In one of those rare moments of perusing the latest ecological literature, I stumbled across an absolute gem, and one that has huge conservation implications. Now, I’m really no expert in this particular area of ecology, but I dare say the paper I’m about to introduce should have been published in Nature or Science (I suspect it was submitted to at least one of these journals first). It was still published in an extremely high-impact journal in ecology though – the Journal of Ecology produced by the British Ecological Society (and one in which I too have had the honour of publishing an article).
Before I get into specifics, I have to say that one thing we conservation biologists tend to bang on about is that MORE SPECIES = BETTER, regardless of the ecosystem in question. We tend to value species richness as the gold standard of ecosystem ‘health’ and ‘resilience’, whether or not there is strong empirical evidence in support. It’s as if the more-is-better mantra strikes an intuitive chord and must, by all that’s ecologically right in the world, be true.
Of course, measuring what is ‘better’ is a difficult task, especially when we are talking about complex ecosystems comprising thousands, if not millions, of species. Does ‘better’ refer to the most temporally stable, the most genetically diverse, the most resilient to perturbation, or the provider of the greatest number of functions and hence, ecosystem services?
It’s up to you, but all these things tend to be difficult to measure for a large number of species and over time scales of sufficient duration to measure change. So the default for plants (i.e., the structural framework of almost all ecosystems) I guess has come down to a simpler measure of success – ‘productivity’. This essentially means how much biomass is produced per unit area/volume per time step. It’s not a great metric, but it’s probably one of the more readily quantifiable indices.
Enter the so-called ‘diversity-productivity relationship’, or ‘DPR’, which predicts that higher plant species diversity should engender higher net productivity (otherwise known as the ‘net biodiversity effect’).
Nice idea, and one that is especially attractive in this day of carbon accounting (forest carbon sequestration as offsets to industrial and transport greenhouse gas production). If true, the net biodiversity effect is adequate justification for maximising species diversity in carbon plantings such that both carbon sequestration and biodiversity value are maximised – the best bang for your planting buck.
The problem is that it’s not even close to being a theory in ecology, let alone a law. In fact, the hypothesis has very mixed empirical support, and the majority of it is in manipulated and simplified grasslands, with no net effect, or even negative relationships, reported for natural forest stands. Not a good sell really if you’re trying to convince a policy maker that more species are better to maximise productivity, and hence, carbon uptake.
Enter Yu Zhang and colleagues and their recently published (online) paper: Forest productivity increases with evenness, species richness and trait variation: a global meta-analysis. Now, I’m a big fan of meta-analyses to answer big questions; there’s nothing like a lot of disparate studies collated to provide insight into broad-scale pattern (e.g., see our recent meta-analysis on the value of primary forests for tropical biodiversity published in Nature last year).
And that’s exactly what they did – Zhang and colleagues collated diversity-productivity data from 54 forest studies to determine the overall ‘average’ direction of the diversity-productivity relationship. Using boosted-regression trees (and slick way to deal with complex multi-variate data), their overall conclusion was that ‘polycultures’ (i.e., many species) were more productive than ‘monocultures’ (single-species stands). So much for the DPR debate in forest ecosystems.
More interestingly perhaps was that the biodiversity metric used dictated the strength of effect found. Even though there was a positive richness relationship (more species = higher productivity), it plateaued after a certain threshold number of species. However, when using ‘evenness’ as the measure of ‘diversity’ (i.e., a metric which includes relative abundance of each species in system), stands that were more even (i.e., not dominated by a few common species) were more productive, and this metric explained a much higher component of the variance in productivity than richness alone. This means that a simple list of species doesn’t really indicate the full potential of the stand because of things like functional redundancy. Finally, they found no evidence for a deviation of the relationship among biomes, suggesting that the effect is global and real.
Now there are a lot of fine details in the analysis that complicate matters, but I think the take-home message is clear. This is a hugely important finding and one that all conservation ecologists should be able to cite when justifying the essential role of diversity in ecosystem function.