Undetected extinctions – the dark figures of biodiversity loss
Conservation status and extinctions of many charismatic species are well monitored and documented by scientists. On the global level, the IUCN Red List provides the most comprehensive assessment of species extinction risks. In 2015, the list of globally threatened species comprised more than 23000 species while 903 species have been assessed as extinct or extinct in the wild. However, the documented extinctions are always a very incomplete list. Firstly, only a small fraction of known species is actually assessed; in 2015 merely 5% of all described species were covered by the IUCN Red List. Secondly, a huge number of species remain still undiscovered and even go extinct before scientists take any notice of their existence. Due to this dark figure of extinctions it is very difficult to produce an accurate estimate of how many species have been lost globally or in a particular area as a result of human activities.
The interesting problem of undetected extinctions was the topic of a seminar given by our invited speaker Ryan Chisholm, a theoretical ecologist from the National University of Singapore. He and his colleagues developed a model to estimate how many undiscovered species have been driven to extinction in a given time and area. Specifically, they studied extinctions of breeding birds in the small city-state of Singapore. Located on a densely populated, highly urbanised island, Singapore has lost most of its tropical forests in the past 200 years. Not only have tigers been driven to extinction but also about 30 % of the bird fauna known to be native to the city-state have fallen victim to the rapid development of the city. Chisholm and his colleagues wanted to find an answer to the question of how many bird species went extinct, without any knowledge that they had ever lived on the island. Previously, another research group had suggested that the total number of undetected extinctions in Singapore could comprise up to 129 bird species. This estimate, however, was based on the questionable assumption that the small island of Singapore originally harboured the same number of bird species as the adjacent 200 times larger Peninsular Malaysia.
During our workshop we got to explore and try out Chisholm’s model, which only requires historical records of discovery and extinctions of known species as an input – data that have been thoroughly recorded in Singapore over the last 200 years. From the data, the model infers detection and extinction rates that vary with time. This variability is a strength of the model since neither research nor habitat destruction have been carried out with constant efforts in Singapore. Chisholm explained that the model assumes that extinctions of detected and undetected species occur at the same pace and that at the present time no more undetected native bird species are breeding anywhere hidden in Singapore’s urban jungles. This latter assumption seems reasonable considering the high enthusiasm of Singaporean birdwatchers. And after all, there haven’t been any detections of new bird species for 20 years. More controversial, we thought, was the former assumption: Do detected and undetected species really disappear at the same rate? Some of us argued that undetected species are usually small and inconspicuous and therefore not as much affected by hunting as known ones, which are usually bigger and more conspicuous and thus at higher risk. Others pointed out that conservation programs can only specifically target species that have already been discovered – how can you protect something that you do not even know of? This would mean that undetected species are more susceptible to extinctions. We realised that there is probably no general answer to this question; the answer probably depends on the group of species, area and threat studied. At least in the case of Singaporean birds the assumption of equal extinction rates seems to hold. In their paper, Chisholm et al. justify their assumption by showing that extinction rates of recently discovered bird species do not differ from those that have been long known.
Their estimate of undiscovered extinctions, which we also reproduced during the workshop, was about 10 species in the last 200 years. This number is about an order of magnitude lower than the alarming figure the other renowned researchers had previously come up with. We learned from Chisholm’s talk that the correct quantification of undetected extinctions is a crucial step for an unbiased estimation of human driven extinctions in the past. This information is very important if we want to predict how biodiversity will be affected by human activities in future. Singapore, as a highly urbanised and deforested island, can serve as an example for other regions in southeast Asia that are expected to develop rapidly in the 21st century. The fact that Singapore still harbours a high proportion of its native bird species (a number that had been highly overestimated previously), suggests that the number of predicted extinctions is probably highly inflated for many places in southeast Asia. During his workshop Chisholm did not only impart a lot of knowledge about ecological concepts and theories (most importantly Species-Area-curves) and challenged our R skills, but he also strongly encouraged us to think critically and double-check widely stated dramatic facts and numbers. It was very inspiring seminar on a highly interesting topic!
Chisholm’s work on the undetected bird extinctions in Singapore was published in the following article in the journal “Conservation Biology”:
Chisholm, R. A., Giam, X., Sadanandan, K. R., Fung, T. and Rheindt, F. E. (2016), A robust nonparametric method for quantifying undetected extinctions. Conservation Biology. doi: 10.1111/cobi.12640
E. Thore, IMAE 2015-2017