If the dinosaurs were still around today, a massive asteroid hurtling in from outer space might not be necessary to cause their extinction. An increasing number of scientists believe that human activities are causing environmental changes that may drive the rates of species extinction just as high as during prehistoric mass extinctions. It is uncertain how losses of earth’s biological diversity could affect the dependability of the many services the environment provides, so using science to forecast possible outcomes has introduced itself as a new priority for many people. The USGS, for instance, states in its future science goals that, “achieving a sustainable future is dependent upon science to aid in understanding, quantifying, and predicting how ecosystem functions will change as a consequence of new patterns of resource and land use.” This perspective represents a new role for ecologists in society. Governments and agencies are frequently turning to them for background and advice regarding decisions on policies that address biodiversity loss (the loss of many different species) and other human driven environmental change.

However, “many ecologists are leery of prediction because they feel that there is much uncertainty about their models, and that the predictions are going to be wrong,” explained Jim Clark of Duke University after a consortium calling for ecologists and policy makers to work together more closely. And, when it comes to biodiversity - one factor that some research suggests could be very useful for making predictions – policymakers find an added dimension of difficulty. Amongst ecologists there has been controversy about to what extent and in which situations biodiversity really matters. There is also conflict about what aspect of biodiversity actually affects how an ecosystem functions. The debate has even been politically charged at times, perhaps - as some ecologists like Stanford’s Harold Mooney have pointed out - because of the pressure from policymakers for timely answers. Shahid Naeem of Columbia University claims that, in his experience, if your results support the ideas of conservationists you may be accused of being biased and manipulated by politics. You have an “extra burden” he believes.

Dr. Naeem has been conducting biodiversity experiments at Cedar Creek Natural History Area in southern Minnesota since the mid 1990s. Though most of his results seem to support a particular viewpoint, he considers himself a voice of moderation in the debate, generally looking for common ground. Laughing, he says he had never had the opportunity to become an environmentalist anyways, since he grew up in the middle of New York city where nature was something you went looking for on a day trip. “I’m not an environmentalist, I speak as a scientist,” he asserts. However, in his opinion the biodiversity controversy has prevented important policies from being made, which he finds regrettable. So, recently he, along with Justin Wright of Cornell University, turned a synthesis mindset toward looking at many major studies on how biodiversity affects the way ecosystems function. They reviewed how the studies were performed and what the results were. Using this overview they assessed the challenge of expanding the scope of biodiversity experiments to address the large-scale issues that interest policymakers and to make the studies more useful for predicting environmental changes.

A couple members of Dr. Naeem’s team heading out to collect data.	A sapling has a different function from a full-grown tree, even of the same species.

Studies suggest, the review points out, that in trying to predict how an ecosystem will function by looking at its biodiversity, it may be less important to consider how many different species are there than to consider what jobs are performed by these species and to group them according to the roles they play. This is an aspect of biodiversity called functional diversity. If this is the case, the review continues, ecologists must come to an agreement on an effective and reliable way to divide species according to how they function within an ecosystem.

Sounds simple? Well, it may not be. Sometimes species have particular traits that they only exhibit when the environmental conditions are right, sometimes the role a species plays changes throughout its lifetime, and there are still other complications besides these. The review suggests that one of the most important divisions to make in the functions species perform would be to separate ‘effect traits’ and ‘response traits’. That is to say, does a trait affect a function of the ecosystem (nitrogen-fixing bacteria affect how much nitrogen is in the soil, for example), or does the trait determine how a species will respond if there is a change in the environment (like a jack-pine that responds to a forest fire by releasing its seeds)?

Biogeography: physical differences within an ecosystem determine how likely it would be to find a particular species there.	Wielding some tools for collecting data on diversity.

Dr. Naeem and Dr. Wright propose a framework for predicting how a function that an ecosystem performs might change in response to a change in the environment. The first step, they say, is to determine what species make up the ecosystem. This would probably involve making many lists. Then, they believe a scientist should take into consideration the differences in the physical characteristics of parts of the ecosystem, such as which portions receive the most rainfall, which are higher in altitude, or which have recently been affected by a fire. This is called the biogeography. Using knowledge of the biogeography, they think that a scientist could come up with mathematical equations that would predict how rare or common the different species in the ecosystem are.

After this, the scientist could decide which environmental change she wishes to consider, which ecosystem function she thinks might be affected, and which species traits seem important to this ecosystem function. For example, if she were interested in greenhouse gasses, she might ask in what way higher carbon dioxide levels (the change) will affect how much can grow in an ecosystem (the function). In this case, some traits she may believe important could be: how the plants in the ecosystem use CO2, how fast plants are decomposed, how well they use nutrients in the soil and the water, and so on. Once the scientist has made all these decisions, the framework suggests that she use existing mathematical models to make predictions.
Computer science and statistics are often used to create models, but not all ecologists believe they are the best method for prediction. Danish ecologist Claus Beier has disagreed with using mathematical modeling to predict how ecosystems will respond to stress. In a commentary he wrote, “I do not believe that any model alone or any single factor experiment will provide the answers we need.” Even though they are very costly and difficult to run, he believes it is necessary to perform massive experiments, on the scale of whole ecosystems, that take into account many different environmental factors all at once. “Do we have any alternative?” he wonders. “I doubt it.”

Dr. Naeem finds that statistics and experiments both have positive and negative sides. “We use statistics to give a best guess,” he explains. “If you do an experiment, you know what’s going on.” In the experimental vein, he is currently the principle coordinator of a five-year-long National Science Foundation funded project known as BioMERGE. BioMERGE is a group of scientists, from different disciplines of biology, who are working together on large-scale biodiversity research across diverse ecosystems including prairie, marine systems, grasslands, and the Amazonian rainforest.

Biodiversity research, out in the field.	From the Biomerge collaborative research project.

Until now, a majority of biodiversity experiments have been done using grassland plant species because they grow quickly and they won’t run away from the researchers, Dr. Naeem says. He hopes that BioMERGE will yield findings that can be applied to a wide variety of systems and that will be more suitable for use in large-scale prediction. Some results have already been published from an estuary in Ireland, but the rainforest team has encountered a difficulty. They have so many species to work with their computer keeps crashing!

Even if new methods and experiments make it more viable to predict how ecosystems will respond to human-driven changes in the environment, it is important to remember that predictions are not certainties. Until there is certainty Dr. Naeem advocates protecting biodiversity as much as possible. “After all,” he says, “unless you believe in things like Jurassic Park, once a species is lost you can’t get it back.”