Video transcript: Untangling nature’s food webs

You may wish to play the video in another window to watch it side by side with the transcript below. Alternatively, you can watch the video on our YouTube channel with captions.

August 2012

Video shows a 3D model of a food web.

Professor Jane Memmott, University of Bristol
One of the fascinating things about food webs is you do see the same patterns come up again and again. So for example, if I had a parasitoid network from my back garden, it wouldn't look that different to a parasitoid network that we collect from the tops of mountains in Hawaii. So you do see the same patterns coming out again and again. In general, you'll find there are a few well-connected species in a network and a lot of species that are much more kind of peripheral to things.

I'm going to show you how we actually visualise a food web.

Video shows food web visualisation on a computer screen.

So starting at the bottom of a food web we've got all the plants. So what we've done here is we've counted leaves, or flowers, or seeds, or some part of the plant. So we add the caterpillars in this particular case. So this is the herbivore part of the community. This is a common caterpillar. This is a rare caterpillar. And these are all of the caterpillar species in this particular plot. So up here we've got the parasitoid community, a common parasitoid and a rare parasitoid. And the different colour codes in this case tells us something about the biology of the organisms. The thing that makes the food webs really useful though is we know who is feeding on who. So for example, that particular caterpillar species here is feeding on this particular plant species. A parasitoid up here is feeding on that caterpillar species. And it's feeding on more than one species though as you can see if I fill in the rest of the community it looks like this. This parasitoid here is feeding on two species of caterpillar. This one here, which is the commonest one in the community, is actually feeding on five different species, one, two, three, four, five. But it is mostly feeding on these two caterpillars.

So this is the architecture of the network. You can plot these networks in various ways. This is the way that I use it because it is the way I am used to reading these things, but you can plot them in 3D and twizzle them around in space and they look just fantastic.

Studying food webs is really very useful when considering pollination because traditionally people look at the crop and what visits it. If you look at that in a network context you need to know if you have a key pollinator to consider the pollinator you've got to consider what is happening outside of the crops flowering season.

Video shows bumblebee visiting lavender.

If it's a bumblebee it is going to emerge in March/April, you need food plants around then so if you look more holistically at the whole picture of the lifespan then you can see how the pollinator and the crop fits into a much bigger picture. Then ideally you can manage that network to actually favour the insects you're after for the pollination services and make sure they are there year after year and you'll get sustainable pollination that way.

Video shows bumblebee visiting a flower.

It's important to save pollinating insects because an awful lot of the food we depend on as humans comes from, is the result of pollination. So while wheat might be wind pollinated, say, all the soft fruits, beans, and a whole variety of other plants are dependent upon pollination. We can't, as people, recreate that service any other way, we're dependent on wild and managed pollinators. So without pollinators we'd get rather hungry!

We need to know which insect species are pollinating which crop species and which insect species can pollinate them. So can we manage these communities better? Rather than just sitting back and hoping we're doing the right thing, we need to know more about which environmental schemes really work. We know that the pollen and nectar mixes can be really rather effective, but there is still an awful lot we need to know. So what can we do to make life better for the pollinators? So we can then actually reap the rewards of this in terms of good pollination services.

Video shows moths, bees and hoverflies collecting nectar from flowers.

It could well be disastrous for UK agricultural loss, major pollinating species. I mean if things like the bumblebees all disappeared, solitary bees, some of the hoverflies can be surprisingly important. If those insects start disappearing, so basically there are no spare parts in the system, we'd see food productivity potentially drop, potentially fairly catastrophically.

Video shows a 3D model of a food web.

Food webs are an excellent tool for modelling extinctions at the community level because what you can do is actually, when you've got your food web, when you've spent a long summer putting together this data, you can load it up onto the computer and then you can actually model the effect of removing a species. So you can take out species one by one according to various rules. So for example, you could remove the rare species as those are the ones most likely to go extinct. Or you could take some of the really well-connected species out like the bumblebees in a pollination network or take out the honey bee and actually ask how many linked extinctions are there? How many plants, how many crops or native plants are left high and dry with no visits at all? And there is good theory to suggest that taking out just a handful of well-connected species can actually lead to the collapse of a network.

In pollination networks you get patterns whereby plants that actually have a very highly general light, where lots of things visit them, tend to attract the really rare species in the community. So again, in pollination networks, my pollination networks from local farms don't look that different to pollination networks from urban habitats or from nature reserves. There are subtle differences and one of the million dollar questions is asking, well you know, is there a fingerprint of some sort of environmental problem that you can see on a network? Is there some subtle pattern that tells you it's about to kind of start having serious problems? If you can predict that before it happens then that's really interesting.



This video may be reproduced in its entirety with due credit to BBSRC.

All media copyright BBSRC unless otherwise stated.

Image, music and video credits:

  • Food web animations courtesy Professor Neo Martinez of the Pacific Ecoinformatics and Computational Ecology lab PEaCE lab: www.foodwebs.org
  • Insect footage from Science Photo Library www.sciencephoto.com
  • Music 'In the End' from www.cinephonix.com