2. Base Counts and Timings for Honey-buzzard Movement in UK in September 2008

2.1 Base Counts in UK with Flight Direction

Table 1 shows the counts of Honey-buzzard reported in the UK on BirdGuides for September 2008. BirdGuides is thought to the most useful resource as the reports are given in full as given by the observers without any interpretation. The BirdGuides data is therefore base data, ideal for further analysis. Of course County Bird Reports have also been consulted for other purposes in this report but identification criteria vary considerably from county to county, submission rates are unknown and important facets such as ageing of birds and flight directions are often treated cursorily or even not covered at all. Table 1 is presented in 3 main parts. The first deals with the 3 main regions, itemising the counts daily through September, for North-east England, East Anglia and South-east England. The second part deals with 3 more regions: Midlands, South-west England and Scotland. The third part deals with the remaining significant region of North-west England and counts elsewhere, for Wales and the Channel Islands. This part also summarises the direction of flight for all regions; direction S-E means the birds were recorded as moving towards compass points between S and E, and SW-W towards compass points between SW and W. Trek means that no direction was given in terms of compass points, the birds being recorded as flying ‘over’, ‘in-off’ or between geographical locations; it is in principle possible to analyse the last named but this has not been attempted yet. The ‘in-off’ category is considered further below. Other includes birds flying towards compass points with a northerly component, basically NE, N or NW, or simply resting. Clearly it would be useful to break down the sizeable Trek component into compass points and the current high numbers of birds in this category must be regarded as a limitation of the data set.

2.2 Summaries of Counts and Directions of Flight through the UK in September 2008

Gross Volume

Table 2, derived from Table 1, shows by region the gross numbers of Honey-buzzard recorded on BirdGuides through September 2008 in the three main periods, identified as 1/9-12/9 with few birds moving before the big exodus, 13/9-21/9 when by far the bulk of the movement occurred and 22/9-30/9 when a steady but smaller outflow occurred. Only very obvious duplicates have been removed from the counts.

Table 2: Gross Numbers of Honey-buzzard recorded on BirdGuides in September 2008 by Region of UK

The highest numbers of Honey-buzzard are usually found in SE England, East Anglia and NE England. Clearly regions further south have a tendency to record higher totals as the birds from further north usually pass through them. 2008 is remarkable therefore for the counts in East Anglia and NE England being considerably higher than in SE England. In 2000 SE England recorded the highest counts as birds departed from the south coast of England to France. From the coverage point of view NE England is covered well on the coast but poorly inland. East Anglia and SE England are presumed to be better covered, certainly in inland areas. Looking at the individual days from 13/9-21/9 for these 3 regions, we see the unusual regional pattern in Table 3.

Table 3: Gross Numbers Recorded on BirdGuides in the three Main Regions from 13th-21st September 2008

So it appears that high numbers passing through East Anglia on 13/9-14/9 did not follow through into SE England. We can see some individual examples of this. For instance where did the 18 moving S at Minsmere on 13/9 from 10:25-13:25 go? Indeed although there may be some duplication, where did the 184 birds passing though East Anglia from 13/9-14/9 go? On the figures above, they did not pass through the well-watched SE corner. Later in the period from 19/9-21/9 we see the pattern more as expected with numbers in SE England lagging slightly and matching the counts in East Anglia. Comparative numbers in NE England and East Anglia are well-matched throughout, suggesting a synchronised movement (perhaps weather-related) rather than a rippled one through the two regions.

Table 4 shows the directions in which the birds were moving in September 2008 for the UK as a whole. The terminology is defined above.

Table 4: Direction of Honey-buzzard Movements in the UK in September 2008

The reason for classifying the compass points as above is to distinguish between birds moving in a westerly direction and those moving in a southerly direction S or a continent-leaning direction of SE/E, together S-E. The results show that the numbers moving S-E heavily outnumber those moving SW-W by slightly over 4:1 with 81% of the birds in the month going in this direction. Further the proportion moving S-E is quite constant throughout the month being lowest at 65% in the first 12 days and highest at 86% on 14/9, 87% on 15/9, 93% on 18/9 and 86% from 22/9-30/9. Thus at no time is there any indication of an influx of birds over the North Sea spreading westerly over the country.

By region the %S-E is 83% for NE England, 88% for East Anglia, 63% for SE England, 72% for the Midlands, 94% for SW England, 50% for Scotland and 83% for NW England. The regions where a westerly movement would be expected if birds were coming in off the North Sea are NE England and East Anglia but 83% and 88% respectively of birds in these areas are moving S-E. The almost total absence of ‘in-off’ records in NE England also supports the idea that the birds here were not coming off the North Sea.

A further breakdown was made for East Anglia showing more precise directions, as shown in Table 5. The table shows that East Anglia was the only region where ‘in-off’ birds were recorded on a significant scale (about 5% of total). Other records were in NE England on the south side of the Humber, hence crossing this estuary from further N, with an isolated record in County Durham.

Table 5: Breakdown of Direction for Honey-buzzard Movement in East Anglia in September 2008

The 16 birds recorded as 'in-off' records on the coast included (from BirdGuides):

13/9

Norfolk (north): Blakeney Point (09:12), East Runton (10:57), Cley (11:35);

Norfolk (north-east): Horsey (moving inland, 10:35), Waxham (11:10), Trimingham (mid-morning), Waxham (moving inland, 11:30), Waxham (12:32);

Norfolk (east): Caister-on-Sea (early morning, 2 birds);

Suffolk (north-east): Lowestoft (moving inland, 08:43), Corton (12:45, 2 birds).

14/9

Norfolk (north): Wells Woods (06:50).

15/9

Norfolk (north): Sheringham (09:50).

20/9

Norfolk (north) : Titchwell (no time).

Six of the 16 birds were recorded from north Norfolk where it is most likely that they had crossed the Wash and made landfall on the southern side. The records on 13/9 from further east in Norfolk and in north-east Suffolk are much more interesting. It is possible that these birds have also crossed the Wash but are pushing out further to the east and making landfall again as they skirt the coast.

On 14/9 the direction through East Anglia seems quite unusual with 19 birds moving E and 4 SE. These included 6 E at Welney, Norfolk, at 12:10, at least 9 E at Burnham Overy, Norfolk, from 10:30-11:15, and 3 SE at Cantley Beet Factory, Norfolk, at 13:00. On 18/9 5 birds moved SE at Beetley, Norfolk, at 11:45. Overall in the month there is a definite eastward lean in the East Anglia data with slightly over half as many birds going E/SE (57) as going S (112). Almost 3 times as many birds (57) are going E/SE as are going SW/W (20).

It is necessary to consider the presence of quite large numbers of Common Buzzard in the movement. This is a surprising feature as it is too early for the migration of significant numbers of Common Buzzard from Scandinavia. British-bred Common Buzzard are said to be sedentary. However, some clues come from Prytherch [2013]. The population has been rising rapidly: “The population [in Avon] rose from 13 pairs in 1982 to 105 pairs in 2012” … “The study population was probably self-supporting throughout and latterly it appeared to produce more adults than it could absorb” [Abstract]. Later [p.276] he comments: “On maturing they might have tried to settle locally, but were forced to move away. Some other juveniles disperse widely and either settle without returning or move back to their natal area ... Therefore, many new adult birds in my study area (or in other similar areas) now have little choice but to move elsewhere in order to establish territories. Most 'empty' ground was likely to be to the east.”

So Common Buzzard migrants in UK in autumn, particularly in eastern areas, are likely to be British birds, irrupting from breeding areas. The rise in Common Buzzard migrants in eastern areas has nothing to do with a Scandinavian origin as is supposed in some quarters. It's obviously more romantic to treat a Common Buzzard migrant as of continental origin but the reality is that it's British, seeking space for a territory. There's very little evidence for migration of continental birds through Britain from ringing returns; further, dramatic increases in UK passage are correlated with increased UK populations, not with unchanged populations on the continent.

It is likely that a number of observers are not aware of the similar appearances of juvenile Honey-buzzard and Common Buzzard. So Common Buzzard, as the safer default identification, may have been over-estimated and Honey-buzzard juvenile underestimated.

2.3 Timing of Honey-buzzard Movement within UK from 13-14 September 2008

Using the Honey-buzzard records on BirdGuides on 13/9 and 14/9 for which times were noted in the submission, a number of calculations were made:

• the numbers of Honey-buzzard moving, on an hourly basis, each day

  • through each county

  • through each region

• the average time at which the Honey-buzzard moved through each county, each day

The time was taken as the time reported for the observation. Occasionally interpolation was used to itemise individual sightings e.g. 4 birds from 10:30-11 would be treated as sightings at 10:30, 10:40, 10:50 and 11:00. Multiple sightings at a given time would be weighted accordingly e.g. 3 birds at 11:15 would be treated as 3 sightings at 11:15. 11:15 and 11:15, respectively. If the only time available was that at which the sighting was reported to BirdGuides, then the observation was ignored; similarly vague times like 'morning' and 'afternoon' were also ignored.

Table 6(a): Timing of Honey-buzzard Movement in UK on 13 September 2008, by region

Table 6(b): Timing of Honey-buzzard Movement in UK on 14 September 2008, by region

Tables 6(a) and (b) show the hourly pattern during each day of the major movement, 13/9 and 14/9, respectively. The peak for all regions is from 10:01-15:00 on the 13/9 and 09:01-13:00 on the 14/9. These migration times are just what would be expected in normal movement with birds avoiding the start and finish of each day, when visibility may be poorer, and taking advantage of the higher temperatures in the middle of the day, which facilitate improved buoyancy. There is no difference in the peak between NE England and East Anglia on 13/9, which suggests that the birds are following a standard migration pattern, starting their flights early morning and moving for 5 hours. On the 14/9 the pattern is similar except that the peak runs one hour earlier, perhaps because the weather was better on the morning of 14/9 than on 13/9.

What is very obvious is that the vast majority of birds moving through East Anglia disappear from the UK at this point and do not exit from the south coast, as in the 2000 movement. It is perfectly possible that, on both 13/9 and 14/9, the great bulk of the birds moved directly from East Anglia to Benelux, crossing the southern extreme of the North Sea and extending their daily trek by at least 3 hours. While most birds crossed the southern North Sea, there was still a small, steady movement through SE England, taking place in the afternoon from 12:01-17:00 on 13/9 and running all day on 14/9 from 07:01-18:00. This movement is thought to be of birds moving through East Anglia earlier and keeping due S, rather than moving E/SE across the North Sea. Another subtlety is that on 14/9 the movement continues at a significant level in the afternoon in NE England, perhaps because of the movement of Scottish birds from further N. So there is apparently some ripple effect from Scotland to NE England and from East Anglia to SE England, the discussion of which is dealt with below when the individual county information is considered.

Table 7 shows the average time, and earliest and latest times, at which Honey-buzzard passed through each county recording area on 13/9 and 14/9. The prime reason for looking at events on a finer-grained county basis is that this might show more clearly the ripple effects, where birds moving through one county, were subsequently recorded in another county further S. On the 13/9 there is very little evidence for ripple effects: the average times in counties far apart such as Northumberland and Durham, about 13:15, and Norfolk and Suffolk, about 12:20, are quite similar. Indeed throughout NE England the average times range only from 12:09-13:18 for Northumberland, Durham, Cleveland, N Yorks and Lincs, with E Yorks earlier at 10:47 and S Yorks later at 15:11. This pattern is consistent with a synchronised movement from the hinterland to the coast, with birds commencing their migration inland in their breeding areas on improving weather conditions and moving SE/E to the coast, as observed at Stocksfield by NR. The pattern is inconsistent with a ripple effect whereby the birds start say in Northumberland and move S through each county in turn. It could be argued that the birds had arrived over the North Sea in a broad front from the E; however, in this case some of the birds would be expected to be moving W, rather than the observed E/SE, and the behaviour would be expected to be much more chaotic after such an arduous journey of say 600 km from Denmark, with birds clearly seeking resting and feeding areas; as discussed elsewhere claims for a small number of birds moving 'in-off' are unconvincing, such records being much more likely to simply reflect the layout of the coast, or even observer wishfulness. There is one obvious ripple effect – between Norfolk and Suffolk – with average times of 11:57 and 12:43 respectively. So the birds moving through Suffolk may have already been seen in Norfolk, around an hour earlier.

On 14/9 there is an early rapid rush across Lincs (in NE England) and Norfolk (in East Anglia). 20 birds move through Lincs with average time 08:38 and 58 through Norfolk with average time 10:50. This movement is likely to be of birds moving through NE England on 13/9 that had roosted in S Yorks/E Yorks/Lincs overnight (13/9-14/9), having travelled perhaps 150-250 km on 13/9 from their breeding areas in 4-5-hours flight at 40-50 kph; for instance Newcastle-Doncaster is 165 km, Hexham-Lincoln is 217 km. This suggests that the early birds flying through Lincs and Norfolk were not new birds: they may have already been observed the day before in NE England; further the Norfolk birds on 14/9 may have been seen earlier in Lincs. In SE England passage is later than in East Anglia with average time of 13:30 against 11:00; further the passage had apparently diminished with 23 birds timed in the well-watched SE England as against 73 in East Anglia. Again this looks like a ripple effect from East Anglia to SE England. But as on 13/9 the flow seems to have mainly moved off SE/E, again to Benelux to join the major exodus there. The birds passing through SE England are those that chose not to cross the southern North Sea, maybe mainly juveniles who relied on their instincts to adopt a southward drive. Further north, in NE England in Northumberland, Durham, N Yorks, E Yorks and S Yorks, the later average time varying from 12:25-15:58 than in Lincs of 08:38 indicates a further fresh arrival from the hinterland and from the north in Scotland.

Table 7: Earliest, Average and Latest Time of Honey-buzzard passing through each County Recording Area on 13 and 14 September 2008

An obvious question is: how many birds were observed in the UK movement? Taking the ripple effects literally, then the Suffolk totals should be subtracted from the total for 13/9 and the Lincs, Norfolk and Suffolk totals should be subtracted from the total for 14/9. However, it is not thought such precise calculations are warranted: there are so many variables in the data set such as proportion of actual movement passing over a station being observed on the ground, the proportion of the actual movement passing over unobserved areas, and the proportion of the actual movement being accurately identified at a station. If the movement was partially inland, then the proportion detected will be substantially reduced as in September most observers are on the coast; compounding this problem is the perhaps controversial view that observers inland, say patch-workers or vismig specialists, are less likely to identify Honey-buzzard than rarity seekers on the coast. There is also the question of whether the totals should reflect numbers reported on BirdGuides or the numbers established after County Records Committee review. While the latter would be the normal preference, the politics surrounding the Honey-buzzard, the uncertain standards for identifying Honey-buzzard and the reluctance of observers to submit formally records to their County Records Committee, make this a difficult choice. This matter will be discussed again.

Summary: The large movement of Honey-buzzard in the UK in September 2008 peaked from 13/9-21/9 when, on BirdGuides, 776 of the total of 886 for the month were reported. Passage was light earlier in the month from 1/9-12/9 with just 35 noted but was still significant late in the month from 22/9-30/9 when 98 reported. The highest numbers regionally in the peak period were found in East Anglia (271), NE England (241) and SE England (110). Because of the ripple effect, regions further south typically report larger numbers than those to the north, so the relatively low count in SE England is surprising and a reversal of the situation in 2000.

For the UK as a whole, 81% of the birds were moving between S and E in the entire month where the direction of flight was reported. In the regions where a W movement would be expected if the birds were coming off the North Sea, 88% were moving between S and E in East Anglia and 83% in NE England. The only region where a significant number of in-off records was reported was East Anglia with 16 noted. It is thought that these are of birds coming off the sea on the north side of Norfolk from further north in England. On 14/9 the eastward lean is very marked in East Anglia with 19 birds moving E and 4 SE.

The migration times are as expected in normal movement with birds avoiding the start and finish of each day, when visibility may be poorer, and taking advantage of the higher temperatures in the middle of the day for lift. On 13/9 there is no difference in the peak between NE England and East Anglia, suggesting that the birds are following a standard migration pattern, starting their flights early morning and moving for 5 hours. Indeed throughout NE England the fairly constant average times suggest a synchronised movement from the hinterland to the coast, with birds commencing their migration inland in their breeding areas on improving weather conditions and moving SE/E to the coast. On 14/9 the pattern is similar except that the peak runs one hour earlier, perhaps because the weather was better. Also this day there is an early rapid rush across Lincs (in NE England) and Norfolk (in East Anglia), likely to be of birds moving through NE England on 13/9 and roosting in Yorks/Lincs overnight. On both days passage in SE England is later than in East Anglia indicating a ripple effect but the flow is much diminished further south with most birds moving off SE/E from East Anglia.