Overall, one-third of the earth's land area is classed as desert (Figure 6.2). The term covers a range of arid landscapes, from the sparsely vegetated regions of western North America to totally barren rock and sand typical of much of the Sahara. Providing there is some vegetation, some specialist bird species can survive and breed in such arid habitats, but most birds cannot find sufficient food or water, or remain active at such high temperatures. Outside the tropics, most deserts are at their best in spring, after winter rains have promoted plant growth and flowering. They are much less use in autumn, when plants have withered under the desiccating heat of summer. The problems of desert flights include the long distances without food or water, searing daytime temperatures and low humidities, which can lead to risks of overheating and dehydration. Birds can avoid the extremes of heat by migrating at night or at high altitudes, but crossing the biggest deserts requires more than 12 hours of flight, even by the fastest birds.
The main sources of water in deserts are rivers flowing through from wetter regions, which provide green bands of riparian vegetation in otherwise parched landscapes, and the scattered oases which form wherever groundwater breaks the surface. These features are visible from a long distance, being marked by taller green bushes and trees than the surrounding desert, and provide habitat, food
and water for a range of birds. However, oases and their associated vegetation typically cover small areas, of up to a few hectares, and are usually few and far between, up to several hundreds of kilometres apart. Although at migration times they can seem packed with birds, they are likely to be visited by only a tiny proportion of those passing over. Moreover, oases are normally used only by those species adapted to the habitats they provide.
The largest and most severe desert in the world sits right across one of the main bird migration routes, between western Eurasia and the Afrotropics. The Sahara stretches from the Atlantic coast in the west to the Saudi Arabian peninsula in the east, extends 1000-1500 km from north to south, and covers an area of 12 million km2, about the same as the USA. Much of it is totally bereft of vegetation, lacking even the sparsest of desert scrub. Daily maximum temperatures during the autumn and spring migration periods average 25-38°C, but sand surface temperatures during the day may reach 70°C. Relative humidity can drop below 10%. The desert varies greatly in severity in different longitudes. The Atlantic coastal strip in the west provides a continuum of scrubby vegetation running north to south. In places, this coastal strip is more than 200 km wide, beyond which the scrub becomes progressively sparser into the desert. In the east, the Nile Valley provides a narrow green corridor for migrants, the only route that is abundantly supplied with both food and water. Over the rest of the desert, vegetation is mostly sparse to non-existent, apart from the few isolated oases. Yet despite its inhospitability, this desert is crossed and re-crossed at all longitudes by an estimated 5000 million birds of 186 species every year (Moreau 1972). Revised estimates, made 40 years later and based on radar observations at sites across the Mediterranean region, suggest figures of 3500-4500 million birds (B. Bruderer, personal communication), but these recent estimates (unlike the earlier one) exclude birds that enter Africa via Arabia and follow several decades of population decline among many migratory species in Europe (Chapter 24). Taking these facts into account, the two estimates, found in different ways, are remarkably consistent.
Birds that follow the Nile Valley in the east, with its narrow border of swamps and cultivated land, or the Atlantic coast in the west with its broad belt of sparse desert scrub, could in theory migrate with short flights and frequent stops, as they could find food at various points en route. The concentrations of migrants are greater at the west and east ends of the desert than in the middle section, at least in autumn, as revealed by radar, but concentrations seem unexceptional in the Nile Valley itself, apart from waterbirds (Moreau 1961, Biebach 1990, Bruderer & Liechti 1999). The option of flight from oasis to oasis is possible only along a row of mountains and oases that stretches southeast from Morocco. Elsewhere, oases are too few and far between. Furthermore, most of the Saharan oases are no longer in a natural state, the native vegetation having been largely removed or replaced by date palms, which offer little sustenance for birds. Most species apparently migrate on a broad front over the most hostile parts of the desert, but the same species may behave differently in spring and autumn, or at different points in the crossing (Moreau 1961).
The greater density of migrants at the west and east ends of the desert in autumn may be as much a consequence of the funnelling effect of land areas to the north as of the desert conditions themselves. Most of the west European migrants cross the Mediterranean at its narrowest, around Gibraltar, from which they reach the western edge of the desert, while the east European and Asian migrants cross the desert in its eastern sector or cross Arabia instead. Moreover, throughout its breadth, the desert would have offered better conditions a few thousand years ago than it does today. The difficulty of this journey is increasing all the time, as the desert expands north and south under human influence. The annual rate of desertification near the southern edge has been calculated at about 0.5%, which corresponds to an area of at least 60 000 km2 added each year by degradation. In these times of climate change, other areas on the migration route are also becoming drier, including parts of Spain, much of North Africa and the Sahel zone to the south. These various changes mean that, over the centuries, birds have had to cross ever widening areas without access to food or water, while suitable staging areas have become smaller and more widely spaced. These trends are likely to continue into the future.
The scorching daytime temperatures, combined with lack of food and water, make crossing the Sahara one of the most arduous journeys undertaken by migratory birds. For some birds, it also follows (in autumn) or precedes (in spring) a crossing of the Mediterranean Sea at its widest points. The west and east parts of the North African coastal region support abundant green habitat extending hundreds of kilometres to the south, but in the central part little vegetated habitat is available between the sea and the desert, offering only limited opportunities for refuelling. This central coastal strip over most of its length is typically less than 20 km from north to south, consisting of sparse perennial scrub, with the addition in spring of ephemeral herbage dependent on winter rain. Beyond this to the south lies barren desert.
Many species cross the Sahara by accumulating the necessary body reserves beforehand (Figure 6.3; Chapter 5). Most passerines increase their body weight by more than 50%, and some by around 100%, effectively doubling their weights before departure (Ward 1963, Fry et al. 1970). For example, the Garden Warbler Sylvia borin weighs about 18 g in the breeding and winter seasons, but increases its body mass up to 37 g before crossing the Sahara, whether on outward or return journeys (Bairlein 1991b).
In some bird species, at least some individuals seem to cross both the Mediterranean Sea and Sahara Desert in a single non-stop flight, while others break their journey for refuelling in North Africa (Moreau 1961, Casement 1966, Bairlein 1992, Biebach 1998). The sea-plus-desert crossing would require a flight of 1500-2500 km, depending on the route taken, and the desert crossing alone would require a flight of 1000-1500 km. The Sahel zone, to the south of the desert, is at its driest in spring, which adds another 500 km to the desert crossing. This is offset to some extent by the greening of about 300 km of the northern Sahara as a result of spring rains in the Mediterranean region (see below). The net effect of these seasonal changes is therefore to increase the desert crossing by about 200 km in spring compared with autumn. Moreover, these distances refer to north-south crossings, and any bird traversing the desert diagonally could face an even longer flight.
A combined sea and desert crossing in autumn has been claimed for various waterfowl and waders and for some slower flying species, such as Sedge Warbler Acrocephalus schoenobaenus (Gladwin 1963), Yellow Wagtail Motacilla flava (Wood 1992) and Common Cuckoo Cuculus canorus (Moreau 1972). Evidence stems not
Figure 6.3 Body weights of Willow Warblers Phylloscopus trochilus at different stages of their autumn migration over the Mediterranean Sea and Sahara Desert. Of all the species that cross the Sahara, the Willow Warbler is most numerous. The fat-free body weight of this tiny bird is about 7 g. The mean weights of trapped samples at various locations on the autumn migration route reveal weights of 1011 g on Cyprus and other islands in the north Mediterranean, decreasing gradually through localities en route to 7-7.4 g in the south of the Sahara. Body weights taken from the same places in different years showed little variation; but weights of birds at different sites in the same general region were often markedly different, depending on habitat. From Biebach (1990).
only from the relative scarcity of these species on the ground in North Africa in autumn, but also from the body reserves accumulated, which are calculated to be sufficient for a journey of this length. A combined sea and desert crossing may also be performed by some hirundines, swifts and bee-eaters, which could pick up food on the way, at least over vegetated areas.
While some species accumulate most of the body reserve necessary for the journey close to the desert, others accumulate most of their reserve beforehand, up to several hundred kilometres north of the desert in autumn, or several hundred kilometres to the south of it in spring (Chapter 5). This further increases the distance that must by flown without substantial further fuelling. For example, Pied Flycatchers Ficedula hypoleuca fatten mainly in northern Iberia in autumn, and may then migrate without feeding until they reach the south of the Sahara. Sedge Warblers Acrocephalus schoenobaenus are even more extreme, accumulating sufficient body reserves as far north as Britain to get them to the south of the Sahara. In the spring, Blackcaps Sylvia atricapilla, Garden Warblers S. borin, Whitethroats
S. communis and other species fatten well south of the Sahara, in the more mesic Guinea zone (Hjort et al. 1996, Ottosson et al. 2005), while Bar-tailed Godwits Limosa lapponica migrate directly from West Africa to the Dutch Waddensea, a distance of 5000 km (Drent & Piersma 1990). In effect, the Saharan crossing is much longer for forest-dwelling passerines than for open-country ones which can find foraging areas at the northern and southern edges of the desert. In one study, forest-dwelling species, such as the Golden Oriole Oriolus oriolus, Wood Warbler Phylloscopus sibilatrix and various flycatchers, were found to arrive in spring on Ventotene Island (off southern Italy) almost devoid of fat reserves, whereas open-country species still had enough fat, on average, for more than 300 km of further flight (Pilastro & Spina 1997). It is clear, therefore, that some species migrate without refuelling over substantially longer distances than are needed to cross only the Mediterranean and Sahara. Different species adopt different fuelling patterns, depending on the routes taken and locations of suitable feeding areas (Chapter 5).
Large birds flying non-stop north-south at 80 km per hour (say) in still air would require around 19-31 hours for the sea and desert journey, and around 13-18 hours for the desert alone; while small birds flying non-stop at half this speed would require twice as long (e.g. up to about 62 hours). The most numerous species to make this journey is the Willow Warbler Phylloscopus trochilus, which is also one of the smallest (fat-free weight about 7 g) and slowest fliers (still-air speed on migration 34 km per hour). This species would take 44-74 hours for the sea and desert journey, and 29-44 hours for the desert alone. In all such species, however, these times could be substantially shortened with the help of tailwinds, or lengthened by headwinds or stops, or by crossing the desert diagonally. In the daytime, frequent thermals could also provide lift and help to speed the flights of any species active then at low enough altitude. Nevertheless, this seems a staggering achievement for birds weighing as little as 12-40 g at take-off.
Conditions for trans-Saharan migration differ greatly between autumn and spring. When the migrants move south in autumn, the Sahel zone immediately to the south is near the end of its wet season. Vegetation is green, flood pools are common and insects are plentiful. This reduces the distance that must be flown without feeding, and favourable southward winds on most days help to speed the birds on their journeys. But in spring, when the migrants move north, the Sahel zone is reaching the end of the dry season, when conditions are by no means ideal for refuelling, and when the same southward winds prevail, hampering northward progress. According to Moreau (1961), birds crossing the Sahara in spring must expect to meet headwinds unless they are travelling diagonally northeast at altitudes greater than 2 km. At these higher altitudes, they meet the anti-trade winds blowing in the opposite direction to those below, and it seems likely that most birds take advantage of them (radar having revealed higher flight altitudes in spring than in autumn). Otherwise, the only way to reduce the effect of headwinds would be to fly low, say within 500 m of the ground.
So, while in autumn many small birds are apparently able to cross the Mediterranean and Sahara without replenishing their reserves, they are less likely to do so in spring. This seasonal difference in conditions may explain why many more migrants are seen in North Africa and around Mediterranean coasts in spring than in autumn, even though overall population levels are lower in spring (Nelson 1973, Shirihai 1996). As indicated above, conditions in the Mediterranean region are much better in spring than in autumn. As a result of winter rains, the migrants in spring encounter a flush of fresh vegetation and insects, together with areas of standing water. The implication is that many more birds make a combined Mediterranean-Saharan crossing in autumn than in spring, and that many more stop to refuel in North Africa in spring than in autumn.
In crossing the desert, some small passerines may make a single non-stop flight, while others have been found to break their journeys in the daytime, but mostly without feeding. At least some individuals of many passerines species (notably Garden Warbler Sylvia borin, Lesser Whitethroat Sylvia curruca and Willow Warbler Phylloscopus trochilus) descend to spend the day sheltering motionless near oases, or even in the open desert in the shade of rocks, continuing their journey in the cool of night (Bairlein 1987, 1988, 1992, Biebach 1990, 1992, Biebach et al. 1986, Bruderer 1994). In this way, they could cross the desert in a few nocturnal flights. The majority of trapped migrants at such sites showed high body mass and fat loading, with sufficient reserves for onward flight, so had not been grounded through lack of fuel. In resting by day and flying by night, such birds travelled in cooler conditions, and may thus have reduced their dehydration risks, but they would have added at least one or two nights to their journey time (giving 2-4 nights for the total journey across the desert). Yet other individuals of these same species stopped at oases where they fed and drank, usually remaining for 2-4 days, but occasionally for up to three weeks at a time, if they needed to rebuild their body reserves. However, observations at oases indicate that the birds stopping are only a tiny proportion of the numbers that must be passing. Strictly diurnal migrants, such as broad-winged raptors dependent on daytime thermals, may stop every night (as shown by radio-tracking of several species, Chapter 8), but would normally obtain no food then.
Further studies are needed to find the relative frequencies of non-stop Mediterranean-Saharan flights, two-step flights with a stop in North Africa, or multi-stage flights, stopping in North Africa and at points in the Sahara (with use of oases or of stopping points along the Nile Valley). The relative frequencies of these different patterns clearly differ between autumn and spring, linked with different winds, ground conditions and time constraints. From autumn radar studies at sites in the Egyptian Sahara, Biebach et al. (2000) concluded that about 20% of all migrants detected were involved in non-stop migration and 80% in intermittent migration, with stopover at the coast (70%) or in the desert (10%). Other studies have suggested that in Mauritania an even greater proportion of individuals descend and rest by day (B. Bruderer, personal communication).
On the face of it, stopovers in the desert without any possibility of refuelling would seem to waste time and energy, but they may help to conserve water, because the bird can seek shade and remain inactive, so as to minimise internal heat production. The possibility that water rather than energy might restrict the flight lengths of birds has long been appreciated, because of early findings that, at laboratory temperatures, the water loss of birds exceeds their metabolic water production during daily activities (Hart & Berger 1972). The need to retain water could therefore be a more important constraint than fuel for small birds crossing the Sahara. Only about
15-25% of the energy expended by muscle is converted to mechanical power, and the rest to heat. The body temperatures of birds measured during flight are generally greater than 41°C, up to 4°C higher than normal. To avoid overheating, the bird must pant, which leads to water loss and eventually to dehydration (Chapter 4). In the absence of drinking water, this risk can be offset by metabolic water production, or by flying at high altitudes where the air is cooler, or at night when temperatures are lower than during the day.
The metabolism of fat, carbohydrate or protein results in the production of 1.1, 0.90 and 0.82 g of water per g of wet tissue expended, or 0.03, 0.23 and 0.16 g of water per kJ of energy released (see Table 5.1). For flying birds, metabolic water derived in this way has been estimated as sufficient to offset the danger of overheating in ambient temperatures up to around 10°C (Biebach 1990, Nachtigall 1990). Small birds flying at air temperatures up to this level should therefore have no additional water loss, even on long flights; but above 10°C further evaporative cooling would be needed, rapidly leading to dehydration. In the Sahara in autumn, if one assumes an air temperature at ground level of 30°C by day and 8°C by night, and a decline of 7°C with each 1000 m rise in elevation, an air temperature lower than 10°C would be found by day at more than 3000 m above ground, and by night at more than 1000 m above ground. In spring, when the ground would typically be cooler than in autumn, an air temperature of 10°C would be found at an altitude greater than 1750 m by day and 500 m by night. At the time of writing, however, doubt hangs over the 10°C estimate, as subsequent work (still in progress) suggests it may be higher, which would alter the altitude estimates. At very high altitudes, reduced humidity could raise water loss, with the thinner air leading to increased ventilation, and hence to further increase in respiratory water loss (Carmi et al.1992, Carmi & Pinshow 1995, Klaassen 1995, Klaassen et al. 1999). Overall, the best option for a small bird may be to fly at night at moderate altitude (say 500-2000 m) and rest in the shade by day.
Birds are unlikely to fix their flight altitudes in terms of water conservation alone, however, for they must also take account of other factors, notably wind conditions, which affect the energy costs of flight. Precise radar measurements of the height distributions of migrants over southern Israel in autumn and spring, combined with simultaneous altitudinal recordings of weather variables, offered an opportunity to examine the presumed importance of wind conditions, energy and water constraints on flight range (Liechti et al. 2000). Predictions were made of the optimal flight altitude, in the conditions prevailing, if birds took account only of tailwinds, only of energy conservation, or of both energy and water conservation, in achieving maximum flight range. These predicted flight altitudes were then compared with the actual height distributions recorded in the field. The authors concluded that wind profiles, and thus energy rather than water limitations, governed the altitudinal distribution of nocturnal migrants. However, this conclusion, drawn from large samples of birds spread over a wide range of altitudes, does not exclude the possibility that individuals may adjust their altitude to whatever was most limiting for them at the time, including water if they were becoming seriously dehydrated.
Regarding fuel type, calculations for Willow Warblers Phylloscopus trochilus and Eurasian Golden Orioles Oriolus oriolus revealed that, under most flying conditions, if the birds used fat alone as fuel, water could impose the main limitation to their total flight range, but if they used 70% fat and 30% muscle (which is two-thirds water), then energy rather than water could become the main constraint (Klaassen & Biebach 2000). The bird would then gain only 74% as much energy as from fat alone, but 26% more water. This situation is probably closer to reality, and it would also allow birds to fly at relatively very high altitudes (in cooler but thinner air) than if they used fat alone without incurring a water debt.
Water loss could also be reduced by increasing the oxygen extraction efficiency, which would allow reduction in the throughput of air, and hence in water lost during respiration.
Further insight can be gained from studies of body composition. The proportion of water within a bird's body decreases with increase in fat content, because fat is stored in anhydrous state. However, the fat-free component of a bird's body usually contains around 67% water, which can reduce by about one-third to around 55% before the bird expires from dehydration (Haas & Beck 1979). For Willow Warblers Phylloscopus trochilus, Biebach (1990) calculated reduction in body water over one night from 67.5 to 62.9%, which is well within the range of dehydration tolerance, and over two rest days, further reduction to 57.9%, which is still above the value at which death becomes likely. Moreover, the water contents of birds caught in the Sahara Desert were in general around 67-69% of the fat-free weight (equivalent to a water:lean dry mass ratio of around 2:1), which is within the usual range of healthy birds (Chapter 5). In addition, the levels of Na+ ions and urea in the bloodstream, measured as further indicators of the state of body hydration, were normal in most birds, and high only in lean ones (Biebach 1990, 1991, Bairlein & Totzke 1992). Furthermore, Willow Warblers and other small passerines found dying in the Libyan Desert showed normal water levels, but had run out of fat, so in these birds, fuel rather than water seemed to have been limiting (Biebach 1991).
Most of the evidence cited for lack of water stress in migrant birds derives from the absence or scarcity of birds found with lowered water contents, even near the end of a long migration (Biebach 1990, 1991, Gorney & Yom-Tov 1994, Landys et al. 2000). This is not true of all samples, however, for 78% of 409 birds caught on the Egyptian coast in autumn, after crossing the Mediterranean Sea, had water:lean dry mass ratios less that 2:1, including 12% with ratios less than 1.4:1 (Fogden 1972a, Chapter 5). The latter were close to the level at which death would be expected to occur. To some extent, as emphasised already, birds could offset increasing water loss by catabolising body protein to yield water. The finding of migrants in the Sahara Desert with high fat but low protein reserves might reflect effects of dehydration.
Migrants clearly show behaviour that could be construed as anti-dehydration, such as shade-seeking, remaining immobile during the hottest part of the day, and drinking heavily on arrival at stopover sites (Biebach 1990, Klaassen 2004). At such a site in the Negev Desert, Blackcaps Sylvia atricapilla put on weight more rapidly when they were provided with water than when not (Sapir et al. 2004). This could be because, in the absence of drinking water, they had to catabolise some food to provide for body water needs, reducing the amount of food that could be stored as fuel. In contrast, water availability had no effect on fuel deposition of Lesser Whitethroats Sylvia curruca in the same place. Differences between these species in their adaptation to arid conditions may explain their differential response to water provision.
To conclude, these various considerations suggest that crossing the Sahara even by non-stop flight is practicable with a balanced water budget if the bird flies at an appropriate altitude. This is difficult during the day in autumn because of the need for tailwinds which occur mainly below 1000 m. This may be why many birds in autumn, in order to conserve water, fly by night and rest by day (see also Carmi et al. 1992). Of course, birds resting in the desert by day are also confronted with high temperatures and low humidity, but they can seek shade and remain inactive, so as not to generate additional body heat. Given sufficient fat, small warblers could easily endure three nights of flight and two stopover days without feeding or drinking, as calculated by Biebach (1991). It seems, therefore, that with sufficient body reserves, small migrants are able to avoid both overheating and dehydration in the hot desert mainly by suitable behaviour. Nevertheless, evidence from the body composition of birds obtained on the Mediterranean-Saharan crossing suggests that, while fuel energy could be limiting in some, water could be limiting in others (Chapter 5).
Migrants travelling between eastern Asia and the Afrotropics also face a difficult and even longer journey than those migrating between Europe and the Afrotropics. They include eastern populations of the Willow Warbler Phylloscopus trochilus, Sedge Warbler Acrocephalus schoenobaenus, Garden Warbler Sylvia borin, Common Redstart Phoenicurus phoenicurus and Northern Wheatear Oenanthe oenanthe, among others. Birds from these populations have overall journeys 1.5-2.0 times longer than their European equivalents. They have to cross a difficult succession of hostile areas, including the deserts and semi-deserts of western Central Asia, the almost treeless areas of the Iranian highlands, and the deserts of the Arabian peninsula. Some also have to cross the world's highest ground in western Asia, including the Tien Shan, Pamiro-Alay and the Himalayas. For forest passerines, inhospitable areas in Central Asia alone span 2000-3000 km, and for eastern Siberian populations of some species (Willow Warbler and Northern Wheatear), the total one-way migration distance may exceed 15 000 km.
The Arabian peninsula offers an alternative route from the Sahara into the more genial parts of Africa (Bourne 1959). Not only is this desert less wide than the Sahara, but the hills and mountains around its borders are vegetated. It supports a number of oases, and is generally better vegetated than the Sahara, because it receives more rain. Not surprisingly, therefore, the Arabian peninsula is crossed not only by the migrants from Asia heading to Africa, but also by many birds from central and eastern Europe (including Blackcap Sylvia atricapilla, Lesser Whitethroat S. curruca and Red-backed Shrike Lanius collurio), which head southeast on the first part of their migration (Chapter 22).
Migration in the Asian deserts and mountains was studied over 12 years by a large-scale programme of observation and trapping at more than 20 different sites scattered between the Caspian Sea in the west and the Hindu Kush, Pamir and Tien Shan mountains in the east (Figure 6.4; Dolnik 1990). These sites together spanned the region from 37-48°N and 53-78°E. This region represents a crossroads of migration routes involving two main wintering areas, so that two main axes of migration were detected: northwest-southeast for Siberian birds wintering in southern Asia, mainly India, and northeast-southwest for Siberian birds wintering in the Afro-tropics. Hence, many of the birds that cross the Asian deserts also cross the eastern Sahara or Arabian Peninsula on the same migration, for which greater fat levels are needed. Of birds captured, 25 species (75.6% of all captures) winter in Africa, and 23 species (9.5% of captures) in Asia, mostly in India, while eight species (14.9% of captures) winter in both Africa and India.
The deserts of Central Asia present a much less arduous journey than the Sahara, because they support more rivers and lakes and generally more vegetation, especially in spring after winter precipitation. An average of 1.5 billion birds (85% passerines) are estimated to cross these deserts on a broad front each autumn and about 0.75 billion each spring, flying mainly at night (Dolnik 1990). The oases, desert lakes and rivers do not provide adequate foraging sites for all these birds and, because of the intense competition at such places, the rate of fat deposition there is low - about the same as in the open desert. Nonetheless, in general, the basic strategy used by passerines migrating across this arid region centres more on effective refuelling during daily stopovers than on maximal fuel storage before starting the journey. The birds thus maintain a moderate rate of travel across the desert, limited by the opportunities for refuelling, but with no net fat loss over the journey.
Many birds of forest and other habitats have little option but to feed in unfamiliar arid habitats with little vegetation. On average, a passerine migrant foraging in the Asian deserts accumulates during one day enough fat to support an estimated 1.4 hours of migratory flight in spring and 0.6 hour in autumn. At both seasons, the average fatness of small passerine migrants initiating nocturnal flight in Asian deserts was 20% of fat-free mass, and in those trapped just at the end of their nocturnal flights it was around 10%, a loss expected over at least 4 hours of night flight (Dolnik 1990). In the Asian deserts, as in the Sahara, birds found by day in shady places that offer little food tended to have high fat levels, whereas those found at potential foraging sites near water tended to have low fat levels (Dolnik 1990). Evidently, lean birds were more specific in their choice of resting sites than fat ones.
The numbers of birds seen and caught in the western deserts were much higher in spring than in autumn, about 2.6 times higher according to data obtained by 'moon-watching' (Chapter 2). This was partly because birds took mainly different routes at the two seasons, but also because many birds stopped in spring but flew over in autumn (Bolshakov 2003). In spring, practically all nocturnal passerine migrants avoided crossing the highlands of Central Asia, which at that time were still snow-covered, and took the desert route instead, whereas in autumn, mountain crossing was much commoner, especially among species wintering in India (Figure 6.4).
In addition, captures at stopovers and long-distance recoveries of ringed birds suggested that, in autumn, most transit passerines from the forest zone of central and eastern Siberia made a detour round the north and northwest edges of the deserts. By heading westward towards Europe, before veering southward to Africa, they migrated through more hospitable areas but lengthened the journey by another 2000 km, bringing the total to 8000-10 000 km. It seems, therefore, that, in spring when the deserts are at their most propitious, many migrants cross them, feeding en route, but in autumn birds largely avoid the deserts by migrating over the mountains to the east, or by heading westward for 2000 km or more, before veering south towards Africa. By taking somewhat different routes in autumn and spring, the birds make the best of prevailing conditions at both seasons.
Was this article helpful?