Avian Demography Unit
Department of Statistical Sciences
University of Cape Town
ADU home     UCT Home page

BIRD NUMBERS Volume 9 Number 2, December 2000

21. Changing suburban birds – terrestrial species

J. A. Harrison
Avian Demography Unit, UCT, Rondebosch, 7701

In comparison with other vertebrates – fish, amphibians, reptiles and mammals – birds do relatively well in man-made environments, both in variety of species and numbers of individuals. The mobility of birds is a key factor, enabling them to avoid the threats which human beings pose, and allowing them to move easily between areas which offer opportunities to feed, breed and roost.

As an ever greater proportion of South Africa’s land surface is transformed from natural to man-made habitats, the impact that human activity has on bird populations becomes more dominant. In western Europe there has long been a situation in which virtually all the land has been altered and conservation actions are focused more on intensive land management and the regulation of human activities than on the preservation of the original natural habitats and ecosystems. Observation of the birds in our cities, suburbs and farms can give us insights into how we are affecting bird populations, and what the future may hold under different development scenarios.

In this article I attempt to classify the types of population change, in terms of their causes, and examine some recent changes in the terrestrial birds of the suburbs of our cities. (A future article will focus on waterbirds.) I hope that this exercise may assist birders in becoming more aware, not only of the birds in their gardens and neighbourhoods, but also of the ecological factors which cause bird communities to change continually.

A classification of causes of change in suburban bird populations

A) habitat changes
  • habitat destruction
  • habitat creation
B) community changes
  • avian changes
  • non-avian changes
C) behaviour changes
  • avian behaviour
  • human behaviour

A) Habitat changes

Development of a new suburb always involves changes to the existing habitats, and initially these changes are mainly destructive. Many of our largest cities – Johannesburg for example – are situated in the grassland biome. Development of a new suburb destroys the natural grassland and, not surprisingly, the grassland birds move out as the bulldozers move in.

Slightly less obvious is why the grassland birds do not return after the suburb has been established and there are gardens with plenty of grass. The answer, of course, lies in the details of the habitat that has been created. The grasses are of different species and are kept short and moist. Also, apart from the buildings themselves, the open nature of the original grasslands is destroyed by the planting of numerous trees and shrubs.

On the other hand, these features are attractive to woodland birds. Our bias towards trees with showy flowers and fruits probably tend to make suburban woodlands especially attractive to herbivorous and insectivorous birds. The result is that the suburbs of Johannesburg, and other towns on the highveld, have become increasingly characterized by woodland species, including Crested Barbet, Blackcollared Barbet, Redbilled Woodhoopoe, Grey Lourie and Olive Thrush. The influx and establishment of these species as breeding residents can be seen as the result of habitat creation, in this case a habitat quite different from the original grasslands.

Occasionally, destruction of habitat occurs outside of the suburbs themselves, creating pressure on birds to find alternative habitats, especially for foraging. Such a situation occurred in 1999 when severe wildfires destroyed much of the natural mountain fynbos habitat on the Cape Peninsula. Many suburban birders noticed a sudden influx of sunbirds, Cape Sugarbirds and Cape Bulbuls which had presumably been displaced by the fires and the complete loss of their natural montane habitats. It may be that some such ‘forced’ influxes have resulted in a few individuals remaining in new areas, thus forming the nuclei of new, permanently resident populations, but in many cases they are merely temporary and short-lived irruptions.

Drought is another catastrophic event which can drive birds into the suburbs in search of food and water. Even outside of drought periods, the availability of water is usually greater around human settlements and this is no doubt an important attractant to species which need to drink regularly.

B) Community changes

As gardens become more densely vegetated, several of the more adaptable species will begin to breed in gardens, e.g. doves. However, certain predatory species are attracted by the denser cover, and perhaps by the presence of the breeding birds themselves, e.g. coucals, shrikes and hawks. The presence of these predators can radically depress the breeding success rates of birds like doves and thereby reduce their numbers. A balance between predators and prey may take several years to establish, or an ongoing cyclical pattern of increase and decrease of predators and their prey may develop.

Competition between species which occupy similar niches can also be significant. For example, suitable holes for hole-nesting birds can be a limiting factor, causing such species to compete for available holes. The less dominant species will tend to be driven out and may even become locally extinct.

Another change which is likely to follow the establishment of breeding populations is the influx of the relevant brood parasites, namely cuckoos, whydahs, honeyguides and Cuckoo Finches. Non-avian changes in the community include the occurrence of species which are commensal with humans, such as rats, and species deliberately introduced by humans, such as domestic cats. However, these examples are perhaps more appropriately dealt with under the heading of behaviour changes.

C) Behaviour changes

This, in my view, is where things become interesting, because it is in changing behaviour that one is able to see adaptation in action. However, it is not necessarily easy to distinguish between an adaptive change in behaviour and a predictable response to a change in habitat or other factors in the environment – any exploitation of habitat which is substantially different from the ‘natural’ can be viewed as adaptive change in behaviour. There is, therefore, an overlap between this category of behaviour change and the categories of habitat and community changes described above.

Notwithstanding this caveat, the adaptations made by some species, although understandable in retrospect, could not have been predicted in advance and are therefore more usefully explained in terms of behaviour change than a simple response to the availability of suitable habitat. An example of such a behavioural adaptation would be the Pied Crow’s exploitation of roads as a source of carrion, involving as it does, a whole new set of behaviours to avoid becoming road-kill itself!

The concept of ‘pre-adaptation’ is relevant to behavioural changes. One would be surprised to find an eagle trying to extract insects from a garden lawn, but one would be less surprised to see a hoopoe do the same thing. The reason is that, although garden lawns are not the natural feeding habitat of either species, the hoopoe can be said to be pre-adapted for probing lawns with its long, decurved bill, while an eagle with its short, hooked bill is clearly not pre-adapted for this style of feeding.

Pre-adaptations can refer to behavioural as well as physical traits. For example, a species which nests on cliffs is better pre-adapted for nesting on high-rise buildings than a species which nests in grass tussocks. A species is said to be pre-adapted because it already has adaptations which give it the potential to exploit a particular new set of conditions. By asking ‘How is this species pre-adapted?’, one can perhaps predict how it will respond to a new set of circumstances.

The species whose behaviours and attitudes are most influential is, of course, Homo sapiens. Human attitudes determine, for example, that we choose to plant trees and lawns in our gardens instead of having expanses of tall, waving grass. We also choose to keep predators such as dogs and cats as pets, and to drive high-speed missiles, called cars, down vast expanses of tar. And so on and so on. All of these human behaviours have an impact on birds.

What makes human behaviour so unusual is its extreme malleability – we constantly change our habits, sometimes for the better and sometimes for the worse. In many suburbs the tradition of using catapults and airguns against birds is now a thing of the past, as is the once-popular hobby of egg collecting. On the other hand, the domestic use of insecticides, herbicides and other poisons increased dramatically since the Second World War, and is probably only now beginning to decrease again as people become more aware of their negative impacts. In societies where birds are popular, feeding tables, birdbaths and nest boxes have become common features of the suburban garden, and these have their own effects on the dynamics of suburban bird populations.

Selected examples

An informal survey of changes in suburban bird populations was conducted through SABIRDNET, and it yielded 47 responses covering all the major metropolitan centres – Johannesburg, Pretoria, Durban, Pietermaritzburg, East London, Port Elizabeth and Cape Town – as well as a number of smaller towns. The species selected here were chosen mainly because they generated several corroborating reports. Reports of increases and decreases in several other species were also received, but space does not allow for a discussion of them all – perhaps in future articles. To preserve readability, I have not cited all of the personal communications in the body of the text, but have listed all contributors under acknowledgements, below.


The Grey Lourie Corythaixoides concolor is perhaps the most famous of the recently established woodland birds on the Witwatersrand. The initial influx occurred in the late 1970s, and in the 1980s the species became common in well-wooded suburbs. This was undoubtedly made possible by the creation of artificial woodland habitat and especially the availability of fruiting trees such as the alien Seringa Melia azedarach (Johnson ASAB1).

There are now indications that the Grey Lourie may be decreasing in some suburbs (Albert Froneman pers. comm.). The probable reason for this is that the woodlands of some suburbs are now becoming too dense for a species which prefers open woodland. The suburban ‘savanna’ is becoming a ‘suburban forest’. The Purple Lourie Tauraco porphyreolophus, a species of dense moist woodlands and coastal forests, already frequents the lush parks and gardens of some cities such as Durban and Harare; perhaps it will eventually also spread to the ‘forests’ of the Witwatersrand.

Barbets, honeyguides and woodhoopoes

In greater Cape Town (CT), the Pied Barbet Tricholaema leucomelas became established as a breeding bird during the 1960s and 1970s, as a result of its being able to exploit extensive thickets of invasive alien acacias for nesting sites (Macdonald 1986). In the 1980s, the Lesser Honeyguide Indicator minor also became regular as it exploited the Pied Barbet as a brood host (Underhill & Underhill 1992).

The Blackcollared Barbet Lybius torquatus and the Crested Barbet Trachyphonus vaillantii are also widely associated with parks and gardens in the northern and eastern parts of the country. These species are also prevalent in the man-made woodlands of highveld suburbs. Increases in the Blackcollared Barbet have been reported from Johannesburg, Klerksdorp, Secunda, Pretoria and Port Elizabeth (PE), and in the Crested Barbet from J’burg, Secunda and Pretoria. In Klerksdorp it appears that the first Blackcollared Barbets arrived in the late 1980s.

A decrease in Crested Barbets was reported from Bedfordview in J’burg, together with an increase in Blackcollared Barbets. This and another report from J’burg suggested that Crested and Blackcollared Barbets actively compete for nest sites (Geoff McIlleron & Bob Cullen pers. comm.). It does not appear that one species always dominates the other, however. Possibly there are other, as yet unknown factors which tip the balance and cause one species to be locally dominant. Increases in honeyguides, especially the Lesser Honeyguide, which specializes in parasitizing treehole-nesting species, were reported from J’burg and CT (Durbanville).

Another treehole-nesting and roosting species, which has increased in Pretoria and J’burg, is the Redbilled Woodhoopoe Phoeniculus purpureus. This species presumably increased in numbers some time after the arrival of the barbets because it cannot excavate its own holes and depends on natural cavities or holes excavated by woodpeckers or barbets (Du Plessis ASAB1).

It is probable that, as suburban trees grow larger and older, dead wood and cavities suitable for nesting become increasingly available, encouraging the establishment of breeding populations of treehole-nesting birds. The provision of nest logs by birders is also believed to be an important factor in boosting local populations.


The arrival of the Rameron Pigeon Columba arquatrix in the artificial woodlands of the highveld might have been predictable, but the timing is significant. This itinerant species daily flies considerable distances between its roosts and the fruiting trees on which it feeds. The establishment of a population in an area is dependent on there being sufficient large fruiting trees in the vicinity to sustain the birds, in all seasons. This condition has probably only recently been met as a sufficient variety and abundance of suburban trees have matured and begun to bear fruit.

Increases in the occurrence and numbers of the Rameron Pigeon appear to have been taking place from the mid-1980s, and were reported from Bedfordview, Bryanston, Fairland, Randburg, Linden, Parkview and Linksfield Ridge in J’burg, and also from Pretoria. Breeding was reported for the Witwatersrand Botanical Garden (Albert Froneman pers. comm.).

Palm Swift Cypsiurus parvus

This species underwent a number of range expansions during the 20th century (Brooke ASAB1). In South Africa it was originally restricted to the Limpopo Valley, but its adaptation to nesting in alien palms, especially Washingtonia spp., and man-made structures, has enabled it to spread steadily southward and westward. It has recently become a common resident of East London and is now being seen occasionally in Port Elizabeth. An increase in numbers was reported from Pretoria.


It is apparent that the Redwinged Starling Onychognathus morio has recently, and rapidly, established itself as a suburban bird in many places. Reports of marked increases in abundance were received from Pretoria, Pinetown, New Germany, PE, Oudtshoorn and CT (northern and southern suburbs). In Cape Town the species became common in the southern suburbs in the early 1990s, while colonization of suburbs on the Witwatersrand appears to be in its early stages, with sightings in most areas still being a novelty.

The Redwinged Starling, originally an exclusively cliff-nesting species, has a long history of nesting on buildings, allowing it to expand its range considerably (Craig ASAB2). What is mysterious is why it should now be establishing itself as a garden bird, and how it is managing to do this with relative synchrony in so many parts of the country. For a species not known to move long distances, the synchrony is especially baffling. The colonization of towns is probably a step in a progression from initially using structures and buildings in rural environments, thereby becoming more accustomed to the presence of humans, and gradually spreading into more densely populated towns and cities. In PE, irruptions have been linked to droughts (Dave Brown pers. comm.), but it seems likely that we are witnessing a stage in a longstanding and ongoing process of colonization of man-made habitats. If this is so, it is less surprising that many widely separated populations are at similar stages of the process at the same time.

Adaptation to new sources of food is probably also a necessary step. These birds have been reported raiding dogs’ feeding bowls, and I have seen them go right inside large refuse bins to find scraps on UCT campus. Their choice of sites for roosting and nesting is also remarkable. I have seen them roost on shop awnings on the Main Road in CT, despite bright lights and noisy traffic, and have seen successful breeding take place in nests situated on top of much-used loudspeakers, inside the students’ union at UCT!

The Palewinged Starling O. nabouroup is much less associated with man-made habitats, but it has been recorded roosting on buildings in Windhoek, Namibia (Craig ASAB2). It may be that this species will, in future, mirror some of the trends of its congener.


The Pied Crow Corvus albus has much in common with the Redwinged Starling in that it also has a long history of association with man-made habitats and structures in rural settings, and it now appears to be rapidly making inroads into urban environments. In Zimbabwe the species has been common to abundant in cities for decades (Jenkins & Underhill ASAB2), but this has apparently not been the case in South Africa until very recently. Marked to spectacular increases in the occurrence of Pied Crows have been reported from the suburbs of Pietermaritzburg, PE and CT, but a decrease was reported from Bedfordview in J’burg. In PE the arrival of the crows appeared to be correlated with a drought in the late 1980s and early 1990s (Dave Brown pers. comm.). The decrease in Bedfordview is likely to have been an initial response to greater density of development.

I suspect that a greater tolerance by city dwellers of large birds, and especially predatory birds, is an important factor in the current increases which may otherwise have occurred much earlier. It is noteworthy that in CT the Pied Crow has become abundant in relatively affluent suburbs and does not appear to be especially associated with the refuse-rich informal settlements on the Cape Flats where, in contrast, the alien House Crow C. splendens has made its stronghold. It will indeed be interesting to see how these two species, both newcomers to CT, continue to partition the urban environment, or learn to coexist. It is also interesting to note that the Black Crow C. capensis has not made the transition from rural to urban habitats; competitive exclusion by the larger Pied Crow may be a factor.


The Cape Sparrow Passer melanurus and House Sparrow Passer domesticus are generally thought of as ubiquitous and abundant, but this is an oversimplification. Declines in the House Sparrow were reported from the southern suburbs of CT, PE, and Pretoria, and in the Cape Sparrow from the southern suburbs of CT. Increases in Cape Sparrow occurrence were reported from the northern suburbs of J’burg and P’mburg. Interestingly enough, a >60% decline in rural House Sparrow populations has been recorded in the UK (Cannon 2000).

Both Cape and House Sparrows feed on the ground in relatively open habitats. Seeds of weeds and grasses are their staple, especially in the case of the Cape Sparrow. During the early stages of development of an area, there tend to be plenty of open lots with weedy growth which make ideal foraging areas for these birds. Later, when development is more dense and gardens more closed in with trees and shrubs, they may run out of places to forage and hence decline. (Another common garden bird which tends to decline with the maturation of gardens is the Bokmakierie Telophorus zeylonus.) However, this explanation does not seem to suffice in all cases. It may be that predation on adults and on nests becomes limiting when owls, hawks, shrikes and coucals move in. There may also be competition with larger and more aggressive species for nest sites.

South African Cliff Swallow

See Bird Numbers 8(2) (December 1999) for a discussion of the behavioural responses of this species to changes in the availability of suitable man-made nest sites.


In compiling examples I have been greatly assisted by reports from subscribers to SABIRDNET who responded to an informal survey on changes in suburban bird populations. My thanks and appreciation go to all of these birders for their observations. Contributors, in alphabetical order, were:

K. Allen (Secunda), C. Boix-Hinzen (CT), H. Bouwman (Potchefstroom), M. Bowker (P’mburg), D. Brown (PE), M. Brown (P’mburg), B. Burchell (CT), B. Cairns (Pretoria), B. Cullen (J’burg), M. Delport (CT), W. du Buy (CT), R. du Preez (Bloemfontein), J. Freer (J’burg), A. Froneman (J’burg), F. Furstenburg (Pretoria), R. Geyser (Pretoria), P. Greaves (J’burg), B. Groom (Edenvale), B. Horton (CT), J. Hughes (CT), A. Kemp (Pretoria), M. Kriek (Pretoria), G. Leisegang (Durban), K. Mahood (Secunda), E. Marais (Pretoria), P. Martin (PE), R. Martin (Pretoria), M. McCall (CT), G. McIlleron (J’burg), D. McKenzie (Vryheid), C. Mitchell (Durban), G. Mulholland (J’burg), S. Muller (J’burg), D. Oschadleus (Pretoria), L. Oxley (J’burg), M. Raum (Edenvale), S. Smit (Vryburg), N. Smith (E. London), T. Snyman (Durban), C. Spengler (CT), S. Terblanche (Pretoria), M. Tripp (CT), K. van Dyk, R. van Staden (Somerset West), D. Whitelaw (CT), J. Wilson (CT), and J. Wilson (Klerksdorp).

Cannon A. 2000. The Garden BirdWatch Handbook. British Trust for Ornithology, Thetford.

Macdonald I.A.W. 1986. Range expansion of the Pied Barbet and the spread of alien tree species in southern Africa. Ostrich 57: 75–94.

Underhill G.D. & Underhill L.G. 1992. First capture dates of Lesser Honeyguides at two localities in the southwestern Cape Province. Safring News 21: 7–10.


The Editor invites you to respond to this article on changing suburban birds. Send a letter to Bird Numbers describing a change in your area, or elaborate on some aspect of this article. Letters should not exceed 300 words and should preferably focus on one species or aspect of change. All letters which are good enough to publish will be rewarded with a copy of one of the Bright Continent Guide series, published by the ADU. – Ed.

ADU Home page

Office Avian Demography Unit
Enquiries/More Information: adu-info@uct.ac.za
Document posted: 10 January 2001