• Change text size
• Skip to content

Australian Natural Resources Atlas

Natural Resource Topics

• About us
• Contact us

Search

• ANRA home
  • Map maker
  • Natural resource topics
  • Data Library

• Natural resource topics
  • Agriculture
  • Coasts
  • Dryland Salinity
  • Invasive species
  • Irrigation
  • Land
  • Natural resource economics
  • People
  • Rangelands
  • Soils
  • Vegetation and biodiversity
  • Water
  • Publications

• Vegetation and biodiversity
  • Overview
  • Assessment
    • New South Wales
    • Victoria
    • Queensland
    • South Australia
    • Western Australia
    • Tasmania
    • Northern Territory
    • Australian Capital Territory
  • Landscape health
  • NVIS framework
  • Extent
    • New South Wales
    • Victoria
    • Queensland
    • South Australia
    • Western Australia
    • Tasmania
    • Northern Territory
    • Australian Capital Territory
  • Clearing
    • New South Wales
    • Victoria
    • Queensland
    • South Australia
    • Western Australia
    • Tasmania
    • Northern Territory
    • Australian Capital Territory
  • Management

5. Birds

Summary

This chapter examines a range of patterns and recent changes in bird species composition and abundance at both continental and bioregional scales. These patterns include species richness, endemism, threatened species, national and core habitat and the condition and trend of some of these attributes for both species and bird guilds.

The assessment relied principally upon the analysis of 6 million records of 744 bird species that were collected during two Bird Atlas periods, 1977-1981 and 1998-2001. As a consequence, only changes in distribution or abundance that occurred since the first Atlas survey are described in these analyses. Of the total number of records in the two Bird Atlases, there was adequate information for the detailed bioregional analysis of trends in 495 species (1.7 million records).

The analyses showed that there were considerable regional differences in the patterns emerging. Three principal interacting factors contribute to the observed trends: changes in landscape health (particularly in highly cleared areas), climatic fluctuations, and the biases resulting from changes in survey methods used between the two Atlases.

Many bird species appear sensitive to declining landscape health caused by broad-scale vegetation clearance, urbanisation, intensified agriculture and introduced predators. Overall, the clearing of land for agriculture appears to have had the greatest non-climatic influence on bird abundance.

Twenty-nine of the 497 species analysed in agricultural Australia showed significant decreases in reporting rates where increased proportions of the landscape have been cleared. Grassland, woodland and groundnesting guilds were particularly affected.

Changes in fire regime (sometimes in association with intensification of pastoralism) have been implicated in the decline of some species, particularly the granivores.

Twenty-six species appeared to decline with decreasing landscape health, a measure that includes abundance of feral animals and weeds as well as intensification of primary production. There was a negative trend in reporting rate in 61 species in the intensive land use zone, when compared to Australia's rangelands, with at least part of this relationship being explained by climatic differences between survey periods. In contrast, species found in guilds that inhabit rainforest, temperate forest and mangrove systems showed significant increases in reporting rate. The reason for this trend is unknown but might be related to more intensive surveys in the second Atlas study.

Some native species are favoured in intensively developed landscapes. For example, the trend in reporting rates of 21 species increased with increasing land clearance and declining landscape health. Up to 15% of the bird fauna is exotic in south-eastern Australia, including eastern Tasmania, and in south-western Australia. This includes some native species that have been introduced outside of their natural range.

Long-term trends in bird abundance or distribution are much more difficult to identify in areas such as the tropical savannas of northern Australia and the arid zone. Separating the effects of climate in these regions from other factors is difficult because the climate, particularly the patterns of water availability across central Australia, differed greatly between survey periods. Two land use factors that vary within bioregions, and are therefore difficult to assess at a bioregional level, are rangeland grazing intensity and changed fire regimes. However, reporting rates of some grazing-sensitive species such as Singing Bushlark (Mirafra javanica) and Flock Bronzewing (Phaps histrionica) have declined.

Interacting with all of these factors were major differences in rainfall, and, in some regions, temperature, between the two Atlas periods. Significant changes in distribution (both increases and decreases) caused by different rainfall patterns were observed for around 6% of species. A number of other species were unusually common near the coast during the first Atlas, but appear to have contracted to the centre, particularly the Channel Country, in the second. However, a number of less mobile species probably increased and decreased at a local scale, resulting in a similar pattern of change.

Details of selected bird species and information on the special values, condition, trend and recommended management responses are provided for each bioregion on the Australian Natural Resources Atlas (www.environment.gov.au/atlas). On-going monitoring of key species identified for each bioregion will allow rapid appraisal of landscape and avian health. It will increase our understanding of the trends in bird distribution and abundance, and the processes affecting them, particularly over similar climatic conditions.

Introduction

Australia has over seven hundred species of birds, found across a diversity of habitats ranging from sub-alpine forest to the spinifex grasslands of central Australia. Birds have a special place in the hearts of most Australians, which is demonstrated by the large number of amateur bird watchers across the continent. Many of these individuals volunteered to collect survey data on birds for a nation-wide project coordinated by the Royal Australasian Ornithologists Union (now known as Birds Australia). The first 'Bird Atlas' survey was conducted between 1977 and 1981 and has been followed up by a second Atlas survey conducted between 1998 and 2001.

While there was a number of differences in the way birds were surveyed in the two Atlases, there is no comparable national scale data set for Australian birds or any other Australian taxa. This chapter uses the data from the two Atlases to examine a range of patterns and recent changes associated with bird species at both continental and bioregional scales. Features assessed include species richness, endemism, threatened species distribution, national and core habitat and the condition and trend in reporting rates of both species and bird guilds.

Similar 'audits' have been undertaken in the United Kingdom, where trends in the abundance of birds is considered one of the primary indicators of the health of the nation, along with the rate of unemployment and the gross domestic product. The broad analysis of the Atlas data presented in this chapter therefore provides significant insights into the environmental well-being of Australia.

The Assessment

The data set used in this report was obtained from the Bird Atlas projects of Birds Australia. The Atlas projects surveyed the birds of the Australian continent and its near-shore islands from January 1977 to December 1981 (referred to as Atlas period 1) (Blakers et al 1984), and from August 1998 to August 2001 (referred to as Atlas period 2). This means that only changes in distribution that have occurred since the first Atlas survey are detected by these analyses. Approximately 41,000 record forms were used from those submitted in Atlas 1 and 47,000 for Atlas 2. Each record form contains one or more sighting. The analysed sheets contained a total of 768,000 sightings in Atlas 1 and 955,000 in Atlas 2. Overall, 6 million records were collected for 744 species during the two bird Atlas periods, 1977-81 and 1998-2001. Of the total number of records, there was adequate information to analyse for trends in 497 species (1.7 million records).

Because of the different ways data has been collected in the two Atlases, their analysis presents numerous challenges, as summarised in the detailed report on methods for this chapter (see the Australian Natural Resources Atlas for details).

The approach taken was to identify statistically significant patterns in the data, and then discuss which factors might contribute to these patterns. In doing so, correlations between changing reporting rates and environmental factors across the bioregions are considered more reliable indicators of the fate of a species than are the absolute changes themselves. The patterns identified are presented as relationships that are worthy of further investigation and highlight areas where more detailed analysis, both of the current data set, and of further specifically designed studies, is warranted.

The data sets were used to describe the distributional characteristics and trends for species at a national and bioregional level. Analysis of national reporting rate change was undertaken for bird populations across the continent. Using reporting rate enabled an unbiased comparison of trends in intensively surveyed areas like south-eastern Australia with those in more remote bioregions. Analysis of core bioregion reporting rate was undertaken for bioregions that are important for individual species only. Species were also grouped into guilds based on habitat and mode of feeding. Analysis of guilds assessed changes in bioregions at both Australia-wide and bioregional scales.

Findings

Patterns in Bird Distribution and Abundance

Number of Species

The bioregions with the highest number of bird species occur along the length of eastern Australia, from Cape York Peninsula to Cape Otway in Victoria and inland through the lower Murray- Darling Basin (Figure 5.1).

Within this band, the highest numbers of species have been recorded from south-eastern Queensland (429 species) and north-eastern New South Wales (428 species), followed by Sydney Basin (423 species) and the Wet Tropics (395 species). These numbers are inflated by seabirds, seasonal migrants that breed outside Australia, and vagrants. The highest numbers of species that breed in Australia (resident species) are found in the Brigalow Belt (BBN 332 species, BBS 342), Wet Tropics (325 species) and South East Queensland (330 species) bioregions. The last two of these bioregions are also in the top four for the total number of bird species, so make a particularly important contribution in terms of the overall number of species. The lowest number of species was recorded from Tasmanian bioregions (e.g. Tasmanian Central Highlands 129 species; Tasmanian Northern Highlands 140; Tasmanian Southern Ranges 151), the arid interior and western half of the continent (e.g. Gibson Desert 133; Little Sandy Desert 146; Central Ranges 151), although some of these bioregions were poorly surveyed.

Many Australian resident species also occur outside Australia, particularly in northern neighbouring countries. Correspondingly, the highest concentration of species only found in Australia is found in southern Australia, particularly inland. The semi-arid zone of southern Western Australia has the highest proportion of these most distinctive elements of the Australian bird fauna, which are adapted to harsh conditions. These are quite unlike species found elsewhere in south-east Asia or the Pacific.

Rarity

As elsewhere in the world, island species are disproportionately represented on the Australian list of threatened species. Both offshore islands under Australian jurisdiction, such as Christmas, Norfolk and Lord Howe, and nearshore islands, including Tasmania, have the highest levels of extinction and rarity in the country. For example, King Island has 13 threatened species, including four that are critically endangered and Kangaroo Island has 7 threatened species and 2 critically endangered. Some of the Tasmanian bioregions also have very high numbers of threatened species, particularly critically endangered species (Garnett & Crowley 2000) that are close to extinction eg. Tasmanian Southern Ranges - 9 threatened species, 2 critically endangered, Tasmanian West - 8 threatened, 2 critically endangered) (Figure 5.2).

The biogeographical effect of islands also occurs on the contintent proper. Patches of vegetation surrounded by crops or cleared land can be viewed as islands of habitat in a contrasting matrix. Bioregions where these land-use patterns are found include the Murray Darling Depression - 11 threatened, 1 critically endangered, Naracoorte Coastal Plain - 9 threatened, 1 critically endangered, the heaths and rainforest patches of South Eastern Queensland - 12 threatened, 2 critically endangered, NSW North Coast - 11 threatened, 2 critically endangered, and South East Coastal Plain - 13 threatened, 1 critically endangered. The southwestern bioregions east of Albany (Esperance Plains, Jarrah Forest) would have had a higher threatened species score for birds but for the highly successful recovery programs that have been conducted in the area for Noisy Scrub-bird (Atrichornis clamosus), Western Bristlebird (Dasyornis longirostris) and Western Whipbird (Psophodes nigrogularis nigrogularis), which have recently been upgraded from endangered to vulnerable.

Centres of Endemism

The number of species occurring in ten or fewer bioregions was considered the most appropriate measure of centres of endemism within Australia. Approximately 25% of the 555 resident species analysed are recorded from fewer than 10 bioregions (Figure 5.3). The largest numbers of these limited range species are found in Cape York Peninsula and the adjacent Wet Tropics.

Fourteen limited range species (ie. that occur in 10 or fewer bioregions) are found only on Cape York Peninsula, although all of them also occur across Torres Strait in New Guinea. The Wet Tropics has a higher number of Australian endemic species, with 13 species largely confined to its upland rainforests between Paluma and Big Tableland, only one of which, the Blue-faced Parrot-finch (Erythrura trichroa), also occurs outside Australia. At a broader scale, over 30% of restricted range species are confined to the combined bioregions of north-eastern Queensland (Cape York Peninsula, Einasleigh Uplands, Gulf Plains and the Wet Tropics), with a further 14% extending down the east coast to northeast New South Wales. Three other areas also containing a disproportionately high number of restricted range species are: Tasmanian bioregions with 14 terrestrial and 4 marine species (12%), southwestern Australia (Avon Wheatbelt, Coolgardie, Esperance Plains, Geraldton Sandplains, Jarrah Forest, Mallee, Swan Coastal Plain and Warren bioregions) with 12 and 3 (10%), and the Top End and Kimberley (Arnhem Coast, Arnhem Plateau, Central Arnhem, Central Kimberley, Daly Basin, Darwin Coastal, Northern Kimberley, Ord Victoria Plain, Pine Creek and Victoria Bonaparte bioregions) with 10 terrestrial species (7%) and no marine species. These areas broadly coincide with the Endemic Bird Areas identified for Australia (Stattersfield et al. 1998).

Introduced Species

The relative abundance of introduced bird species is a measure of habitat alteration, which in turn has been related to changes in bird abundance and distribution. Large metropolitan areas are the main source of introduced species. The highest abundance of introduced species is found in southern Victoria, eastern Tasmania and parts of South Australia making up to 15% of all birds recorded (Figure 5.4).

Most are species introduced from overseas, such as the Rock Dove (Colombia livia), House Sparrow (Passer domesticus) and Common Starling (Sturnus vulgaris). However, there have also been introductions of Australian species, like the Long-billed Corella (Cacatua pastinator) that now flourishes in nearly every capital city due to aviary-releases in the 1980s. Eastern Tasmania has a particularly high loading of native introduced birds, as has south-western Australia, particularly near Perth. One new species, the Barbary Dove (Streptopelia risoria), has become established since the first Atlas.

Trends in Abundance

The national reporting rate of 169 species increased, that of 68 species decreased and 219 species showed no net change. This generally positive trend was found despite a tendency for many species to be seen less frequently using the methods of the second Atlas.

This is demonstrated by the observation that for about 1.5% of species, there were no records during the second Atlas in bioregions that had been estimated to contain at least 10% of their population in the first Atlas. All but 5 of these (0.03%), however, were vagrant or cryptic birds or in poorly surveyed bioregions, suggesting these species were more likely to have been missed than absent. This figure is balanced by the 1.2% of species where greater than 10% of their population in the second Atlas was in bioregions where they had not been recorded in the first. Overall, there were greater apparent changes in abundance of bird species between the two Atlas periods than changes in their distribution.

For much of arid and semi-arid Australia, the overall reporting rate of Australian breeding species was relatively stable or rose more than 20% (Figure 5.5). This pattern contrasts with decreased reporting rates in the south-east and south-west of mainland Australia, eastern Tasmania, and much of the Top End of the Northern Territory.

Trends among birds associated with different habitats varied greatly between guilds. Significant increases in reporting rate were recorded for rainforest, temperate forest, tropical and temperate woodland, and mangrove birds, although the reasons underlying this pattern are unclear. In contrast, significant decreases in reporting rate were detected for grassland birds and facultative woodland species.

Among grassland birds the greatest decline in reporting rate was in the south-east (Figure 5.6), despite relatively stable climatic conditions. There was a similar pattern of decline for woodland birds, with the greatest declines being where introduced birds are most prevalent. Overall, reporting rates of freshwater birds and migratory shorebirds also decreased.

Garnett (1992) identified two nesting guilds and two feeding guilds with higher than expected numbers of threatened species. Trends in reporting rates of these guilds were analysed here. Hollownesting species showed virtually no change in reporting rate, whereas the reporting rate of groundnesting species decreased, again particularly in southeastern Australia, but also in the Top End of the Northern Territory (Figure 5.7).

Obligate granivores were recorded at a higher rate in the second Atlas, particularly in semi-arid areas. However, ground-feeding insectivores decreased, again particularly in the south-east (Figure 5.8).

Factors that Contribute to Patterns in Bird Distribution and Abundance

Climate

At a continental level, both Atlas periods were wetter than average, with the second period wetter than the first. Only a few areas, particularly Tasmania and parts of Victoria, were drier in the second period (Figure 5.9). A number of species appear to have been advantaged by the wetter conditions, with fifteen species showing a significant increase in reporting rate with increasing rainfall.

The first two years of the first Atlas project coincided with an inland drought that had followed three years of exceptional inland rainfall. As a result, many mobile species, particularly waterbirds, had dispersed in large numbers as the inland dried out, thus inflating their numbers at the edges of the continent. A wet phase was underway during the second Atlas project, which did not include the subsequent drying of the inland. Twelve species showed a reporting trend that responded negatively to increasing rainfall.

For many of the remaining 468 species for which no statistical relationship with rainfall was evident, distributional patterns of change suggest both an increase in wetter inland bioregions, particularly the Channel Country, and a decline, possibly back to more normal densities, along the eastern margin of the continent. During the first Atlas, arid zone bioregions were dry compared to the wetter period that had preceded the second Atlas. During the second Atlas the arid zone was wetter than the period before it. Populations of these species appeared to have increased near the coast during the first Atlas, and in the centre in the second. For the more mobile species, these changes can be accounted for by movement between regions.

In contrast, populations of some of the more aridadapted species, such as Inland Dotterel (Charadrius astralis) and Gibberbird (Ashbyia lovensis), which may also have increased during the mid-1970s, had lower reporting rates during the second Atlas period in bioregions where those of more wide-ranging species increased.

For more sedentary species, such as Hooded Robin (Melanodryas cucullata) or Jacky Winter (Microeca fascinans), increased reporting rates inland and a decrease in the intensive use zone is likely to result from localised population changes, with the wetter climate producing favourable conditions in the inland, and more intensive land use closer to the coast being detrimental. Hence, the trends from the semiarid zone appear to mainly reflect climatic variation, but those in the intensive use zone are more likely to be responding to changes in landscape health.

The decreased reporting rates for many freshwater birds and migratory shorebirds may also have been driven by climate. However, specialised surveys suggest that the decline may also be related to longterm declines in the health of many catchments. The decline in reporting rate of shorebirds, despite increased attention to shorebirds in recent decades, may be indicative of long-terms trends within the flyway and warrants more detailed analysis.

Another possible climatic effect is a decrease in reporting rate of several species with distributions centred on the uplands of south-eastern Australia. These trends are counter-intuitive given the protection of most alpine habitat and the adaptation of species like Flame Robins (Petroica phoenicea) and Gang-gang Cockatoos (Callocephalon fimbriatum) to agricultural and suburban landscapes during winter. Their apparent decline, therefore, may have been caused by a general increase in minimum temperatures in south-eastern Australia, a trend most likely to first affect species from the highest altitudes.

Threatening Processes

In the agricultural regions of Australia, there is a range of responses to changes in type and intensity of land use. For 29 of the 495 species analysed, reporting rates decreased significantly with increased proportions of the landscape cleared. Overall, the clearing of land for agriculture appears to have had the greatest non-climatic influence on bird abundance in Australia. Some species have been lost as soon as habitat is destroyed, but others persist for decades in remnants. Species in the latter group will generally decline over time because of loss of habitat or lack of recruitment from other areas. These findings are consistent with the findings of Saunders & Ingram (1995) in south-western Australia and Robinson & Traill (1996) and Reid (1999) in southeastern Australia who recorded major changes in the bird faunas of Australia's main agricultural regions.

In the case of grassland birds, the gradual effects of increases in land use intensity, including fragmentation, agricultural intensification, urbanisation and, possibly, an increase in introduced predators are all likely to have played a role in their decline. Land use intensification is also the most likely reason for the decline of ground-feeding insectivores and ground-nesting species, at least in south-eastern Australia.

Twenty-six species showed a trend in reporting rate that responded negatively to decreases in landscape health (see Figure 5.10 for an example), a measure that includes abundance of feral animals and weeds as well as intensification of primary production. Sixty-one species had significantly lower reporting rates in the intensive land use zone compared to Australia's rangelands, with part of this relationship being explained by climatic differences between survey periods. Although the Atlas was not designed to determine trends in rare species, little correspondence was found between the areas of declining landscape health and the concentration of limited range species.

Two land use factors that vary within bioregions, and are therefore difficult to assess at a bioregional level, are rangeland grazing intensity and changed fire regimes. However, reporting rates of some grazing-sensitive species, such as Singing Bushlark (Mirafra javanica) (Figure 5.11) and Flock Bronzewing (Phaps histrionica) have declined, consistent with trends in grazing pressure on grasslands. Changes in fire regime (sometimes in association with intensification of pastoralism) have also been implicated in the decline of some species, particularly the granivores.

For a smaller suite of species, there was an increase in reporting rate with declining landscape health. Ten species increased with land clearance levels. Eleven species increased in regions with the worst continental stress class. Fifteen species also apparently fare better in the extensive use zone than in the intensive use zone. Many, like the Galah (Cacatua roseicapilla) had already expanded well into the agricultural areas at the time of the first Atlas (Blakers et al. 1984). By the time of the second Atlas, this movement towards the coast was complete (Figure 5.12). Other species, like Crested Pigeon (Ocyphaps lophotes) and Little Corella (Cacatua sanguinea), still seem to be taking advantage of the change from woodland to fields of granivore food, although for others the links are less readily explained.

Conclusions

The analysis of Bird Atlas data has identified some important patterns in distribution and abundance of bird species and the factors that drive their condition and trend. For birds that range across the vast arid centre of Australia, climate is a key driver of their movements and abundance.

Centres of diversity, range restriction and rarity, lie along the perimeter of Australia, particularly in the east and south of the continent. These coincide with the main concentration of people in Australia and regions of intensive land use. Birds are responsive to the threats associated with these areas.

Many bird species appear to be sensitive to declining landscape health caused by broad-scale vegetation clearance, urbanisation, intensified agriculture and feral predators.

Overall, the clearing of land for agriculture appears to have had the greatest non-climatic influence on bird species, which is reflected in the reduced abundance of a number of species in the agricultural regions of Australia. The most urgent actions identified by this and other studies are to end the clearing of native vegetation, impose more appropriate fire regimes (see also Woinarski & Recher 1997), control feral and native animals whose presence threaten native species, and restore functional ecosystems.

Actions are not only required in the more intensively used regions of Australia. Even though human settlements are sparse in areas such as the tropical and sub-tropical savannas of northern Australia, and the intensity of pastoralism is relatively low, altered fire and grazing regimes have major impacts on bird species (Franklin 1999). These land use factors vary within bioregions, and were therefore difficult to assess at a bioregional level in this study.

This analysis highlights a number of species that appear unable to tolerate landscape modification for agriculture, particularly clearing. More detailed analysis of the results of the two Atlases, particularly subdivision of guilds into increaser and decreaser species, and finer-scaled spatial analysis of individual species within core habitat, would further define the effects of threatening processes and lead to better management of our biodiversity.

• Next
• Contents
• Previous