By Emily Nicholson (This article was first published in the October 2012 issue of Decision Point, The Monthly Magazine of the Environmental Decisions Group)
Biodiversity around the world has undergone dramatic declines over the last century. In response to this, the Convention on Biological Diversity (CBD) was signed in 1990s by 168 countries, including Australia. Each country committed itself to bringing about a significant reduction in the rate of loss of biodiversity by 2010. A range of indicators reveals the global community has failed to meet this goal. At the tenth meeting of the Conference of the Parties, held in Nagoya in 2010, a new set of targets was adopted. These are known as the Aichi Targets (Nagoya is located in Japan’s Aichi Prefecture) and they aim to produce more concrete results. As part of this, the targets are accompanied by a suite of indicators to measure progress.
Two key questions arise from the CBD targets and indicators:
- Can the targets be achieved, and with which policies?
- Are the CBD indicators capable of signalling changes in biodiversity as a result of these policies?
We examined these two questions in a recent analysis (Nicholson et al. 2012). We argued that policies and indicators should be evaluated before being implemented, to determine if the policies and indicators being proposed will actually produce the results that we’re after. We tested our ideas with two case studies where we used models to project forwards the potential impacts of the policies on biodiversity and the indicators used to track it. The first looked at protected areas in Africa. The second examined the impacts of bottom trawling on marine ecosystems.
Expanding ‘paper’ parks
The case study on African protected areas focussed on Aichi Target 11. This target calls for at least 17% of terrestrial areas to be placed in effective and well managed protected areas by 2020. Unfortunately, protected area coverage alone gives little information on how well they are performing in protecting biodiversity.
Recent research has found that the population abundance of large mammals in 78 African protected areas declined on average by 59% between 1970 and 2005 (Craigie et al. 2010). We compared different policy options, modelling the effects of policy on the fate of 53 mammal species, including lions, zebras, impala and African wild dogs. We compared: 1. doing nothing; 2. increasing the reserve system to 17% (but with the current population declines); 3. halting declines in the current reserve system; and 4. increasing the reserve system to 17% AND stopping declines. We fed the population trends into the Red List Index, an index of extinction risk for species of plants and animals based on the IUCN Red List status of species (See Decision Point #59, p8).
“We need to determine if the policies and indicators being proposed will actually produce the results that we’re after.”
We found that the Red List Index could differentiate between the policy options. This is handy as it suggests we could pick up positive changes due to such policies when we are monitoring biodiversity using the index. It is also shows that the Red List Index can provide a useful way to draw together the outputs of scenario modelling for some policies to compare options, particularly when presenting them to decision-makers.
Our second main finding was that management effectiveness was a much better option for improving the fate of the 53 mammal species than expanding protected areas (without changing the way we manage them). In fact, expanding to 17% provided little benefit over doing nothing, while expanding and improving management was only a bit better than just improving management of the current protected area network. Keep in mind that our modelling was pretty simple (just projecting current trends), and we assumed that all the species responded similarly – their population trends all improved with effective management, and there were no interactions (e.g., lions and zebras both went up in better managed reserves; in the real world, however, if the lions increased, the number of zebras would probably decrease!).
Scraping the bottom
In our second case study, we looked at stopping bottom trawling – one of the most destructive fishing methods currently used. This fits with the very general CBD target on fisheries management, Aichi Target 6: “By 2020 all fish and invertebrate stocks and aquatic plants are managed and harvested sustainably, legally and applying ecosystem based approaches…”.
We modelled the effects of halting or halving the amount of bottom trawling in six ocean systems using 10 ecosystem models. We applied a modelling framework called Ecopath with Ecosim which models the interactions of food webs. We examined the effects of the policy as it trickled down through the food web: when trawling is halted, the fish species that are the main target of the fishery will increase. Their main food species will decrease in number, as will species they compete with for food or other resources. By contrast, anything that eats targeted fish will have more food available and will increase in numbers too. We fed the changes in biomass from the scenarios into the Living Planet Index, another key CBD indicator, which measures changes in abundance in vertebrate species.
What did we discover? Mainly that it’s hard to find consistent and simple results using such complex models with a single indicator. The big problem was that not all species in the ecosystem models are used by the Living Planet Index, which relies on available time series data collected for other purposes.
For example, people like birds, and therefore there is a lot of bird data in the Living Planet Index – the majority in the marine systems we looked at. This means that our results in many regions were driven by what happened to the birds, and for some regions birds actually decreased in number. This was particularly the case in the Mediterranean and the North Sea, where birds eat the discards of trawling – stop the trawling and the birds lose a big chunk (up to 30%!) of their food source.
The focus on birds in the indicator was exacerbated by the fact that the Ecopath foodweb models we used were not built to assess how seabirds are faring–they were mostly aimed at the assessing fisheries. In all cases, all seabird species were modelled as a single group. The differences between species, the threats they each face, and the impacts of policies on them were ignored, partly because of the reasons for which the models were built, but also because every complexity cannot be included in already very complex models.
Other than birds, the rest of the species that make up the Living Planet Index are a rather patchy lot. In some regions the key beneficiaries of stopping bottom trawling, such as skates and rays, were simply not included in the Living Planet Index, so there was no bump to the index from the policy, while species who were recipients of knock-on effects, whether positive or negative, were represented to varying degrees. As a result, the index was all over the place depending on which species were included, and not consistent with the overall changes in biodiversity, which were generally positive as a result of reducing bottom trawling.
Our main conclusion is that the complexity of the responses to a policy such as stopping bottom trawling can’t be readily reflected in one indicator, especially if it doesn’t contain a very representative set of species. In other words, more indicators are needed, and these indicators need to target particular aspects of biodiversity that we want to keep an eye on.
Surprisingly, although biodiversity indicators are used to show how biodiversity is trending, they are rarely tested to see if those trends are telling the story we think they are telling. Within fisheries science, the performance of indicators is tested more thoroughly. This is something we in conservation could learn from.
The risk of being ineffectual
The Aichi Targets are a good start in trying to reduce declines in biodiversity, but without testing whether the policy options that stem from the targets work, they run the risk of being as ineffectual as the CBD goal to slow the loss of biodiversity by 2010.
Our framework for testing the impacts of policy and indicators of change is also very relevant in other local and global policy situations, such as the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) (see Decision Point #61, p12), where scenario modelling will be critical to developing and testing policy.
Expanding ‘paper parks’ doesn’t do much to stop declines in key species, just as indicators won’t tell you much about stopping bottom trawling if the right data are not included in them. To be useful and relevant, conservation scientists must make testable predictions about the impact of global policy on biodiversity to ensure that targets actually achieve anything.