The scene was typical for an international gathering of governments: bureaucrats, sat behind nameplates and speaking through interpreters. But the less than typical result of the votes cast at this 1992 Convention in Nairobi was an ambitious international agreement to “halt the loss of biodiversity”.
Exceptional rates of species extinctions, rapid habitat loss, and increasingly variable climate meant the deck were stacked against the newly formed Convention on Biological Diversity achieving its goal to protect the biodiversity of life on Earth. Realising this, the convention established the Global Strategy for Plant Conservation as a taxonomic barometer of progress towards their broader goals.
Plants were a natural choice. They are exceptionally important. Consider, for a moment, the food you eat, clothes you wear, medicine that makes you well, materials used to construct your house, not to mention those that brighten its interior and garden. They are the critical infrastructure that supports the diversity of life on earth.
The first challenge the set was to complete “a widely accessible working list of all known plant species”. By 2010, the major botanic gardens announced they had, through The Plant List, fulfilled the first goal. In what became known as the Aichi Targets, the convention laid down - and governments supported - 20 further aims by 2020.
Among others aims, the body has a target to protect 17% of the earth’s terrestrial surface: a pragmatic, ambitious, but still achievable target, given the proportion of global land already nominally protected stands at around 13%. Another target is to conserve 60% of plant species in national parks and other protected areas. Are these compatible goals, is there a way which they can be achieved together? In our recent paper in the journal Science, we sought to show that, while there are major impediments, it is possible.
To us, the question posed an optimisation problem: how to accumulate the most species into a given amount of area. We started with the extraordinary database assembled by the Royal Botanical Gardens at Kew, England, which holds up-to-date details of 109,000 species of plants, including their distribution across more than 350 separate countries and regions. We applied to this dataset various computational methods, greedy algorithms, which try to create a full set of objects from the smallest number of constituent parts, and genetic algorithms, which mimic evolutionary trends to provide optimal solutions to problems.
The result was an unprecedented illustration of how species are concentrated in some of the world’s most biodiverse places. Our algorithm worked by choosing the region with the most species in the least area. Then it continually added regions that, when added to the regions already chosen, brought together the maximum number of species in the least amount of area. We continued until all species were included in our set of regions. This “greedy” approach quickly showed that it was possible to include at least a portion of 86% of all species’ distribution within just 17% of land surface. This makes one inclined to think that the CBD’s ambitions won’t be so hard to realise after all.
But more often conservation biologists are concerned with where endemic species are concentrated. Endemism describes species that live within a region and nowhere else - Madagascar for example, is the only place on the planet where many creatures such as lemurs and loris are found. So we took another measure, and were able to piece together regions of the world that contained the entire distribution of 67% of the world’s plant species within the 17% target.
These regions are the priorities for conservation. Tropical and subtropical islands, moist tropical and subtropical forests (especially those on mountains such as the Ecuadorian Andes), and Mediterranean ecosystems are examples. With the set of regions in hand, we could then ask if our results are representative of other taxonomic groups. An overwhelming majority of distributions of bird species (89%), amphibians (80%), and mammals (74%) overlapped with the regions we identified as important.
Finally, we asked if the regions that we identified as important were being protected preferentially, as doing so would seem to be the most optimal conservation strategy. But in fact the regions we identified were protected just barely (14%) above the global average of 13%.
Our results are just the beginning of determining how to best protect the most important biodiversity hotspots in the world. We think of this work as “strategic” — giving insights into the regions where conservation actions must concentrate. But these are too large - we have to map species on much finer scales to offer recommendations for practical conservation actions. This is something my co-authors Stuart Pimm and Clinton Jenkins are working on through the conservation NGO SavingSpecies.
Our results highlight how national implementation strategies make it difficult to efficiently achieve international conservation goals. Protecting the most species in the least area would place disproportionate pressure on just a few of the world’s countries - as has been seen recently, in the case of the pressure put on Ecuador not to drill for oil in Yasuni National Park. Is this fair? Resolving this issue will likely require many more meetings, and many more bureaucrats. I’m not sure we have the time to wait.