The authors have declared that no competing interests exist.
Conservation areas are critical for biodiversity conservation, but few citizen science studies have evaluated their efficiency. In the absence of thorough survey data, this study assessed which species benefit most from conservation areas using citizen science bird counts extracted from the Atlas of Living Australia. This was accomplished by fitting temporal models using citizen science data taken from ALA for the years 2010–2019 using the INLA approach. The trends for six resident shorebird species were compared to those for the Australian Pied Oystercatcher, with the Black-fronted Dotterel, Red-capped Dotterel, and Red-kneed Dotterel exhibiting significantly steeper increasing trends. For the Black-fronted Dotterel, Masked Lapwing, and Red-kneed Dotterel, steeper rising trends were recorded in conservation areas than in other locations. The Dotterel species’ conservation status is extremely favourable. This study demonstrates that, with some limits, statistical models can be used to track the persistence of resident shorebirds and to investigate the factors affecting these data.
Shorebirds account for around 10% of Australia’s bird species. The majority of Australian shorebirds are declining due to risk factors such as habitat degradation and human activity [
Tidal flats in Australia have been damaged by or lost to reclamation, altered water regimes, pollution, sea-level rise and weed invasion [
Bird Observation and Conservation Australia (now BirdLife Australia) [
For these reasons, we examined changes in the resident shorebird population on Mornington Peninsula, which is located near Melbourne, Australia. From the mainland to the Bass Strait, this peninsula is bordered on the east by Western Port Bay and on the west by Port Phillip Bay. As explained above, these species exhibit a range of vulnerabilities, with the establishment of conservation zones proving to be an effective strategy of mitigating some of these vulnerabilities. We hypothesised that seven common resident Mornington Peninsula shorebird species which all nest on the ground, Red-capped Plover (
Conservation areas are a key strategy for preserving biodiversity, but few citizen science studies have evaluated the effectiveness of this strategy. Citizen science data from the eBird platform [
A major contributory factor to shorebird decline is habitat loss [
The conservation areas of Mornington Peninsula Shire was obtained from the Mornington Peninsula Shire Biodiversity Conservation Plan 2019 [
Citizen science data plays an increasingly important role in the conservation domain. Continually updated, reliable and comparable biodiversity data is necessary to implement international conservation policy [
In addition, there has been increasing citizen science effort in recent years which means that trends in absolute numbers of sightings must be treated with caution. The Annual Report 2018–19 for the Port Phillip and Western Port Catchment Management Authority provides an example of the power of citizen science for monitoring the persistence of birdlife in the vicinity of Melbourne, Australia. Between the 2005–06 and 2016 analyses the number of wildlife sightings increased from 437,845 to over 3 million dues to a proliferation of citizen science survey programs. The large amount of data provided by citizen scientists means that more advanced statistical modelling techniques can be applied [
The data presented in this study were obtained from Atlas of Living Australia (ALA) which is a validated citizen science database [
The development of species distribution models has benefited in the protection of biodiversity by connecting science to policy and decision-making processes. These models have evolved to generate future landscape scenarios based on known and anticipated environmental conditions. However, spatial, or temporal scales can confound inference about changes in species observation data when used to draw conclusions about potential impacts on a different scale. Our data is confined to a relatively small spatial area over a short period of time (2010–2019), and for this reason we only consider temporal models. Annual bird counts are assumed to follow the Poisson distribution [
The Integrated Laplace Approximation (INLA) method is an approximation tool for fitting Bayesian models for species abundance. INLA is a more resilient alternative to the Markov Chain Monte Carlo (MCMC) based Bayesian analysis technique [
The purpose of this study was to examine the health of resident shorebird species on the Mornington Peninsula using ALA citizen science data and INLA modelling. We examine the effect of conservation reserves on shorebird abundance for seven shorebird species in this study. It is anticipated that the consequences of these variables will vary according to the vulnerability of each shorebird species. Our analysis is separated into two major components. First, we tested whether there are significant differences in the trends for different species, to determine which species are more/less vulnerable. This is done using the entire data set using the species variable with seven categories. Then, we tested the effects of conservation areas for each of the species. This is done for each species using a binary variable to indicate the level of protection for the location of each recorded sighting.
Records of annual data for resident shorebirds for the Mornington Peninsula from 2010 to 2019 were obtained from the Atlas of Living Australia database. The Atlas of Living Australia database gathers basic data on bird abundance and distribution at a variety of temporal scales. Much of this geocoded observational data were gathered during systematic surveys made by trained volunteers or qualified biologists [
To obtain only the validated data, we have filtered the data as follows. Excluded spatially suspect records, records based on scientific name quality, records with additional spatial quality issues, duplicate records, records based on location uncertainty, records with unresolved user annotations, records that are environmental outliers, records based on record type and records pre-1700. We have included present-only records from eBird Australia [
The resident shorebird species included in this study are: Black-fronted Dotterel, Masked Lapwing, Red-capped Dotterel, Red-kneed Dotterel, Australian Pied Oystercatcher and Sooty Oystercatcher. In this study, considered the total number of bird sightings recorded in each year for each of the above seven resident shorebird species, separately for conservation and other areas.
The Mornington Peninsula is a unique place when it comes to biodiversity. It is home to a wide range of plants and animals, including species of regional, state, national and international significance. The Mornington Peninsula is built of complex geological formations, resulting in diverse landforms and habitat types. The major habitat types on the Peninsula include central hills, waterways, wetlands, north central plains, sandy beaches and dunes, cliffs and headlands, rocky shores, mudflats, saltmarsh, mangrove swamps and estuaries. Approximately 10% of the peninsula land is protected within parks or reserves, including a national park, a state park, state conservation reserves, local bushland reserves and foreshore coastal reserves. These remnant areas of bushland provide an important refuge for the diverse range of plants and animals on peninsula.
The Mornington Peninsula forms part of the Western Port Biosphere Reserve (WPBR), which covers five Local Government Areas around Western Port Bay and carries out projects to test ways of balancing conservation with development. The Western Port wetlands are included in the Ramsar List of Wetlands of International Importance and are the primary reason why the United Nations Education, Scientific and Cultural Organisation (UNESCO) declared the Western Port catchment as one of only four active biosphere reserves in Australia, and one of only 701 in the world [
The Mornington Peninsula National Park is the largest reserve on the peninsula with inland and coastal components. Mornington Peninsula National Park is the most visited National Park in Victoria, with intensively used recreation nodes at Portsea, Sorrento, and Cape Schanck [
We fitted temporal models for the number of annual sightings recorded for the above resident shorebirds on the Mornington Peninsula between 2010–2019. A Bayesian hierarchical modelling approach was used to conveniently account for parameter uncertainty and potential temporal dependence. We have considered temporal models with Poisson and Negative Binomial distributions [
For a categorical variable, such as species or level of protection, the general hierarchical model can be described as follows, where
We include the categorical variables in the model by way of dummy variables (υ), with one of the categories chosen as the reference category. The interaction between year and these dummy variables tells us about the difference in slope between these categories and the reference category and allows a test of significant.
For species
Depending on whether or not a sighting occurred in a conservation area the general link function for the expected annual total shorebird counts is given below with
As before
Species | BirdLife | eBird | Victorian Biodiversity Atlas | Total counts | Estimated growth per annum (%) |
% Sightings in conservation areas |
---|---|---|---|---|---|---|
3 | 84 | 103 | 190 | 4.7 | 10% | |
110 | 519 | 15 | 644 | 26.6 | 54% | |
1715 | 2328 | 213 | 4256 | 9.2 | 58% | |
8 | 129 | 137 | 14.6 | 18% | ||
8 | 279 | 287 | 44.5 | 88% | ||
293 | 1632 | 1 | 1926 | 6.4 | 87% |
Coefficients | Mean | SD | 95% Credibility Interval | |
---|---|---|---|---|
0.025quant | 0.975quant | |||
2.701 | 0.293 | 2.147 | 3.32 | |
-0.03 | 0.203 | -0.424 | 0.371 | |
0.04 | 0.044 | -0.053 | 0.124 | |
0.042 | 0.025 | -0.008 | 0.091 | |
0.015 | 0.026 | -0.036 | 0.066 |
There are many ongoing conservation programs on the Mornington Peninsula. Our results in
Species | Coefficients | Mean | SD | 0.025quant | 0.975quant |
---|---|---|---|---|---|
Intercept | 1.589 | 1.034 | -0.529 | 3.676 | |
year | 0.107 | 0.16 | -0.22 | 0.433 | |
conservation area | -3.519 | 2.9 | -9.771 | 1.612 | |
year: conservation area | 0.192 | 0.325 | -0.392 | 0.885 | |
Intercept | 2.613 | 0.808 | 1.151 | 4.452 | |
year | 0.078 | 0.121 | -0.199 | 0.296 | |
Intercept | 4.841 | 0.286 | 4.35 | 5.517 | |
year | 0.04 | 0.042 | -0.056 | 0.117 | |
conservation area | 0.111 | 0.073 | -0.032 | 0.254 | |
Intercept | 1.786 | 0.677 | 0.397 | 3.102 | |
year | 0.121 | 0.092 | -0.068 | 0.301 | |
year: conservation area | 0.036 | 0.117 | -0.187 | 0.272 | |
Intercept | 2.918 | 3.517 | -4.796 | 9.547 | |
year | -0.405 | 0.484 | -1.375 | 0.602 | |
Intercept | 2.991 | 0.25 | 2.452 | 3.448 | |
year | 0.041 | 0.039 | -0.031 | 0.125 | |
year: conservation area | 0.03 | 0.023 | -0.017 | 0.076 |
The purpose of this study was to examine the effect of species and conservation areas on the numbers of the Mornington Peninsula’s common locally nesting shorebird species. This was accomplished through the use of data from the Atlas of Living Australia and Poisson temporal models fitted using the INLA approach. We were interested in determining whether the rate at which reported numbers change over time varies by species and protection areas. Our findings pose several critical conservation challenges.
While the Australian Pied Oystercatcher population has expanded dramatically in Western Port over the last 40 years [
Declines in population were reported for both the Masked Lapwing and the Red-capped Dotterel up to 2009 [
Coastal environments are the most vulnerable for resident shorebirds on the Peninsula, particularly for shorebirds with limited distributions such as the Sooty Oystercatcher. We discovered a strong favourable effect for the locally threatened Sooty Oystercatcher in our conservation area models. The increased numbers of Sooty Oystercatchers may be explained by their locally threatened status, which necessitates regular surveys and the establishment of conservation areas.
In addition, this study has found that the percentage increase for the numbers of Black-fronted Dotterel, Red-kneed Dotterel and Red-capped Dotterel has been stronger in conservation areas than other areas on Mornington Peninsula since 2010, and the same has been true for the Masked Lapwing. Contributing factor are likely to include the control of introduced predators such as Red Fox (
Among other habitats, the Australian Pied Oystercatcher, Masked Lapwing, and Red-capped Plover make extensive use of tidal mudflats. Previously, only peripheral data existed on species that frequent freshwater habitats (Black-fronted and Red-kneed Dotterels), as well as rocky or sandy coastlines (Sooty Oystercatcher). The majority of conservation areas on the Mornington Peninsula that are important to shorebirds are located on picturesque beaches, tidal mudflats, and reserves with freshwater habitats and sandy coasts. This habitat disparity between the Mornington Peninsula and previously published research could account for the majority of the changes mentioned above. The ALA sighting data used in this study do not corroborate previously reported positive trends for the Australian Pied Oystercatcher by [
Citizen science data is recommended for large geographic area studies, such as Australian shorebird analysis, and seasonal data, such as pest eradication analysis, but is less reliable for small geographic area studies, such as Mornington Peninsula analysis. Through citizen science databases, observer data is made publicly available with the collaboration of the observers. The majority of sensitive data and structured survey data are not accessible to the general public. As a result, our study utilised validated citizen science ALA data. However, there are legitimate worries about the detection bias inherent in citizen science data [
Along with the nearby Edithvale-Seaford Wetlands, the Eastern treatment plant is part of the Carrum Wetlands Important Bird Area (IBA) and supports many bird species of regional, state, national and international conservation significance. This has not included as a conservation area and may be affect for the species performance. We have not included this Eastern treatment plan which is also a limitation of this study.
Temporal models fitted with INLA provide versatile and useful frameworks for statistically modelling count data. While citizen science data has limitations, particularly when comparing relative citizen science bird counts of various species, there are evident benefits to using citizen science data to monitor endangered species citizen science counts. Citizen science data are often of higher quality for these species, at least in terms of survey data obtained in conservation areas, and it would be difficult to obtain sufficient data from other sources. However, increasing counts of other birds in conservation areas, such as Dotterels, indicates an increase in citizen science effort for less endangered species, which bodes well for future conservation efforts.
According to the findings of this study, local shorebirds benefit from the presence of conservation areas. However, it appears improbable that conservation areas would be established at many of the sites with high levels of human activity, suggesting that better conservation outcomes may result from achieving desirable shorebird abundance targets. It has been suggested by the findings of this study that, in the future, statistical models may be useful for addressing challenges associated with monitoring the persistence of resident shorebirds and for exploring factors that have an impact on these data.
(CSV)
(CSV)
(CSV)
(R)
Port Phillip and Westernport Catchment Management Authority (PPWCMA) for their interest in this project. In addition, we would like to thank Parks Victoria to provide us the conservation areas in Mornington Peninsula and the citizen scientists who collected the data and the Atlas of Living Australia for making the data available.