SPRING MIGRATION OVER PUERTO RICO AND THE WESTERN ATLANTIC, A RADAR STUDY

Migration over Puerto Rico was recorded by time-lapse filming of the display of a long-range surveillance radar on 40 days and 37 nights in the period 2 March-29 May 1971. Moderate density movements occurred every night; low density movements occurred on most days. Many birds, primarily passerines, took off from Puerto Rico each evening at 20–45 minutes after sunset.
Almost all birds flew to the west, NW or north. Birds were seen approaching from the direction of the Windward Islands and Venezuela, over Puerto Rico, and departing towards the Bahamas and eastern coast of the U.S. Uni- and multivariate analyses showed that the number of birds departing W-N each evening was positively correlated with following winds.
There is less night-to-night variation in the amount of migration at Puerto Rico than in eastern North America. However, this is apparently the result of less variable weather in the tropics, not the result of any lesser degree of meteorological selectivity by the migrants.
The tracks of the birds were correlated with wind direction. Birds moved WNW-NW with NE side winds but NW-NNW with SE following winds. The tracks were rarely exactly downwind. The variance amongst the directions of individual birds at any given time was usually small and not correlated with cloud cover or magnetic disturbances. The estimated headings of the birds varied from day to day in a pattern suggesting adjustment of headings to compensate at least partially for lateral wind drift.
In autumn many birds approach Puerto Rico from the north or even east of north; in spring few birds moved in the opposite directions. This difference in routes takes advantage of prevailing wind patterns.     

A RADAR STUDY OF THE AUTUMN MIGRATION OF WOOD PIGEONS COLUMBA PALUMBUS IN SOUTHERN SCANDINAVIA

The migration of Wood Pigeons in southern Scandinavia was studied from 21 September to 10 October 1971 and from 16 September to 15 November 1972 using radar stations supplemented with observations from an aircraft and a network of ground observers. By far the largest quantities of Wood Pigeons migrated after cold front passages with northwesterly to northeasterly tailwinds. Most birds departed on a few days, apparently as a consequence of strong preference for tailwind situations. With northwesterly winds a proportionately high migratory activity was recorded in the Kattegatt area. With northeasterly winds activity was higher in the Baltic area. This allowed the Wood Pigeons to make maximal use of the tailwind component, and their ground speed usually exceeded 80 km/h. The calculated mean air speed was 60 km/h. Their dependence on tailwind was particularly strong when the birds were engaging in long sea-crossings, such as across the Kattegatt. Different coastlines affected the geographical pattern of migration in different ways. Frequently Wood Pigeon flocks flew almost parallel to the coast but some distance off shore, until they finally departed. The deflective force of coastlines was greatest when the birds' ground speed was low, that is, under headwind conditions or in calm weather. Mean track directions measured over two areas in northern Skane, called Inland W and Inland E, situated about 60 km apart, differed by 11, those over the western area being directed more to the south than those over the eastern. No significant correlation with wind directions was found in these areas. Combining data from the whole land area, however, track directions were found to vary from day to day in significant correlation to the wind direction. Mean track directions over the Baltic agreed with those over Skane, but both differed significantly from those over the Kattegatt. Both over the Baltic and over the Kattegatt directions were significantly correlated with wind directions, and showed greater variation than track directions over land. Daily track differences over the Baltic resulted both from differences taking place over the land, and from real wind deflection (drift). Both over the land and over the sea heading directions were correlated with wind directions, suggesting compensatory efforts on the part of the birds. On three days extensive fog covered much of the study area. Wood Pigeons continued to migrate, but certain aberrations in their behaviour were noted. Over land migration was relatively heavier in the west with northwesterly winds and in the east with northeasterly winds. The correlation demonstrated between wind direction and the mean track direction was based upon the fact that populations with different inherent primary directions made up different proportions of the migrating cohorts under different wind conditions (pseudo-drift). The incomplete compensation for wind deflection over the sea is ascribed to the lack of visual orientation cues. The more accurate orientation possible over land suggests one reason for the birds' reluctance to flights across the open sea. When mean track directions of Wood Pigeons in different parts of southern Scandinavia were related to the migratory goals of these birds, it was found that they have to change their primary direction in the course of their journey from breeding to wintering areas.     

DETECTION BY RADAR OF AUTUMN MIGRATION IN EASTERN SCOTLAND

Radar observations through one autumn in Aberdeenshire are described showing that more migration takes place in anticyclonic weather than transitional weather, and even less migration takes place in disturbed weather. More migration is detected by radar with following winds than with opposed winds.
Comparison of data collected in Norfolk and Aberdeenshire during one season at each shows that more migration is detected by radar in Norfolk, and westward movements, which are a feature of migration into Norfolk, appear in Aberdeenshire only when birds are first drifted northwards.
Dawn ascent and reorientation movements of birds in the northern North Sea are described, showing that changes in heading were consistently between S.W./S.S.W. and S.S.E. during the autumn studied.
Weather data, radar data and ground observer data from Aberdeenshire and the Isle of May Bird Observatory were analyzed and show again that the normal migration pattern is to fly high with following winds and low only with opposing winds as in Norfolk. However, the weather in Aberdeenshire was more disturbed than it had been in Norfolk, with the consequence that weather factors, other than wind, which affect the normal migration pattérn were found to be much more common.     

THE MIGRATION OF RAPTORS AND STORKS THROUGH THE NEAR EAST IN AUTUMN

Between 1963 and 1965 three expeditions have investigated the autumn migration of raptors and storks, on two occasions in southeast Turkey, and once in Lebanon.
Nearly all the soaring birds leaving Europe by the Bosphorus cross Asia Minor and turn south at the Gulf of Iskenderun. The commonest of these migrants are White Stork Ciconia ciconia , Honey Buzzard Pernis apivorus , and Eagles Aquila spp.
Other species such as Common Buzzard Buteo buteo are not usually seen crossing the Bosphorus, but occur in large numbers in the flocks seen south of the Gulf of Iskenderun. It is suggested that these are birds from Russia and north Turkey. Common Buzzards also occur on Cyprus in autumn, but their point of origin is not clear. Short-toed Eagles Circuetus gullicus and vultures join the migration from their breeding grounds in Turkey and the Levant.
Different species tend to migrate at different times of year. White Storks, Egyptian Vultures Neophron percnopterus and Black Kites Milvus migruns move chiefly between late August and mid-September, while eagles Aquilu spp. start in late September and continue until November. Other species are intermediate, or spread their migration out over a longer period.
The volume of migration at different times of day is discussed. No general conclusions are possible but in 1964 a correlation was obtained between cloud cover and the start of the migration.     

THE AUTUMN MIGRATION OF SOARING BIRDS AT THE BOSPHORUS

The migration of soaring birds was observed at Küçük Çamlica at the southern end of the Bosphorus between 14 July and 8 November 1966. Simultaneous watches were also carried out at other points on the Bosphorus on a number of dates. The largest movements of birds of prey occurred on days of light northeasterly winds, the largest movements of storks on days of light winds with a southerly component. On most days the stream of migrants appeared to be concentrated over the southern end of the Bosphorus. Migration frequently occurred right throughout the day, though the peak period was usually not spread over more than three hours. Figures are given for the daily times of migration of the commonest soaring birds. Daily counts of soaring birds (storks, raptors and Cranes) migrating over the Bosphorus at Küçük Çamlica are given. The main species found migrating were (with total number recorded in brackets) White Stork Ciconia ciconia (207,145), Black Stork C. nigra (6,194), Honey Buzzard Pernis upivorus (8,997), Buzzard Buteo buteo (12,949), Spotted Eagle Aquila clanga/pomarina (4,309) and sparrowhawk Accipiter nisus / brevipes (5,224). The autumn migration of 1966 is discussed in detail in a systematic list. Buzzards B. buteo were recorded in large numbers for the first time at the Bosphorus, and were the commonest bird of prey. Cranes Grus grus were also recorded for the first time. Comparison is made between our results and those of previous workers, though differences of coverage rule out any firm conclusions.     

MIGRATION ACROSS THE MEDITERRANEAN OBSERVED BY RADAR

Bird migration across the Mediterranean was studied by radar from an aircraft carrier for brief periods during the springs of 1961, 1962 and 1963 and in the autumn of 1963. The radar displays were photographed and analysed to give intensity (scales 0 to 6) and tracks have been plotted on the accompanying Maps together with Tables setting out the weather data and records of bird sightings. Most echoes that could be measured were between 4,000 and 6,000 ft. Small (passerine) echoes usually moved at 20–30 knots, large ones (presumed to be waders, ducks, etc.) at up to 47 knots. The analyses of each of the nine sets of observations are discussed in detail from the Atlantic to the south of Crete, including the Aegean. There is clear evidence, within the limitations of the stated methods used and the dates of observations, that migration takes place along the whole length of the Mediterranean at almost uniform density except across the Ionian Sea, where it is greatly reduced. This is possibly because the long desert crossing and the long sea crossing are separated by only a very narrow coastal strip of vegetation and there is obvious survival value in crossing further east or west. There is no evidence for concentration across the Sicilian narrows, through Malta or across the Straits of Gibraltar. The direction of migration west of the Ionian Sea is N.E./S.W. and to the east is N.N.W./S.S.E. A large-scale movement was observed coasting 20 miles offshore from Portugal in autumn tracking S.W. off Cape St. Vincent. This area in particular deserves further investigation as also does the area south of Crete. Migration is considerable at least from 18 August-24 October. The main departure of small migrants from land starts 40–50 minutes after sunset and tails off with the dawn. The diurnal pattern of intensity observed at sea is noted and it is suggested that the shape of the build-up and the time of the peak can provide a clue as to the point of departure. There are indications that in spring, especially south of Crete, some migrants do not rest in the narrow coastal strip but cross the Sahara Desert and the Mediterranean Sea non-stop. The significance of the results is discussed in the light of further visual observations made in 1964 and 1965.     

NOTES ON AUTUMN MIGRATION IN THE MIDDLE EAST

Notes were made on autumn migrants in Jordan and Iraq in the late autumn of 1956. Three movements were distinguished, a trans-desert passage of passerines and water birds, local movements of larks and sand-grouse, and incursions of northerly winter visitors. Some factors affecting the movements are discussed.     

OBSERVATIONS ON THE AUTUMN MIGRATION IN THE AREA BETWEEN THE SEA OF AZOV AND THE CASPIAN

Bird-migration was observed 5 October-14 November 1942 along 150 km. of railway line between Salsk and Stalingrad (between the Sea of Azov and the Lower Volga). The typical steppe-birds of South Russia were recorded—larks, harriers, bustards etc.—and also astonishing numbers of crows, starlings, finches, thrushes, pigeons and others. Magpies, Yellow Buntings (many), sparrows, Crested Larks, Great Tits and Partridges also appeared as migrants.
Most birds were travelling southwest or south but the geese were going west (except for one flock), as were Redwings and part of the Great Bustards and Yellow Buntings. Eastward movement was rare.
The immediate destination of the birds observed can only be presumed, but those flying west would reach the coasts of the Sea of Azov and very likely continue further. Birds flying southwest and south were headed for the eastern half of the Black Sea, or the Caucasus. This range, which forms a very pronounced barrier, is well known to be crossed by migrants.
After the autumn sojourn another period of observation followed, 1–18 January 1943 (mostly between Proletarskaja and Salsk).
In all, 45 species were observed alive, and under telegraph wires birds of 25 species were found dead, including 10 not otherwise recorded.     

QUANTITATIVE STUDY OF MIGRATION WITH 23-CENTIMETRE RADAR*

A quantitative study was made of the displays of "angels" on the screen of a 23-centimetre radar installation on Cape Cod, Massachusetts. The birds responsible for the displays were nocturnal migrants, mainly passerines.
Other things being equal, the density of angels decreased roughly exponentially with distance from the centre of the radar screen, being halved about every eight miles. The rate of decrease varied only slightly with total migration density, and could be used to estimate the total number of bird targets in the air.
Estimates obtained by this means were compared with estimates of migration density obtained from "moon-watching" observations, which were reduced by a new method of calculation which incorporates radar measurements of the height of flight. Each angel was found to correspond to between two and twelve birds. After considering and rejecting other explanations, it is concluded that many birds migrate in groups at night. There is evidence that many species (including Parulidae, Turdidae and Emberizidae) migrate in small groups, rather than a few species in large groups. Groups of passerine night-migrants are probably spread over a wide area, since it is rare to see two birds cross the moon together.
Once the average group-size has been determined in this way, radar can be used to measure the density of migration over a large area. Estimates of low migration density have a standard error about 25%, but estimates of high migration density are less accurate, and very high migration densities cannot be measured at all, although they can be accurately identified as such.
The results in this paper cannot be applied directly to other radar stations, nor to other species of birds.