Home » Biology » Environment » “No Ecosystem is an Island”

Want to Get Involved In Research?

[su_heading size="15" margin="0"]The BioInnovation Group is an undergraduate-run research organization aimed at increasing undergraduate access to research opportunities. We have many programs ranging from research project teams to skills training (BIG-RT) and Journal Club.

If you are an undergraduate interested in gaining research experience and skills training, check out our website (https://bigucd.com/) to see what programs and opportunities we have to offer. In order to stay up to date on our events and offerings, you can sign up for our newsletter. We look forward to having you join us![/su_heading]

Newest Posts

“No Ecosystem is an Island”

By Daniel Friedman, Genetics ’14

For years, ecologists have modeled the biodiversity of natural forests as if they were oceanic islands, adrift in an unlivable sea of humanity. However, research published in April in Nature by C. Mendenhall et al. suggest that this is not the most accurate or predictive way to think about these pockets of nature. By comparing bat diversity on countrysides and oceanic islands, they find that fragmented land ecosystems behave markedly different than their oceanic counterparts. They find that forest “islands” maintain species at higher overall levels of biodiversity than ocean islands, and also gain/lose species in unique patterns. This has relevance to humanity’s actions to support biodiversity on land, and suggests the need for new models, metrics, and strategies of conservation.


Mendenhall, C., Karp, D., Meyer, C., Hadly, E., Daily, G., “Predicting biodiversity change and averting collapse in agricultural landscapes”, Nature, 2014.

Image from Abu Shawka, 2009. Creative Commons.


2 Comments

    • “No ecosystem is an island” summarizes what I learnt from this paper. Here is a fuller explanation:

      Studying the biodiversity patterns of ocean islands has been tremendously fruitful for ecology research. Islands are of defined size, have limited migration rates for most species, and often appear in clusters (“a natural experiment” with replicates). This allows ecologists to perform elegant studies of speciation/extinction rates, invasive species biology, and niche competition, for example.

      To model fragmented forests, terrestrial ecologists commonly use the mathematical models that were generated to describe ocean islands. This metaphor is intuitive — an “island” of rainforest, surrounded by a “sea” of agricultural land. As humans continue to make the “islands” of rainforest smaller and more isolated via habitat destruction, ecologists want to know how biodiversity patterns will look in 10, 20, 50 years. Should we have a few big “islands”, or many small ones? Can logging ever be sustainable?

      Mendenhall et al. 2014 takes a step back, and finds that the biodiversity dynamics of forest “islands” are quite different than dynamics observed in ocean islands. This then provides a step towards better modelling of terrestrial ecosystems.

      Lastly, when studying ecology, we always need to keep in mind that isolated systems don’t exist. Closed systems are merely scientific conveniences, not deep statements about modularity in Nature. Thus, “no ecosystem is an island” for the same reason “no man is an island” — because islands don’t exist!

Leave a comment

Your email address will not be published. Required fields are marked *