CCAMLR

Broad-scale survey of the abundance of colonial breeding penguins requires locating all, or the great majority, of colonies as the first of several survey stages. Given the remoteness of Antarctica and the sub-Antarctic islands, satellites offer obvious potential for such a task in this region. Past evaluations of the utility of satellites for the detection of penguin breeding sites are reviewed. Despite the obvious potential for such use, very few evaluation studies have been undertaken. The studies indicate great potential, but also caution on or allude to the need for further evaluation or consideration with respect to the following issues: spectral response of surrounding material, variability in the spectral response of guano due to environmental features, inadequate or ambiguous signal from guano, and spatial resolution of the technology and penguin breeding sites. Developments in satellite technology since the time of the studies will have alleviated some issues such as spatial resolution. Some directions for further evaluation work, and possible survey design options for addressing deficiencies in current satellite technology, are discussed. Some specifications of current satellite sensors that may be useful for this purpose are given.

WG-EMM is currently investigating the feasibility of undertaking broad-scale surveys of land-based predators in Antarctica. Such surveys are likely to rely heavily on recent technological developments, such as satellites and unmanned aerial vehicles (UAVs), which may allow cost-effective survey in large and remote regions. We describe some specifications and assess the advantages and disadvantages of one such UAV, the ‘Aerosonde’, which is manufactured by Aerosonde™ in Melbourne, Australia. The Aerosonde is designed largely for long-distance, high-speed flights with data collection and real-time transmission back to a flight control centre, but could also serve as a platform for aerial photography. Its advantages in this role would be long-distance capability and low noise levels. Its disadvantages include cost, the need for a launching and landing runway, difficulty in operating around mountainous terrain, and likely instability in strong winds.

There are unresolved issues surrounding the use of power analyses for examining the ability of CEMP data to detect change. The effect size that should prompt a response if observed in a parameter and the likely response of parameters such as arrival or fledgling weights are two of these issues. Understanding the source of the variability required to generate ‘noise’ for the power analysis simulations is another and it has major ramifications on power estimates. Increasing the variability associated with power analysis estimates results in a decreased level of power to detect a trend. Similar power analysis results are obtained for a fixed co-efficient of variation of the temporal variability SD in relation to the initial value irrespective of the magnitude of the initial size. This means that power analysis results based on occupied nest counts with increasing levels of temporal variability (% CV) are applicable to other parameters that are suitable for trend detection. The four CEMP parameters considered in this paper have similar power estimates because their estimates of temporal variability are comparable (5.2 – 6.7%). For example, with a 10 year monitoring program it is possible to detect fixed increases or decreases larger than 2% each year with more than 80% power. Increasing the duration of the monitoring program has a positive impact on power estimates. There is very little difference between results generated using an exponential model compared with a linear model for short term monitoring programs of up to 10 years duration.

We examine the magnitude of a variety of sources of variability associated with CEMP parameters. The CEMP parameters considered in this paper include: A3: breeding population size, A1: arrival weights, A5: duration of foraging trips, and A7: fledgling weights. Sources of variability can generally be considered as either sampling variance or process related variance, although sometimes the two can be confounded. Temporal variability, that is the between-season variability, was consistent across parameters at around 5-7% CV. The smallest source of variability was associated with measurement error although this source of variability is difficult to assess. The largest source of variability was related to the timing of data collection which we term as the within-season variability. This ranged between 4.6% for population size up to 19.8% for fledgling weights. It would be useful to explore ways of reducing the magnitude of within-season variability because of the potential influence it has on detecting between-season changes in these parameters. This could potentially be achieved by either pooling several 5-day CEMP periods to reduce the time span over which the data are collected while simultaneously maintaining a sufficient sample size or by standardising summary statistics against some chronologically relevant event and reassessing temporal variability. Partitioning the sources of variability further than we were able to for this paper may be useful because our estimates of temporal variability incorporate within-season variability and measurement error. However, any refined estimates of variability would only affect the ability to detect long term changes in a parameter if sampling variance can also be reduced.

An important issue in regional-scale trend detection is the degree of concordance in trend between sites in relation to an average regional trend. If there is much or more variability between sites than within sites over time, inter-site variance can overwhelm the effects of other sources of variation in a system, resulting in low power in trend detection across that scale, despite precise methods of measurement and long time series of data. We used published data from repeated regional-scale surveys of Adelie penguin breeding population size in east Antarctica to assess the degree of spatial concordance in population trends within the scale of notional small-scale management units, and given these estimates of trend concordance, then used the power analysis program MONITOR to predict power for various multiple-site monitoring scenarios. Under the conservative, hypothetical scenarios we examined, monitoring at only 2 or 3 sites provided insufficient power to detect a trend. At the other extreme, monitoring at all sites in a region (a census) may be an unnecessary use of resources because the gains in power over a design using a sample of sites were very marginal. Reasonable power was achieved by monitoring around 6 sites in a region under conservative criteria of a 2-tailed test and alpha = 0.10. However, fewer sites were required for less conservative criteria; using a 1-tailed test instead of a 2-tailed test meant that a few less sites were able to achieve the same power for a fixed duration for detection, and increasing the significance or Type I error level from 0.10 to 0.20 improved power such that a few less years or sites were required to detect a trend. Monitoring every 3 years instead of annually reduced power only very marginally.

An assessment of the environmental processes influencing variability in the recruitment and density of Antarctic krill (Euphausia superba DANA) is important as variability in krill stocks affects the Antarctic marine ecosystem as a whole. Naganobu et al. (1999) had assessed variability in krill recruitment and density in the Antarctic Peninsula area with an environmental factor; strength of westerly winds (westerlies) determined from sea-level pressure differences across the Drake Passage, between Rio Gallegos (51°32’S, 69°17’W), Argentina, and Base Esperanza (63°24’S, 56°59’W), at the tip of the Antarctic Peninsula during 1982-1998. Fluctuations in the westerlies across the Drake Passage were referred to as the Drake Passage Oscillation Index (DPOI). They found significant correlations between krill recruitment and DPOI. Additionally, we calculated a new time series of DPOI from January 1952 to May 2003.

This paper responds to a request (WG-EMM-02 paragraphs 3.46 to 3.47) for standard methods for determining demographic parameters. It is noted the methods (CEMP A4) as published in CEMP Standard Methods 2003 appear adequate. Revision may be required in the future following consideration of how aspects of predator demographics may be used for management in the CCAMLR context. The considerable tagging and search effort required to obtain demographic data and the period over which such effort needs to be sustained is described.

Parameters measured under the CCAMLR Ecosystem Monitoring Program (CEMP) for Adélie penguins at the Béchervaise Island CEMP site were compared between seasons of contrasting krill availability. Krill biomass estimates were derived from shipboard surveys carried out within the penguins’ normal foraging range during the 2001 and 2003 breeding seasons. More than three times as much krill was present during the two-week survey period in 2001 than in 2003. Penguin parameters that showed significant differences between the two seasons included A5: foraging trip duration, A6: breeding success and A8: meal mass and dietary composition. Penguins travelled further to forage in 2003 than 2001, stayed away longer and brought back smaller meals. Fish (mostly Pleuragramma antarcticum) contributed significantly to the diet in 2003 but was only a minor component in 2001. Differences between years were particularly apparent during the late guard to early crèche stages of chick rearing, coinciding with the timing of the krill survey. Chick mortality peaked during this period also. The findings illustrate the sensitivity of parameters A5 and A8 to prey availability during the short time scale of the chick rearing period. Data on meal mass and foraging trip duration were combined to provide an index of provisioning rate, analogous to the functional response referred to in predator-prey theory. This showed the expected concave monotonic relationship to krill biomass over the period of investigation. These results are discussed in relation to aspects of foraging behaviour, monitoring programs and management issues.

The ability to use upper-trophic level species as ecosystem indicators is determined by the ability to relate changes in indices of their performance to changes at lower trophic levels. Using indices of predator performance from four species of krill-eating predator together with independent ship-based acoustic estimates of krill abundance from South Georgia the relationship between a range of indices of predator performance and krill abundance was examined. There was a distinct relationships between the variability of indices and the biological processes that they measured; body mass parameters had the lowest variability (CV 50 %. Predator parameters that reflected processes occurring during the summer showed the closest relationship with krill abundance, especially those for species with foraging ranges similar to spatial scales at which krill surveys were undertaken. Population size parameters showed no functional response relationship with annual krill abundance estimates. Combining the summer parameters into a single combined index provided a better fit with the krill data than any of the individual parameters.

Mackerel icefish have been harvested in the CCAMLR Region for over thirty years. In the Atlantic sector of the Southern Ocean they feed preferentially on krill and are themselves preyed upon by fur seals and several avian species. These attributes make them suitable for consideration within the CEMP. With that in mind the following indices are outlined that might be incorporated in the programme: Standing Stock, Cohort Strength and Recruitment, Natural Mortality, Length at Age 1+ and 2+ years, Condition, Gonad Maturity and Diet