Human population size: speeding cars can’t stop quickly

28 10 2014

Stop breeding cartoon-Steve Bell 1994Here at ConservationBytes.com, I write about pretty much anything that has anything remotely to do with biodiversity’s prospects. Whether it is something to do with ancient processes, community dynamics or the wider effects of human endeavour, anything is fair game. It’s a little strange then that despite cutting my teeth in population biology, I have never before tackled human demography. Well as of today, I have.

The press embargo has just lifted on our (Barry Brook and my) new paper in PNAS where we examine various future scenarios of the human population trajectory over the coming century. Why is this important? Simple – I’ve argued before that we could essentially stop all conservation research tomorrow and still know enough to deal with most biodiversity problems. If we could only get a handle on the socio-economic components of the threats, then we might be able to make some real progress. In other words, we need to find out how to manage humans much more than we need to know about the particulars of subtle and complex ecological processes to do the most benefit for biodiversity. Ecologists tend to navel-gaze in this arena far too much.

So I called my own bluff and turned my attention to humans. Our question was simple – how quickly could the human population be reduced to a more ‘sustainable’ size (i.e., something substantially smaller than now)? The main reason we posed that simple, yet deceptively loaded question was that both of us have at various times been faced with the question by someone in the audience that we were “ignoring the elephant in the room” of human over-population.

Read the rest of this entry »





Biodiversity Hotspots have nearly burnt out

10 07 2014

dying embersI recently came across a really important paper that might have flown under the radar for many people. For this reason, I’m highlighting it here and will soon write up a F1000 Recommendation. This is news that needs to be heard, understood and appreciated by conservation scientists and environmental policy makers everywhere.

Sean Sloan and colleagues (including conservation guru, Bill Laurance) have just published a paper entitled Remaining natural vegetation in the global biodiversity hotspots in Biological Conservation, and it we are presented with some rather depressing and utterly sobering data.

Unless you’ve been living under a rock for the past 20 years, you’ll have at least heard of the global Biodiversity Hotspots (you can even download GIS layers for them here). From an initial 10, to 25, they increased penultimately to 34; most recently with the addition of the Forests of East Australia, we now have 35 Biodiversity Hotspots across the globe. The idea behind these is to focus conservation attention, investment and intervention in the areas with the most unique species assemblages that are simultaneously experiencing the most human-caused disturbances.

Indeed, today’s 35 Biodiversity Hotspots include 77 % of all mammal, bird, reptile and amphibian species (holy shit!). They also harbour about half of all plant species, and 42 % of endemic (not found anywhere else) terrestrial vertebrates. They also have the dubious honour of hosting 75 % of all endangered terrestrial vertebrates (holy, holy shit!). Interestingly, it’s not just amazing biological diversity that typifies the Hotspots – human cultural diversity is also high within them, with about half of the world’s indigenous languages found therein.

Of course, to qualify as a Biodiversity Hotspot, an area needs to be under threat – and under threat they area. There are now over 2 billion people living within Biodiversity Hotspots, so it comes as no surprise that about 85 % of their area is modified by humans in some way.

A key component of the original delimitation of the Hotspots was the amount of ‘natural intact vegetation’ (mainly undisturbed by humans) within an area. While revolutionary 30 years ago, these estimates were based to a large extent on expert opinions, undocumented assessments and poor satellite data. Other independent estimates have been applied to the Hotspots to estimate their natural intact vegetation, but these have rarely been made specifically for Hotspots, and they have tended to discount non-forest or open-forest vegetation formations (e.g., savannas & shrublands).

So with horribly out-of-date vegetation assessments fraught with error and uncertainty, Sloan and colleagues set out to estimate what’s really going on vegetation-wise in the world’s 35 Biodiversity Hotspots. What they found is frightening, to say the least.

Read the rest of this entry »





More species = more resilience

8 01 2014

reef fishWhile still ostensibly ‘on leave’ (side note: Does any scientist really ever take a proper holiday? Perhaps a subject for a future blog post), I cannot resist the temptation to blog about our lab’s latest paper that just came online today. In particular, I am particularly proud of Dr Camille Mellin, lead author of the study and all-round kick-arse quantitative ecologist, who has outdone herself on this one.

Today’s subject is one I’ve touched on before, but to my knowledge, the relationship between ‘diversity’ (simply put, ‘more species’) and ecosystem resilience (i.e., resisting extinction) has never been demonstrated so elegantly. Not only is the study elegant (admission: I am a co-author and therefore my opinion is likely to be biased toward the positive), it demonstrates the biodiversity-stability hypothesis in a natural setting (not experimental) over a range of thousands of kilometres. Finally, there’s an interesting little twist at the end demonstrating yet again that ecology is more complex than rocket science.

Despite a legacy of debate, the so-called diversity-stability hypothesis is now a widely used rule of thumb, and its even implicit in most conservation planning tools (i.e., set aside areas with more species because we assume more is better). Why should ‘more’ be ‘better’? Well, when a lot of species are interacting and competing in an ecosystem, the ‘average’ interactions that any one species experiences are likely to be weaker than in a simpler, less diverse system. When there are a lot of different niches occupied by different species, we also expect different responses to environmental fluctuations among the community, meaning that some species inherently do better than others depending on the specific disturbance. Species-rich systems also tend to have more of what we call ‘functional redundancy‘, meaning that if one species providing an essential ecosystem function (e.g., like predation) goes extinct, there’s another, similar species ready to take its place. Read the rest of this entry »





Rise of the phycologists

22 09 2011

Dead man's fingers (Codium fragile) - © CJA Bradshaw

I’ve had an interesting week. First, it’s been about 6 years since I was last in Japan, and I love coming here; the food is exquisite, the people are fantastic (polite, happy, accommodating), everything works (trains, buses, etc.) and most importantly, it has an almost incredible proportion of its native forests intact.

But it wasn’t for forests that I travelled to Japan (nor the sumo currently showing on the guest-room telly where I’m staying – love the sumo): I was here for a calcareous macroalgae workshop.

What?

First, what are ‘macroalgae’, and why are some ‘calcareous’? And why should anyone in their right mind care?

Good questions. Answers: 1. Seaweeds; 2. Many incorporate calcium carbonate into their structures as added structural support; 3. Read on.

Now, I’m no phycologist (seaweed scientist), but I’m fascinated by this particular taxon. I’ve written a few posts about their vital ecological roles (see here and here), but let me regale you with some other important facts about these amazing species.

Some Japanese macroalgae - © CJA Bradshaw

There are about 12,000 known species of macroalgae described by phycologists, but as I’ve learnt this week, this is obviously a vast underestimate. For most taxa that people are investigating now using molecular techniques, the genetic diversity is so high and so geographically structured that there are obviously a huge number of ‘cryptic’ species within our current taxonomic divisions. This could mean that we’re out by up to a factor of 2 in the number of species in the world.

Another amazing fact – about 50 % of all known seaweed species are found in just two countries – Japan and Australia (hence the workshop between Japanese and Australian phycologists). Southern Australia in particular is an endemism hotspot.

Ok. Cool. So far so good. But so what? Read the rest of this entry »





No substitute for primary forest

15 09 2011

© Romulo Fotos http://goo.gl/CrAsE

A little over five years ago, a controversial and spectacularly erroneous paper appeared in the tropical ecology journal Biotropica, the flagship journal of the Association for Tropical Biology and Conservation. Now, I’m normally a fan of Biotropica (I have both published there several times and acted as a Subject Editor for several years), but we couldn’t let that paper’s conclusions go unchallenged.

That paper was ‘The future of tropical forest species‘ by Joseph Wright and Helene Muller-Landau, which essentially concluded that the severe deforestation and degradation of tropical forests was not as big a deal as nearly all the rest of the conservation biology community had concluded (remind you of climate change at all?), and that regenerating, degraded and secondary forests would suffice to preserve the enormity and majority of dependent tropical biodiversity.

What rubbish.

Our response, and those of many others (including from Toby Gardner and colleagues and William Laurance), were fast and furious, essentially destroying the argument so utterly that I think most people merely moved on. We know for a fact that tropical biodiversity is waning rapidly, and in many parts of the world, it is absolutely [insert expletive here]. However, the argument has reared its ugly head again and again over the intervening years, so it’s high time we bury this particular nonsense once and for all.

In fact, a few anecdotes are worthy of mention here. Navjot once told me one story about the time when both he and Wright were invited to the same symposium around the time of the initial dust-up in Biotropica. Being Navjot, he tore off strips from Wright in public for his outrageous and unsubstantiated claims – something to which Wright didn’t take too kindly.  On the way home, the two shared the same flight, and apparently Wright refused to acknowledge Navjot’s existence and only glared looks that could kill (hang on – maybe that had something to do with Navjot’s recent and untimely death? Who knows?). Similar public stoushes have been chronicled between Wright and Bill Laurance.

Back to the story. I recall a particular coffee discussion at the National University of Singapore between Navjot Sodhi (may his legacy endure), Barry Brook and me some time later where we planned the idea of a large meta-analysis to compare degraded and ‘primary’ (not overly disturbed) forests. The ideas were fairly fuzzy back then, but Navjot didn’t drop the ball for a moment. He immediately went out and got Tien Ming Lee and his new PhD student, Luke Gibson, to start compiling the necessary studies. It was a thankless job that took several years.

However, the fruits of that labour have now just been published in Nature: ‘Primary forests are irreplaceable for sustaining tropical biodiversity‘, led by Luke and Tien Ming, along with Lian Pin Koh, Barry Brook, Toby Gardner, Jos Barlow, Carlos Peres, me, Bill Laurance, Tom Lovejoy and of course, Navjot Sodhi [side note: Navjot died during the review and didn’t survive to hear the good news that the paper was finally accepted].

Using data from 138 studies from Asia, South America and Africa comprising 2220 pair-wise comparisons of biodiversity ‘values’ between forests that had undergone some sort of disturbance (everything from selective logging through to regenerating pasture) and adjacent primary forests, we can now hammer the final nails into the coffin containing the putrid remains of Wright and Muller-Landau’s assertion – there is no substitute for primary forest. Read the rest of this entry »





Tropical forests cooking their biodiversity

5 05 2011

Another ‘hot’ essay by Bill Laurance recently published online by Yale Environment 360 (a publication of the Yale University School of Forestry & Environmental Studies). Bill asked me to relay it on ConservationBytes.com, so here it is in full:

Much attention has been paid to how global warming is affecting the world’s polar regions and glaciers. But a leading authority on tropical forests [that would be Bill] warns that rising temperatures could have an equally profound impact on rainforests and are already taking a toll on some tropical species.

On Jan. 12, 2002, in the Australian state of New South Wales, biologist Justin Welbergen was observing a colony of flying foxes for his Ph.D. research. The temperatures that day on Australia’s subtropical, eastern coast reached record highs, soaring to 42.9 ° C (109 ° F) at the weather station closest to Welbergen’s study site — nearly 8 ° C higher than the average summer maximum temperature.

The flying foxes, or giant fruit bats, normally just doze in the treetops through the day, but on this afternoon they were fanning themselves, panting frantically, jostling for shady spots, and licking their wrists in a desperate effort to cool down. Suddenly, when the thermometer hit 42 ° C, the bats began falling from the trees. Most quickly died. Welbergen and his colleagues counted 1,453 flying foxes that died from the heat in one colony alone. The scorching heat that day killed at least 2,200 additional flying foxes in eight other colonies along a 250-kilometre stretch of coastline. All the deaths occurred in colonies where temperatures soared above 41.7 ° C. Read the rest of this entry »





The rarity of commonness

18 08 2009

I’m attending the 10th International Congress of Ecology (INTECOL) in Brisbane this week and I have just managed to find (a) an internet connection and (b) a small window to write this post.

I have to say I haven’t been to a good plenary talk for some time – maybe it’s just bad luck, but often plenary talks can be less-than-inspiring.

Not so for INTECOL this year. I was very pleased to have the opportunity to listen to biodiversity guru Professor Kevin Gaston of the University of Sheffield give a fantastic talk on… common species (?!).

clones

If you have followed any of Kevin’s work, you’ll know he literally wrote the book on rarity – what species rarity is, how to measure it and what it means for preserving biodiversity as a whole.

Now he’s championing (in a very loose sense) the importance of common species because it is these taxa, he argues, that provide the backbone to the persistence of all biodiversity.

Yes, we conservation biologists have tended to focus on the rare and endemic species to make certain we have as much diversity in species (and genetic material) as possible when conserving habitats.

There are a lot more rare species than common ones, and the most common species (i.e., the ones you most often see) tend to have the broadest distributions. We know from much previous work that having a broad distribution reduces extinction risk, so why should we be concerned about common species?

Kevin made a very good point – if you turn the relationship on its head somewhat, you can state that the state of ‘commonness’ is itself ‘rare’. In fact, only about 25 % of the most common species account for about 90-95 % of ALL individuals. He used an interesting (and scary) example to show what this can mean from an extinction perspective. Although very back-of-the-envelope, there are about 2000 individual birds in a km2 of tropical forest; we are losing between 50000 and 120000 km2 of tropical forest per year, so this translates into the loss of about 100 to 240 million individual birds per year; this is the sum total of all birds in Great Britain (a bird-mad country). Yet where do we have the best information about birds? The UK.

Commonness is also geologically transient, meaning that just because you are a common species at some point in your evolutionary history doesn’t mean you have always been or always will be. In fact, most species never do become common.

But it is just these ‘rare’ common that drive the principal patterns we see globally in community structure. The true ‘rare’ species are, in fact, pretty crap predictors of biodiversity patterns. Kevin made a good point – when you look at a satellite image of a forest, it’s not all the little rare species you see, it’s the 2 or 3 most common tree species that make up the forest. Lose those, and you lose everything else.

Indeed, common species also form most trophic structure (the flow of energy through biological communities). Take away these, and ecosystem function degrades. They also tend to have the highest biomass and provide the structure that supports all those millions of rare species. Being common is quite an important job.

Kevin stated that the world is now in a state where many of the so-called common species are in fact, “artificially” common because of how much we’ve changed the planet. It is these benefactors of our world-destroying machinations that are now in decline themselves, and it is for this reason we should be worried.

When you start to see these bastions of ecosystems start to drop off (and the drop is usually precipitous because we don’t tend to notice their loss until they suddenly disappear), then you know we’re in trouble. And yet, even though once common, few, if any, once-common species have come back after a big decline.

So what does this mean for the way we do biodiversity research? Kevin proposes that we need a lot more good monitoring of these essential common species so that we can understand their structuring roles in the community and more importantly, be able to track their change before ecosystem collapse occurs. The monitoring is crucial – it wasn’t the demise of small companies that signalled the 2007 stock market crash responsible for the Global Financial Crisis in which we now find ourselves, the signal was derived from stock data obtained from just a few large (i.e., ‘common’) companies. All the small companies (‘rare’) ones then followed suit.

A very inspiring, worrying and somewhat controversial talk. Watch out for more things ‘Gaston’ on ConservationBytes.com in the near future.

CJA Bradshaw

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to Furl








Follow

Get every new post delivered to your Inbox.

Join 6,871 other followers

%d bloggers like this: