Size matters in Southern Ocean ecosystems models

A collection of fish and other mesopelagic creatures on trays.
Some of the incredible variety of fishes, squids and other invertebrates that live in the top 1000 m of the Southern Ocean. These animals are an important alternative food source (energy pathway) to krill, for higher predators. (Photo: Rowan Trebilco)
Rowan Trebilco in front of an iceberg.A generic Southern Ocean food web showing the different energy pathways from primary producers (phytoplankton, diatoms) to higher predators.

Marine ecologist Rowan Trebilco, working within the Antarctic Climate and Ecosystems Cooperative Research Centre (ACE CRC), received the second RJL Hawke Postdoctoral Fellowship in Antarctic Environmental Science in 2015. During his three year fellowship Dr Trebilco advanced development of size-based models, to help understand the role of ‘mesopelagic’* fish and squid in shaping Southern Ocean ecosystems, now and in the future.

Southern Ocean ecosystems provide us with many things of value, from economically important krill and finfish fisheries, to charismatic megafauna, such as penguins, seals and whales, and smaller, yet (arguably) equally charismatic planktonic creatures. The Southern Ocean also provides valuable ecosystem services, like producing oxygen and sequestering carbon; around 40% of the global ocean’s uptake of anthropogenic carbon is thought to occur in the Southern Ocean.

These values are fundamentally underlain by the movement of energy (and material) through food webs. A pressing question is: what impact will changing environmental conditions and patterns of human use have on the provision of these values and services? However, it is almost impossible to observe these processes directly across large scales. Instead, we need models that we can combine with observations, to build a picture of how things work and understand how they may change. No one model is perfect though – different models have different strengths, weaknesses, and gaps.

A solution is to have a ‘toolbox’ of different models. This helps ensure that gaps are filled, and if several different models tell the same story and are congruent with available observations, it provides confidence that the story is realistic. Meteorologists have been doing this for years, and models are at the point where we’re pretty good at forecasting the weather in the short to medium term (something thought impossible not so long ago). We’re just starting down this path in the field of ecosystem modelling.

At the ACE CRC we are developing such a toolbox of models for Southern Ocean ecosystems†, to better understand how they work, how they may change, and how they might be best managed under different scenarios of future climate and fishing. Contributing to this is a major part of my Hawke Fellowship and my role within the CRC.

A big gap in Southern Ocean ecosystem models (and ocean ecosystem models globally) has been the poor representation of mesopelagic fishes and squids. We know that mesopelagic fish are the most abundant vertebrates in the biosphere, and that they account for more biomass than any other group of fishes. We also know that mesopelagic fishes and squids are key prey for penguins, seals and other large marine predators. In fact, they provide a major alternative energy pathway (a route by which energy can move from phytoplankton to top predators, and be exported to deep water) to the better-understood, krill-dominated food chains. But detailed understanding of their ecology – their diets and what drives their distribution and abundance – and their relative importance in food webs, compared to krill, is lacking.

The overarching goal of my project for the Hawke Fellowship was to address this gap by developing size-based models to understand the role of mesopelagic fishes and squids in Southern Ocean ecosystems (Australian Antarctic Magazine 29: 8-9, 2015). Size-based ecological models differ from traditional species-focused approaches in that they focus on body size rather than species identity as the principle descriptor of an individual’s role in a food web. This means that they can quantify and predict ecosystem structure and change without detailed species and stage-specific dietary data.

I’ve been fortunate to have several opportunities to connect this work with larger research initiatives nationally and internationally, which has helped broaden the scope and impact of the work.

In 2015-16 I participated in the design and planning of the Kerguelen Axis (K-Axis) voyage (Australian Antarctic Magazine 29: 2-3, 2015), then co-led the fish and squid sampling program at sea. During the voyage we sampled the mesopelagic community from the surface to 1000 m at 36 stations. The locations were chosen to maximise our ability to disentangle how key environmental drivers (such as sea ice, frontal features and water depth) drive ecosystem characteristics. At each station, the ‘mid-water opening-closing’ device we used on the net allowed us to split the catch into 200 m depth layers. At the same time, we used the ship’s acoustic echosounders to build up a picture of how biomass was distributed in the water column.

We sorted and photographed the catch at sea, then froze everything for further processing and analysis on return to Australia. Since the voyage, we have been measuring all the individual fish and invertebrates (including squids) in the photographs. We are also analysing samples from individual fish to get information on food-web linkages (to see who is eating whom), using biochemical tracers such as stable isotope analysis of muscle tissue and genetic analysis of gut contents. This has yielded an incredible new dataset for fishes and squids that will inform the development of models and allow us to evaluate their performance. This dataset has also sparked new collaborations with colleagues in France, the UK, China and Japan.

While the original plan for my fellowship had been to focus only on mesopelagic fishes and squids, there was a lot of interest in extending the models to include zooplankton and higher trophic levels, such as seabirds, marine mammals and toothfish. To facilitate this I coordinated a group of experts to pull together information across these other trophic levels. Development of a model that includes these groups is currently underway.

In addition to local collaborations I’ve also developed international partnerships. In 2016 I was awarded a Scientific Committee on Antarctic Research Fellowship to spend three months developing collaborations with two groups of colleagues in France.

The first group, at the Museum of Natural History in Paris, has been studying the natural history of mesopelagic fishes for many years, particularly on the Kerguelen Plateau. I worked with them to identify the fish we caught on the K-axis voyage, and determine how to best represent mesopelagic fishes in my models.

The second group at CLS (Collecte Localisation Satellites – a subsidiary of the French space agency), has developed a model (SEAPODYM) to understand how mesopelagic taxa move energy through the water column and make it available to higher trophic levels. This model was developed for tropical seas, with a focus on tunas, but there is clear potential for Southern Ocean applications. Together we worked on a Kerguelen Plateau implementation of SEAPODYM, which will be incorporated into the ACE CRC ecosystem model toolbox, and will provide the foundation for a PhD student project next year.

My time working at CLS also led to my inclusion on an EU-funded project called MESOPP (MEsopelagic Southern Ocean Predators and Prey). This project aims to develop standardised methods and datasets for assimilating estimates of mesopelagic biomass from active acoustic data, in ocean ecosystem models. The size-based models I have been developing for the Hawke project will be one of three main groups of models used for this project.

The Hawke Fellowship has been an amazing opportunity, and being able to combine it with my position in the ACE CRC has allowed me to ‘think big’ and develop large, collaborative, long-term research projects. Much of this work is set to come to fruition over the next 6–12 months and it will substantially advance our understanding of how mesopelagic fishes and squids fit into Southern Ocean ecosystems, how they support key predator populations, and how things are likely to change into the future.

Rowan Trebilco
ACE CRC and UTAS

*The mesopelagic zone is normally defined as the top 200–1000 m of ocean. The animals that live here are broadly called ‘mesopelagics’, although many of them migrate vertically through the water column, moving into the top 200 m (the epipelagic zone) at night, then back to deeper water during the day. Similarly, some inhabitants of the upper parts of the bathypelagic zone (1000–4000 m) move to mesopelagic depths during the night and may also be caught at mesopelagic depths.

†Australian Antarctic Science Project 4366