Hobart

Trophodynamic Workshop - Abstracts

Wednesday 14 April 2010, 9am - 5pm (Tas time)
CSIRO Auditorium, Hobart


Olly Berry
CSIRO Marine and Atmospheric Research
Floreat

Genomic Tools for Trophodynamic Questions

Species identification by means of DNA sequencing technologies is a burgeoning field, and has the potential to provide unparalleled resolution to marine trophodynamic studies. In this presentation I will introduce the genomics platforms best suited to this analysis, and contrast them in terms of cost, scale, transferability between systems, and their abilities to provide quantitative dietary information.


SJM Blaber
CSIRO Marine and Atmospheric Research
Cleveland

Predation studies from northern Australia and beyond

A brief overview is given of studies of the diets of more than 1000 species of fishes from northern Australia, Papua New Guinea, the south Pacific, south-east Asia and south Asia. This research formed part of broader projects undertaken by the Tropical Fish Ecology program and encompassed various fisheries projects funded by CSIRO, FRDC, ACIAR, DAFF, GBRMPA, as well as other relevant government agencies and conservation organizations. In most cases all dietary analyses were quantitative and in some projects they were linked to experimental food consumption studies.


Cathy Bulman
CSIRO Marine and Atmospheric Research
Hobart

From Guts to Glory (or Mush to Models)

Traditional stomach sampling studies have provided most of the information that underpins current ecosystem models based on a trophic structure. While these studies are invaluable and sometimes expansive, the data still have limits, and gaps in our knowledge remain. New techniques that can offer a broader or different perspective of trophic interactions will be a valuable tool in gathering trophic data, particularly if they offer a broader but cost-effective coverage of species.


Grace Chiu
CSIRO Marine and Atmospheric Research
Acton

Statistical Social Network Modelling of Trophic Relations in Food Webs: Current State and Beyond

At the recent Techfest 2010, we presented the current state of this statistical methodology for understanding predator-prey relationships in food webs. It differs from classical social network analysis techniques by explicitly modelling the random link between any given pair of species, and the complex dependence structure on these links. The resulting statistical inference can address (i) predation activity, (ii) what makes a given species a prey or a predator, (iii) the tendency for predator-prey role reversal, (iv) trophic clustering, and (v) predation preference (latent clustering). However, modelling complications arise for various reasons, from the lack of data on various aspects of the food web, to the perhaps non-random nature of links viewed from certain perspectives. Through its application to the famous Benguela ecosystem (Yodzis 1998 in JAE), we highlight the usefulness of the current methodology and what is required to bring out the full potential of this ongoing research.


Gavin Fay and John Field
CSIRO Marine and Atmospheric Research
Hobart

Including data on predation mortality into stock assessments: ecosystem and single-species approaches for longspine thornyhead (Sebastolobus altivelis)

Stock assessment models rarely consider the impacts of changes in the levels of predation-related mortality. We present methods of explicitly accounting for changes in mortality rates, by including data on the abundance, consumption, and size-specific preferences of predators into the currently applied stock assessment framework for longspine thornyhead. The merits of incorporating these data directly into stock assessment models are discussed.


Mike Fuller
CSIRO Marine and Atmospheric Research
Hobart

Oh Data, Data, Wherefore art thou Data?

Trophic studies, or any analysis requiring information from more than one source, benefit greatly from easy access to the data feeding them. When analyses are using their own data or information from a small number of sources the overheads of integration are relatively low. As the complexity of data requirements increase (as demanded by ecosystem models) so do the issues associated with sticking it all together. Strategies data managers find useful for overcoming the burden of data integration include development of standard repositories and adoption of common keys such as CAAB codes that are shared between as many datasets as possible reduce the burden of integration. And – time for feedback. Where do the data users want their data to go, and what methods are the most convenient for getting it back?


Beth Fulton
CSIRO Marine and Atmospheric Research
Hobart

Ecosystem models and the desperate need for diet data

Release of computing constraints has seen the burgeoning of trophodynamic ecosystem models over the last 25 years. Unfortunately parameterisation of diet matrices has been more magic than science in those models. Even the best parameterised models miss connections are either based on snapshots or average away much variation and space and time. Moreover, within models diets can also vary to differing degrees depending on which model is being used. Given the different models can lead to different results, or that even that different trophic connections have different implications for management, how do we (i) remove the gaps, (ii) capture spatial and temporal variation and other sources of variability or uncertainty.


Shane Griffiths, Gary Fry, Tonya van der Velde, Fiona Manson and Richard Pillans
CSIRO Marine and Atmospheric Research
Cleveland

A synopsis of trophic studies on tropical neritic pelagic fishes in northern Australia and their importance for quantitative ecosystem models

Over a 3-year period (Aug 2002 and Sept 2005) a small CSIRO-funded study supported the opportunistic collection of neritic pelagic fishes from recreational and commercial fishers and scientific independent prawn trawl surveys in northern Australia. The primary aim of the study was to cost-effectively collect dietary information for an Ecopath model being developed for the Northern Prawn Fishery, which lacked species-specific and regionally-specific information on most pelagic fishes in the region, ranging from small planktivorous scads to large billfish. Stomachs and other biological material (e.g. otoliths, gonads, muscle tissue) were collected from 3250 fish representing 41 species (30 teleosts and 11 elasmobranchs). Stomachs were processed and the Biomass and frequency occurrence of each prey in each stomach was recorded, as well as the length of each prey item. Non-metric Multidimensional Scaling (MDS) using a Bray-Curtis similarity measure based on biomass data was initially used to allocate each species to an ecological functional group for the Ecopath model. However, Multivariate Regression Trees Analysis was later found to be a more appropriate method to explore similarity of diets among and within species. The value of having species-specific and regionally specific diet data for the development of ecosystem models will be discussed.


Geoff Hosack
CSIRO Marine and Atmospheric Research
Hobart

Towards data driven qualitative analysis of food web stability properties: even qualitative models need data

Qualitative models predict how ecosystems can respond to future changes using two stability properties: (1) resilience, which determines if the ecosystem is robust to temporary perturbations, and (2) resistance, which determines how species respond to long-term sustained pressures. Inputs for the qualitative model are ideally derived from empirical data that provides information on the functional forms of trophic relationships, such as functional responses, predator saturation, or donor control. Important processes that mediate trophic relationships, such as modified interactions, may also be included. The qualitative model thereby depicts the mechanistic links that drive ecosystem function and uses this information to qualitatively predict how the ecosystem will respond to future perturbations. Model validation requires long-term observations to match the time scale of predicted changes in the ecosystem.


Rhys Leeming
CSIRO Marine and Atmospheric Research
Hobart

Elucidating Trophodynamic Uncertainty and Change with Nitrogen Isotopes of Amino Acids

Stable isotopes of carbon and nitrogen have been used successfully in many lake and marine trophic studies to elucidate, not only trophic positions, but also trophic shifts induced by disturbance and migration. However, results have had to be interpreted with caution as baseline signatures can vary both seasonally and spatially and these variations can propagate up the food chain. At least, 85% of the nitrogen measured in bulk muscle tissue is integrated inside amino acids. It has been discovered that one sub-set of amino acids have their isotopic signature “set” by primary production and are not fractionated through the food chain. Thus, sampled at any trophic step of a food web they have the potential to yield values for the base of the food chain. The other sub-set of amino acids are repeatedly fractionated at resynthesis and therefore yield a significantly higher resolution estimate of trophic enrichment than bulk isotope values alone; after all, the bulk values are an average of the amino acid values. We suggest that selective analysis of amino acids in tissue samples for isotopic content could greatly augment broader isotopic and gut contents studies of ecosystems. This technique has the ability to distinguish between nutrient and trophic dynamics at an integrated scale. It will be possible to test whether groups of the same species forage at similar or different trophic levels and whether they are likely to be from one or more regional groups. There is the real prospect of looking for changes in trophic status (climate, fishing pressure) from archived to present day samples. And it should be possible to use these data to constrain ecosystem model outputs and to measure and distinguish variability across the food web.


Peter Nichols
CSIRO Marine and Atmospheric Research
Hobart

Signature lipids – where are we at and what is next?

The application of signature lipid methodology is being increasingly demonstrated and applied for a range of fish, squid and higher predators, as well as in environmental studies. In food web studies, the signature lipid approach can complement traditional stomach content methodologies, and also is thought to generally provide a longer term dietary signal, as well as information fish condition. Protocols include lipid extraction, fatty acid methylation followed by gas chromatography (GC) and GC-mass spectrometry (GC-MS) analyses. Case studies have highlighted the application of this approach to various species, with e.g. digestive gland (DG) fatty acid profiles of squid indicating that these species were reliant on a myctophid-based diet. It is also possible to examine the effect of regional, temporal and other differences including maturity on diet and the overall role of lipid in marine species. The use of signature lipid data for the determination of the diet of lower tropic levels including zooplankton, molluscs and higher predators such as cetaceans and pinnipeds is also possible. The case studies undertaken have been informative to date, and the approach is now available for integration with other researchers including modelers.


Andy Revill
CSIRO Marine and Atmospheric Research
Hobart

Physics to fish, Isoscapes to food webs: can we use isotope landscapes to help model fisheries ecosystems?

Marine food webs are only one compartment of a much wider series of interactions which include broad scale oceanographic influences on the environment in which the food web operates. Often these far field influences are ignored but as we develop new tools for tracing energy flow we also have the opportunity to investigate how these broader processes influence and impact upon the smaller scales. A technique now routinely used in ecological studies is that of stable isotopes and this has found wide ranging application in marine food webs, particularly for estimating trophic levels. What is now becoming evident is that we must take into consideration the broader isotopic "landscape" (Isoscape) when interpreting these results. In this talk I'll present some of the isoscape concepts and illustrate important aspects for food web studies and how we might use these to our advantage when modelling ecological interactions.


Mat Vanderklift
CSIRO Marine and Atmospheric Research
Floreat

Certainties and uncertainties in use of bulk stable isotope data to inform fisheries ecology

Stable isotope ratios of organisms have now been used for several decades as a way of understanding food webs, and their use in now quite commonplace. In recent years there have been significant advances in mathematical models that use known relationships between consumers and their diets to determine the relative contributions of different diets. There have also been advances in understanding of the physiological pathways that lead to the observed patterns. But how much certainty is there in the conclusions we can draw from stable isotope data, and what are the reasons for uncertainty? In other words, how robust are the inferences we are making. I will (hopefully) identify at least some of the things we do know, some of the things we think we know, and some of the things we are still unsure about. I will also identify some alternative ways of using stable isotope data.


Jock Young and Petra Kuhnert (CMIS)
CSIRO Marine and Atmospheric Research
Hobart/Glen Osmond

The power of many!

A major limitation of diet studies is the difficulty of collecting samples over large distances and long time frames. I explore here the advantages (and hurdles) of combining data sets from different regions and oceans, and the power that that can bring.

[back]

[Top of Page]

Location:
CSIRO = Marine Laboratories Auditorium, Castray Esplanade, Hobart

For further information, or to schedule a seminar, contact:
To schedule a seminar, contact:
Clothilde Langlais, (Oceanographic seminars) CSIRO Marine and Atmospheric Research (03) 6232 5399
Natalie Kelly, (Biology/Modelling seminars) CSIRO Marine and Atmospheric Research 0438 452 483
Jillian Enraght-Moony, (seminar administrator) CSIRO Marine and Atmospheric Research (03) 6232 5320
Communications Manager, Antarctic Climate and Ecosystems CRC (03) 6226 2265
Margaret Hazelwood, Institute of Antarctic and Southern Ocean Studies (IASOS) University of Tasmania (03) 6226 2971

Last updated 11/05/10