third attempt (3 in 2002). 5 pairs bred again following earlier successes (4 in coconut shells).

4. 3 pairs used their first nest for second broods. Failed pairs moved 25m to 250m for a further attempt - but often still within a 'feeding territory'. 2 pairs made a second attempt in nest sites where other pairs had recently failed.

5. 34 of the 39 nests were at, or within a few feet of, previous sites.

6. 20 pairs used either nests from previous years or boxes placed close by. 23 nests were in boxes.

7. 21 nest sites were sited against house walls, 2 in house rainwater gutters, two in Yew trees, one in Scots Pine and one on a garden fence.

8. Average nest failure rate 44% (22.7% in 2002 and 27% in 2001). 43% of nests in boxes failed. Predation by Domestic Cat, Grey Squirrel and Jackdaw was proved. Predation by small rodent, House Sparrow and Woodpigeon (the last two by damaging nests) was strongly suspected. Tawny Owl, Magpie and Jay were also suspects.

9. 36 nests were deemed suitable to include in detailed data analysis. Details are as follows:

Comparing 2003 with previous years' data:

10. Average clutch size (inc. 2nd attempts and 2nd broods) 3.97 (4.07 in 2002 and 4.09 in 2001).

11. Average brood size (young fledged per nest inc. failures) 1.80 (2.68 in 2002 and 2.2 in 2001).

12. Average success rate (no. young fledged per egg) 63.7% (63% in 2002 and 54% in 2001)

13. For second broods only:

a.. There were no failures.

b. The 5 pairs produced 10 young from 17 eggs - average clutch size 3.4, average brood size 2.0, average fledging success 59%.

14. 55 of the 56 adults appeared to have survived up to the fledging of their young.

15. In addition to survey work JC made 222 site visits.

16. Apparently many local Spotted Flycatchers delayed setting off on migration this year. Groups of up to 12 birds were seen during the first week in September - the last sighting on the 9™

Weather - Bad weather in early May compounded the problems that the birds had faced on migration. Cool, wet and/or windy weather seriously reduce the amount of food available to Spotted Flycatchers. Many other migrants relying on flying insects for food were apparently also affected. However birds such as Swalllow and House Martin often feed at some distance from the nest site but Spotted Flycatchers do not - thus further restricting the availability of food.

One adult was found freshly dead in her nest. There was evidence of severe muscle wasting and the bird was very thin. The pair were first reported at the nest site on 21st May and did start to re-build last year's nest. The bird was found dead on 9th June in the unfinished nest.

At a second site, a pair was seen commencing nest building on 10th May - the earliest nest record for this year. However, in a stop/start way it took them until 3r June to complete the nest. During that time they would disappear for up to a day (presumably looking for food). On 31 * May the house-owner found the nest on the ground beneath the coconut shell box. . It is thought that House Sparrows had pulled out the material whilst building their own nest nearby. She quickly replaced the nest and the Flycatchers resumed building later in the day. The birds appeared doomed to failure as later on the young, desperately begging for food were clearly in danger of knocking each other off the nest. Once again the house-owner stepped in, placing a wicker basket lined with a blanket just below the nest. On two occasions a youngster was found in the basket and each time it was revived and replaced in the nest. All four young eventually fledged over a period of two days, the smallest seen to fly up into a nearby tree to join its waiting siblings. The date was 5th July - some 57 days after nest building began! They did not attempt a second brood!

At one site a single egg hatched from a clutch of four. The hatching date was 16 days after the last egg laying date - the average is 12-14 days. However the youngster fledged after 10 days, 2-4 days earlier than the average.

Grey Squirrel - Long-suspected of predating SF nests, one individual was actually watched by JC as it ate the eggs whilst the adults watched helplessly.

However, another pair was observed by JC attacking a Grey Squirrel near their nest. The animal had approached along the top of a 2m high wall. The birds constantly harried it by diving vertically from several metres. Eventually, one bird was seen to actually strike the squirrel - it was not clear whether with beak or claw but the result was dramatic. The animal seemed to convulse and then ran at full speed for the cover of an overhanging shrub, where it 'preened' itself for a few minutes before creeping away.

Jackdaws almost certainly predated several Flycatcher nests. These predators nest in chimneys of houses now converted to central heating. It would appear that the young Jackdaws reach the fledging/post fledging period at the time when Spotted Flycatchers are starting to nest. Perched on rooftops the adults have a clear view of the gardens below. Jackdaws have been seen killing freshly-fledged Robins, Blackbirds and Thushes in gardens. Once the post-fledging groups and flocks of Jackdaw have moved away (usually to local farmland), the incidence of nest predation noticeably drops. Some house owners are wiring off the chimney pots to prevent access by birds.

A failed nest from a coconut shell was removed and checked. The wall thickness was an average of 1.5cm-increasing to l.Scmatthe 'lip'. It was estimated that the material used amounted to 60-70% of that found in a 'natural' nest.

One of the two pairs located in Bredons Norton nested precisely in a site that had been used for eight years during the 1990's but which had been unoccupied for the past four years.

Spotted Flycatchers have been reported feeding much later than other diurnal birds - often well after Corvids and thrush spp. have gone to roost. In late August JC watched a bird still feeding when it was too dark to see it perching - it was only visible when it flew up and was silhouetted against the sky. Its only companion was a bat!

During the first two years of the survey, butterflies were occasionally reported as being part of the Spotted Flycatcher diet. However, apart from one report of Marbled White, only dark-coloured species, such as Small Tortoiseshell, Red Admiral and Peacock were recorded. In 2003 an observer in Overbury and another in Conderton saw a Flycatcher take a Small White. On 2nd September JC was watching a group of around 12 Flycatchers feeding in his garden and he noted that on at least 8 occasions birds chased down and caught Small Whites.

For several days in early September the birds around JC's garden were observed feeding on insects and Elderberry fruit. At one time a bird was seen to regurgitate several fruit -possibly because it had gorged on them.

Nobody reported Spotted Flycatchers drinking or bathing in garden ponds or bird-baths. However, during a heavy downpour of rain in late August, whilst other birds had taken shelter, JC observed a group of Flycatchers perched on dead twigs and vigorously bathing -including tail-fanning, wing quivering etc. In late August'/early September Flycatchers were regularly observed drinking at the bird-bath.

Small flocks of Spotted Flycatchers have been noted post final fledging and prior to migration. There are few observations on this but groups have been noted at Bredons Norton, Conderton, Kemerton and Overbury. It seems likely that the local populations congregate to feed up before setting out on their migration.

Flycatcher enthusiasts found various ways of supporting the birds and the project:

a. Several supplied photographs of particularly 'tame' birds.

b. One found a way of increasing the local food supply. She placed a dish of old cat food about five metres from the nest. This was quickly colonised by flies - which in turn were caught by the Flycatchers!

c. At least two households set up telescopes in their gardens - at a safe distance from the nest. Theirs and other local families were able to watch the adult birds feeding their young and to see the young finally fledge.

d. One lady got up one morning to find a newly-fledged youngster on the kitchen floor - brought in overnight by the cat! She and the neighbours quickly put up a ladder and replaced the youngster in the nest. It flew out again but this time landed safely in a nearby tree, to be immediately collected by its parents!

e. Another lady returned home after a morning's shopping to find an adult Flycatcher in the sitting room - it must have flown in via the front porch and door which had been left open earlier. The bird was captured, checked and released, whereupon it flew straight back to the coconut shell nest six feet from the porch and resumed incubating. The pair successfully reared that and a second brood.

f. Several people, upset by the rate of predation of the nests (of Flycatchers and other birds) took various actions including the culling of Grey squirrels and Magpies and by 'deterring' cats and Jackdaws. The latter by wiring off chimney pots (see earlier).

2003 seems to have been a disastrous year for the Bredon Hill population of Spotted Flycatcher. From discussions with observers elsewhere in Britain it seems likely that this situation was not unique - with a complete absence of birds being reported by several observers. It would be wonderful if research coverage were such that we could discover that the birds had survived the winter and their migration, but had been forced by the weather to move elsewhere to breed. With luck, information supplied by BTO recorders might give some indication. If so then hopefully this year was just a 'blip' and more birds will return in 2004.

The coconut shell nest box experiment was affected by the drop in numbers of the Spotted Flycatcher, in that fewer boxes than expected were occupied. From the data collected it was possible to use 19 nests in coconut shells and 17 natural nests for direct comparison. This is clearly a small sample to work from. In addition, other factors such as predation and local food availability will not be constant for every site. Nevertheless the data may provide an indication as to initial success or failure.

In the event the figures do not show a clear-cut advantage of one nest type over another. Results for nest success rate, clutch size and brood fledging are very similar. Coconut shell sites, whilst clearly attractive to Spotted Flycatcher do not appear to significantly affect breeding success. The fact that 3 out of 4 second broods were in Coconut shells may be important - or may just reflect particular local conditions. What may be significant is that the birds do not need to expend as much time and energy on nest-building in coconut shell sites - and this could affect their ability to breed twice -even in a 'late' season. Further research within a wider, enlarged project would be required to fully assess the value of providing artificial sites.

However, the nest-box scheme has other implications in that it has clearly helped to widen and improve public awareness of the Spotted Flycatcher - and indeed, other, lesser-known species of bird.

After three years of the Bredon Hill Spotted Flycatcher survey it is worthwhile reminding participants of how much they have helped to achieve. We have located a total of 100 pairs and 126 of their nests. 82 pairs and 116 nests have been successfully monitored

So many local people were supportive of the scheme that they cannot all be mentioned here. Grateful thanks to all who allowed complete freedom of access over a period of four months and to those who contributed to the huge number of phone calls received. Special thanks to all the coordinators and supporters - especially Fred Wood, John Brunt and Georgina Millway.

Thanks once again to Pamela for leg-work, monitoring, fielding phone calls and reading drafts.

Thanks to the British Trust for Ornithology for the grant towards expenses.