7 The Honey-buzzard Movement across Southern Europe 11-20 September 2008

From the Internet, counts were compiled as far as possible on a daily basis for S Europe. Sources used were: France, Spain Trektellen, Gibraltar GONHS 2008 Report, Malta BirdLife-Malta Report. The results are shown in Table 11, together with those from the UK for comparison. Note that for Malta the 700 count for 19/9 comes from Sammut, Fenech, & Pirotta (2013); BirdLife recorded 400 this day.

7.1 Overview

The movement through Benelux is reflected very quickly in the figures available for France. It appears that the birds exiting Belgium on the morning of 14/9 travelled the 800km to the Spanish border, east of the Pyrenees, at speed with 503 noted there on 15/9 and 124 on 16/9 with very few thereafter. Conditions were very favourable for a fast crossing of France with strong N/NW winds and clear skies. The birds would have flown around 400km each day, perhaps doing an 8-hour day at 50km/hour. Of course coverage in France is only partial but the lack of passage due south of Belgium after 16th suggests that this source was largely exhausted.

In Spain the main coverage was near the Straits of Gibraltar. The late movement was not conspicuous here with the last significant passage noted on 12/9 with 451 birds passing Mirador de las águilas (Andalusia) and 1718 noted at Gibraltar (GONHS), probably with some duplication. It is 1068 km from Narbonne in the eastern Pyrenees to Gibraltar, at least 3 days flying under good conditions at 400 km a day. So the movement near Narbonne in France from 12/9-16/9 would not be expected to reach the Straits until 15/9-19/9 at the earliest. From 16/9-21/9 587 birds were counted at the Straits but this is significantly less than the 1441 counted near Narbonne and Perpignan from 12/9-16/9. It is notoriously difficult to produce comprehensive counts at the Straits because of the variety of tactics that are adopted by migrating raptors so it is likely that many birds escaped being counted, when the late wave passed through. Very few were noted at the Straits for the rest of the month, with 147 from 21/9-30/9.

Some additional interesting figures come from Malta where an exceptional count of 700 was made on 19/9. In their paper on raptor migration through Malta, Sammut, Fenech, & Pirotta (2013) noted, as an exceptionally heavy passage involving probable night movement, that 700 Honey-buzzards were counted on 19th September 2008, with over 300 roosting within the immediate vicinity of the watchpoints, yet fewer than 50 left the roost the following morning from the whole area. There is a fascinating possible correspondence here between the last heavy movement through Denmark of 801 on 11/9, mostly in the east of the country in Sjælland, and the count at Malta. The distance from Copenhagen to Malta is 2200 km, giving a migration speed of 275 km per day over the 8 days from 11/9-19/9, which is reasonable for such a period. However, as said above, too much reliance should not be placed on recorded counts per se. The interest in the Malta counts is that it suggests that a significant number of the birds from Denmark did pass due S through Italy and Malta on their way to Africa. Over a wider period from 13/9-28/9 at the Buskett observation post in Malta, 676 Honey-buzzard were noted, including 400 on the 19/9 [BirdLife 2008 Report]. Some Honey-buzzard migrate directly to Libya without passing by Malta but no counts across the Straits of Messina could be found for 2008.

Again, looking at Table 11, the figures for the UK are extraordinarily late in their timing, even in a southern European context. Almost as many birds (440, 49.7%) move in the UK in the second part of the month, defined as 15/9-30/9, as in the first part (446). Elsewhere only about 11% of the birds move in the second part, that is only 2492 (11.1%) out of the month's total of 22534 in France, Spain and Malta move from 15/9-30/9.

Because of its pivotal role, the data available for France country is discussed in more detail below.

7.2 France

An obvious question is where did the birds go after their strong arrival in the southern parts of Benelux on 13/9 and 14/9. Information for France is very incomplete on Trektellen but there are a number of stations in NW France in Picardy and Pas-de-Calais, at least one station in SE France in Rhône Valley and at least one station in S France, E of the Pyrenees. Heavy passage in France was recorded on 1/9 (2851 on passage) and 7/9 (806 on passage), Although we are more concerned with the movement around and just after 13/9-14/9 to check on the progress of the birds arriving to the N in Benelux and earlier in Germany, we need to cover a broader time-spell to be relatively sure of the course of the movement through France. Therefore the passage data has been collated from 11/9-20/9 for the 5 Trektellen sites in France reporting Honey-buzzard. At 2 of these sites, Banc de l'Ilette (coast, Picardy) and Cerdagne (near Perpignan), birds were only noted on one day, But at the remaining 3 sites, birds were noted on 7 days at Roc de Conilhac (near Narbonne), 6 days at Andance (near Lyon) and 3 days at Dune Marchand (coast, near Belgian border). The results for these 3 sites are tabulated below in Table 12, with details given of the counts of Honey-buzzard and any associated Marsh Harrier, Common Buzzard and Osprey, the weather for each day and the time of the watch.

* +1 at Banc de l'Ilette (coast, Picardy) on 11/9; ** +44 at Cerdagne (near Perpignan) on 15/9; ^† MaH Marsh Harrier, CBuzz Common Buzzard, Osp Osprey.

Table 12. Counts of Honey-buzzard and some other Raptors in France 11-20 September 2008 (Trektellen)

The movement through Benelux from 13/9-14/9 had only a very faint follow-through in north-west France with just 3 recorded on these two dates. It is therefore clear that in France the birds moved decisively away from the North Sea coasts, with a bias away from the SW towards the S in continuing their journey. On the broader time-scale, 6 birds were noted from 11/9-20/9, suggesting this is not a significant route in general.

Much more significant movement was noted in the Rhône Valley with 117 counted from 11/9-20/9 of which 82 were on 14/9. The distance from Ghent-Lyon is 624 km. At the Trektellen stations, weather conditions included N winds in the Rhône Valley on 14/9-15/9. Weather Underground gives more detail. In Lyon, winds were NNW/N/NNE at 18-28 km/hour from 09:00-17:00 on 14/9 in dry weather WU; N at 20-35 km/hour from 09:00-17:00 on 15/9 in continuing dry weather WU. In such good conditions for migration the birds might cover overland 400 km per day, doing 8 hours flying at 40-60 km per hour. However, it would still take at least 2 days for the birds to reach the Rhône Valley from Benelux so the birds leaving Benelux on 14/9 cannot have reached the Rhône Valley the same day. Unfortunately no counts were made from 15/9-16/9 when the birds from Benelux might have been expected to pass but low counts from 17/9-18/9 give no evidence for this being a significant route from Benelux. The birds in the Rhône Valley on 14/9 may have been following a SW route from Denmark to the east of the Pyrenees. The 375 km from Lyon to Perpignan could be covered in one day, under good conditions, meaning these birds would arrive at the Pyrenees on 15/9.

Data from Switzerland was also examined to check routes on the eastern edge of France. These show that the last significant passage (p.13) at Défilé de l’Ecluse, Haute-Savoie, was on 5/9, totalling 570 birds, with further much smaller numbers up to 10/9. The heavy passage on 5/9 of 570 birds was followed closely by 261 at Andance in the Rhône Valley on 7/9 and 545 at Roc de Conilhac on the same day. So it appears that the bulk of the late flow at Haute-Savoie had moved through southern France quickly, making this at best a minor route for the middle of the month.

The highest counts were in south France, east of the Pyrenees, where 1524 were noted from 11/9-20/9 with 312 on 12/9, 295 on 13/9, 207 on 14/9, 503 on 15/9 and 124 on 16/9. The distance from Ghent-Marseilles is 827 km, which is just achievable in 2 days flying with wind assistance. At the Trektellen stations, weather conditions included strong NW winds in the E Pyrenees from 12/9-15/9. Weather Underground gives more detail. In Perpignan, winds were WNW/NW at 32-59 km/hour (gusts to 87 km/hour) from 09:00-17:00 on 14/9 in dry weather WU; NNW/N at 24-43 km/hour (gusts to 67 km/hour) from 09:00-17:00 on 15/9 in dry weather WU; light and variable at 4-20 km/hour from 09:00-17:00 on 16/9 in dry weather WU. So the winds were extremely favourable for rapid migration on the whole route, enabling birds leaving southern Benelux early on 14/9 to reach the extreme south of France from late on 15/9 through to 16/9, following the movement of the low pressure over Benelux to the SSE. These birds appear to have moved due S, keeping west of the Rhône Valley, where numbers were much smaller than at the Pyrenees.

For the birds from Britain, the move over the southern North Sea had paid dividends by enabling a switch to be made away from the Atlantic to more settled weather (and more thermals) on the continent.

This simple picture of exit from Benelux early on 14/9 and arrival east of Pyrenees on 15/9 and 16/9 needs some adjustment when a broader time-scale is considered. Birds were moving in high numbers to the east of the Pyrenees from 12/9-14/9, clearly too early to be those who had migrated through Benelux from 13/9-14/9: even with the very favourable conditions for migration on 14/9, it is most unlikely that Honey-buzzard migrating from Benelux would achieve over 800 km in 1 day. It is likely that these birds were part of the late exodus from Scandinavia, recorded in Denmark from 5/9-11/9, totalling 2945 (274 Jutland, 2671 E Denmark). The distance from Copenhagen to Marseilles is 1469 km, so this would take around 4-7 days in a variety of weather conditions at 200-400 km per day, in line with the peak late period counts east of the Pyrenees from 12/9-16/9. The total recorded east of the Pyrenees in this period is 1524 birds (814 12/9-14/9, 710 15/9-20/9), 57% of the total noted in Denmark, but the timing is of more significance than the recorded numbers, which will be imprecise due to the vagaries of counting techniques, including duplication of counts at different sites, coverage being patchy and birds being missed. In Denmark the counts, except for the spectacular 524 on 11/9 at Skansebakken, came from a variety of stations, so the same bird may have been counted several times and the overall totals will be an overestimate of the numbers moving. There are also some further systematic effects. In particular, reducing the numbers expected at the Pyrenees, some birds from Denmark are likely to have flown due S through Italy and Malta (as noted above). Increasing the numbers expected at the Pyrenees, some birds will have come from Britain.

It is suggested that these early birds east of the Pyrenees from 12/9-14/9 were from a second stream which left Denmark and moved SW across Germany and France from Schleswig-Holstein in a relatively direct route to the Pyrenees. These birds are likely to have left Denmark from 5/9-8/9 as a minimum of 4 days is needed for the trek. Finally we can speculate about the fate of any birds that took part in the second stream after crossing into Denmark on 11/9. These would have drifted less to W, moving on a more SW course on 12/9 through Germany (The Honey-buzzard did Cross the North Sea: from England to Benelux). Ahead to the SW, conditions were not good for passage on 13/9. For instance at Dijon it rained all day with winds N 20-25 kph WU; at Strasbourg it rained all day with winds N/NW at 10kph WU. However, it is most unlikely that the birds would have travelled the 1010 km from Lubeck to Dijon in two days so delays were probably minimal. Conditions improved on 14/9 enabling birds arriving from the NE to progress: at Dijon it was dry with winds N at 20-30 kph WU; at Strasbourg it was dry with winds NE at 22-28 km/hour WU. It is 449 km from Dijon to Marseilles so it is just possible for the birds travelling through Dijon on 14/9 to arrive the same evening. However, some would arrive at least one day later so birds in the second stream would be at best one day earlier than those in the first stream, which had been drifted W towards the North Sea coasts of Benelux, where they were also temporarily blocked.

A map showing the routes across southern Europe is given here. There is considerable uncertainty in the routes because of the incomplete coverage of France by Trektellen. Finding further sources might increase the confidence with which the picture is painted.

7.2 Summary

The Honey-buzzard appeared to make a rapid exodus from Benelux through France on favourable N/NW winds as the blocking low pressure over Benelux moved SSE. The birds moved due S, making exceptional speed at 400 km per day, following the movement of the low pressure to the SSE, mainly to the west of the Rhône Valley, arriving east of the Pyrenees from 15/9-20/9 with 710 birds noted. This stream included the birds from Britain: the move over the southern North Sea had paid dividends by enabling a switch to be made away from the Atlantic to more settled weather (and more thermals) on the continent. There appears to have been very little movement SW onto the north-west coast of France. Other birds were noted east of the Pyrenees from 12/9-14/9. These were thought to have taken a relatively direct SW route from Denmark through Germany, moving to the east of Luxembourg and then S towards the eastern Pyrenees. Birds in both movements were drifted by the easterly winds but the ones moving W were slowed down by the blocking low pressure over Benelux. It appears that the last major movement in Denmark continued south through Italy and Malta. The data confirms that the figures for the UK are extraordinarily late in their timing, even in a southern European context.