Quelea quelea (weaver bird)

Red-billed quelea are small, short-tailed, sexually dimorphic weaver birds about 12 cm long and weighing 15 to 26 g. The sexes differ in appearance according to season. In their breeding plumage males are polymorphic, with different amounts of black, yellow, pink, purplish or white in the head and neck region and different amounts of colouration on the belly (Ward, 1966). The colour patterns of the males’ plumage are not signals of quality (Dale, 2000) but are thought to be involved in individual recognition processes (Dale et al., 2001). However, the redness of the males’ bills is an indicator of condition and quality (Dale, 2000) and social dominance (Shawcross and Slater, 1984). Males of the nominate race typically have a black face ‘mask’, consisting of a black forehead, lores, cheeks and upper throat surrounded by varying amounts of pink, yellow, purple or black. Sometimes the mask is white. The surrounding colour may end on the lower throat or stretch as far as the belly, with the remainder of the underparts being light brown or white with some dark striations. The upperparts are light brown with dark longitudinal stripes, particularly centrally, with those on the rump paler. Tail and upper wing dark brown; flight feathers edged greenish yellow. Eye has narrow red orbital ring and brown iris. Legs orange. Bill bright red. The non-breeding plumage of males lacks any bright colour, with much of mask, forehead and crown becoming grey-brown with dark streaks, chin, throat and faint supercilium white; bill becomes pink and legs flesh-coloured. Females in breeding plumage resemble non-breeding males but have yellow bill and eye-ring. Non-breeding female has a pinkish bill.

Eggs are laid in clutches of 1-5, usually 3, measure 18 x 13 mm and are bluish or greenish, occasionally with some dark spots. Clutches of 6 have been noted but these may be the result of egg-dumping by females that are not the nest-owners. Nestlings are born naked with wisps of down on shoulders and crown, with white bills. Eyes open and feathers appear on day 4.

Within 2-3 months of hatching, juvenile birds complete a post-juvenile moult to resemble non-breeding adults, but with grey head, whitish cheeks and buff edges to flight feathers and wing coverts, followed 1-2 months later by a pre-nuptial contour moult, when they begin to assume the adult breeding plumages. Nestlings have a lavender-tinged horn-coloured bill that turns orangey purple before the post-juvenile moult, then pinkish-purple and finally red (Jones et al., 2002).

The different subspecies are primarily distinguished by the colours of the adult male breeding plumages. Adult breeding Q. quelea quelea have a buff crown, nape and underparts and a broad band on the forehead. Adult breeding males of Q. quelea aethiopica generally lack the black band across the forehead and the buff wash on their underparts, which may be replaced with a pink wash. Adult breeding males of Q. quelea lathamii have a broad band on the forehead and mainly white underparts. However, there is much variation and some individuals cannot be ascribed to a subspecies based on morphology alone. Some interbreeding between subspecies may occur at transition zones.

All three subspecies move in association with the rains. In West Africa the movements of Q. quelea quelea are generally south-north and then north-south, but the opposite strategy may also be adopted (Manikowski et al., 1989). For instance, Ward (1965; 1971) proposed that in Nigeria birds travel 300-600km southwards during their early-rains migration, at the start of the rains in June-July, when the rain causes their grass-seed food to germinate. Thus, they reach areas such as the Benoue River valley, where the encroaching rains have already caused grasses to set seed. Then, after six weeks, the birds set off northwards until they find a suitable breeding habitat, breed, and then repeat the process in progessive moves further north, as ‘itinerant breeders’. Others have pointed out that some populations travel northwards at the start of the rains to exploit remaining ungerminated seeds (Manikowski et al., 1989). The situation in Senegal and Gambia also differs but remains to be understood, although southeastwards moves followed by northwestward return movements seem likely.

In eastern Africa, the movements of Q. quelea aethiopica are more complicated, especially where there are bimodal rain patterns which may allow individual birds to breed up to four times in a year. It is thought that there are two sub-populations. The ‘intermedia’ group move from Tanzania in the early rains migrations to southern Somalia, from where they return to breed in central Tanzania in February-March, followed by itinerant breeding migrations to successively more northern sites, usually culminating in central Kenya in May. Meanwhile, the ‘aethiopica’ group moves from its dry season areas in its early rains migrations from northern and central parts of Sudan and Ethiopia in May-June, to then breed in the south of these countries and in South Sudan, returning north during August-October (Jaeger et al., 1989).

Although the southern African subspecies Q. quelea lathamii is not differentiated genetically (Dallimer et al., 2003), it has been shown to have a migratory divide (Dallimer and Jones, 2002; Dallimer et al., 2003). Birds in dry season quarters in Zimbabwe in October-November will move in their early rains migrations, either south towards South Africa or Mozambique or west-north-west towards Angola. Those reaching southern South Africa and Mozambique will have flown over the rains to where conditions will soon permit breeding, which will be followed by breeding migrations progressively further northeast. Those reaching Angola will perform similar movements in the opposite directions. Additional populations that spent the dry season in either southern South Africa or Angola will have been forced by the first rains to travel northwestwards or southeastwards, respectively, in September-October. That this description conforms with the birds’ known activities was confirmed by observation of breeding in areas predicted by a forecasting model based on the above scheme (Cheke et al., 2007).

Ringing studies have confirmed movement patterns from South Africa. The longest distance between ringing and recovery sites is 2545 km, for a bird ringed in South Africa and recovered in the Democratic Republic of the Congo (Oschadleus, 2000).

After fledging successfully, most juvenile birds are deserted by their parents, which move on generally in separate flocks, in which there is some evidence of sexual segregation and male-biased dispersal away from the natal group (Dallimer et al., 2002). There is also evidence that some birds return to the same dry season quarters in successive years (Borello and Cheke, 2011).

The population in Réunion was probably due to intentional introduction.

Extra-pair fertilizations sometimes occur with quelea, with 31% of offspring being unrelated to at least one behavioural parent (Dallimer and Jones, 2007). All of the southern African populations of Q. quelea lathamii are a single inter-breeding population (Dallimer et al., 2003). Despite this lack of population structure, there is some sub-structure, with evidence for male-biased dispersal (Dallimer et al., 2002). The male colour patterns are inherited but the mask and breast colours are independently assorted (Dale, 2000). Genes coding for quelea carotenoid colours have been investigated by Walsh et al. (2012).

The red-billed quelea is a communal breeder, building nests in colonies that are usually found in thorn trees such as Acacia tortilis, A. mellifera and Dicrostachys cinerea, but occasionally in reeds or sugar cane, or other plants listed by Jarvis and Vernon (1989). Some of these colonies are massive, stretching to up to 20 km in length and 1 km wide and involving millions of birds (such as at Malilangwe, Zimbabwe), with nests at densities of 30,000 nests per hectare. More than 6000 nests have been counted in one tree (Craig, 2010).

Nests are built over a 2-3 day period and only by males. The nests of Q. quelea lathamii are oval balls of interwoven grass, 20-130 mm tall, 100 mm wide and 89-06 mm deep. The nests are unlined. The entrance hole is 39-58 mm wide and 24-28 mm high with an overhanging porch of 22-26 mm (Jarvis and Vernon, 1989b). Males display at their nests to attract females. The clutch size varies from 1-5, usually 3, and is determined by the reserve protein levels of the females (Jones and Ward, 1976). Clutches of 6 have been noted but these may be the result of egg-dumping by females that are not the nest-owners. Colonies are synchronized and incubation takes 10-12 days. Nestlings are born naked with wisps of down on shoulders and crown, with white bills. Eyes open and feathers appear on day 4. Both parents feed the nestlings during a fledging period of 10-11 days and for a few days after they leave the nest.

Before migration, quelea build up their fat reserves by feeding on insects such as termites, with the amount of fat laid down linearly related to the distance to be travelled (Ward and Jones, 1977). Before breeding, protein-rich food such as termites and green grass seed is eaten, and reserve protein levels – the amount of labile stores in muscle sarcoplasm (Kendall et al., 1973) – reach a peak before declining, as they are used up by the efforts involved in reproduction (Jones and Ward, 1976).

The red colouration of the eye-ring, legs and bill of breeding males is carotenoid-based. The red in bills and feathers relies on the presence of enzymatically derived keto-carotenoids (Dale, 2000).

Wild birds tend to live for 2-3 years (Bird Trader, 2013), although they can live longer. A captive bird lived for 18 years and 9 months (Butler, 1913).

Quelea are thought to exchange information on locations of food sources when gathered at roosts or colonies (Ward and Zahavi, 1973). Huge flocks of breeding birds leave colonies and roosts in undulating streams at or just before dawn, apparently in directions of food or water sources, making a distinct noise due to numerous wing-beats. Small flocks return to roosts via a drinking site from 30 minutes before dusk, with numbers of arriving birds increasing as dusk approaches, sometimes in attenuated streams. Stragglers may continue to arrive up to 15 minutes after dusk. The roosting birds move about within the roost and are noisy for a further 30 minutes or so before they settle down. Behaviour at roosts and preferred vegetation for roosting were described by La Grange (1989).

After breeding, adults and juveniles form partially separated flocks, with most juveniles remaining in the natal area after parents have moved on. It is these flocks of juveniles that are responsible for most crop damage. In non-breeding season, birds roost communally in huge flocks, often at sites used in successive years. They also form ‘day roosts’ at the hottest time of day, when they congregate in vegetation near food or water.

Flocks feed on the ground in rolling motions, with trailing birds constantly leap-frogging leaders to exploit the next source of fallen seeds, behaviour which minimizes foraging in areas that have already been exploited. They also take seeds directly from plants.

Red-billed quelea drink daily and flocks will appear at water-holes, usually perching on surrounding vegetation before going down to the water.

The red-billed quelea is considered to be the most abundant terrestrial bird on earth, with a peak post-breeding population estimated at 1,500,000,000 (Craig, 2010), and as many as 170,000,000 in southern Africa alone (Yeld, 1993). The population in Kruger National Park, South Africa, was estimated to be 33,000,000 (Craig, 2010).

The populations of Q. quelea lathamii in southern Africa comprise one inter-breeding population (Dallimer et al., 2003) and this is probably also true of Q. quelea quelea in West Africa and Q. quelea aethiopica in eastern Africa.

Quelea mainly eat grass and cereal seeds from the plant and from the ground, preferring grains 1 x 2 mm in size. For crops eaten, see Hosts/Species Affected.

When red-billed quelea need to build up protein, such as before migrations or breeding, quelea will feed on insects including Orthoptera, Hymenoptera, Coleoptera, Hemiptera, Dermaptera, Lepidoptera, Isoptera and Diptera, as well as spiders and snails. Up to half of the diet fed to nestlings consists of insects. Breeding females eat snail shells and calcareous grit, seemingly to enhance calcium levels for egg production (Jones, 1976).

A colony of more than 12,000 nests per hectare was estimated to have a monthly consumption of 1845 kg of seeds per hectare and 214 kg per hectare of insects (Craig, 2010).

When in small groups, quelea often associate with weavers Ploceus spp. and bishops such as the red bishop Euplectesorix, and in West Africa they may join flocks with the golden sparrow Passer luteus and various species of estrildids (Craig, 2004). Sometimes red-billed quelea form mixed roosts with weavers, estrildids and the barn swallow Hirundo rustica.

The red-billed quelea is found in semi-arid vegetation in Africa, from sea level to 3000m, usually with thorn bushes and not far (up to about 30km) from water. It is frequently found on agricultural land. It usually nests and roosts in thorn bushes, but a variety of other plants may be used for communal gatherings including sugar-cane, reeds and eucalypts.

Red-billed quelea may be influential in the destruction and creation of acacia bush, as their nitrogen-rich droppings promote grass growth but also raise fertility levels above those tolerated by trees, which then die or, if they survive, may be destroyed by subsequent fires, leading to a more open habitat (Vernon 1989).

Control actions against quelea, both spraying and the use of explosives, sometimes have deleterious environmental consequences (McWilliam and Cheke 2004; Cheke et al., 2012, 2013).

Trees may be destroyed if excessive numbers of red-billed quelea roost in them. Red-billed quelea can also foul waterholes (Elliott, 1989a).

Red-billed quelea positively impact biodiversity by being food for a variety of mammalian, avian and reptilian predators, as seed distributors, and for their roles in nutrient re-cycling. However, breeding colonies usually force other birds to leave.

Control actions against quelea, both spraying and the use of explosives, sometimes have deleterious consequences to biodiversity (McWilliam and Cheke 2004; Cheke et al., 2012, 2013).

The economic value of red-billed quelea has not been adequately evaluated. Mullié (2000) estimated that trappers in Chad were selling up to 10,000,000 birds, valued at US $75,000 at 1994 prices. Birds caught at Kondoa, Tanzania, were sold for the equivalent of 0.75 – 2 GB pounds per 100 birds in 2009.

Red-billed quelea can be beneficial as food. In Kondoa, Tanzania, communities use traps woven from star grass Cynodon nlemfuensis to catch hundreds of quelea a day (Cheke, 2011). The bird is also eaten in many other parts of Africa (Jaeger and Elliott 1989; Mullié 2000). Quelea guano is collected in Nigeria (Ward, 1965). Huge flocks of the birds are spectacular and a sight for tourists, such as in Kruger National Park. Quelea have been seen eating pest insects such as Helicoverpa armigera, Spodoptera exempta and Locusta migratoria (Elliott, 1989a), but they provide food for the pest armoured bush crickets Acanthoplus discoidalis (Walker) (Cheke et al., 2003).

These have not been sufficiently studied but red-billed quelea very probably play major roles in seed distribution and nutrient re-cycling, and as food for other organisms.

No sanitary and phytosanitary measures specific to quelea are known, although minimization of grain spillage is advisable.

The reliance of the birds’ breeding activities on rainfall allows for some early warning, based on experience of the timings, quantities and locations of rain associated with quelea infestations in the past. This was formalized into a forecasting model by Cheke et al. (2007) for the southern African subspecies Q. quelea lathamii, driven by satellite-derived rainfall estimates. They utilized knowledge that 60 mm of rain within a two week period would stimulate grass-seed germination, and hence the initiation of the early rains migration, and that only if a further 240 mm falls within the following six weeks will conditions allow breeding. By showing colour-coded maps depicting where and when the rainfall had led to particular conditions, a system forecasting where and when quelea could or could not breed was devised. By comparing predictions with observed events, the system was shown to be 85-99% reliable. The forecasts were made available online from 2002 until 2008, when funding to support maintenance of the forecasts stopped.

Possible if the locations of breeding colonies and roosts are reported to control teams promptly.

Most farmers in affected areas are aware of the danger to their crops of quelea. Information sections of Ministries of Agriculture may publicize quelea outbreaks when they occur. In Botswana, South Africa and elsewhere there is concern about the negative impacts of quelea control measures on non-target organisms and the environment in general.

Given the enormous numbers of red-billed quelea, eradication is not a realistic possibility, even though in an average year 173 control operations may kill 50,000,000 birds in South Africa alone (Willemse, 2000).

Attempts have been made to attract quelea to roost in purpose-grown patches of vegetation, such as napier grass Pennisetum purpurum, which act as ‘trap roosts’ where the birds can be controlled (Jarvis and La Grange, 1989).

In some years it is possible to harvest crops before quelea populations reach an area. For instance, in some years if rice is harvested during mid-May to mid-June in Chad and Cameroon, quelea damage may be reduced to less than 1%. However, sometimes the birds arrive within the above period, leading to crop losses of 13-26% (Elliott, 1979). Similarly, strategies such as planting as early as possible or the planting of fast-maturing strains of millet and sorghum are advised. If rainfall is sufficient to sustain the crop, the planting of maize instead of small-grain crops can be successful, as red-billed quelea are unable to eat maize on the plant.

If nesting birds with newly built nests or unhatched eggs are consistently disturbed they may desert colonies and move elsewhere. The harvesting of quelea for food, either directly from nests (Jaeger and Elliott, 1989; Mullié 2000) or indirectly by catching flying adults (Elliott et al., 2013), may succeed in protecting crops if the colonies involved are not very big.

Chemical control involves spraying with organophosphate avicides such as fenthion  or cyanophos. The targets are concentrations of birds at roosts or breeding colonies. Spraying takes place at dusk, using either vehicle-mounted sprayers or aircraft. Dosages of 2-4 l/ha of fenthion are usual but occasionally they range from only 0.5 l/ha up to as high as 14 l/ha.

Some varieties of sorghum such as Ark-3048 have low concentrations of tannins but high concentrations of the cyanogenic glycoside dhurrin, which is avoided by quelea (Tarimo, 2000). See also Bullard and Gebrekidan (1989).

Combinations of cultural practices and harvesting of chicks or adults for food, or destroying nesting colonies, are possible IPM strategies. Traditional methods such as bird-scaring were reviewed by Bashir (1989), as were repellents by Bruggers (1989) and agronomic practices by Bullard and Gebrekidan (1989).

So far no remote sensing methods have been used other than the use of satellite derived rainfall data for forecasting models (Cheke et al., 2007), although satellite monitoring of habitat suitability has been attempted (Wallin et al., 1992). Thus, monitoring and surveillance relies on scouting, aerial surveys or reports by affected farmers.