Lauderdale

Tasmania, Australia

-42.912673, 147.481715

November 2008

Modelling

MORPH

Abstract

Reproduced with permission.

The Walker Corporation Pty Ltd has proposed the development of a marina village at Lauderdale in southern Tasmania (i.e. the Project). The proposed development site includes a large area of tidal flats (70 ha) that is used by a wide range of migratory waders and a significant proportion of the world population of Pied Oystercatchers. Lauderdale forms part of a complex of lagoons and bays situated within the Derwent/Pitt Water region that collectively support approximately 8% of the estimated global population of 110,000 Pied Oystercatchers. Averages of 2.6% and 2.4% of the global population of Pied Oystercatchers occur at Lauderdale in summer and winter respectively (Aquenal 2008b), satisfying Criterion 6 of the Ramsar Convention of 1% of the global population being present at a site. Within the Derwent/Pitt Water region, the South Arm Neck site in Ralphs Bay also exceeded this 1% criterion. The effect of loss of bird habitat (foraging, roosting and breeding) has been identified as an issue that needs to be addressed as part of the Integrated Impact Statement (IIS) for the Project.

The current study assessed one aspect of habitat loss by using a behaviour-based model to determine the likely impact of the proposed development on food resources available to Pied Oysterctachers. This report details the modelling studies and background analyses needed to parameterise the model. A total of seven sites located in south eastern Tasmania were included in the model, and were selected on the basis of supporting large numbers of Pied Oystercatchers and having potential links to the proposed development site. Two of the sites, Orielton Lagoon and Barilla Bay, fall within the existing Pitt Water/Orielton Lagoon Ramsar site. The seven sites included a total of 34 ‘patches’ identified for the model and had a total area of 616 ha. Field data required as input to the model were collected during spring, summer and winter seasons to correspond with important life stages of wader species and hence variation in food availability. Data collected related to bird numbers, invertebrate prey species, and bird foraging activity, while information on sandflat exposure, energy demands by the birds and the energy content of their food were also input to the model.

Surveys by Harrison (2008) provided foraging data input to the model, and identified 20 prey species consumed by Pied Oystercatchers during full surveys. Polychaete worms and two species of bivalve (Katelysia scalarina and Anapella cycladea) accounted for approximately 90% of prey consumed in numeric terms. Inclusion of additional less common benthic infauna bivalves, as well as the gastropod Salinator fragilis and epibenthic bivalves Crassostrea gigas and Mytilus galloprovincialis, meant than data input to the model accounted for 98% of food items. The largest single numerical contributor to the diet was polychaete worms, which accounted for more than 40% of prey during each season.

The benthic invertebrate survey (Aquenal 2008a) provided the densities and sizes of the invertebrate prey available and targeted the species identified as important components of the Pied Oystercatcher diet from the above foraging studies. Energy intake by the birds was determined by calculating the relationship between the length of the organism and the ash-free dry mass (AFDM) for eight species (four bivalves, three polychaetes and one gastropod) for the seven sites and three seasons surveyed. These relationships were applied to the invertebrates observed to have been consumed by Pied Oystercatchers in the foraging surveys and were converted into the ‘common currency’ of AFDM. Analyses found that bivalves comprised on average 88% (range 70-95%) of the energy intake expressed as AFDM across all sites.

Birds at Lauderdale had the highest intake rates (c. 6g AFDM per hour) and, in terms of biomass (expressed as AFDM) also consumed the highest proportion of bivalves. In terms of foraging, Lauderdale was therefore the highest quality site surveyed.

In the absence of habitat loss (i.e. no development) the model predicted that there was a seasonal minimum of four times the amount of food needed to sustain the current number of birds at Lauderdale (mean across seasons of 189 and a maximum of 252 birds). This minimum occurred during the winter season when the quality of food was lowest. European Oystercatchers require between two to eight times the amount of food required to maintain low mortality rates, depending on the dominant prey types. At Lauderdale and surrounding sites, where birds feed largely on dispersed rather than aggregated bivalve species, the birds would suffer little intra-specific competition and we initially expected that the amount of food required in the environment would be closer to two, than eight times the amount the birds needed to consume.

After removing the exact area that is covered by the proposed development, the model predicted no deaths or loss of condition of Pied Oystercatchers on the basis of reduced food availability. To ensure removal of any potential bias in the data, a sensitivity analysis was applied by increasing the number of birds to the seasonal maximum and not allowing birds to wade for food, however the same result was achieved. This scenario assumes that the development does not (a) impact on the sediments (and therefore food resources) to the south of the proposed development or (b) impact on the behaviour of the Pied Oystercatchers in the remaining area. These predictions do not take into account the presence of other waders but we expect the interaction between these species and Pied Oystercatcher to be relatively low for two reasons. First, there were relatively low numbers of other wader species present at the sites. Second, other wader species will not depend on bivalves to the same extent as the oystercatchers. As polychaetes only made up 11% of the Pied Oystercatcher diet (in terms of AFDM) overall, there was only a small amount of overlap in their diets.

To allow for the scenario of the proposed development having an adverse impact on areas to the south of its ‘footprint’, additional model simulations were conducted whereby all of the habitat was removed from Lauderdale and birds were forced to move elsewhere. On the basis of food resources, birds were predicted to move to South Arm Neck and Pipeclay Lagoon and, due to the large amounts of food available at those sites, no increase in mortality, or loss of condition, were predicted. The conservative approach of increasing the numbers of birds in the model to their seasonal maximum in each site, and not allowing them to wade for food, also resulted in no increased mortality or decrease in body condition.

The above default habitat loss simulations also assumed that the Pied Oystercatchers were the only source of prey mortality, whereas other factors may also be important. Additionally, the invertebrate survey may have overestimated the amount of food available to the birds. To account for these possibilities, a further sensitivity analysis was conducted by re-running the model, retaining the above assumptions about the seasonal maximum number of birds present and an absence of wading, but also assuming that only 50% of the observed food supply was actually available to the birds. Given the measured error (5th percentile of the total AFDM in the site) in the invertebrate survey varied seasonally between 31 and 43% of the total AFDM actually measured, and the fact that the preferred prey populations were relatively stable through the year (implying that mortality rates of these prey were relatively low), a 50% reduction of prey density was considered relatively extreme. The combined assumptions of this sensitivity analysis therefore provided a highly conservative assessment, with model output likely to reflect a worst case impact on the basis of reduction of food resource availability alone. For the partial habitat loss scenario (development area only), no increase in mortality or loss of condition were predicted with the 50% reduction in food. However when the other sectors to the south of the proposed development area at Lauderdale were made unsuitable, the model run predicted high mortality rates of 55% on average per year. In this scenario, it was therefore predicted that adjacent sites would not be able to support all birds displaced and mortality would occur.

The model predicted that the current foraging resources at Lauderdale could support a potential maximum of up to 1,000 Pied Oystercatchers, with the area directly affected by the development’s footprint supporting up to 200 individual Pied Oystercatchers. Making all of Lauderdale unavailable would therefore reduce capacity for 1,000 birds from the network of sites considered. It is unlikely that the development’s impact would be limited to its immediate footprint and it is likely that birds would be forced to move to other sites. Given the current situation (i.e. current Pied Oystercatcher population level and food supplies), the most likely scenario is that there would be sufficient food available to these birds at other sites. If all the habitats at Lauderdale were made unsuitable by the proposed development then, again, the most likely scenario is that there would be sufficient food. It is only in the worst case scenario that mortality dramatically increases.

Based on the current surveys, food does not therefore seem to be the factor limiting Pied Oystercatcher population numbers in south eastern Tasmania. The model indicates that sites such as Lauderdale, South Arm Neck and Pipeclay Lagoon are key sites for maintaining high survival for Pied Oystercatchers in the region. Data obtained by Birds Tasmania (see Aquenal 2008b) show no evidence of significant declines in the population.  Based on 1983-2005 biannual wader count data, the Pied Oystercatcher population in the Derwent/Pitt Water region increased in the 1990s although there was a greater variability in numbers in the later years. Although there was an excess of food, if the oystercatcher population undergoes a similar increase in future then food, particularly that available in winter, may become a limiting factor.

Funding and Collaboration

British Trust for Ornithology (originally from an organisation in Australia)

Related Paper:

Stillman, R.A. & Wood, K.A. (2013). Predicting food requirements of overwintering shorebird populations on the Solway Firth. A report to Scottish Natural Heritage and Marine Scotland. Bournemouth University, Poole. 37 pp.