Counting Parrots by Counting Bats (or Snakes, or Frogs)

Or, all about cross-taxon congruence.

My co-authors and I recently published a scientific paper titled “Cross-taxon congruence of terrestrial vertebrates across Philippine Key Biodiversity Areas”. This is my summary of it, written using more everyday terms. Link to the paper here. Alternative link here.

From left to right: a guaiabero (Bolbopsittacus lunulatus), a lesser short-nosed fruit bat (Cynopterus brachyotis), and a vine snake (Ahaetulla prasina). Photos © Rio Constantino

Imagine you’re a biologist trying to count how many vertebrate species are found in a forest (remember that vertebrates are animals with backbones: birds, reptiles, amphibians, and mammals like you and me). You’re doing this because it’s one way of gauging the forest’s health. In general, the more species you find in a jungle, the healthier it should be.

Upon reflection, you realize that the job involves checking for the presence (or absence!) of possibly hundreds of different kinds of animals. It’s a difficult task, requiring considerable investments of both time and manpower. But unfortunately, like many biologists, you don’t have lots of funding. You don’t have the budget to hire ornithologists, mammologists, and other experts to help out in the field. You may not even have the money to pay for all the supplies needed for extended fieldwork.

The job needs doing. But how?

What if, to make the most of limited resources, you focused on birds instead? Rather than trying to account for every species in the area, you just count those with feathers, under the assumption that if you find lots of parrots and eagles and orioles and the like, there will also be plenty of other creatures with fur and scales and wet slimy skin.

If you decide to do so, congrats! You’re using what we call surrogates, where you use one relatively easy to study metric as a stand-in for other aspects of biodiversity which are more difficult to measure. Biologists rely on surrogates a lot, given how overwhelmingly complex nature can be.

It's important to remember, however, that when we use surrogates, we also assume that they actually correlate with whatever it is they’re supposed to replace; that if you find lots of birds in the forest, you’ll similarly find plenty of bats. In more technical terms, we’re assuming that one metric is congruent with the other. But is this assumption really correct?

A map of all terrestrial KBAs in the Philippines. Data from Ambal et. al. (2012).

That’s what our recently published paper, titled Cross-taxon congruence of terrestrial vertebrates across Philippine Key Biodiversity Areas, wanted to test out, using decades of data collected by generations of scientists across more than a hundred Key Biodiversity Areas (KBAs) in the Philippines. KBAs are places identified by scientists that have lots of biodiversity, and are often important targets for conservation.

We made a tally of all the different vertebrate species found in each KBA, while also identifying which species are threatened with extinction, and which are endemic. A species is endemic if it's only found in the Philippines. We then investigated whether any consistent patterns would emerge if you took all those tallies and compared them to each other.

And it appears that the assumptions behind the use of surrogates hold true. Species richness is indeed congruent across all vertebrate orders, at least in the Philippines. It’s not a mathematically exact correlation, and you can’t calculate that just because there are three types of frog in the forest there must also be six different kinds of mice. But if you do find that a forest is greatly species rich for one type of vertebrate, then you can reasonably conclude that it must also be species rich in all the rest.

We made several other findings which have important implications for conservation. First, reptiles performed the best as surrogates compared to other taxa, while birds performed the weakest. This is significant, because birds have historically been the most used surrogate by biologists while crafting conservation plans, as they are often the easiest animals to observe in the field.

Second, the distributions of endemic species are also congruent across vertebrates. In other words, places with plenty of species in total, will likely also have lots of species that can be found nowhere else in the world.

And lastly, the proportion of threatened species is not congruent with species richness. To put it plainly, just because a forest has lots of threatened eagles, deer, and the like, doesn’t mean it will have lots of species in total, period.

Many protected areas have been put in place specifically to safeguard the habitats of threatened species. However, it seems that in doing so we may be inadvertently leaving out many other species that are not yet on the brink of extinction, yet may still be slowly approaching it, given the pace at which many ecosystems are being destroyed around the world. If there’s any lesson to be gained from this, it’s that we should be cautious of focusing on one slice of the world’s biodiversity at the expense of all the rest.

All that said however, and going back to the scenario at the beginning of this article – if you’re a cash-strapped biologist trying to monitor a forest’s entire fauna, rejoice! You can safely use surrogates instead. (Hopefully the next round of funding will give you an opportunity for more detailed analyses).