Methods & Site Selection

Development of a tractable, improved method for counting shorebirds A protocol has been proposed which will establish the procedures and program for shorebird monitoring in Australia. This Activity has been recently completed as part of Birds Australia’s “Developing a Programme for Population Monitoring of Shorebirds in Australia” Project.

Below is a brief summary of the recommended protocol, see Haslem et al. (2008) for further and more detailed information.

Site Selection

Two hundred and thirty-four sites, termed “shorebird areas”, have been identified as areas of important habitat for both resident and migratory shorebirds in Australia (see Clemens et al. 2008). Many of these shorebird areas are associated with coastal tidal flats and beaches; others are located in non-coastal wetlands.

It is within these predefined shorebird areas that shorebird counts for ongoing population monitoring will primarily be undertaken. By targeting these areas, those known to be regularly used by large numbers of shorebirds, monitoring efforts will have the greatest chance of detecting population changes at a national level.

Defining Count Areas

Each shorebird area has been identified to include all areas of habitat, contiguous and non-contiguous, used regularly by the same group of shorebirds. Non-contiguous habitats have been combined in the same shorebird area when evidence (e.g. local shorebird expertise, patterns in shorebird count data) suggests that birds exhibit regular movements between them. Thus, multiple roosting and feeding areas may be included within the same shorebird area. This approach has been taken with the objective of ensuring that, as far as possible, each shorebird area represents a closed shorebird population.

Depending on how counts are traditionally undertaken in different shorebird areas, each has been broken up into one or more “count areas”. In many shorebird areas, count areas represent known high tide roosts. Separate shorebird counts should be undertaken within each of these count areas. Data collected in each individual count area will then be aggregated to give the total number of birds observed in the entire shorebird area. To reduce the amount of geographic bias in current shorebird sampling, a further 330 count areas located in wetland or coastal habitats have been randomly selected. Selection of these additional count areas has been stratified across 11 biogeographic regions encompassing all of Australia. Thus, these randomly-selected count areas increase the geographic representativeness of survey coverage. They should help reduce any sampling bias introduced by changes in shorebird distributions due to the abundance of ephemeral wetlands in Australia.

Count Frequency

At least two shorebird counts should be planned over the summer months and a third count in winter should be planned in areas with historic winter data, or where Double-banded Plover are known to occur. January or early February is the preferred time for undertaking the summer count in most shorebird areas.

July is the preferred month in which to undertake winter counts at shorebird areas. For each of these three counts, a national count date will be recommended, but it is understood that local factors (i.e. variations in tides) may prevent all areas being surveyed on the same date. Therefore, it is recommended that counts be undertaken as close to the national count date as possible. In shorebird areas near the coast, counts should generally be conducted in the hours either side of high tide (shorebirds are often more tolerant of observers on a falling tide).

Helping an organised approach to data collection

When undertaking population monitoring counts in shorebird areas, counters are encouraged to use the newly-updated datasheet. One datasheet should be completed for each count area. This datasheet prompts counters to record all the important details relating to each visit (such as location, date, weather conditions and observer contacts), as well as the number shorebird individuals/species observed during the count. Additional information on changes in habitat condition, disturbance and apparent threats to shorebirds, observed at shorebird areas is also requested. Such information will allow more detailed investigations into the cause of population changes at shorebird areas over time. These scores only need be estimated once a year.

Summary

Population Monitoring Protocol

A protocol is proposed which will establish the procedures and program for shorebird monitoring in Australia. Populations of many shorebird species are declining throughout the world (International Wader Study Group 2003, Committee for Holarctic Shorebird Monitoring 2004, van de Kam et al. 2004), and here in Australia (Barter 1992, Creed and Bailey 1998, Wilson 2001, Minton et al. 2002, Reid and Park 2003, Olsen and Weston 2004, Gosbell and Grear 2005, Gosbell and Clemens 2006).

The extensive loss of key stop-over habitats in the flyways of migratory shorebirds has been identified as a key threat to many species (Milton et al. 2005). Shorebird monitoring in the Australian over-wintering habitats of migrants provides valuable information with which to assess the status of these species.

Australian non-breeding grounds are ideal for monitoring migratory species as, unlike in their breeding grounds, these shorebirds congregate together and their populations are likely to be more stable than in staging areas. Further, shorebird monitoring is essential for understanding how local habitat changes are affecting numbers of both migratory and resident shorebirds.

Australia is a signatory to several international agreements on shorebird conservation (JAMBA, CAMBA & ROKAMBA) and both migratory and threatened resident species are federally listed under the Environmental Protection and Biodiversity Conservation Act (1999). For successful conservation management of shorebirds, it is critical that shortand long-term trends in population numbers are accurately identified. Thus, the following protocol has been developed to guide the monitoring of migrant and resident shorebird populations in Australia in a scientifically rigorous manner.

It is understood that the program under this protocol will only be able to at best identify changes in the populations of shorebirds in or entering Australia; and further study in the breeding range would verify declines seen here.

Where to count shorebirds

Two hundred and thirty-four sites, termed “shorebird areas”, have been identified as areas of important habitat for both resident and migratory shorebirds in Australia (Clemens et al. 2008). Many of these shorebird areas are associated with coastal tidal flats and beaches; others are located in non-coastal wetlands. It is within these pre-defined shorebird areas that shorebird counts for ongoing population monitoring will primarily be undertaken. By targeting these areas, those known to be regularly used by large numbers of shorebirds, monitoring efforts will have the greatest chance of detecting population changes at a national level. Each shorebird area has been identified to include all areas of habitat, contiguous and noncontiguous, used regularly by the same group of shorebirds.

Non-contiguous habitats have been combined in the same shorebird area when evidence (e.g. local shorebird expertise, patterns in shorebird count data) suggests that birds exhibit regular movements between them. Thus, multiple roosting and feeding areas may be included within the same shorebird area. This approach has been taken with the objective of ensuring that, as far as possible, each shorebird area represents a closed shorebird population. Depending on how counts are traditionally undertaken in different shorebird areas, each has been broken up into one or more “count areas”. In many shorebird areas, count areas represent known high tide roosts. Separate shorebird counts should be undertaken within each of these count areas.

Data collected in each individual count area will then be aggregated to give the total number of birds observed in the entire shorebird area. In most cases, it is this aggregated count data that will be used in further analyses. However, counting shorebirds in these smaller spatial units is valuable because it ensures that counts are undertaken in the same location every year. This spatially explicit data enables tracking of changes in abundance at a finer scale which can allow us to identify local impacts more easily. Further, data collected at the level of “count areas” allows corrections to be made when the total shorebird area coverage varies.

From the full number of shorebird areas (n = 234) located across the country, 150 have been identified as being of high priority for shorebird monitoring (Figure 2). Collectively, these 150 shorebird areas form some of the most abundant areas for a range of resident and migrant species. Many have already been regularly monitored for a number of years by organised teams of shorebird counters. For others, more effort will be required in order to facilitate/reinitiate regular shorebird counts.

The ability to successfully detect national trends in shorebird populations is dependant on regular monitoring in as many of these areas as possible. To guard against bias in sampling, a further 330 count areas located in wetland or coastal habitats have been randomly selected. Selection of these additional count areas has been stratified across 11 biogeographic regions encompassing all of Australia. Thus, these randomly-selected count areas increase the geographic representativeness of survey coverage. For example, they should help reduce any sampling bias introduced by changes in shorebird distributions due to the abundance of ephemeral wetlands in Australia.

Figure 2. Location of 150 priority shorebird areas for ongoing population monitoring across Australia.

When to count shorebirds

Migratory species spend the summer months (October – March) in Australia and so summer counts are essential for recording visiting migrants. January or early February is the preferred time for undertaking the summer count in most shorebird areas. We recommend planning a second count date in summer at each shorebird area to provide a back-up count if conditions are not suitable during the first count. If conditions are suitable on the first count, and the number of shorebirds observed is thought to be a fair reflection of the populations using the area, the second count will not be needed for national monitoring.

Winter counts are important for recording Double-banded Plover (a winter migrant to Australia), and overwintering shorebirds. However, until an assessment of the value of these winter counts is more complete, priority should be given to areas which have historically been counted in winter and areas where Double-banded Plover is recorded. July is the preferred month in which to undertake winter counts at shorebird areas. For each of these three counts a national count date will be recommended, but it is understood that local factors (i.e. variations in tides) will not permit all areas to be surveyed on the same date. Therefore, it is recommended that all counts are done as close to national count dates as possible.

In shorebird areas near the coast, counts should generally be conducted in the hours either side of high tide (shorebirds are often more tolerant of observers on a falling tide). Where possible, undertaking additional counts in these shorebird areas is recommended.

Shorebirds are mobile species that often show frequent movements within shorebird areas and are also known to move between different areas. For this reason, undertaking more than one count per season will increase the potential for counts to most-accurately represent the number of shorebirds using a given area. This can be especially valuable when trying to detect trends at the level of each shorebird area, or to accurately reflect the larger numbers of shorebirds that may use an area outside the summer months. Further, data collected on a greater number of occasions throughout the year will provide a better understanding of the use of shorebird areas by different species, as it has been shown the maximum number of shorebird species using an area is only reflected if multiple counts are conducted throughout the year (Alcorn 1994).

Such information will result in a clearer picture of the importance of these areas to shorebirds. How to count shorebirds When undertaking population monitoring counts in shorebird areas, counters are encouraged to use the newly-updated datasheet (Oldland et al. 2008). One datasheet should be completed for each count area. This datasheet prompts counters to record all the important details relating to each visit (such as location, date, weather conditions and observer contacts), as well as the number shorebird individuals/species observed during the count. All these data have been found to be the minimum required to meet the programs objectives, and the more observers that adopt these new sheets, the more time it will save when processing and storing data.

Additional information on changes in habitat condition, disturbance and apparent threats to shorebirds, observed at shorebird areas is also requested. Such information will allow more detailed investigations into the cause of population changes at shorebird areas over time. These scores only need be estimated once a year.

Finally, recording ‘component counts’ that identify the size of separate groups / individual birds observed is valuable as it provides a way to estimate the error associated with counting large numbers of birds. This information is particularly useful when collected at shorebird areas where many birds are observed. Why follow this monitoring protocol? Monitoring a large number of shorebird areas is important to successfully identify shorebird population trends in Australia.

Collecting regular count data at these priority shorebird areas will result in adequate data to detect national population changes of the magnitude of interest (i.e. 50% decline in 5 years) in most shorebird species. For example, count data will need to be collected at around 40 shorebird areas in order to detect a 30% decline, over 10 years, in the national Great Knot population, should it be occurring. Such ability to detect changes in shorebird populations is a vital component of successfully managing for their conservation.

Further, the additional information being requested here is all potentially valuable in either increasing chances of detecting trends by identifying and accounting for other variables, or beginning to understand what kinds of things are impacting shorebird numbers throughout the country.

Finally, the detection of trends for species within any one shorebird area will often require more rigorous monitoring as previously discussed than what is being recommended to detect national changes.