More than garden plants: extending the conversation of urban gardens as an important refuge for Australian birds

Domestic dogs (Canis familiaris) or cats (Felis catus) are the pets of choice for millions of Australians with approximately 6.38 million dogs and 5.33 million cats living in households across the continent (Animal Medicines Australia 2022). Unfortunately, the presence of both dogs and cats can pose a risk and a threat to birds (Doherty et al. 2016). Since the relatively recent introduction of cats to Australia (i.e. 1788), they have become a significant predator of the native and ill-adapted birds across Australia, including in gardens (Legge et al. 2020). Research has conservatively estimated that domestic roaming cats kill as many as 61 million birds across Australia each year (Woinarski et al. 2017) with cats preying on birds both day and night. However, the presence of domestic dogs has also been demonstrated to have a significant impact on bird behaviour and habitat selection in urban areas (Paker et al. 2014). Importantly, whether or not the dog actually catches birds may be irrelevant (but see below regarding window strikes); the perceived risk associated with the mere presence of a dog making it less likely that a bird will visit or breed in a garden where a dog is present (Paker et al. 2014).

The best way to mitigate the pet-related threat to birds is containment. While some owners of larger dogs may find this challenging, a small dog can be more easily contained indoors. Likewise, if urban residents want the best outcomes for birds in their gardens, cats should be kept indoors or fully contained at all times. Cat owners can provide their pets with the mental stimulation derived from ‘outdoors time’ by installing a cat run or enclosure (De Assis and Mills 2021). The benefits of cat containment are not just limited to the safety of birds but also the welfare of the cat (De Assis and Mills 2021; Calver et al. unpublished). Furthermore, pet cat containment wholly removes the risk of car strike and other potentially fatal interactions with people and other pets.

The threat posed by dogs to birds also be concerning outside of residential gardens but in nearby urban areas. Australia’s towns and cities are often situated in coastal zones with extensive intertidal areas providing a recreational extension to the urban garden in many places. These intertidal areas often serve as resting and foraging habitat for many threatened East Asian Australasian Flyway migratory waders (i.e. shorebirds) during the Austral summer. Taking the family dog to the beach and estuaries of our coastal urban areas is a popular and long-standing use of coastal habitats. However, dogs that are allowed to roam off leash and flush foraging or roosting shorebirds in intertidal areas create a costly disturbance to these migratory species during the critical rest and refuelling stage of their life history (Weston and Stankowich 2014; Steven et al. 2011; Perera et al. 2022). Repeated bouts of human disturbance during this period may have detrimental flow on effects to an individual’s ability to reproduce or indeed complete a northward migration back to the Arctic (West et al. 2002; Lilleyman et al. 2016). Dog owners can reduce or even mitigate this threat through awareness of the times of year that coastal intertidal zones are inhabited by shorebirds (either migratory or resident) and avoid dog walking in these areas seasonally. Switching to inland dog parks and sports fields can mitigate this conflict without depriving their dogs of important exercise and socialisation.

Despite many Australian birds being nectarivores (i.e. honeyeaters), many of the most appreciated birds across the continent are insectivores. Fairywrens (Malurus spp.), willie-wagtails (Rhipidura leucophrys), Australian magpies and even honeyeaters during breeding season all rely on robust populations of insects for food. There is global recognition and concern among ecologists regarding declines in the abundance of insects (Sands 2018; Wagner 2020; Wagner et al. 2021). In Australia, there has been recent calls for attention to be directed towards this issue and the implications for numerous woodland bird species, many of which are insectivorous (Braby et al. 2023). The drivers behind these declines may not be simple, with climate change, insecticide use, artificial light and habitat loss all likely drivers (Wagner et al. 2021). While many would advocate for conservation of insects in their own right, the flow on effects these population declines have on the broader ecosystem, including birds, cannot be understated. Fortunately, native Australian bees at least, have experienced a surge in research interest and public awareness about nature in the urban environment. Blue-banded bees (Amegilla spp.) are often a centrepiece in many urban art murals and social media influencers and groups showcasing native bee content are growing in their reach. While research has shown that urban remnant patches hold more native bee diversity than private gardens (Prendergast et al. 2022), the urban gardener can do a lot to help bolster these remnant patch populations in the urban garden.

Some of the threats to insects are not easily controlled for by the urban gardener; global climate change and altered phenology or large-scale land clearing for example. However, some smaller scale threats can be mitigated at a local level by individual land holders. Avoiding the use of insecticides, especially systemic formulations, in gardens is one of the most immediate behaviour changes that can improve the viability or the urban environment as a habitat for insects (Sánchez-Bayo et al. 2013). When gardeners choose to cultivate locally native plant species, they also remove the need to use insecticides as Australian native plants are chemically and physically adapted to deter herbivory from insects. Artificial light has been implicated in elevated insect mortality and reduced breeding success (Owens et al. 2020; Kehoe et al. 2022). Limiting the use of artificial lighting outdoors throughout the night can prevent these impacts at local scales.

Most urban residents are familiar with the ever-present population of rats and mice in and around houses and local businesses. In an effort to supress these species’ populations, rat poisons are often deployed containing second generation anticoagulants (SGARs). SGARs work by attracting consumption by the rodents and then stimulating a thirst response that motivates the rats to go outside in search of water. SGARs kill rodents via catalysis of internal bleeding. The impact these products has on birds, specifically raptors and birds of prey, is known as secondary poisoning (Mendenhall and Pank 1980). Secondary poisoning occurs when a bird preys upon a rat or a mouse that has consumed the SGAR subsequently accumulating the toxic chemicals in their liver and tissues, which after repeated exposure, results in a long and painful death to the bird. There is now extensive evidence that SGARs are having widespread impacts on urban birds across Australia; incidence of secondary poisoning in boobook (Ninox boobook), powerful owls (Ninox strenua) and barn owls (Tyto alba) is growing (Lohr 2018; Cooke et al. 2022, 2023).

The use of rodenticides may be unavoidable for some urban properties, particularly those which are businesses serving food and beverages. However, these businesses, along with private landholders can choose to use alternative products such as first-generation anticoagulant rodenticides (FGARs) which pose a reduced risk of secondary poisoning. However, even FGARs are not considered the best practice approach to rodent control. The RSPCA in Australia states that the most humane method of rodent control is the use of snap traps. Snap traps usually result in an instant kill of the target species and have zero risk of secondary poisoning.

Raptors make an ideal flagship for the urban adaptor category of birds. With the right conditions in the urban environment, the population of raptors (especially owls) would have adequate roosting and nesting sites and serve as a naturally occurring biocontrol of pest rodents. However, if their populations continue to be impacted by the use of SGARs, we will likely see localised extinctions of raptors and miss out on this ecosystem service potential before it is fully realised among the broader community.

Although not well documented in Australia, birds striking windows in urban areas is a cause of mortality for many bird species globally. For some regions, the risk is amplified for birds unfamiliar to an area if they are undertaking seasonal migration, either latitudinal or altitudinal in nature. For North Americans, this can be an issue for buildings in the flight path of the numerous hummingbird species that migrate between central America and North America (USA and Canada). In Australia, both resident and migrant species can be affected. Species with directional and fast flight patterns such as fruit-doves, kingfishers and even noisy pittas (Pitta versicolor) can be particularly vulnerable to window strike. This is a concerning issue for areas east of the Great Dividing Range in eastern Australia where these species undertake altitudinal migration between higher altitude and coastal habitats, many of which are now highly developed. If a window strike doesn’t result in direct mortality, there is a secondary indirect risk of cat or dog predation as a result of stunning (Rebolo-Ifran et al. 2021).

The hazard of window strikes often arises from glass reflecting the sky and vegetation nearby which gives birds the false impression of extent or depth of habitat. The most frequently suggested mitigation strategies include making modifications to the surface of glass windows using etching, UV patterning on glass or decals, placing netting or screens on the external surface of the windows (Birds Queensland 2024; The Cornell Lab 2024; The Humane Society of the United States 2024). Several bird conservation organisations around the world suggest placing water sources such as birdbaths relatively close to windows, thereby encouraging them to approach windows slowly, thereby reducing the chance of window strike. In the USA and UK, where bird feeding is promoted among many conservation organisations and supported with best practice guidelines for those regions (Audubon International 2019; RSPB 2024), hanging bird feeders in front of windows also seems to lessen the likelihood of strikes.

Choosing the right species to cultivate in urban gardens has been the most important commonly spruiked method of making the urban environment a viable habitat for birds. Keeping dogs and cats contained will serve little purpose in improving the bird friendliness of the urban garden if the food resources or nesting habitat available is inadequate. Two key aspects of plant species selection need to be considered when choosing an approach: species composition and plant structure (Green 1984; Sewell and Catterall, 1988; Parsons et al. 2006). Species composition is often linked to the food resources available for birds, while the plant structure can dictate the nesting habitat viability and protection from predation, although these are not mutually exclusive relationships. Species selection ideally include plants that are locally native and thus ecologically suited to the birds found in a given area. Experimental research supports this notion with native plants favoured by native birds as opposed exotic plant species (French et al. 2005; Paker et al. 2014) – but see Van Helden et al. (2020) for an example that does not support this for native mammals in Australia.

Nectar-bearing plant species in a given location are likely to be co-evolved for the honeyeater species found within the same natural distribution. In Australia, where honeyeater diversity is high, bird species range from very small in body size to relatively large boisterous. Correspondingly, the flora of Australia comprises small nectar-bearing flower species delicate enough to be visited by small birds through to larger more robust species with inflorescences that can withstand the weight of the larger honeyeaters. However, some researchers and conservation practitioners caution gardeners selecting for a nectar-heavy plant species, with some gregarious honeyeater species (e.g. noisy miners) tending towards aggressive and despotic behaviours in landscapes that exhibit high nectar availability (Sewell and Catterall 1998; Barth et al. 2015; Catterall 2004). Alternatively, the urban gardener could think critically about whether the plant species selected are appropriate for the birds they are seeking to attract and support. For example, to attract insectivorous birds that are urban adaptors, the plant species selected need to attract insects, usually as pollinators (White et al. 2005) and select against providing nectar-rich, year-round flowering cultivars.

The structure of the garden is also a critical aspect dictating habitat suitability for different species of birds, especially with respect to breeding habitat. Substantial research indicates that urban vegetation structure is as important as species composition for supporting a diversity of Australian bird species (Sewell and Catterall 1998; Garden et al. 2006; Davis and Wilcox 2013; Threlfall et al. 2016, 2017). The trend towards simplistic and homogenous gardens, often marketed as ‘low maintenance’, has resulted in an urban environment that lacks the ability to support many bird species we would term urban adaptors. Even when native plant species are selected, few bird species will thrive in a garden that lacks adequate shelter and complexity. Small Australian birds, such as fairywrens, pardalotes (Pardalotus spp.) and silvereyes (Zosterops spp.), that vulnerable to ‘bullying’ and even exclusion by larger, more aggressive birds, need refuge in structural complexity in gardens (Garden et al. 2006; Parsons et al. 2006; Humphrey et al. 2023). This is usually provisioned through the inclusion of medium sized shrubs and dense multi-layered canopies that provide adequate shelter and refuge for smaller birds.

Retaining old trees wherever possible is the best way to ensure cavity nesting species such as pardalotes, parrots and owls can continue to sustain their populations in the urban environment (Sewell and Catterall 1998; Barth et al. 2015). There is also compelling evidence for the myriad benefits trees in urban areas provide to humans (Salmond et al. 2016; Locosselli and Buckeridge 2023). Challenges to large tree retention include the perceived risk associated with public safety and damage to property, as well as other structural issues related to the need for irrigation and root invasion (Salmond et al. 2016; Locosselli and Buckeridge 2023). Despite this, the risks posed by having large trees in urban areas can be managed, by specialist arborists who favour tree retention, without complete removal (e.g. see habitec.org.au). Unfortunately, risk-averse state governments tend to enact policies that are counter to a constructive approach to retaining large native trees in urban areas, where risk mitigation usually supports complete removal (Phelan et al. 2019). Fortunately, the private landholder still holds much of the right to decide whether trees are retained or not and can address external concerns through monitoring and management of large trees.

The concern regarding declining insect populations is an issue for birds and the wider urban ecosystem both in Australia and internationally (Braby et al. 2023). To create a functioning ecosystem in an urban garden that attracts and supports insects, requires an understanding of plant-insect relationships. Plants that have small flowers, usually white, green, blue or yellow are generally considered the most attractive to insects such as flies, bees and other small flying pollinators (Garcia et al. 2022). Insect-attracting plants may be herbaceous (small plants with fleshy stems) or small to large shrubs (with woody stems). Trees with small flowers have the ability to attract huge numbers of insects seasonally and canopy foliage adds to this insect load available to certain bird species. Smaller insectivorous birds, which are some of the most relevant in the urban adaptor conversation, are more likely to be found in the lower to mid strata of a garden, emphasising the importance of insect-attracting shrubs.

While birds need food and nesting sites in gardens for their populations to thrive in urban areas, so too do insects. As such, and with growing interest in insect conservation (Fig. 1a, b), there has been a recent surge in the promotion and installation of bee and insect ‘hotels’ or houses in many urban gardens. Artificial bee and insect hotels are comprised of a number of materials that mimic the substrate or structures that many insects live in. Despite the popularity and potential benefits of provision bee and insect hotels, some have cautioned that they may disproportionately favour introduced insects in certain locations (i.e. North America and Europe) (MacIvor and Packer 2015; Geslin et al. 2022). As such, their use should be employed with a sound knowledge of what is and is not native to the local area.

Most of Australia’s bees and wasps, many of which are favourite food for birds, reproduce in a diverse range of physical structures in nature (Batley and Hogendoorn 2009; Heard 2016). Small tubular cavities are favoured by leaf cutter bees and many wasps, while many solitary bees roost or nest in soil or clay (Heard 2016). Some Australian bees will even excavate nesting tunnels in crumbly mortar between bricks. The co-benefits of encouraging locally native invertebrates into the urban garden extends well beyond providing food for birds. Wasps are excellent companions to the vegetable garden, with many actively searching for cabbage moth larvae to take back to their nesting tubes to entomb with their eggs. Additionally, blue-banded bees are effective pollinators of popular vegetable garden plants including tomatoes, eggplants, zucchini and cucumbers through their ability to perform buzz pollination (Cooley and Vallejo-Marín 2021). Buzz pollination differs from normal pollination where bees use their flight muscles to vibrate at a higher frequency, resulting in better pollen release from floral anthers and thus better pollination.

There is a long-standing literature on the positive effects of bird feeding on the relationship between birds and people in the Northern Hemisphere, especially for the UK (Cox and Gaston 2016) and USA (Wilcoxen et al. 2015; Dayer et al. 2019). However, the debate around feeding Australian wild birds is complex and controversial (Jones and Reynolds 2008; Reynolds et al. 2017). Advocates, in the UK for example, write of the connection between people and the birds they feed and support the notion that this connection can enhance pro-conservation attitudes and behaviours (Clark et al. 2019). Conversely, opponents here in Australia, cite issues such as, disease outbreak associated with congregations of birds gathering in gardens to be fed (Reynolds et al. 2017), the provision of inappropriate foods for certain species (Gillanders et al. 2017) and the ecological implications of feeding birds whose presence and boldness has negative impacts on the remaining community of birds as a result of feeding (Parsons et al. 2006). Furthermore, the species selected for by urban residents are often large and rambunctious species (i.e. urban exploiters) that actively exclude species that lean towards urban adaptor status, thereby exacerbating the very issue the bird feeding public think they are redressing (Parsons et al. 2006). Clearly there are complexities and exceptions to most generalisations you can make regarding wild bird feeding (Jones 2011) and a lack of policy pertaining to it in many locations (Baverstock et al. 2019). There is a consensus that the best scenario for Australia’s wild birds is to protect the remaining remnant habitats wherever possible and enhance suboptimal habitats through careful restoration or cultivation, depending on the tenure of the land (BirdLife Australia 2024). In the case of private land, and urban gardens to be more specific, cultivation of plants with bird habitat in mind for ‘providing food’ for birds is considered the best practice approach (BirdLife Australia 2024).

Less controversial is the provision of water via birdbaths; a common practice in Australia (Cleary et al. 2016) and around the world (for example see Gibbons et al. 2023). Urban areas tend to be impoverished of reliable water sources, where creeks and ponds have often been infilled or drained to make way to housing development (Gibbons et al. 2023). Many birds rely on access to fresh water, and some have obligated drinking patterns inherent in their biology, such as most, if not all black-cockatoo species, that drink every afternoon shortly before settling to roost (McGrath 1991; Johnstone and Kirkby 2016). Clean and well-maintained birdbaths will provide an important source of water for drinking and bathing in urban Australia where access to fresh water from rainfall is predicted to become less reliable as a result of anthropogenic climate change (for example, Broderick and McFarlane 2022; Zarch et al. 2017).

Finally, provisioning of nesting sites will be another important intervention for urban birds as hollow-bearing trees become a rare feature of the landscape (Treby et al. 2014). Many urban bird species are cavity nesters, including, cockatoos, lorikeets, parrots and laughing kookaburras (Dacelo novaeguineae). However, there are a suite of less encountered urban adaptor species that also rely on hollows for reproduction including owls, owlet-nightjars and pardalotes. Each of these birds will have different requirements in terms of the size, shape and position of a nesting box. Nest boxes can be problematic if they are positioned in a manner that exposes any potential occupants to predation from domestic pets. Ensuring ample height from the ground can be one mechanism to minimise this risk. Additionally, nest boxes can be exploited by unwanted species such as feral honeybees or introduced bird species (e.g., common myna [Acridotheres tristis] in the eastern states, laughing kookaburras in the west). Measures should be taken to preclude these species from taking up residence in a provisioned nest box.