Dangers of forcing regressions through the origin

17 10 2017

correlationsI had an interesting ‘discussion’ on Twitter yesterday that convinced me the topic would make a useful post. The specific example has nothing whatsoever to do with conservation, but it serves as a valuable statistical lesson for all concerned about demonstrating adequate evidence before jumping to conclusions.

The data in question were used in a correlation between national gun ownership (guns per capita) and gun-related deaths and injuries (total deaths and injuries from guns per 100,000 people) (the third figure in the article). As you might intuitively expect, the author concluded that there was a positive correlation between gun-related deaths and injuries, and gun ownership:



Now, if you’re an empirical skeptic like me, there was something fishy about that fitted trend line. So, I replotted the data (available here) using Plot Digitizer (if you haven’t yet discovered this wonderful tool for lifting data out of figures, you would be wise to get it now), and ran a little analysis of my own in R:


Just doing a little 2-parameter linear model (y ~ α + βx) in R on these log-log data (which means, it’s assumed to be a power relationship), shows that there’s no relationship at all — the intercept is 1.3565 (± 0.3814) in log space (i.e., 101.3565 = 22.72), and there’s no evidence for a non-zero slope (in fact, the estimated slope is negative at -0.1411, but it has no support). See R code here.

Now, the author pointed out what appears to be a rather intuitive requirement for this analysis — you should not have a positive number of gun-related deaths/injuries if there are no guns in the population; in other words, the relationship should be forced to go through the origin (xy = 0, 0). You can easily do this in R by using the lm function and setting the relationship to y ~ 0 + x; see code here). Read the rest of this entry »

Massive yet grossly underestimated global costs of invasive insects

4 10 2016
Portrait of a red imported fire ant, Solenopsis invicta. This species arrived to the southeastern United States from South America in the 1930s. Specimen from Brackenridge Field Laboratory, Austin, Texas, USA. Public domain image by Alex Wild, produced by the University of Texas "Insects Unlocked" program.

Portrait of a red imported fire ant Solenopsis invicta. This species arrived to the southeastern USA from South America in the 1930s. Specimen from Brackenridge Field Laboratory, Austin, Texas, USA. Public domain image by Alex Wild, produced by the University of Texas “Insects Unlocked” program.

As many of you already know, I spent a good deal of time in France last year basking in the hospitality of Franck Courchamp and his vibrant Systematic Ecology & Evolution lab at Université Paris-Sud. Of course, I had a wonderful time and was sad to leave in the end, but now I have some hard evidence that I wasn’t just eating cheese and visiting castles. I was actually doing some pretty cool science too.

Financed by BNP-Paribas and Agence Nationale de Recherche, the project InvaCost was designed to look at the global impact of invasive insects, including projections of range dynamics under climate change and shifting trade patterns. The first of hopefully many papers is now out.

Just published in Nature Communications, I am proud that many months of hard work by a brilliant team of ecologists, epidemiologists and economists has culminated in this article entitled Massive yet grossly underestimated costs of invasive insects, which in my opinion is  the first robust analysis of its kind. Despite some previous attempts at estimating the global costs of invasive species1-4 (which have been largely exposed as guesswork and fantasy5-10), our paper rigorously treats the economic cost estimates and categorises them into ‘reproducible’ and ‘irreproducible’ categories.


Gypsy moth (Lymantria dispar) adult. Dimitri Geystor (France)

What we found was sobering. If we look at just ‘goods and services’ affected by invasive insects, the annual global costs run at about US$70 billion. These include agricultural, forestry and infrastructure damages, as well as many of the direct costs of clean-up and eradication, and the indirect costs of prevention. When you examine that number a little more closely and only include the ‘reproducible’ studies, the total annual costs dip to about US$25 billion, meaning that almost 65% of the costs recorded are without any real empirical support. Scary, especially considering how much credence people put on previously published global ‘estimates’ (for example, see some citation statistics here).


Formosan subterranean termite Coptotermes formosanus by Scott Bauer, US Department of Agriculture, Agricultural Research Service

There’s a great example to illustrate this. If you take it at face value, the most expensive invasive insect in the world is the Formosan subterranean termite Coptotermes formosanus estimated at US$30.2 billion/yr globally. However, that irreproducible estimate is based on a single non-sourced value of US$2.2 billion per year for the USA, a personal communication supporting a ratio of 1:4 of control:repair costs in a single US city (New Orleans), and an unvalidated assumption that the US costs represent 50% of the global total.

Read the rest of this entry »

Ice Age? No. Abrupt warmings and hunting together polished off Holarctic megafauna

24 07 2015
Oh shit oh shit oh shit ...

Oh shit oh shit oh shit …

Did ice ages cause the Pleistocene megafauna to go extinct? Contrary to popular opinion, no, they didn’t. But climate change did have something to do with them, only it was global warming events instead.

Just out today in Science, our long-time-coming (9 years in total if you count the time from the original idea to today) paper ‘Abrupt warmings drove Late Pleistocene Holarctic megafaunal turnover‘ led by Alan Cooper of the Australian Centre for Ancient DNA and Chris Turney of the UNSW Climate Change Research Centre demonstrates for the first time that abrupt warming periods over the last 60,000 years were at least partially responsible for the collapse of the megafauna in Eurasia and North America.

You might recall that I’ve been a bit sceptical of claims that climate changes had much to do with megafauna extinctions during the Late Pleistocene and early Holocene, mainly because of the overwhelming evidence that humans had a big part to play in their demise (surprise, surprise). What I’ve rejected though isn’t so much that climate had nothing to do with the extinctions; rather, I took issue with claims that climate change was the dominant driver. I’ve also had problems with blanket claims that it was ‘always this’ or ‘always that’, when the complexity of biogeography and community dynamics means that it was most assuredly more complicated than most people think.

I’m happy to say that our latest paper indeed demonstrates the complexity of megafauna extinctions, and that it took a heap of fairly complex datasets and analyses to demonstrate. Not only were the data varied – the combination of scientists involved was just as eclectic, with ancient DNA specialists, palaeo-climatologists and ecological modellers (including yours truly) assembled to make sense of the complicated story that the data ultimately revealed. Read the rest of this entry »

Demonising the hellbender

19 09 2014
Not 'Hellraiser', FFS - 'hellbender'

Not ‘Hellraiser’, FFS – ‘hellbender’

Here’s one by my new PhD student, Leah Collett:

I have never heard of the hellbender before. “Brilliant name”, I thought. Then I saw it mentioned again a few days later, in company with honey badgers and blue-footed boobies, in a recent article on why we need to see nature as useless in order to ‘save’ it.

So what is it? The hellbender is a species of salamander found in eastern North America, the only species of its genus it turns out, and one of only three left in its family (Cryptobranchidae – ‘hidden gills’). It is assessed as Near Threatened with possible extinction in the near future by the IUCN Red List. Along with the, sadly all too often, habitat degradation it is suffering due to it being highly sensitive to environmental change, it is still believed to be poisonous and that it will kill other fish and their eggs with this poison. In this day and age, some people will kill these salamanders when they come across them when they are out fishing, due to inaccurate held beliefs. I think I found it more incredible that this still occurs in a country where people have quick access to research and information denouncing such a myth.

This wonderful document on the biology and ecology of the hellbenders includes a section titled ‘Man and Hellbenders’ and highlights the likely culprit for starting the ‘hellbenders are poisonous – kill them all’ campaign:

“Dr N. Bayard Green (1971) related how a former editor of the Pocahontas Times (Pocahontas, West Virginia) in 1926 continually tried to vilify the hellbender as a destroyer of game fish and their eggs. Throughout West Virginia many sportsmen’s groups attempted to eradicate many of the so-called enemies of fish and game.”

(I could not read the original paper of Dr Green’s, although it might be available in the collections section). Read the rest of this entry »

Eye on the taiga

24 03 2014

boreal damageDun! Dun, dun, dun! Dun, dun, dun! Dun, dun, daaaaah!

I’ve waited nearly two years to do that, with possibly our best title yet for a peer-reviewed paper: Eye on the taiga: removing global policy impediments to safeguard the boreal forest (recently published online in Conservation Letters).

Of course, the paper has nothing to do with cheesy Eighties music, underdog boxers or even tigers, but it does highlight an important oversight in world carbon politics. The boreal forest (also known as taiga from the Russian) spans much of the land mass of the Northern Hemisphere and represents approximately one quarter of the entire planet’s forests. As a result, this massive forest contains more than 35% of all terrestrially bound carbon (below and above ground). One doesn’t require much more information to come to the conclusion that this massive second lung of the planet (considering the Amazon the first lung) is a vital component of the world’s carbon cycle, and temperate biodiversity.

The boreal forest has been largely expanding since the retreat of the glaciers following the Last Glacial Maximum about 20,000 years ago, which means that its slow progression northward has produced a net carbon sink (i.e., it takes up more atmospheric carbon that it releases from decomposition). However, recent evidence suggests that due to a combination of increased deforestation, fire from both human encroachment and climate change, mass outbreaks of tree-killing insects and permafrost melting, the boreal forest is tipping towards becoming a net carbon source (i.e., emitting more carbon into the atmosphere than it takes up from photosynthesis). This is not a good thing for the world’s carbon cycle, because it means yet another positive feedback that will exacerbate the rapid warming of the planet. Read the rest of this entry »

Native invaders divide loyalties

7 09 2012

California sea lion at Bonneville fish ladder. Credit: U.S. National Oceanic and Atmospheric Administration

As if to mimic the weirder and weirder weather human-caused climate disruption is cooking up for us, related science stories seem to come in floods and droughts. Yes, research trends become fashionable too (imagine a science fashion show? – but I digress…).

Only yesterday, the ABC published an opinion piece on the controversies surrounding which species we call ‘native’ and ‘invasive’ (based on a recent paper published in Global Ecology and Biogeography), and in June this year, Salvador Herrando-Pérez wrote a great little article on the topic entitled “The invader’s double edge“.

Then today, I received a request to publish a guest post here on ConservationBytes.com from Lauren Kuehne, a research scientist in Julian Olden‘s lab at the University of Washington in Seattle. The topic? Why, the controversies surrounding invasive species, of course! Lauren’s following article demonstrates yet again that it’s not that simple.

A drawback to the attention garnered by high-profile invasive species is the tendency to infer that every non-native species is bad news, the inverse assumption being that all native species must be ‘good’. While this storyline works well for Hollywood films and faerie tales, in ecology the truth is rarely that simple. A new review article in the September issue of Frontiers in Ecology and the Environment, describes the challenges and heartbreaks when native species run amok in the sense of having negative ecological impacts we typically associate with non-native species. Examples in the paper range from unchecked expansions of juniper trees in sagebrush ecosystems with wildfire suppression, to overgrazing by elk (wapiti) released from predation following the removal of wolves and mountain lions. Read the rest of this entry »

You’re not even remotely concerned enough

31 08 2012

I’ve just returned from a 6-week trip to the United States and I am now dealing with the intensity of things left undone for so long [sigh]. But that trip was interesting for many reasons. First, and as I’ve already posted, I finished a book with Paul Ehrlich that will be out sometime early in 2013 (but I won’t deal with that here). I also attended an interesting, if slightly confusing, conference on ecosystem services. And finally, I had the pleasure of meeting Tony Barnosky in person, and we decided that we should definitely collaborate on a few things.

Another thing that struck me – and this happens no matter how often I visit the U.S., is just how completely insane that country’s politics are. The extremist, libertarian, plutotheocratic bullshit spewed by the far right to the detriment of the very people who support them is enough to make you vomit. And this startling and thoroughly backward world-view is now starting to penetrate more and more into Australian society and politics. From an environmental perspective, it’s a continuation of a downhill slide that started with Reagan’s destruction of environmentalism in the U.S., and Joh Bjelke-Petersen‘s war on the environment in Australia, and will only continue to get worse.

Of course, the main victim of reason in all these polemic politics is that we are doing next to nothing to mitigate horrendous climate disruption. Only yesterday, George Monbiot was lamenting (nay, pleading) that our governments are doing practically nil to avoid what can only be described as the greatest threat to our way of life since the World War II – in fact, the War and its associated holocaust is small bikkies compared to what awaits us.

And this is the most stressing part – even people who choose to use their brains and accept that we have an immense, global problem on our hands generally are not even remotely concerned enough. Read the rest of this entry »