Effectiveness of the protected areas on the Mornington Peninsula for the common resident shorebird species using citizen science data

In the absence of comprehensive survey data this study used citizen science bird counts, extracted from the Atlas of Living Australia, to assess which species benefit most from protected areas. This was done by fitting temporal models using the Integrated Laplace Approximation (INLA) method. The trends for five resident shorebird species were compared to the Australian Pied Oystercatcher, with significantly steeper upward trends identified for the Black-fronted Dotterel, Red-capped Dotterel and Red-kneed Dotterel. Steeper upward trends were observed in protected than unprotected areas for the Black-fronted Dotterel, Masked Lapwing and Red-kneed Dotterel. This work suggests that, with some limitations, statistical models can be used with citizen science data for monitoring the persistence of resident shorebirds and for investigating factors that are impacting these data. The results for the Dotterel species in protected areas are particularly encouraging.

shows the protected areas of the Mornington Peninsula that contribute most to 88 biodiversity conservation based on an analysis of remnant native vegetation cover and quality, 89 landscape context and threatened species habitat obtained from Parks Victoria. These areas will 90 benefit the most from actions that are targeted to enhance biodiversity (Schmidt, 2019). As 91 proposed by Diamond (1975), birds are better protected if reserves are larger, round or square 92 rather than long and thin, with multiple reserves linked by steppingstones or corridors, and 93 containing at least one large reserve to ensure that local extinctions in small reserves can be 94 replenished. The network of reserves shown in Figure 1 exhibits some of these characteristics 95 in some areas. Based on the above it is therefore hypothesised that reductions in the number of 96 shorebirds counts will be greater outside than inside these protected areas. Citizen science data plays an increasingly important role in the conservation domain.

103
Continually updated, reliable and comparable biodiversity data is necessary to implement 104 international conservation policy (Schmeller et al., 2015). Citizen science involving non-105 professionals and professionals as contributors can provide an intensive source of species 106 observation data. In traditional citizen science studies, citizens are trained to collect samples 107 using a standardised method to ensure high data quality. For instance, as a consequence of the 108 continuous, long running Western Port survey, the significance of Western Port for birds has 109 become widely recognised. As a result, it has become a key monitoring site for Palaearctic and

118
The data presented in this study were obtained from a validated citizen science database

153
The key advantages of INLA are the ease with which complex models can be created and 154 modified, without the need to write complex code, and the speed at which inference can be 155 done, even for temporal problems with hundreds of thousands of observations (Sarul, 2015). 156 In this study we have used ALA citizen science data and INLA modelling in order to investigate 157 the health of resident shorebird species on the Mornington Peninsula. In this study, we 158 investigate the effects of climate and conservation reserves on shorebird abundance for seven 159 shorebird species. It is likely that the effects of these variables will be different for each Agriculture, Water and the Environment, 2020). 165 We have divided our analysis into two main sections. First, we tested whether there are 166 significant differences in the trends for different species, to determine which species are 167 more/less vulnerable. This is done using the entire data set using the species variable with seven 168 categories. Then, we tested the effects of protected areas for each of the species. This is done 169 for each species using a binary variable to indicate the level of protection for the location of  The Mornington Peninsula is a unique place when it comes to biodiversity. It is home to a wide 181 range of plants and animals, including species of regional, state, national and international   As before allows for a temporal random effect with first-order autoregressive dependence 244 and the random error term is represented by . All these models were implemented using the

249
Descriptive Statistics 250 Table 1 shows that the Atlas of Living Australia data was mainly sourced from eBird together    Red-kneed Dotterel with estimated growth rates of 19%, 9% and 32% per annum. This 265 significance is indicated by the credibility intervals for their 1 estimates which do not contain 266 zero. Figure 3 illustrates the above trends on a logarithmic scale.   Table 3 show that the protected areas are particularly 276 important for the resident shorebirds including locally threatened Sooty Oystercatcher.

277
However, although the bird counts for the Black-fronted, Red-capped and Red-kneed Dotterels 278 were initially low, the results show that the latest trend is starting to reverse this effect with 279 significantly greater growth seen in protected areas. Figure 4 illustrates these upward trends 280 while also emphasising the obvious growth in Masked Lapwing numbers.

343
The use of unstructured citizen science data in this study is an obvious limitation. As discussed 344 above, the nature of these data and their focus on threatened species makes some degree of bias 345 inevitable. In particular, the ALA growth rates reported in this paper must be treated with 346 caution because to some extent they reflect the increased effort made by citizen scientists.

347
Another limitation of this study is that the analysis uses only annual data. We have used annual 348 shorebird sightings in order to avoid seasonal bias in the data and to ensure adequate sample 349 sizes for the INLA modelling, however, the shortness of the time series considered (only 10 350 years) makes this another limitation of our analysis.

352
Temporal models fitted using INLA provide flexible and useful frameworks for modelling 353 count data using statistical models. While citizen science data has its problems, especially for