Google Knows the Next Extinct Species

All animal species on Earth are part of an animal chain, a network, a relationship between predator and prey, a “food web” and if one species goes extinct, some of the others might suffer too, but which one?

Ecologists Stefano Allesina and Mercedes Pascual also asked themselves this question but they also found a way of answering it: thanks to Google’s Page Rank algorithm.

The search engine is so popular these days that it has become a verb, and the two scientists “Googled” for ways of knowing which species in a certain web is the most likely to cause a wave of extinctions, successfully.

They made a new algorithm based on Google's page ranking method and on a changed “eigenvector” - a ranking algorithm, that allowed them to order species depending on their importance for co-extinctions, and coming out with a “sequence of losses that result in the fastest collapse of the network.”

This is a major achievement as there are species that are more or less important to the functioning of the web, and if they are extinct they can trigger one or several co-extinctions.

This new approach can be applied to any model, based on the theory that if any given prey species disappears it will affect any given predator species.

Also, this algorithm is perfect for identifying extinction sequences, even more accurate than all previous models that were based on the number of connections between species, and it can process millions of sequences.

In a food web there are many links but not all have an important part in the strength of the web, so the advantage of this adapted EIG algorithm is that it will actually find the most efficient route to system collapse.

“The algorithm works in this sense better than all the others previously proposed and lays the foundation for a complete analysis of extinction risk in ecosystems,” say the scientists.

The modified algorithm will become crucial in the future as scientists and ecologists are very likely to use it to identify the areas of highest concern, for funding and ecological research.

The results of the research for this algorithm were published in the journal PLoS Computational Biology, Planetsave reports.