Observations of Herring Gull (Larus argentatus) arrival in the large gull roost at Hilfield Park Reservoir in Hertfordshire during 2004 suggested that a substantial proportion of the birds were arriving from a source well to the south-east. It was hypothesised that this was central London, near Westminster, and by means of observations from locations including Deacons Hill (Borehamwood), Featherstone Hill (Mill Hill), Woodfield Park (Brent Reservoir), Parliament Hill (Hampstead Heath), Hampstead Parish Church (Hampstead) and Primrose Hill (Camden) this roost flight-path, over Paddington, out through Maida Vale, over Brent Reservoir and up through Colindale, was confirmed. Due to the required longer travel distance (13 miles) and greater ascent (over 100m) these birds tended to arrive relatively late in the roost compared to most other gulls of presumed more local origin. Small numbers of other gull species, in particular Lesser Blackbacks (Larus fuscus) and Black-headed Gulls (Larus ridibundus), were also found to follow the same flight-path out of north-west London, though their ultimate departure points remain uncertain.
Of interest is that fact that at the observation points most distant from the Hilfield roost, particularly Primrose Hill and Westminster but also Hampstead, most gulls, including Herring Gulls, were moving in a transverse direction to/from the major roosts in the Lea Valley at William Girling and King George V Reservoirs. These roosts are more accessible from central London, both in terms of distance and altitude, as are those on the Thames Estuary and the roost at Queen Elizabeth II Reservoir, Walton-on-Thames. It might therefore be deduced that the Herring Gulls coming to Hilfield could be deriving some extra benefit from attending this roost, perhaps in terms of increased security or information exchange with birds from other feeding areas, to justify the extra effort in travelling there. The former reason would also appear to be consistent with the observed steady increase in numbers roosting at Hilfield in recent years and seems most likely. Factors such as these could well apply more generally in motivating Herring Gull attendance at different, and less accessible, roosts, but further much more detailed studies would be required to systematically evaluate the reasons. Considering also the continuing status change since the in-depth research of Hertfordshire’s gulls by Sage in the 1950/60’s and Gladwin in the 1980’s the general need for more extensive new studies of large gull movements in the area is now apparent.
Contents
Hilfield Park Reservoir gull roost
Hilfield Park Reservoir (local map) is one of only two major gull roosts in Hertfordshire and currently holds about 20,000 birds at the mid-winter peak [1]. Of these there are usually less than 500 “large gulls”, of which the majority are Lesser Blackback and approximately 100 are Herring Gulls, whilst the small gulls are dominated by Black-headed Gull.
The numbers of each species using the roost has changed considerably over the years since the reservoir was first filled in 1955 [2, 3]. This is particularly true of the Herring Gull, which was once rare in Hertfordshire but which built up substantially in numbers during the 1950’s and 60’s, with a peak count at Hilfield of 7,250 on 11/01/1969 [2] (when large numbers of Great Blackback (Larus marinus) were also present). There was then a very steep decline to about 100 birds at the 1983 BTO Winter Gull Roost Survey [2], and a further fall to only 3 and 9 birds in the 1992 and 1993 (BTO) winter roost counts, respectively [4, 5].
This major decline has been attributed in part to the closure of nearby waste tips and changes in waste disposal practices [2]. However, there has been a renewed upsurge in numbers in recent years with a consistent increase from 2002 onwards [6] and 90 birds were estimated at the BTO 6th Winter Gull Roost Survey (WinGS) in 2004 [1]. These numbers were sustained through into winter 04/05, with a new high of 106 birds on 12/12/2004 [Cripps, personal communication, 2004/HBC Bulletin 235]. There was a further increase to 135 on 6/11/05 [HBC Bulletin 241] and 150 on 27/2/06 [HBC Bulletin 242]. Furthermore, it is likely that these totals are on the conservative side, as discussed below.
Origin of Herring Gulls at Hilfield Park Reservoir
The BTO Winter Gull Roost Survey at Hilfield Park Reservoir is traditionally divided into a number of regional sectors as shown in Figure 1 [4, 5]. Data is gathered on arrival of “small gulls”, i.e. mainly Black-headed Gull and Common Gull (Larus canus), and “large gulls”, as mentioned above, in 15 minute periods, and estimates of total numbers by species are deduced from sample counts of birds on the water.
Figure 1 – Map of Hilfield Park Reservoir showing BTO Winter Gull Roost Survey sectors
On the day of the 2004 BTO roost survey the author was assigned “sector E” in the south-east corner of the reservoir. He was therefore well placed to observe the arrival of a good number of large gulls from a south-easterly direction towards the end of the count, after dark, mainly entering via sectors F (south-east) and E (east). Though visibility was poor at this point, it was apparent that most if not all of these birds were Herring Gulls rather than Lesser Blackbacks, a fact consistent with his own observations of the timings of arrivals in the roost during the previous year [7]. Most were also adult birds.
The author immediately wondered how this observation could be explained. His own observations of Herring Gull movements in north-east Scotland, over 20 years, have shown that this species follows traditional flight-paths to and from roost sites [8]. These tend to be linear, i.e. shortest distance routes, but do accommodate topographical features to some extent, following valleys for instance, and sometimes also following obvious landscape features such as roads. In this case, with relatively large groups of birds arriving in close succession from the same approximate direction it seemed that they might have travelled together from an essentially small region of origin by a traditional route. Being a very late arrival it might also be inferred that the departure point was more distant than that of most of the other birds in the roost. The following areas were initially considered:
Brent Reservoir
The most obvious explanation was that the birds were arriving from a pre-roost at Brent Reservoir, just over 5 miles south-east. Some brief investigation confirmed that large gulls were regular visitors to Brent [9] and did indeed often come in for a pre-roost [Self, personal communication, 2004], though there was no regular overnight roost at the site in winter (which, though of similar surface area to the Hilfield reservoir, is relatively shallow).
However, flight time from this location would only be of the order 10 to 15 minutes which does not fit very well with the late arrival. Moreover, neither would this origin explain the observed species discrepancy, i.e. Herring Gulls arriving later than the Lesser Blackbacks, unless the former had some general disposition towards a later arrival at roosts. The latter hypothesis seemed unlikely, thus attention turned further afield.
Docklands/Lower Thames Valley
Recognising that the winter distribution Herring Gull is much more concentrated towards coasts and rivers than that of Lesser Blackback, which is found at a fairly uniform density across much of central England [10, 11], it seemed worth considering origins from the Thames valley or even the coast. In particular, it made some sense to consider the Docklands area and the lower Thames valley, which also lay in the approximate direction from which the birds were arriving.
The immediate objection to this theory was that the major Herring Gull roosts in Greater London, viz The Thames Estuary (4,962 birds in 1993 BTO roost survey [11]) and the William Girling and King George V Reservoirs (2,350 birds in 1993 [11]) are much more easily accessible to birds from these areas. This is both in terms of distance (relatively short journeys down the Thames valley for the major estuary roosts and only c. 8 miles to William Girling from central London, c.f. over 12 miles to Hilfield) and altitude (the main roosts are at sea level and only very slightly above, respectively, whilst Hilfield lies at approximately 100m altitude and to access it from the south or east birds must pass over a summit of above 110m).
West London/Upper Thames Valley
Origins from further west up the Thames valley were considered but also seemed difficult to justify, both in terms of the observed approximate arrival direction, and the fact that there are further major gull roosts in West London area around Staines, notably Queen Elizabeth II Reservoir at Walton-on-Thames [11, 12]. Indeed, locations in the Thames valley to the west of Westminster would certainly be nearer to the latter roost than Hilfield, and also involve a much smaller altitude difference.
River Thames in central London
A source somewhere between the latter two locations, i.e. the River Thames in central London, though certainly lying nearer to most, if not all, of these other major roosts, seemed to be the most reasonable explanation for the observations. The flight-path would therefore be constrained to a fairly narrow corridor, not lying too far east, where it would be in the main “catchment” of the Lea Valley or Thames Estuary roosts, or too far west, where the influence of the south-west London reservoirs would be significant. Considering topography, it was also considered that the lower-lying areas in the two “valleys”, Maida Vale in Paddington and Colindale in Edgware, might be followed (c.f. observations in north-east Scotland [8]). A route out from the Westminster area approximately along the line of the Edgware Road (A5) therefore seemed most likely. No evidence of a previous knowledge of this route was found in the literature and this was confirmed by Sage [personal communication, 2005].
Thus, the hypothesis for the study was defined as follows:
A substantial proportion of the Herring Gulls currently roosting at Hilfield Park Reservoir originate from a relatively small area in central London, near Westminster, thus traversing a distance of approximately 13 miles and height of over 110m to and from the roost each day. Their late arrival in the roost is due in part to that fact that they have travelled further and ascended to a greater height in order to get there than the majority of the other gulls present, which are of presumed more local origin.
This possible roost flight-path, together with the others mentioned above, is shown in Figure 2:
Figure 2 – Schematic map of Greater London, focussed on the north-west, indicating possible evening flight-paths of gulls originating from the River Thames at Westminster, to the major roosts at: Hilfield Park Reservoir (NW); Lea Valley Reservoirs (NE); Thames Estuary (E); Queen Elizabeth II Reservoir (SW)
The main sources of data consulted for information about movements of gulls in Hertfordshire, and the status of Herring Gulls in the county, were “The Birds of Hertfordshire”, by Gladwin & Sage [2] and the annual Hertfordshire Bird Reports [13]. The former provides a summary of the extensive studies undertaken historically, i.e. the intensive research of the roosting and feeding habits of gulls in Hertfordshire undertaken by Gladwin in the early 1980’s [2] and before that the studies of Sage commencing in the mid-1950’s [3, 14, 15, 16]. These include detailed observations both of feeding movements and general patterns of migration through the county. The Hertfordshire Bird Reports [13] are generally less useful in this regard as they tend only to include information on significant counts at different sites, and certainly in recent years the data has been very sparse (e.g. the highest count from anywhere in the county in 1999 was only 10 birds!). A detailed study of the gulls at Hadham Towers landfill, by Harris, was published in the 1994 report [17], including useful data on Herring Gulls.
Even in Gladwin & Sage [2], very little information could be found regarding Herring Gull roosting/feeding movements, which are the specific interest here. Moreover, most of the records relate to the period from the mid 1950’s to the early 1980’s so are probably now unrepresentative of the species’ current status and habits. Nevertheless, it is interesting to note that these authors considered that the vast majority of Herring Gulls visiting Hertfordshire have traditionally been rubbish-tip feeders and hence the huge reduction in numbers roosting at Hilfield by 1983 could be attributed, at least in part, to waste tip closures and changes in disposal practices in the area including Rickmansworth, Watford, St Albans, Radlett and Borehamwood [2]. The fall in numbers at the Lea Valley roosts from 8,450 in 1969 to 5,000 in 1983 was noted as being much less significant, though the reduction subsequently continued with only 2,350 birds present in 1993 [11]. This difference might be due to the fact that accessible waste sites have remained in operation in these areas, as mentioned below.
In recent years the largest feeding concentrations of Herring Gulls in Hertfordshire have still been found in the Lea Valley. Recent Hertfordshire Bird Reports [13], the Herts Bird Club website sightings archives [18] and sightings reports on the Hertsbirding yahoo group [19] reveal that a significant number of Herring Gulls are associated with the area around Amwell, including Amwell Gravel Pit, where there is now a winter roost, and other sites such as Amwell Roundabout, Rye House, Hoddesdon, Broxbourne Gravel Pit and the nearby land-fill sites at Westmill Quarry (Ware) and Hadham Towers (Widford). Weekly counts at the latter site in 1994 showed a strong seasonal variation with double figure numbers only between late November and early February, but a notable peak of 248 birds in mid-January [17], which is consistent with earlier studies in the Midlands [20]. It was adjudged that in winter at least 70-80% of these birds were argentatus-race individuals from Scandinavia [17], c.f. [21]. An exceptional peak count of 1300 birds was also recorded nearby Westmill Quarry on 10/2/90 [13], but peak winter counts in the area have usually been in the range 300-450 individuals, from 1987 to date. Significant but generally smaller numbers have also been found in the Colne Valley, at water bodies such as West Hyde, Troy Mill and Tilehouse Gravel Pit. There were several three-figure counts between 1991 and 1996, and a high count of 500 at the last-mentioned location on 29/12/96. Curiously, no counts have been published since this final observation, suggesting a significant subsequent decline. Just outside the county significant numbers of Herring Gulls also visit the land-fill site at Hedgerley near Gerrards Cross in Buckinghamshire, about 15 miles south-west of Hilfield.
The birds recorded feeding at the two landfill sites near Ware were known to roost at both Nazeing Gravel Pit (Essex) and at the Lea Valley reservoirs in 1994 [17], whilst the latter was identified as the main roost for birds feeding at waste tips east of Hatfield in the early 1980’s [2]. No mention is made of any roosts on the Hertfordshire side of the border in Gladwin & Sage [2] so any which are now regular have perhaps developed in importance in recent years. The species breakdown at the current roost at Amwell Gravel Pit, seems to be much more biased towards large gulls than the traditional Hilfield and Tring roost compositions [18, 19], though the total number of birds present is relatively small. Black-headed Gull is still the most numerous species but amongst larger gulls several hundred Great Blackbacks have been reported on a few occasions in recent years, and this is the only part of the county where such numbers can still be seen. It is also likely now to be the largest Herring Gull roost in the county, but no regular roost count data could be accessed to confirm this. A particular problem here is that where count data are included in the Hertfordshire Bird Reports no mention is normally made of whether these relate to roosts, pre-roosts or feeding flocks.
As far as the author is aware, no more detailed information has been published regarding the feeding movements of any of these birds. Sage [personal communication, 2005] confirmed that his own studies never established the presence in the Hilfield roost of birds known to have come out of the London area itself. However, as significant feeding flocks continue to be reported from the above-mentioned landfill sites in the west of the county, both in the Hertfordshire Bird Reports [13] and the website recent sightings reports [18, 19], it appears that a substantial proportion of these birds are still rubbish-tip feeders and the majority probably continue to use the local roosts in the Lea Valley.
For Lesser Blackback, a very relevant and interesting comment is the observation of Gladwin, for the autumn and winter of 1982-83, that 86% of the birds roosting at Hilfield were found feeding at Hertfordshire waste tips, and that the remaining 14% “flew daily into the London area to feed” [2]. The total number roosting at Hilfield peaked at 2,200 that autumn, and was just over 100 in mid-winter, so the latter group could have been a significant number at times. However, no precise information about deduced or observed feeding locations in the London area is mentioned.
Other interesting information about gull movements around London is described for Black-headed Gulls. Gladwin’s studies showed that in the winter of 1982-83, 82% of the birds roosting at Hilfield fed in the area of Hertfordshire approximating to the catchments of the Rivers Ver, Gade, Chess and Colne, extending north to Harpenden, north-west to Berkhamsted, and west to Rickmansworth [2]. As with the large gulls, relatively few birds were therefore found to be feeding in the areas on the London side. Nevertheless, an interchange between the major gull roosts at Hilfield, the Lea Valley and the Staines Reservoirs was also established, with small parties of birds seen departing from Hilfield heading off towards the Staines and Lea Valley roosts on most afternoons, and different birds seen arriving back from these directions [2]. By following the movement on successive days interchange between these three sets of reservoirs was proved. Sage, 1959 [14] and Siva-Jothy, 1982 [22] had described similar observations at Hilfield, and Sage also noted that flight lines tend to be "traditional" [16].
As discussed below, such interchanges might be a factor in bringing gulls from different areas to Hilfield. However, these observations seem not to have been recorded for the large gull species [Siva-Jothy/Sage, personal communication, 2005], though given the much smaller numbers of birds involved such movements could easily have been overlooked.
It is noted by Gladwin that most Common Gulls from the major Hertfordshire roosts feed on grassland (>92%, [2]); the Lea Valley roosts are larger, but most of these birds move out to feed across central Essex [2], and it seems unlikely that many of these birds would move into the London area to feed where there is bound to be less extensive grassland/arable habitat available.
There has also been some more recent research on gull movements in the London area undertaken by the Central Science Laboratory [23], c.f. colour-ring observation described for 16/12/04 below. These studies are reportedly concerned with establishing the potential for bird strike hazards due to gulls in the Heathrow area. As far as the author is aware, nothing has been, or is intended to be, published on these studies.
In conclusion, very little information was found in the literature on Herring Gull feeding movements in Hertfordshire, with data on movements and roost flocks from recent years being particularly patchy. More specifically, no direct evidence was uncovered regarding a possible central London source for some of the Herring Gulls roosting at Hilfield, or indeed for any gull species roosting in Hertfordshire.
The first step in pursing the study was to make a more thorough analysis of the data recorded from the Hilfield roost itself, from both the provisional results of the 2004 BTO WinGS roost survey [1] and the author’s own roost counts in 2003 [7].
Figure 3 shows the arrival trends for small and large gulls from the BTO survey, both as a cumulative total and a derived rate of arrival (birds per hour):
Figure 3 – Comparison of time of arrival of small and large gulls at 2004 BTO roost survey [1], each data point being a cumulative total (thick curve) or a derived rate (thin curve)
Clearly the large gulls’ arrival exhibits a much different pattern to that of the small gulls. Though both groups had a similar early peak arrival at around -45 minutes on this day, the large gulls show a sustained relatively high arrival rate thereafter culminating in an obvious second maximum at the final sample point after dark.
The large gull data can be analysed further by considering the breakdown of arrival by sector, or direction, c.f. Figure 1. The total net arrivals (i.e. incoming minus outgoing) for each 15 minute period are presented in full in Table 1 [1]:
| Direction | Sector | -137 | -122 | -107 | -92 | -77 | -62 | -47 | -32 | -17 | -2 | 13 | 28* |
| NW | A | 0 | 0 | 3 | 7 | 5 | 8 | 6 | 14 | 8 | -1 | 2 | 0 |
| NNW | B | 0 | 0 | -5 | -2 | 14 | 0 | 11 | 1 | 0 | 3 | 0 | 0 |
| NNE | C | 0 | 0 | 3 | 2 | 26 | 34 | 85 | 0 | 15 | 4 | 0 | 0 |
| NE | D | 0 | 0 | 0 | 0 | 5 | 13 | 1 | 2 | 2 | 0 | 0 | 0 |
| E | E | 0 | 0 | 1 | 0 | 1 | 1 | 7 | 10 | 8 | 7 | 28 | 0 |
| SE | F | 0 | 0 | 2 | 1 | 3 | 3 | 15 | 9 | 28 | 27 | 57 | 0 |
| S | G | 0 | 0 | -1 | -1 | -2 | 0 | -1 | 0 | -8 | 0 | 0 | 3 |
| W | H | 0 | 0 | -1 | 2 | 0 | 0 | 2 | 5 | 4 | -5 | 0 | 0 |
| All | All | 0 | 0 | 2 | 9 | 52 | 59 | 126 | 41 | 57 | 35 | 87 | 3 |
Table 1 – Net arrival totals of large gulls according to direction/sector for each sample period, time after sunset (minutes) [1] (* - counting ceased early in this final period, hence the totals relate to less than 15 minutes)
By accumulating data from neighbouring sectors, and plotting net arrival rates for each 15-minute period, Figure 4 shows graphically that the early peak is mainly due to birds arriving from the north (sectors B to D) whilst the latter peak is almost entirely due to birds arriving from the south and east (sectors E to G). Relatively few birds apparently arrived in the roost from the west (sectors A and H), perhaps due to the use of Colne valley reservoirs by large gulls from these areas.
Figure 4 – Comparison of net arrival rates, per 15 minute period, of large gulls from different directions [North (B to D), South&East (E to G) and All] during 2004 BTO roost survey [1]
Note that data for the final sample interval (+28 minutes) is not shown on Figure 4 as counting had been terminated early in this period – hence no large gulls were recorded in sectors E and F despite the large arrivals in the previous 15 minutes. The author considers it is likely that if this period had been counted in full, and visibility was still sufficient, then continued large gull arrival from the south-east would have been logged. As many of these birds were clearly Herring Gull this suggests that the total derived from sample counts for this species may be on the conservative side.
Hence it is clear that a substantial number of large gulls, at least a third of the total, arrive relatively late in the roost from the south-east. But these data do not reveal anything about the species breakdown in each time period, since the overall estimates were based on sample counts of birds on the water and hence averaged over the whole duration. However, the author’s data from observations of the roost during the preceding year [7], from the nearby M1 footbridge at TQ153955, spanning 9 dates (29/1, 7/2, 15/2, 22/2, 23/2, 1/3, 3/4, 4/4/03 and 16/1/04), had revealed a systematic difference between Lesser Blackback and Herring Gull arrival times, as shown by the cumulative totals in Figure 5 (note the logarithmic scale):
Figure 5 – Comparison of time of arrival of large gulls during 2003 gull-roost counts [7], each data point being the cumulative total number of each species present in the roost
The data shown are derived from all dates mentioned. There is considerable scatter in the totals due both to seasonal changes in numbers and the influence of the weather. It had been determined that time of arrival in the roost with respect to sunset can vary very considerably according to the latter, with early arrival under heavily overcast conditions, particularly if cold, and very late arrival under bright and mild conditions. This variation was typically +/- 30 minutes amongst the smaller gull species and essentially mirrored by the larger gulls though the latter were not averse to moving well after dark under favourable conditions, and individuals were occasionally seen leaving the roost well after sunset.
Curve fits to average values are shown which make clear that there is a systematic difference in arrival times between Lesser Blackbacks and Herring Gulls – with the former arriving in larger numbers earlier, whilst the majority of the latter usually arrived after sunset. The late arrival of Herring Gulls can be appreciated more clearly from Figure 6, which uses an ordinary vertical scale:
Figure 6 – Time of arrival history of Herring Gulls during 2003 gull-roost counts [7]
A final point to note is that the total numbers of both large gull species logged in the 2003 counts [7] are lower than those of the 2004 BTO roost survey, which itself may well have under-estimated Herring Gull numbers due to the reasons given above. Part of this difference is probably due to birds arriving late in the roost after dark having been missed, but an apparent systematic discrepancy between counts of birds arriving and birds on the water, with the latter being significantly smaller, was also noted in the official BTO survey, and had also occurred at the previous count in 1993 [5, and Cripps/ Fearnside, personal communication, 2004].
The reasons for the observed disparity remain unclear, though it is perhaps true that once down on the water, especially with reducing visibility, Herring Gulls are more likely to be overlooked than Lesser Blackbacks. Thus, a late arrival of this species might not be well represented on the basis of sample counts made rather earlier under conditions of good visibility. Under this scenario, the estimated number of Lesser Blackbacks, derived from the sample count ratio of large gull species times the net total of large gulls seen entering by observers on the perimeter, would be correspondingly higher than the true number which might still be accurately counted on the water, leading to an apparent discrepancy. More work would be needed to clarify this.
Observations along main roost flight-path
In order to prove the above hypothesis it was necessary to make observations along the roost flight-path. This was not a small task due to the potentially large area involved, the difficulty of accessing suitable watch-points with sufficient views in the right directions and the fact that the movement generally takes place under conditions of poor visibility. The most logical approach seemed to be to start with observations near the roost and work back progressively towards the point(s) of departure.
Full data on the observations is provided in the Appendix. An overview explaining the rationale and summary conclusions for each set of observations is provided below.
Note that the vast majority of the observations were made at dusk, tracking back the inward flight-path. Dawn observations were also considered but are more problematic due to birds leaving the roost very early, normally before sunrise [8], and consequent significant difficulties due to poor visibility.
17/01/04 - Hilfield
This was the main BTO WinGS gull roost survey, as described above. A total of nearly 200 large gulls were observed entering the roost from a basically south-easterly direction (see Table 1 above), with most arriving towards the end of the count and the bulk of the final arrivals being Herring Gulls. Birds entering through sector E (east) were seen to swing round to enter the reservoir from the east after passing over the summit of the hill in the south-east corner.
31/01/04 – M1 and A406 Brent
Observation was made from the car along the length of the M1 from Hilfield southwards and around the A406 through Brent down into West London. No large gulls were seen, though a single Black-headed Gull was moving north-west over Brent Cross at 10 minutes after sunset. This lateness was consistent with the bright conditions. It is likely that the major passage of large gulls also passed over relatively late that day and had not arrived at the time of observation.
7/02/04 – Deacons Hill
These were the first observations from Deacons Hill just south of Hilfield which affords fine views over the town of Edgware down to Colindale. Large gulls could be seen moving up the centre of the valley, over the Edgware hospitals, at a range of up to 3 miles. The birds were travelling at an height of approximately 30m, i.e. below the tops of some of the tower-blocks in the Colindale area, which themselves would have provided excellent views of the roost flight!
A total of 135 large gulls were counted arriving from the south-east in the 15 minutes from 10 minutes after sunset, even though the light was poor.
10/02/04 – Mill Hill
Observations made from Featherstone Hill, Mill Hill, which is 2.5 miles south-east of the Deacons Hill watch-point. This afforded a better view over Colindale and Edgware town centre but views toward the south were rather restricted. The same flight-path was confirmed for most large gulls but at least one bird passed much further east approximately along the line of the A1/M1.
14/02/04 – Parliament Hill
Tracing back the flight-path in the direction of the most easterly birds suggested that Hampstead Heath would be an excellent watch-point, particularly the high ground on Parliament Hill. Prior to the visit enquiries were made via the Marylebone Birdwatching Society [24] who keep bird records for the Heath, but no information on large gull movements had been recorded [Game, personal communication, 2004].
On the day itself, unfortunately only a single large gull was seen, well to the south and moving east (presumably to the Lea Valley roosts). In fact views out to the south-west were significantly restricted due to trees and buildings in the area west of the Heath.
21/02/04 – Deacons Hill
The Hilfield roost was viewed briefly from the A41 and a 1st-winter Great Blackback was noted in the roost together with at least 20 Herring Gulls at 18 minutes before sunset.
From Deacons Hill a relatively early arrival of large gulls was seen, perhaps due to the heavily overcast conditions. There was also a brisk north wind and birds were clearly labouring to gain height at the final steep climb at the head of the valley. Good numbers of Herring Gulls were again present in this group.
22/02/04 – A406 Hendon and M1
Observations were made from the car moving north around the A406 in Hendon and up the M1. 2 Lesser Blackbacks were seen moving north-west towards Hilfield over the A1-A406 junction, i.e. well to the east of the main flight-path through Colindale.
23/02/04 – Brent Reservoir
This was the first visit to Brent Reservoir. With relatively overcast and cold conditions an early passage was expected and groups totalling 45 birds (27 Herring Gull and 18 Lesser Blackbacks) passed over the north-east end of the reservoir just before sunset. These birds were heard calling loudly before they were seen - many of the Herring Gulls were adults in obvious pairs; unfortunately the view towards the south-east is restricted by the higher ground on Dollis Hill, and more immediately by tall trees and buildings along the south-east side of the reservoir, so in retrospect the conclusion that the birds had arrived from the south-east was not very confident and the possibility that birds had circled round from a pre-roost at the south-west end of the reservoir could not be excluded.
26/02/04 – Deacons Hill
Observations were again made from Deacons Hill. This time only 56 large gulls were logged during an hour of observation around sunset, with the majority being Lesser Blackback. By comparison with the records for 7/02/04 this data suggested that there might be a significant daily variation in numbers.
28/02/04 – Primrose Hill
Since no birds had been seen north-west over Parliament Hill it was decided to try Primrose Hill to the south, as this would correspond to a more westerly flight-path. Observations were made for the half-hour period around sunset.
Initially there were at least 10 Herring Gulls frequenting the Regents Canal area to the immediate south-east. The final departure direction of these birds was not clear, due to progressive dispersal at low level, but all large gulls observed in roost flight were travelling north-east, i.e. exactly perpendicular to the flight-path towards Hilfield. These included Herring Gulls and Great Blackback. Some of these birds were directly overhead whilst others were well to the south moving between the skyscrapers in the Marylebone area, such as the BT Tower, making a spectacular sight against a sunset sky.
It is clear that this is a generally excellent location for monitoring gull roost flights across the central London area, and for any birds using Regents Park; however, due to tall buildings on the south-west side it is no good for viewing any birds moving north-west through Maida Vale. Given the excellent view south, the fact that no birds were seen passing over north-west seemed to suggest that origin of the Hilfield was either further north or, more likely, further west, and in either case obscured from view.
18/03/04 – Brent Reservoir
As no birds had yet been seen on the hypothesised flight-path anywhere south-east of Brent Reservoir the possibility that the birds logged at that location on 23/02/04 had come from a local pre-roost could not be ignored. Hence a return trip was made, this time viewing from the higher ground in Woodfield Park.
The reservoir surface was checked in advance and no birds were found at pre-roost. A total of 150 large gulls were then logged moving through north-west from 40 minutes before sunset until after dark and this time there could be no doubt that they were all arriving from the south-east over Dollis Hill.
28/03/04 – Hyde Park
In absence of any other apparently accessible watch-points towards the south-east, beyond Dollis Hill, it was decided to make observations directly from Hyde Park over 5 miles to the south-east. Mitchell [12] mentions that Lesser Blackback and Herring Gull numbers in the park are usually fewer than 10, though the latter has attempted to breed on the Serpentine island, but considering the attraction of the water bodies and ample tourist scraps a reasonable number of birds might be expected to frequent the area on occasion.
This visit was unfortunately mistimed, being an hour early (due to the clocks having gone back!) but it was nevertheless of interest to record just a small number of the larger gulls at Kensington Gardens boating pond, with a few more Herring Gulls over the Bayswater shopping area to the immediate north.
Some of the large gulls on the boating pond are shown in Figure 7:
Figure 7 – Large gulls on Kensington Gardens boating pond, Hyde Park, 28/3/04
This was the final visit that was possible before the author moved north to Scotland; at this point, central London origin had not been proven and it was certainly possible that the birds were originating from an area well to the north-west of Westminster, i.e. urban north-west London. Due to the author’s move away from the area it seemed inevitable that the study would end on this inconclusive note.
16/12/04 – St James’s Park
A day trip to central London allowed observations from St James’s Park in the late afternoon. In the half hour before sunset, there were about 15 Herring Gulls, nearly all adults, together with over a hundred Black-headed Gulls exploiting the ponds and tourist scraps. The former were also feeding in short grass.
One of the Herring Gulls bore an orange colour ring, and had been marked as a 1st-winter, in the Central Science Laboratory (CSL) study of gull movements around the London area [23], at Gerrards Cross landfill on 12/11/03; it had been seen since at Dungeness on 23/01/04 and at Beddington Farm landfill on 18/05/04.
Most of the large gulls departed from sight in an approximately easterly direction, with the exception of a single bird moving apparently north or slightly west of north – possibly towards Hilfield. About 15 minutes after sunset a total of at least 250 Black-headed Gulls passed high overhead in a north-easterly direction, presumably travelling to the Lea Valley roosts.
It was interesting to see that a single adult Herring Gull remained to roost on the main pond, together with the exotic wildfowl. No reason was apparent for this as the bird appeared healthy and active.
29-30/12/04 – Brent Cross
A brief return trip to London enabled some final observations before the year end. Viewing from a tall building at Brent Cross briefly at dusk 29/12/04, no large gulls were seen, and at dawn on 30/12/04, birds were observed arriving to feed in the area well before sunrise from the north and east, but no major passage of birds was discerned.
30/12/04 – Hampstead Parish Church
Another attempt was made to find a suitable viewing point nearer to the presumed origin of the evening flight-path. The first potential watch-point visited was Hampstead Cemetery at Fortune Green; the view from here was good toward the west but blocked by buildings towards the south and east. No gulls were seen.
Hampstead Parish Church turned out to be a better location. The south-west corner of the churchyard afforded excellent views south across Maida Vale towards Earls Court, extending from approximately Notting Hill in the west to Marble Arch in the east, with additional more limited views further west and east.
At 15:45hrs, 15 minutes before sunset, a group of 4 Herring Gulls were seen well to the south but soon lost to view and clearly not in roost flight. Then at 15:53hrs, 7 minutes before sunset, and group of 14 Herring Gulls was picked up distantly to the south moving steadily north-west through the lowest part of the valley, at a flight altitude of about 30m. The location was approximately over central Paddington and moving through into Maida Vale. The birds were extended in a line in obvious roost flight. Though distant, at approximately 2 miles south, the observed flight action left no doubt that these were all Herring Gulls and the essentially pale upperpart coloration indicated that most, or all, were adults or sub-adults. These were surely the target group and a very exciting observation for the author – though they would surely not have merited a second glance from any other casual observer.
Figures 8 and 9 show views south from Hampstead Parish Church towards Kensington and Marylebone:
Figure 8 – View south over Paddington to Earls Court and Empress State Building
Figure 9 – View south-east towards Marylebone
Though the precise origin of the birds was not observed, given the good views south as far as Earl’s Court, and the fact that prior to commencing roost flight birds usually circle gaining height and would have been visible (the whole area was being periodically scanned through a telescope), the group had almost certainly come from south of Regent's Park, and possibly from the areas around the West End, Victoria or Westminster.
Five minutes later a 3rd-winter Herring Gull moved north-west past the hill, to the south, but much nearer, probably within a mile.
Over the next 10 minutes a total of 10 Lesser Blackbacks passed east or north-east round the southern side of the hill, presumably moving towards the Lea Valley roosts. A final large gull went north-west at 15 minutes after sunset, by which time visibility was poor and species could not be determined.
Though there was no opportunity for further visits to the Hilfield roost to check if numbers there had been maintained through into the current winter, information was received that relatively large numbers had still been present there in the autumn (e.g. 82 birds on 6/11/04) [Cripps, personal communication, 2004]. The HBC Bulletin edition 235 subsequently confirmed that 106 birds had been logged there on 12/12/04. Thus there was no reason to doubt that a good number of birds were still normally arriving in the roost from the south-east by late December.
31/12/04 – Hampstead Parish Church
A brief final return visit was made to the site at dawn in an attempt to track the birds back along the flight-path closer to their origin. Unfortunately visibility was very poor due to persistent drizzle but the first large gulls arrived from the east at 21 minutes before sunrise, a Lesser Blackback closely followed by a Herring Gull, navigating closely around the hill. A steady passage of mainly small gulls west round the south side of the hill then became apparent, with 44 Black-headed, 11 Common during the next half hour, together with one more each of Herring Gull and Lesser Blackback.
It is likely that any birds in a roost-flight from the north-west had already passed by before visibility improved at 08:00hrs (6 minutes before sunrise). Two more Herring Gulls moving south-east at 19 minutes after sunrise may very well have re-orientated from another direction as a number of gulls were making short feeding movements in the surrounding area by that stage. Thus, no further useful information on the study flight-path was gathered but this should be easy to obtain from this location in conditions of better visibility.
A summary of the above flight-path observations is shown in Figure 10, below:
Figure 10 – Summary of main gull roost flight-path observations over north-west London, as far as Hilfield Park Reservoir, with direction of evening flight indicated (yellow arrows, scaled to give a qualitative idea of the numbers of birds seen at each location, large arrows being >100 birds, and smallest arrows being just one or two individuals)
Further to the observations described above, supplementary data and comments were obtained from third parties, the most significant of which was information received from Tom Gladwin for West London and Des McKenzie for central London. Also, some further observations were made by the author on a return visit to the area in December 2005.
West London observations, mid-1990's [Gladwin]
During 2005 the author contacted Tom Gladwin who was able to provide some relevant supplementary information from his own unpublished observations in West London in the mid 1990's. Large gull species, including Herring Gulls and Lesser Blackbacks, had been noted heading off from West London in the direction of Hilfield at dusk during the winter months. These included birds feeding and/or loafing on landfill and grassland sites in the Colnbrook area, by the M4 near the M25 intersection, and others were seen moving north-east over Uxbridge. The observations were made in the context of studies of gull movements, and of other large, or flocking, species which are potentially hazardous to aircraft, in the area around Heathrow airport, and occurred in several months between September and February. The distance from this area to Hilfield is very similar to that from central London, i.e. about 13 miles. It is also notable that at the observation points these birds were moving in the opposite direction to the predominant southward evening roosting movement down the Colne Valley to the Langley-Iver area where the latter flow of birds then turns eastwards towards the mosaic of reservoirs and larger complexes of quarry lakes.
Paddington Green observations, 1998 to date [McKenzie]
The author also contacted Des McKenzie (former LNHS Bird Recorder for Inner London) who was able to supply further independent data from his observations gull movements over Paddington Green from 1998 to date, as listed below. This location (local map) is very close to the identified main roost flight path and the movements recorded are therefore strong support for the author's own observations. However, no clear trends in numbers can be discerned and more data is still needed.
19/11/98 - 36 Black-headed, 3 Common and 7 Herring Gull NW at dusk
20/12/98 - 60 Black-headed, 1 Common and 14 Herring Gull NW at dusk
09/01/99 - 14 Black-headed, 1 Common, 2 LBB and 4 Herring Gull NW at dusk
29/03/99 - 15 Herring Gull NW at 16:00hrs
01/01/00 - 36 Herring Gull N/NW at dusk
21/01/00 - 26 Black-headed and 3 Herring Gull NW at dusk
22/01/00 - 30 Black-headed Gull and 5 Herring Gull SE at 09:45hrs
23/01/00 - 13 Black-headed Gull, 3 LBB and 6 Herring Gull N/NW at dusk
03/02/00 - 4 LBB NW at dusk
12/01/02 - 9 Black-headed, 2 LBB and 9 Herring Gull N/NW at dusk
03/12/02 - 12 Herring Gull N/NW at 14:15hrs
25/02/03 - 47 Black-headed, 16 Common and 10 Herring Gull N/NW at dusk
07/01/04 - 22 Herring Gull N/NW at 14:00hrs
12/09/05 - 5 LBB Gull North at 16:07hrs
09/11/05 - 27 Black-headed, 1 Common and 13 Herring Gull NW at 15:00hrs
14/11/05 - c.120 Black-headed Gull S/SW at 07:40hrs
09/12/05 - c.60 Black-headed Gull SW at dawn (with 3 ad Common Gull)
14/12/05 - 6 Herring Gull W/NW at 14:26hrs
17/12/05 - 6 Herring Gull NW at 14:04hrs
Primrose Hill observations, 2005 [Welch]
In late December 2005 the author returned again to Primrose Hill to make another attempt to determine the departure point of the identified roost flight-path. Observations were as follows:
30/12/05 - poor visibility due to continuous rain with only local movements observed (including a single flock of 150 Black-headed Gulls heading NE at 15:53hrs, 7 minutes before sunset). 20 Herring Gulls and 1 Lesser Blackback also flew over NE (towards Lea Valley roosts), up to 16:19hrs.
31/12/05 - much better visibility giving good views over the whole of central London. Large gulls were seen rising up above buildings in the West End from about 15:50hrs, including one group of at least 14 birds. Most of these remained circling and interacting in the same general area until at least 16:15hrs (15 minutes after sunset) when visibility was becoming poor. As on previous visits a good number of birds passed over NE towards the Lea Valley roosts, including a group of 30 birds directly overhead at 16:15hrs. However, between 16:05 and 16:20hrs, several of the birds circling over the West End were clearly seen to move off NW along the line of the previously identified roost flight-path. Allowing for a flight time of 20-30 minutes these would have arrived in the Hilfield roost well after dark - and would also have been very hard to see at any intermediate points along the route, i.e. from Paddington Green or Hampstead Parish Church.
Hence it was finally concluded that the ultimate departure point of at least some of the birds is actually very near central London including the West End, perhaps as far south as Victoria and Westminster. There is undoubtedly generally good feeding for large gulls in those areas, with plenty waste food derived from high densities of eating establishments and tourists, etc. Significant numbers of birds can be seen there during daylight hours, and nearby around Bayswater, c.f. 28/03/04 observations.
One more personal observation from this area: on 22/03/06 15+ birds were observed rising up at dusk from the area near Regent St/Mortimer Street in W1. However, most of them were seen to move off NE at sunset (18:13hrs). Any other records welcome!
Observations from along the roost flight-path confirmed that some proportion of the Herring Gulls in the Hilfield roost in Hertfordshire do indeed travel out from central London, c.f. Figure 10. Records for 23/02/04 and 18/03/04 demonstrated that a good number of large gulls, totalling at least 150 birds, consistently arrive from south-east of Brent Reservoir, whilst the final observations on 30-31/12/04 showed that at least a proportion of these birds, including a decent-sized group of Herring Gulls, travel up over Paddington through Maida Vale. The independent observations over Paddington Green support this and suggest that the route has been in use for a number of years. The flight-path is apparently adapted to the local topography and also follows fairly closely the line of the Edgware Road (A5), which is perhaps used by the gulls as a landmark. The author's 2005 observations suggest that the origin of at least some of the birds is the food-rich areas around the West End, perhaps including a number of birds from Hyde Park (latter mentioned in [12], c.f. observations of 28/03/04) and further south around Victoria or Westminster. The single bird north over St James’s Park on 16/12/04 hints at the latter possibility.
Whilst the main focus of the study was on Herring Gulls, since it was this species which apparently arrived latest in the roost and seemed most likely have travelled furthest, significant numbers of Lesser Blackbacks were also seen using the same route. As this species was often travelling in close association with the Herring Gulls it is quite likely that some, at least, could originate from the same general area as the latter species; alternatively, this species may be feeding over a wider area in north-west London and might have a tendency to join up with other large gulls following the roost flight-path from further south-east. These possibilities could only be evaluated by means of more detailed observations from key vantage points along the route. Yellow-legged Gulls (Larus michahellis) and Great Blackbacks, which are also regular in small numbers in the Hilfield roost, may also travel with these birds.
The observations also showed that a few large gulls move north-west towards the Hilfield roost following more easterly flight-paths, including individuals over Hendon (10/02/04, 22/02/04) and Hampstead (30/12/04). These presumably originate from other parts of north-west or central London. Whilst such birds would certainly boost the total arriving at Hilfield from the south-east sector, they are probably of minor significance compared to the larger core group travelling along the observed main roost flight-path.
Most interestingly, observations from St James’s Park (16/12/04), Primrose Hill (28/02/04) and Hampstead (30-31/12/04) demonstrated that the predominant movement of both large and small gulls over the Paddington-Westminster area is out towards the major gull roosts in the Lea Valley, with some also perhaps travelling east down the Thames valley towards the estuary roosts (see Figure 10). This makes sense as these roosts are both nearer, lying within 10 miles of the area in question, and at much lower altitude than the Hilfield roost; they also normally contain much large numbers of Herring Gulls and usually hold good numbers of all of the common gull species. The birds from the Hilfield roost, moving north-west in the evening, are therefore ignoring the main north-easterly movement over this part of central London which bisects their own roost flight-path. Furthermore, the other major roost in the area, Queen Elizabeth II Reservoir at Walton-on-Thames, also lies closer than Hilfield to the deduced point of departure and is again at a much lower altitude.
The supplementary records of Gladwin from West London, described above under "Additional Observations", are an interesting parallel with the current observation of birds moving across the dominant roost flight paths near central London, and also hint at a possible direct interchange between roosts, with birds perhaps using different roosts on different days, c.f. the Black-headed Gull interchanges noted by Gladwin [2], Sage [14] and Siva-Jothy [22]. However, according to the data obtained in the 2004 WinGS count, very few birds arrived in the Hilfield roost from the west (or the south), amounting to only about 5% of the total arriving from the south-east, and none arrived at the later time periods when the main south-easterly arrival was in progress. In fact, more birds were logged departing (19) than arriving (16) from the south and west sectors as a whole, which does suggest a definite movement in these directions, i.e. towards West London, assuming that birds were not simply recirculating. Despite this suggestion of reciprocal movements, the net contribution of such exchanges with the West London roosts is likely to be small or negligible in the context of the total numbers at Hilfield.
The fact that the main group of gulls coming out from central London to Hilfield are choosing not to travel to any of the three major roosts in the Greater London area, mentioned above, suggests that these birds are not selecting roost site purely in terms of “ease of access” or simple “popularity”. At less than 15 miles each way, travelling to and from the Hilfield roost is not, of course, an abnormally long or arduous journey from central London for this species, which will often travel much further than this on a daily basis (at least 60 miles, c.f. [8, 25]). However, unless there were a good reason to do otherwise it might reasonably be assumed that the birds would tend to join up with the predominant roost flights over their feeding areas, which they clearly are not doing in this case. Considering also the West London observations, with evening departures from an area very close to some major roosts, it might therefore be concluded that there are indeed likely to be significant external factors motivating the birds’ attendance at Hilfield. An associated issue is the possible reason(s) behind the observed steady increase in numbers roosting at Hilfield in recent years (though the current totals of around 100 birds, at peak, are still only a tiny fraction of the historical values!).
Two obvious candidate reasons for these observed behaviours are: 1. the possibility that the Hilfield roost offers increased “security” in some way, e.g. by a relative lack of disturbance and lack of access to ground predators, and 2. the fact that by intermingling with gulls from other areas the birds might somehow be able to exchange information on good feeding, etc. These issues are discussed further below:
Improved security
Regarding pressures from disturbance and predators, it is possible that recent changes in conditions at one or more of the other major roosts in the area has led to them being viewed as less favourable, thereby motivating the birds to travel further to a “better” roost. This explanation would be consistent both with justifying the apparent greater effort expended in order to reach Hilfield and the long-term growth in numbers at Hilfield, the latter being a progressive response to this apparent pressure elsewhere. Hilfield is certainly a sufficiently large area of deep open water that there would appear to be no threat from predators, and there is also currently no water sport or fishing activity there. However, no direct information could be found to justify any of this speculation, and it would no doubt be very hard to obtain such.
Information exchange
It is well-established that the smaller gull species move between major roosts on a daily basis (see section on “Previous studies”, above) and this factor might well also be relevant to large gulls, though easily overlooked due to the relatively small numbers of birds involved. The small movements out of the roost in the south and west sectors in the 2004 WinGS count, c.f. Table 1, together with the observations of birds departing from the area of the West London reservoirs at the time of the evening roost flight [Gladwin, personal communication], suggests that this may well be occurring.
One of the reasons for such movements could be in allowing exchange of information on productive feeding areas local to Hilfield. If this were the case it might therefore be expected that there would be significant daily movements of Hilfield birds to and from other feeding locations, i.e. in other directions. However, relatively few birds apparently travel out to feed in the local areas around Hilfield, c.f. Table 1, and if this reason were the main motivation for birds travelling to Hilfield from central London it would be harder to explain why the progressive increase in the number of birds consistently travelling to and from central London to feed on a daily basis. Also, the potential for information exchange must be greater at the larger established roosts, e.g. those in the Lee Valley, which also contain birds feeding in areas to the north of London. It is presumed that one of the fundamental reasons that gulls roost in large gatherings is to facilitate these exchanges, and as such there would tend to be an opposite pressure inhibiting movement to small and remote roosts.
Hence it seems more likely that reasons under the first category mentioned are the prime motivator. There may of course be many other factors, not considered here, influencing the behaviour of these birds. Whatever the true reason(s) for the birds attending the Hilfield roost, it could also be reasonably speculated that these factors might apply more generally in motivating Herring Gull attendance at different, and less “accessible”, roosts, and hence the question may be of wider interest. Equally, similar knowledge about roost attendance from other areas might shed some more light on the case studied here. Notwithstanding the above speculation, further more detailed study would clearly be beneficial in assessing to what extent each of these factors, and any other significant influences, are actually relevant, and so as to effectively distinguish between them. More general suggestions for further work are described below.
Overall, it is apparent that the roosting and feeding behaviour of large gulls, and Herring Gulls in particular, in Hertfordshire and surrounding areas, is currently not very well monitored or understood. It is interesting to note that concerning Lesser Blackbacks, Gladwin stated back in 1985 that “in view of recent changes, a new study of this species’ movements through the county is needed” [2]. Considering the subsequent period of continued status change there is no doubt a renewed need for detailed contemporary studies of the movements and habits of the larger gulls in Hertfordshire, including Herring Gulls.
This study was clearly cut short due to time constraints and the author’s move out of the area. However, now that the main roost flight-path is established, it should be a relatively easy job to gather more detailed data, particularly by means of careful observations in good visibility from Hampstead Parish Church and Primrose Hill, and perhaps in Hyde Park, together with any other suitable vantage points towards central London. Des McKenzie is now supplying monthly data from Paddington Green, as mentioned above, and this should in due course provide a clearer picture for that part of the flight-path. However, due to apparent day-to-day variation in numbers and in the timing of the roost flight there would also be considerable merit in recording the passage from a number of points along the route on the same day, i.e. with multiple observers. This would serve to confirm the proportion of individuals of each species deriving from each area along the flight-path and help to finally determine how many of the Herring Gulls on the main route originate from the most southerly departure point.
Another aspect meriting further study is the use of the same roost flight path by the smaller gull species. A cursory analysis of the Hilfield data from the 2004 BTO roost survey [1] reveals a very interesting time variation for the directional dependence of the arrival of these species too, as shown in Table 2:
| Time with respect to sunset (minutes) | -107 | -92 | -77 | -62 | -47 | -32 | -17 | -2 | 13 | 28 |
| Total small gulls arriving from south-east | 65 | 410 | 680 | 1050 | 1144 | 1247 | 261 | 114 | 30 | 4 |
| Percentage of small gulls arriving from south-east | 14 | 25 | 25 | 23 | 22 | 40 | 46 | 62 | 70 | 80 |
Table 2 – Small gull arrivals from the south-east (sectors E, F & G), as total and percentage [1]
The final data points involve relatively small numbers of birds, as the total arrival rate had diminished to very low levels towards the end of the count (c.f. Figure 3). However, the almost linear increase in the proportion arriving from the south-east from -47 minutes onwards is striking and whilst overall only 27% of the small gulls arrived from these sectors most of the birds arriving after sunset were from this direction. Since Black-headed Gulls greatly outnumbered Common Gulls in the roost the majority of these arrivals from the south-east may well have been the former species, and indeed the 4 birds recorded as arriving in the last time period definitely were (arriving in sector E, counted by the author). Moreover, whilst Black-headed and Common Gulls were ignored during most of the other counts described here, small numbers of Black-headed Gulls were also logged NW over Brent Cross after sunset on 17/01/04 and moving in the opposite direction before dawn on 30-31/12/04, suggesting that at least a few individuals might indeed follow a similar route. The Paddington Green data reported above also indicates movements of reasonable numbers of this species, and Common Gull, with peaks of 120 and 16 birds along the direction of the dawn and dusk flight paths, on 14/11/05 and 25/2/03 respectively. More work is still needed to confirm these observations and to determine the ultimate departure/destination point for these species.
As mentioned above, it is clear that much more detailed observations would also be required in order to systematically evaluate the underlying factors influencing the birds’ attendance of different roosts, and to consider the associated question of the reasons for the steady increase in numbers roosting at Hilfield in recent years. In addition to further observations along the roost flight-path this would ideally require much closer monitoring of the movements in and out of the roost on a daily basis, and, if possible, at other roosts in the area. As described in “Previous studies”, detailed observations of this type, for Black-headed Gulls, were effectively used by Gladwin [2], Sage [14] and Siva-Jothy [22] to establish interchange between three major roosts in the London area. However, there are significant practical difficulties with such studies, which would preferably involve multiple observers. In particular, the only clear way of establishing whether information exchange is likely to be relevant factor in bringing these Herring Gulls to Hilfield would be to gather detailed information on the directions of movements into the roost in an evening, and back out the following morning. The latter aspect in particular may well prove to be impossible due to the birds moving before it is light enough to identify them with any certainty. It is therefore unlikely that the secrets of these birds will be fully uncovered in the near future.
I would particularly like to thank Bob Cripps, Tony Blake, Jack Fearnside and Steve Murray of the Hilfield Reservoir Bird Report (HRBR) group for very valuable information concerning large gulls at the Hilfield roost. As co-ordinator of the 2004 BTO Winter Gull Roost Survey at Hilfield, Tony also kindly made available the provisional count data [1] which was essential to the study. Herts & Middlesex Wildlife Trust (HMWT) and Three Valleys Water support the HRBR work and permit licensed access to this private site. Also, Chris W. Dee, regional (Herts) WinGS co-ordinator, kindly checked a draft version of this report and provided advice about use of the provisional WinGS data. I was also grateful to receive comments from Bryan Sage and Mike Siva-Jothy, based on their own earlier studies at the Hilfield roost, and Tom Gladwin, including some more recent observations from the West London area.
Further afield, thanks to Andrew Self, Bird Recorder for London (London Natural History Society (LNHS) recording area), for information on large gulls at Brent Reservoir, Des McKenzie, former LNHS Bird Recorder for Inner London, for information on gull movements over Paddington Green, and Dr. Meg Game, an ecologist for The Corporation of London, for information from the Hampstead Heath area.
Finally, the most deserved and heartfelt thanks go to my wife and daughter (aged 15 months in December 2004) who have put up with another year’s apparently inexplicable and obsessive interest in what, to most, is a particularly uninspiring, and often much-disliked, urban pest. Several of the observations described above were made in the context of general family outings, for instance the Parliament Hill trip on 14/02/04 doubled as a Valentine’s Day walk. We're still together. I was particularly grateful to my daughter who settled down quickly at a family party on 30/12/04 allowing me to pop out for one more time, making what turned out to be the critical observations for the whole study! Though, in the end, only a tiny insight was obtained into the lives of these resourceful birds it proved fascinating to me and will hopefully contribute in some small way to inspiring further studies of Hertfordshire’s gulls in the years to come.
1. BTO 6th Winter Gull Roost Survey (WinGS): provisional data for Hilfield Park Reservoir for survey of 17/01/2004 (local (Hilfield) co-ordinator: Tony Blake/regional (Herts) co-ordinator: Chris W. Dee); see also http://www.bto.org/survey/special/winter_gulls.htm
2. Gladwin, T.W. & Sage, B.L. (1986) “The birds of Hertfordshire”, Castlemead Publications/Ward’s Publishing Services, Ware, Herts, UK, ISBN 0-948555-01-7
3. Sage, B.L. (1958) “A new gull roost in the London area”, London Bird Report, vol. 22, pp. 50-51
4. Fearnside, J. (1993) “Gull roost survey, Hilfield Park Reservoir, Hertfordshire, 1 November 1992”, in The Hertfordshire Bird Report 1992, Hertfordshire Natural History Society: Herts Bird Club, Transactions, vol. 31, pt. 6, pp. 454-457, December 1993, ISSN 0375-3409
5. Fearnside, J. (1994) “BTO national gull roost survey, Hertfordshire, January 1993”, in The Hertfordshire Bird Report 1993, Hertfordshire Natural History Society: Herts Bird Club, Transactions, vol. 32, pt. 2, pp. 170-172, December 1994, ISSN 0375-3409
6. Cripps, R. & Fearnside, J. (eds.) (2004) “Hilfield Park Reservoir, Bird report for 2003”, HMWT/Three Valleys Water
7. Welch, S. (2004) “Some observations of the Hilfield gull roost – 2003”, informal report supplied to Herts Bird Club
8. Welch, S. (2004) “Large gulls in the Banchory area: Observations of numbers and movements 1984-2004”, http://geocities.datacellar.net/steve_extra/Banchory_gulls.html
9. Welsh Harp Conservation Group (WHCG) website (2005), including on-line 1996-1997 bird reports: http://www.brentres.com/
10. Lack, P. (1986) “The atlas of wintering birds in Britain and Ireland”, BTO/IWC, T & A D Poyser, Calton, Staffs, UK, ISBN 0-85661-043-7
11. Burton, N.H.K., Musgrove, A.J., Rehfisch, M.M., Sutcliffe, A. & Waters, R (2003) “Number of wintering gulls in the United Kingdom, Channel Islands and Isle of Man: a review of the 1993 and previous Winter Gull Roost Surveys”, British Birds, vol. 96, pp. 376-401, August 2003
12. Mitchell, D. (1997) “Where to watch birds in the London area”, Christopher Helm/A & C Black, London, ISBN 0-7136-3868-0
13. The Hertfordshire Bird Reports (editions for years 1984 to 2001), Hertfordshire Natural History Society: Herts Bird Club, Transactions, volumes 29 to 35, ISSN 0375-3409
14. Sage, B.L. (1959) “A history of the birds of Hertfordshire”, Barrie and Rockliff, London, ISBN/ASIN B0000CKFO4
15. Sage, B.L. (1964) “The gull roosts of the London area”, London Bird Report, vol. 28, pp. 63-68
16. Sage, B.L. (1970) “The winter population of gulls in the London area”, London Bird Report, vol. 33, pp. 67-80
17. Harris, A. (1996) “Gulls at Hadham Towers landfill site, Widford, Herts., 1994”, in The Hertfordshire Bird Report 1994, Hertfordshire Natural History Society: Herts Bird Club, Transactions, vol. 32, pt. 4, pp. 387-396, February 1996, ISSN 0375-3409
18. Herts Bird Club website sightings archive, from March 1999 to date (2005): http://www.hertsbirdclub.org.uk/sightings.html
19. Herts Birding yahoo groups forum (2005): http://groups.yahoo.com/group/hertsbirding/
20. Dean, A. (1987) “Seasonality of Herring Gulls in the west Midlands”, British Birds, vol. 80, p. 632, December 1987 [also available online on the West Midland Bird Club website, here: http://www.westmidlandbirdclub.com/biblio/bb/80-632.htm]
21. Stanley, P. I., Brough, T., Fletcher, M. R., Horton, N. & Rochard, J. B. A. (1981) “The origins of Herring Gulls wintering inland in south-east England”, Bird Study, vol. 28, pp. 123-132
22. Siva-Jothy, M. (1982) “Some observations of the arrival patterns of gulls at a winter roost”, Third year student project, University College, London
23. Central Science Laboratory, Wildlife Ecology and Management, Bird Management Team (2005): http://www.csl.gov.uk/science/organ/wem/bird/
24. Marylebone Birdwatching Society - MBS (2005): http://geocities.datacellar.net/birdsmbs/
25. Rock, P. (2005) “Urban gulls: problems and solutions”, British Birds, vol. 95, pp. 338-355, July 2005
Appendix – All 2004 observation data
Data from timed roost flight-path counts and roost observations
Format:
Date
Conditions
Location
Sunset/sunrise – approximate time
Timed record of observations of all species of note
Totals for each species, where relevant
Report by Stephen Welch, January 2005, updated January 2006