As someone who has grown up in the era of An Inconvenient Truth and World Wildlife Fund infomercials about endangered pandas and the myriad ways in which they can be saved, I have heard a lot about the world’s staggering beauty and diversity. Studying the biological sciences for four years at Hanover certainly helped with that, but even before I had set foot on a college campus I was well aware of the importance of maintaining the entire spectrum of life here on Earth. Whether from National Geographic specials or the ridiculously beautiful series Planet Earth, there is a considerable amount of talk about how valuable all of Earth’s species really are.
What’s interesting then, is how little we really know about the diversity of life on our planet. Estimates as to the total number of species on Earth are often immensely disparate (ranging from 3 million species to 100 million) and heatedly contested, with the basic assumptions they rest upon being often shaky and sometimes ludicrous. A recent study published through the Public Library of Science proposes a new method for reaching an accurate and reliable estimate for the number of species we live with, though, and an estimate that may just carry some weight.
According to the paper’s new findings, there are about 8.7 million eukaryotic species here on Earth, with about 2.2 million of those being marine species (the numbers differ for prokaryotes, whose cellular life is considerably different from ours, and whose species definition is much more liberal). Now we have only described about 1.2 million species thus far in around 250 years of concerted effort, meaning that there is still a whopping 86% of species left to be found on Earth, which includes about 91% of all marine species that still elude our best scientific explorations.
The study used a remarkable new statistical approach which looked at the rate of discovery for new taxonomic groups and used a regression model (along with other assorted statistical feats I don’t have the expertise to suitably explain here) to determine how many total groups there would be at each taxonomic level (i.e. phyla, genera, species, etc.). They applied their model to taxa they considered to be well-researched thus far and found that their model, when applied to a higher taxonomic level, reached an estimate that was very close to the number of species already discovered. This, along with the solid grounding in statistical modeling their method has, supports their new method as a very reliable means for estimating biodiversity, and subsequently affirms their estimate as a fairly accurate representation of the breadth of life on Earth.
Now there are still a number of issues to be taken into account here, and it should be noted that the authors of the paper do a good job of acknowledging them. These issues range from the amount of effort currently being put into discovering new species (which would affect their rates of discovery, and so greatly affect the group’s estimates) to the completeness of current taxonomic knowledge, or basically how much we can rely on what we already know to give us a complete picture from which we can extrapolate anything meaningful. From my perspective, perhaps the most important issue that needs to be addressed with these types of study is the definition of a species. Different disciplines use different criteria to determine where to draw the lines between species. With the advent of genetic sequencing and its increased accessibility, researchers are finding more and more that what were once thought to be separate species are in fact very closely related and vice versa. Phylogenetic research, as this field has been dubbed, is changing the face of our taxonomic structure still, and the dust doesn’t seem terribly close to settling. As such, there is a fair amount of instability to be kept in mind when we talk about establishing estimates of species by extrapolating from current numbers. We aren’t even sure how many species we really know about, much less how many we don’t.
I am not pointing this out to say that these types of studies are in vain or pointless or anything along those lines. Although I can’t say I don’t get frustrated at the number of wild conclusions drawn from incomplete data. I mean 3 million to 100 million? That’s a hell of a spread. I think we can do better. And we need to. Having a reliable estimate of Earth’s biodiversity is necessary to understand what we are losing as human society progresses, even if that estimate can’t be infallible. At our current level of knowledge and rate of species loss, there will be a distressingly large number of species that will undoubtedly go extinct before we even know they exist. There will be, and probably already have been, whole lineages of life on Earth, histories of evolution and adaptation as rich as our own, inextricably tied to our existence and the world we inhabit, that we will snuff out without ever having known they lived. That seems like a big deal to me. And if what we need to get people to understand why this should matter is an estimate with a standard deviation of 1.3 million, then that’s what we’ll get.