Peter Rock

ABSTRACT
This report outlines the status of the large gulls, The Herring Gull Larus argentatus and The Lesser Black-backed Gull Larus fuscus, which breed on roofs within the City limits of Worcester as assessed on 28th April and 6th May 2003. An estimate of breeding numbers and species split are presented.

An overview of roof-nesting by the large gulls in Britain and in the region is also presented. Additionally, observed colour-rings, the presence of a hybrid Lesser Black-backed Gull x Herring Gull and the first colour-ringing of gull nestlings in the city are discussed.

INTRODUCTION
In the wake of an increasing number of complaints about ‘seagulls’, I was asked by the Department of Environmental Health, Worcester City Council to undertake a survey of breeding gulls in the city, advise on possible, but appropriate, future measures.

The assessment of the Worcester population took place on 28th April and 6th May 2003 and was one in a series of 14 assessments which I undertook in the Severn Estuary region in 2003. Essentially, very little is known about urban breeding by the large gulls in the region described by a rough ellipse which has Birmingham at its northern point and arcs through Wiltshire, Somerset, the Gower Peninsula and Hereford. I will discuss the results of the Worcester and other assessments below.

What we DO know, however, is that urban breeding is impinging on human activity. It is now a matter of some urgency that its scale and rates of increase are accurately understood. Between 1976 (Monaghan & Coulson 1977) and 1994 (Raven & Coulson 1997) populations rose at approximately 13% per annum. It is believed that the rate of increase since 1994 has been rather higher. As complaint levels increase concomitantly, it is felt that a pragmatic approach to this issue is necessary, based on sound research.

THE HISTORY OF WORCESTER’S GULL POPULATION
The first recorded breeding of one pair of Lesser Black-backed Gulls in Worcester is estimated to have been in 1982 (G. Phillips, pers comm). This was on a chimney stack immediately opposite the Museum/Library. At that time there were no birds breeding on the Gala Bingo House and no other pairs were noted elsewhere. Thus, in the absence of any other records, this must be regarded as the start date of the Worcester colony. However, early urban colonisation by the large gulls tends to go unnoticed until there are several pairs and some disputes (with the accompanying noisy behaviours) between neighbouring pairs (pers obs). It is possible, therefore, that the Worcester colony started at a slightly earlier date.

RESULTS
From observations of occupied nests and other procedures it is estimated that the Worcester breeding population is between 263 and 291 pairs, with a nominal figure of 277 pairs. This makes Worcester a small to medium-sized urban colony in national terms (Rock 2003a).

Within Worcester the distribution of nesting areas is fairly clear. The vast majority of nests are to be found on city centre roofs, with smaller numbers on industrial and commercial units. Little, if any, breeding activity occurs within residential areas. This is typical of other urban colonies within the defined region.

The highest nesting density is to be found in the city centre, with the most populous roof being the Gala Bingo House. Several other buildings, such as The Abbey National Building Society, support moderate numbers of gulls.

The species split in Worcester is some 4.9:1 in favour of Lesser Black-backed Gulls. There are NO urban colonies in the region where Herring Gulls dominate and this split is above the mean of 4.0:1 for the 14 colonies assessed in 2003.

It is, therefore, concluded that the colony comprises 230 pairs of Lesser Black-backed Gulls and 47 pairs of Herring Gulls.

OBSERVATIONS OF RINGED BIRDS
In general terms, there are two events in a metal-ringed bird’s life - the date and place of ringing and the date and place of recovery (usually of the dead bird). Whilst the information gleaned from these two events is very important in describing movements (Rock, in Wernham et al 2002), it fails to describe the life-histories of long-lived birds (the LBB record is 34 years, 10 months and 27 days). It is precisely for this reason that colour-ringing was devised.

Between 1988 and 1995 some 500 nestlings were colour-ringed in Gloucester with the site identifier G and 88 were marked with individually-coded rings (P. Stewart pers comm). The distance from Gloucester (36 km) is well within the expected recruitment range (pers obs).

Perhaps more interesting are the two birds hatched in wild colonies in excess of 100 km distant. Wild-hatched birds breeding in urban colonies are uncommon.

Urban gulls do not tend to recruit into traditional, wild colonies with less than 2% of Bristol birds doing so (pers obs). Instead, they seek out urban colonies in which to breed once they are old enough. This is because the period from hatching to fledging is some six weeks and after fledging offspring remain attached to the nest site, relying upon their parents until they become independent, for another two to three weeks. During this time they build up a kind of postcard in their minds of what a gull colony looks like and it is to this postcard that they return to breed in three or four years time.

Recruitment from traditional, wild colonies into urban colonies is more difficult to assess because rather few birds have been colour-ringed in the region. Between 1989 and 1994 some 600 birds were marked with a site identifying ring showing the letter F on Flatholm Island (in Bristol Channel) and between 1995 to the present, Flatholm colour-rings have been individually coded and 100 birds per year ringed (various Severn Estuary Gull Group reports).

The incidence of Flatholm birds in urban situations is low - even in Cardiff which is less than 10km distant (Rock 2003b). It is therefore supposed that recruitment levels from wild to urban are low and may be as low as from urban to wild. Thus, it appears that urban colonies are self-sustaining.

Migration movements by Lesser Black-backed Gulls are an important and interesting part of our understanding of roof-nesting gulls, but perhaps more pertinently, four of the eight colour-ringed birds (and one metal-ringed bird) have visited Gloucester Landfill at Hempsted and that WHITE3 CT was seen there on 20th June 2002 - exactly when its nestlings would have been on a rooftop in Worcester (assuming that this bird bred in Worcester in 2002).

Gloucester Landfill is 35 km from Worcester and easily within the feeding range of Worcester gulls. A large number of Bristol-ringed gulls breeding in Bristol has been recorded at Gloucester Landfill over the years and Bristol is 49 km away. The large gulls are perfectly capable of flying long distances very rapidly in search of food and a 70 km round trip is a small journey, probably taking only 2-3 hours - including feeding.

It remains to be seen whether or not Throckmorton Landfill has an influence on urban breeding gulls in Worcestershire and further afield now that 24 gulls have, for the first time, been ringed in Worcester (see below). Birds observed during the 2003 survey are predominantly females.

The 1993 and 1994 Bristol cohorts were sexed by DNA analysis by Oxford University and, from measurements taken in all subsequent years, multivariate analysis predicts the sexes of approximately 3,000 birds. During the survey, it was also possible to determine sexes by observation of pairs, or by close scrutiny of birds’ structure (females are slightly smaller than males, less muscular-looking, rounder-headed, with smaller, finer bills). However, It is not always possible to sex individual birds, especially when they are alone, because there is some overlap in measurements between the two sexes. It is also difficult when observations are brief.

Why so many females?

Typically, colonial breeding birds of many species adopt a strategy of emigration of one or other sex in order to maintain the strength of the gene pool. In the large gulls it is, on the whole, the females which emigrate while the males return to the natal colony (pers obs). Needless to say, there are exceptions and BLACK2 TX (sexed by DNA analysis) is a male!

HYBRID BREEDING IN WORCESTER
One hybrid was noted during the 2003 survey on Reindeer Court. Hybrids result from the cross breeding of the two species and in most other taxa they are usually infertile. Amongst the large gulls, however, this does not appear to be true - there are several viable hybrids breeding in Bristol and Bath (pers obs).

Typically, hybrids are intermediate in mantle colour between Lesser Black-backed and Herring Gulls. There appear to be two types of hybrid (pers obs) which are best described as

(1) dark-mantled Herring Gulls, or (2) light-mantled Lesser Black-backed Gulls. The two types appear to have the physical characteristics (jizz) of the species they are most closely allied to. The leg colour of type 1 is most usually greyish-straw instead of pink, but type 2 has yellow legs. Most usually, too, the hybrids of each type tend to breed with the species they most closely resemble (pers obs).

Those which have yellow legs are something of an identification dilemma. Superficially, they look very much like Yellow-legged Gulls L.cachinnans. The Mediterranean race of this species L.c.michahellis is undergoing a northerly range expansion which has seen it entering colonies in the Low Countries (where it also hybridises) (N. van Swelm pers comm). Yellow-legged Gull was confirmed as a British breeder in Dorset in 1995 (Ashby 1995).

 The hybrid is the bird closest to you in the row of gulls

ROOF-NESTING IN BRITAIN
The populations of all of the large gull species in Europe rose steadily from about 1930 onwards, with some populations increasing at 13% per annum. The reasons for this growth are varied and species-specific, but are primarily to do with food supply and to a lesser extent with national protection measures for these species (Cramp & Simmons 1983, Lloyd et al 1991).

Gull populations in Britain, particularly those of Herring Gull and Lesser Black-backed Gull, began to rise dramatically from the mid-1950's. This coincided with The Clean Air Act of 1956 which required that landfill sites should cover refuse with inert material at the end of each day's tipping instead of burning it. Gulls of all species, being opportunists, were quickly able to take advantage of the opportunity presented to them by feeding whilst tipping was in progress. We are, nowadays, all familiar with this sight. The result was that populations rose dramatically and by fifteenfold and more in the Severn Estuary area (Mudge & Ferns 1980).

Traditional colonies were outgrown and the colonisation of urban areas began. Though there are records of roof-nesting, particularly by Herring Gulls in coastal towns prior to The Second World War (Parslow 1967), urban colonies in Britain seem to have originated in the late 1950’s and 1960’s, but it was not until the early 1970's that colonies began to grow. Since then roof-nesting by the large gulls has spread throughout Britain and most towns and cities in the country nowadays support populations of urban gulls ranging from a few pairs to colonies of national significance.

Britain, of course, is not alone in supporting colonies of urban gulls. There are significant urban colonies in France (e.g. Cadiou 1997) and smaller colonies in Belgium (e.g. Francois 2002) and Netherlands (F. Cottaar pers comm). In recent years, Yellow-legged gulls L. cachinnans have also begun breeding on rooftops in Italy (L.Bembich pers comm), Gibraltar (pers obs) and Portugal (pers obs). It is nowadays a European issue. It also occurs in the United States and Canada (e.g. Belant 1997)

Quite why towns and cities were colonised in the first place remains unclear but, from a gull's point of view, buildings are not so very different from cliff-sided islands. However, towns and cities offer some decided advantages which traditional colonies do not (Rock 1993): namely,

1. Massive food availability. 2. Limitless nest-site availability. 3. An almost total absence of predators and disturbance. 4. Ambient temperatures slightly higher than the surrounding countryside (allowing an early start to breeding). These factors have, without doubt, contributed to the outstanding success of the urban gull.

By contrast, the breeding performance at the traditional colony on Skomer Island in 1990 was 0.6 young per pair (Griffiths 1991), in other words, about one chick per pair every other year. But in 1994, the figure was estimated at 0.2 and in 1998 at 0.1 (Perrins & Smith 2000). This equates to one offspring per pair every 10 years! In declining colonies, such as Skomer Island, chick mortality, as a result of inter- and intra-specific predation, is high. The chicks wander around the territory a great deal more, thus risking attack from neighbours. The parents tend to be less in attendance because they need to spend longer foraging for food, and are therefore unable to afford a suitable level of protection to their offspring. In colonies where breeding success is high, chick mortality is very low (Spaans et al 1994 ). There appears to be very little, if any, inter- or intra-specific predation in Bristol and other roof-nesting colonies (pers obs.).

It appears that urban colonies throughout Britain are increasing, possibly at the expense of traditional colonies (Raven & Coulson 1997). The key to all of this lies in food availability.

ROOF-NESTING WITHIN THE REGION
During the breeding season of 1994 a survey of Larus gulls nesting on buildings and other man-made structures was undertaken in Britain and Ireland (Raven & Coulson 1997). Estimated numbers were 16,900 pairs of Herring Gulls and 3,200 pairs of Lesser Black-backed Gulls. In the Bristol Channel region (which included Worcester) totals were 913 pairs of Herring Gulls and 867 pairs of Lesser Black-backed Gulls with a grand total of 1,780 pairs. It was, however, conceded that underestimation was likely.

In 1994 the species split for the Bristol Channel region was approximately equal. The 2003 totals show a species split of 4 : 1 in favour of Lesser Black-backed Gulls. There are NO colonies in the Severn Estuary Region where Herring Gulls dominate. Some colonies (including Worcester) are well above the mean split and are at the forefront of increasingly widening splits. It is entirely possible that the Lesser Black-backed Gull is rather more suited to urban breeding than the Herring Gull and may also be more aggressive in territorial disputes (pers obs).

Table 3. deals with only 14 locations. There are, of course, many more locations in the region than the 40 listed in Raven & Coulson, but even with only 14 locations, the total number of roof-nesting pairs is four times that of 1994.

Bristol has not been assessed since 1998. It must, nowadays, be around 1,800 pairs. Other significant colonies (i.e. 500-1000+ pairs) in the region certainly exist. These are likely to include Newport, Port Talbot, Aberthaw, Merthyr, etc. There are also many small colonies in the region of 20 to 100 pairs (pers obs). These would include Hereford, Tewkesbury, Stonehouse and so on. Thus, since 1994, it is possible that there has been tenfold increase in numbers of large gulls breeding on rooftops in the region.

Urban gulls breed very successfully at between 2-3 fledglings per pair per year (pers obs). Therefore, this year’s production from Bristol and the 14 listed sites will be of the order of 20,000 fledglings. Factoring in survival rates of 65% for the first year, 80% for the second year and 85% for the third year (P. Monaghan pers comm) the pool of potential first-time breeders in 2007 will be slightly under 9,000 birds. (These figures, of course, do not take into account previous seasons’ breeding success). Adult survival rates are approximately 95%. Thus, relatively few adults will die before 2007 and populations will be added to by young birds seeking to breed.

From the figures above, it can quickly be appreciated why urban breeding by the large gulls is growing so dramatically. Further, if the observed four-fold increase in populations is applied to the national figures from 1994 of 20,100 pairs, the total for 2003 would be 80,000 pairs of urban gulls! It is surely less than this...? It is also, therefore, quite clear that understanding the ecology of roof-nesting gulls is now a matter of considerable urgency.

COLOUR-RINGING IN WORCESTER
Understanding the dynamics of urban breeding within the region is now a pressing ambition and I have extended the Bristol study to include other urban colonies. In 2001 I started colour-ringing in Bath and in 2002, Cheltenham was included. In 2003, Cardiff, Bridgend and Worcester were added. Thus, in 2003, six urban colonies have had birds colour-ringed.

The Bristol (and Worcester) scheme colour-rings are very large (37mm tall) showing a two letter code. A different colour is used each year. These rings are easily observed and, using a telescope, can be read at distances up to 500 metres. Colours and ring codes identify individual birds. Only nestlings are ringed because only their origin is certain.

Using metal rings alone results in a recovery rate of less than 3%, whereas colour-ringing can enable recovery rates of up to 80% (Rock 1999c). Colour-ringing is, thus, a most powerful tool for investigating the lives of these species. Many Bristol-ringed gulls nowadays have very long life-histories, several of which are in excess of 100 records, showing considerable movements of all kinds - including migrations - as well as stay lengths at breeding locations and elsewhere. It is therefore easy to understand the value of such a system for uncovering what is actually happening in the lives of these birds.

The hoped for results will not arrive until 2006, or 2007. However, the incidence of second year birds breeding in urban situations appears to be increasing. Third year birds breed commonly in town and this is a sign of an expanding colony (pers obs).

In the intervening time, much extremely valuable information can be gathered. This includes post-fledging movements and dispersal, winter movements and migrations, colonial attendance during immature years, feeding locations, etc.

It is hoped that members of the Worcestershire readership - now that you have your very own colour-ringed gulls - will pay considerable attention to any gathering of gulls

REFERENCES
ASHBY, V. 1995. Dorset Bird Report.
BELANT, J.L. 1997. Gulls in urban environments: landscape-level management to reduce conflict. Landscape and Urban Planning 38:245-258.
CADIOU, B. 1997. La reproduction des Goelands en milieu urbain: historique et situation actuelle en France. Alauda 65(3) 209-227.
CRAMP, S. & SIMMONS, K.E.L. (BWP) 1983. Birds of the Western Palearctic. Volume 3 Waders to Gulls. Oxford University Press, Oxford.
FRANCOIS, R. 2002. Numbers and behaviour of roof-nesting Herring Gulls Larus argentatus and Lesser Black-backed Gulls Larus fuscus in Belgium. Natuur.oriolus 68(3) 123-126.
GRIFFITHS, R., 1991. Sex-biased mortality in the Lesser Black-Backed Gull at the nestling Stage. Ibis 134.
LLOYD, C. & TASKER, M.L., 1991. The Status of Seabirds in Britain & Ireland. T. & A.D. Poser, London.
MUDGE, G. & FERNS, P. 1980. A Census of Breeding Gulls in the Inner Bristol Channel. Cardiff.
PARSLOW, J.L.F. 1967. Changes in status among breeding birds in Britain and Ireland: part 3. British Birds 60: 177-202.
PERRINS, C.M. & SMITH, S.B. 2000. The breeding Larus gulls on Skomer Island National Nature Reserve, Pembrokeshire. Atlantic Seabirds 2(3/4): 195-210.
RAVEN, S.J. & COULSON, J.C. 1997. The distribution and abundance of Larus Gulls nesting on buildings in Britain and Ireland. Bird Study 44;13-34.
ROCK, P. 1993. Roof-Nesting Gulls in Bristol. Report.
ROCK, P. 1994. Roof-Nesting Gulls in Bristol. Report.
ROCK, P. 1999a. Roof-Nesting Gulls in St Helier. Report to Environmental Services, Jersey.
ROCK, P. 1999b. The Efficacy of the Colour-ringing system used for Herring Larus argentatus
and Lesser Black-backed Gulls L.fuscus in Bristol, 1980-1997. Ringing & Migration.
ROCK, P. 2003a. Roof-Nesting Gulls in Worcester. Survey conducted in May 2003. Report to Worcester City Council.
ROCK, P. 2003b. Roof-Nesting Gulls in Cardiff. Survey conducted in May 2003. Report to Cardiff County Council.
ROCK, P. 2003c. Roof-Nesting Gulls in Gloucester. Follow-up Survey conducted in May 2003 Report to The Planning Department, Gloucester City Council.
ROCK, P. 2003d. Roof-Nesting Gulls in Bath. Follow-up Survey conducted in April 2003 and summarising surveys from 1995. Report to The Department of Environmental Services, Bath & North-East Somerset Council.
ROCK, P. 2003e. Roof-Nesting Gulls in Cheltenham. Follow-up Survey conducted in June 2003 Report to The Environmental Services, Cheltenham Borough Council.
ROCK, P. 2003f. Roof-Nesting Gulls in Bridgend. Survey conducted in April 2003. Report to Bridgend County Borough Council.
ROCK, P. 2003g. Birds of a feather flock together. Environmental Health Journal May 2003 132-135.
ROCK, P. 2003h. Gulls about Town. BBC Wildlife Magazine, Vol 21, Number 6, 32-38.
SPAANS, A.L. et al. 1994. The relationship between food supply, reproductive parameters and population dynamics in Dutch Lesser Black-Backed Gulls. IBN-DLO, Wageningen.
STEWART, P. 1996. First Progress Report. Severn Estuary Gull Group.
WERNHAM, C.V., TOMS, M.P., MARCHANT, J.H., CLARK, J.A., SIRIWARDENA, G.M. & BAILLIE, S.R.(eds). 2002. The Migration Atlas: movements of the birds of Britain & Ireland. T. & A.D. Poyser, London.