So, this is my first time in China, I'm very happy to be here and I want to thanks the
organizers for giving me the opportunity to come here.
So I hope that my talk will be partly complementary to the talk that Andrew just gave to you.
And I try to highlight what, in my opinion, food-led models can help in understanding
the impacts of global change on the southern ocean ecosystem.
As mentioned before, in the past, marine resources in the southern ocean have been exploited.
And for some groups, the effect of this exploitation is still visible.
Baleen whales, the biomass of Baleen whales is still to 10% of pre-exploitation levels.
Several fish stocks have been depleted in the past, and this certainly has affected the
structure of marine food waste in the southern ocean.
Another hypothesis is that, for example, harvesting of whales might have affected the carbon-obtained
capability of the southern ocean.
Whales are considered as good in recycling iron in the upper water column, and iron is
a limiting factor for primary food activity in the southern ocean.
Nowadays, there are major environmental changes that are ongoing in the southern ocean, some
region, as the Western Antarctic Peninsula, where there has been, which is one of the
most rapidly worming regions unheard, which decreased the NCIS extent and duration.
In the last three decades, there has been also a decrease in the biomass of krill stocks
in the South Atlantic, so in the same region that is subjected to rapid climate warming.
And this has been linked to reduced recruitment on the population of krill because of reduction
in NCIS extent and duration.
So one of the goals of Sue's theme six is to understand how the ecosystem will respond
to global change, and both climate impacts and harvesting effect have the potential
to affect the structure of food waste in the southern ocean, and it's useful for policymakers
and managers to be able to disentangle what of these two factors affects and how the structure
of the food waste, the slide that Andrew showed before.
So today, there's a lot of data sets that are already available, biological data sets
available.
More data sets are sampled through the Sentinel program and through Sue's, and the Integrated
Climate and Ecosystem Dynamics program wants to address the modeling part of understanding
how the ecosystem will respond to climate change.
So in the southern ocean, there is interagency in forcing an habitat structure.
I'm thinking about the duration of sea ice, season, the biogeochemistry, but southern
ocean ecosystems are interconnected by the Antarctic Circle Podocore, that is important
in moving plants around the continent, I'm thinking here also, for example, the Antarctic
Krill, and so you have a series of regional food webs, on the left in the bottom you have
a schematic, a conceptual model for a food web in the western Antarctic Peninsula region,
I'm sorry, Europe, South Western Antarctic Peninsula region, or Scotia Sea region.
This food web, we think that it's centered on the Antarctic Krill as the principal intermediate
traffic level, the transfer energy from primary producers to top predators, to the right you
have a schematic for the raw sea, this food web is probably different, you have the crystal
krill and the Plerogramma Antarctic, Antarctic Silverfish, that are probably the two most
important intermediate traffic levels for energy transfer.
So how do you study this food web, there's a series of different modeling approaches,
like inverse models, qualitative models, you'll see here, quantitative net food models, Antoine
models, and the paper that I'm showing you here, it's a nice paper, Marfiata 2015, so
when they put forward some ideas on how to do good practices or best practices in modeling
marine food web, and they also suggest the use of a generic model structure for modeling
marine food webs, that will allow us to more easily compare the regional food webs.
So I just want to show you an example here for two food webs, that I participated in modeling
these two regions, Southwestern Antarctic Peninsula and South Georgia Shelf.
So we assembled the available data in a common model structure for the two regions, and then
calculated energy flows in the food web.
So you can see that the numbers are different in the two food webs, and I don't necessarily
want to remember the number that you see here, but having a common model structure, we can
see in which of the two food webs, and energy flow is more important than the other.
So different traffic pathways in the two food webs.
We said that the grill, it's an important species in the Western Antarctic Peninsula
South Scotia Sea, and so here with colors, here you should see, this is the Antarctic
food, so you have the different food web loops, the size of the box is not proportional to
the volume, I just want you to read the names, but the connectors between the boxes, they
show you the flow of energy, including from the tray of the grill to the grill, and this
used to be rare, so from grill to the predators, so you can see that in the two systems, the
magnitude of the flows between prey and predators is different.
It's also very important to address uncertainty when we do quantitative food web models, so
here we try to, so we build models in which, for each model parameters, we assess the
viability around the model parameter and use Monte Carlo assembly to create food web
modes, and then we looked how energy transfer in the food web changes in what you've shown
in this scenario, so in this simulation here, we simulated a reduction of 50% in Antarctic
food biomass, and on this axis, you have the food web, and on this axis, you can see the
change in productivity in the food web when you decrease the biomass of the Antarctic
food, so that means that in this scenario, we need a change in the amount of energy that
is transferred from in the food web to the fish and cycle of us, and to the secret and
money models, and the fact that these are results for 1,000 simulations, the majority
of them fall below the zero level, there will be no change in respect to a reference simulation,
so there's a viability in the response, and the same exercise for the South Georgia Shelf
model shows much smaller viability, so this is a way to compare also if two ecosystems
can respond in a similar way to a common scenario, and two end models, so here are two types
of end-to-end models, here you have the food web model, in which we add the recycling of
movement by all the ethylophobic components of the food web, recycling of movement back
at the base of the food web, and right now these models are implemented in recycling
of ammonia, but it will be possible in the future to recycle also iron, for example from
plastic such as wheels, and this might be interesting to quantify the amount of recycled
and new production in the food web, on the other case you have a food web model, and
with the green I wanted to show you that the base of the food web can be forced with data
that come from a biogeochemical model, so this is a simple way to integrate a food web
model with results from a biogeochemical model.
I think that an important use of food web models is to integrate data from different
sources and also from other models, so if you have your food web model in the center,
this can be driven by results from physical models, by results from each model which
is a physical model, so this is a start to build a mechanistic model for how the biology
can change, as I said before you can work with biogeochemical models or with data.
So overall I think that food web models can help us develop our conceptual understanding
of the functioning of the ecosystem, how the different structure of an ecosystem can affect
its stability and resilience in face of harvesting effects and climate change, and they can be
useful to identify the essential biological variables that we need to measure in order
to assess the present status of the ecosystem and how this can change in the future.
Thank you.
Thank you.
