Food webs are a core part of biology – we start learning about them in primary school (this eats this, which eats that, and this is the apex predator and so forth, with lots of lines between species forming the ‘web’).
They are the most common and recognisable representations of ecosystems, and food-web theory is a very large and well-established field. Indeed, insights emerging from food-web theory have given rise to many approaches on which parts of the web we should focus on to keep them functioning, and which species we should make priorities.
Dr Eve McDonald-Madden from the University of Queensland (and CEED) and colleagues have applied environmental decision science to food-web theory to see if we can learn anything new. What they revealed, just published in Nature Communications, could revolutionise the field.
“The vast majority of food web research is about ecosystem conservation and management,” says McDonald-Madden.
“As decision scientists, we specifically investigated how food webs can be used to make decisions about ecosystem management. And what we found represents a major change in the way we think about the importance of individual species in food webs.”
McDonald-Madden and colleagues applied a range of Artificial Intelligence (AI) optimisation techniques to develop optimal management strategies for food webs (using both real and hypothetical food webs as case studies). They then compared these optimal solutions with a range of indices emerging from conventional food-web theory.
“We made the surprising discovery that species identified as important for food web persistence using a variety of popular indices frequently represent poor choices for conservation management of those food webs,” says McDonald-Madden. “Interestingly, a modified version of Google PageRank gives the ‘least-worst’ performance for the management of a large number of food webs.”
The Google PageRank algorithm is used by Google to rank websites in their search engine (and ‘Page’ doesn’t refer to web page, it’s actually named after Larry Page, one of Google’s founders). PageRank works by counting the number and quality of links to a page to determine a rough estimate of how important the website is. The underlying assumption is that more important websites are likely to receive more links from other websites.
PageRank is a relatively simple way of cutting through the enormous complexity of the internet and it was first used to address the importance of species in food webs a few years ago by Stephano Allesina from University of Chicago. McDonald-Madden and colleagues were able to demonstrate its potential for informing risk-averse conservation decisions for managing species in food webs.
“Our work provides a robust methodology to extend past and future food-web research from simple assertions of conservation utility to actual tests,” says McDonald-Madden. “Further, this work provides a novel interdisciplinary link between network theory, artificial intelligence and conservation.
“Trying something novel in a well-established and enormous field such as food-web theory is a daunting challenge. However, our results have demonstrated there’s still much we can learn. Hopefully these studies will change the way people investigate food webs and produce tangible conservation outcomes.”
McDonald-Madden E, R Sabbadin, ET Game, PWJ Baxter, I Chade`s & HP Possingham (2016). Using food-web theory to conserve ecosystems. Nature Communications 7, Article number: 10245.