I’d like to introduce the concept of refugia. (Refugia is the plural form; refugium is singular.) It’s not my concept, nor is it a new one. Refugia, in the way I’m using the word, refers to the places where individuals, populations, or species find a haven during times of crisis. Refugia can refer to:

  • places that remain hospitable and survivable for species while the rest of the world goes through a crisis that results in a mass extinction, occasionally called “extinction refugia”
  • an escape from predators, also called “predation refugia” – although this doesn’t always refer to place, but instead can also refer to some other aspect of the organism, like the animal growing too large for the predator to easily bite or grab; this would be called “size refugia.” (This particular type of refugium is not one I’ll be covering much in my discussions).
  • places where the habitat will remain sufficiently stable in the future with respect to temperature, humidity, precipitation, etc., also called “climate refugia”
  • open spaces in sea ice that allow animals to exist, which can last from a season to decades or longer, called “polynyas”
  • places where species are protected from hunting or harvesting, called “hunting refugia”

and many other types of refugia. In the fossil record, we have evidence of species surviving global crises (like those that cause mass extinctions) in refugia. Refugia also exist from present-day environmental and climate crises, although we are often responsible for creating refugia: marine protected areas, national parks, wildlife refuges, wilderness areas, native grassland preserves, etc.

As I said earlier, the concept of refugia isn’t new. Darwin didn’t use the term, but he considered the refugial concept when he considered ecosystems during the last ice age. He imagined that as the ice sheets grew and covered the northern hemisphere, ecosystems shifted southward away from the glaciers, almost moving as a whole. He envisioned intact forests with their trees, plants, and animals moving as one as the forests slowly spread to the south1. This type of species’ movement in response to escaping a crisis is called a “range shift,” when the geographic boundaries of species’ or populations’ distributions move to accommodate factors like cold temperatures, changes in precipitation, or altered resources. The first scientist to use the term “refugia” in print was Packard in 1886. Similar to Darwin’s ideas, he used the term to denote those places where species survived the last ice age in areas south of the continental glaciers.

So why study refugia in the fossil record? First, and probably most important, refugia were places where species successfully survived global crises, and if we can understand the properties that made ancient refugia successful, we can apply that knowledge to modern, man-made refugia, like marine protected areas and wildlife preserves. Second, we can also see where refugia formed but then failed, and study why the failure happened: was the cause behind the failure something we have to consider (like no place for species to move when a refugium becomes inhospitable) or is the failure caused by we don’t need to worry about (like the uplift, or tectonic raising, of the sea containing the refugium existed, so that eventually the sea drained and the species either left or went extinct3)?

If you are familiar with the fossil record, you know that many of our studies delve into rocks that represent sediments and ecosystems that lived millions and even hundreds of millions of years in the past. And might you realize that the further back in time you go, the more incomplete the fossil record is, because not all sediments were preserved and the fossil-bearing rocks that were formed from those sediments have had more chances to erode. So then, you probably wonder, why bother with a fossil record so old and so incomplete? Why not study today’s animals and plants to figure out from them the refugia they need?

One of the major problems with studying present-day ecosystems and species is that we usually lack data on what those systems and habitats were like under pristine conditions. Usually we study a species or an ecosystem already in crisis, in other words, in the midst of change or decline. If we study the recent fossil record – thousands of years, tens of thousands of years – we can get baseline conditions for those ecosystems and species. That gives us information on the conditions that the species and ecosystems thrive in, but it doesn’t give us the type knowledge we need for refugia.

This is where the older, deep-time fossil record comes in. Life on Earth has survived quite a few mass extinctions and even more climate crises, including many ice ages and greenhouse climates. (In fact, the normal state of the Earth is change, not stasis! But more on that in the future.) Perhaps we can’t go back millions of years to study a particular oyster or a special reef ecosystem, but we can look at ancient species and ecosystems in refugia and understand the general principles of what made those refugia successful, or what it is that species require of refugia. And we can compare refugia through space and time to understand those principles that are generalizable for refugia success, no matter the species, system, or habitat. Those sorts of things are directly applicable to creating present-day and future protected areas. We can’t answer specifically “how many square kilometres should be set aside for Marine Protected Area X for this particular fish species?” but we can answer questions like, “which is better, one big refugium or several little, connected protected areas?” or “how long will we need marine protected areas in the future?”

So, that’s the introduction to refugia. Stay tuned for some of the questions I’ve asked the fossil record about refugia, and some of the answers I have investigated!4


1 Today, we know that Darwin’s ideal range shift of entire ecosystems never existed. Instead, the Northern Hemisphere was a mosaic or mixture of species from many ecosystems, and animals moved among that mosaic. If you were to go back in time, you wouldn’t have quite the same forest then as you’d know today; it might have a mixture of northern and southern trees too. Not to mention the mammoths and giant sloths moving among those trees!

Furthermore, my research shows that ecosystems almost never move together as a whole into refugia. In the fossil record, I’ve found only two examples of refugia where the ecosystem, or a particular portion of the ecosystem, stayed intact. Most refugia instead were a mix of species that were present when the refugia formed and other species that moved to refugia from elsewhere in the world, plus a few species that either moved out of the refugia or went extinct. More on this later.

His discussion on southward movement of ecosystems is found in the Origin of Species:

Darwin, C. (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London

2 Packard, A.S. (1886) Geological extinction and some of its apparent causes. American Naturalist, vol. 20, pp. 29–40

3 This was an interesting example from the Serpukhovian mass extinction (also known as the Late Mississippian or Early Carboniferous mass extinction, around 323 million years ago) of a failed refugium. The Serpukhovian mass extinction was the seventh greatest of all time (see Powell, 2008 for references and a good discussion of this extinction). This title has since been challenged, suggesting that it should be one of the “Big 5,” placing the extinction at fifth greatest, knocking the Late Ordovician mass extinction out of the top 5 (McGehee et al., 2012).

During and after the extinction, the Tindouf Basin served as a refugium for many species. The Tindouf Basin was a sea connected to the remnants of the ancient Rheic Ocean. The sediments of the Tindouf Basin – including its fossil record – are now preserved in rocks in mountains crossing the Morocco and Algeria border. The refugium itself was initially successful, and the ecosystem in the basin remained intact and thrived, because it was isolated from the crisis in the rest of the world by ocean currents.

(Here is an example of ocean currents helping to isolate an area for a refugium. Thus, perhaps the study of ocean currents would also help locate good areas for present-day and future refugia, protecting animals and seaweeds from warming, ocean acidification, pollution, etc.)

The Tindouf Basin refugium ultimately failed because it was in a zone of active tectonism. That means that the tectonic plates were actively moving in that region, and as they moved, the Tindouf Basin slowly raised upward so that eventually the ocean water drained from the sea and the animals perished. This process of raising – or uplift – happened over millions of years, so it’s not something we would have to be concerned with in the refugia we make! None of the species escaped that refugium, so it became what I call a “refugial trap” – a place that starts out as a good haven for species to survive, but then ultimately fails and turns deadly.


Cozar, P., Vachard, D., Somerville, I.D., Medina-Varia, P., Rodriguez, S., Said, I. (2014) The Tindouf Basin, a marine refuge during the Serpukhovian (Carboniferous) mass extinction in the northwestern Gondwana platform. Palaeogeography, Palaeoecology, Palaeoclimatology vol. 394, pp. 12-28

Powell, M.G. (2008) Timing and selectivity of the Late Mississippian mass extinction of brachiopod genera from the Central Appalachian Basin. Palaios, vol. 23, pp. 525-534.

McGehee, Jr., G.R., Sheehan, P.M., Bottjer, D.J., Droser, M.L. (2012) Ecological ranking of Phanerozoic biodiversity crises: the Serpukhovian (Early Carboniferous) had a greater ecological impact than the end-Ordovician. Geology vol. 40, pp. 147-150.

4 Schneider, C.L. (2018) Refugia past, present, and future: lessons from ancient geologic crises for modern marine ecosystem conservation. In Tyler, C.L. and Schneider, C.L., eds., Conservation Paleobiology, pp. 163-208