Showing posts with label conference. Show all posts
Showing posts with label conference. Show all posts

Friday, 10 July 2009

Quotes from Orwell and North American Mammals

I found this quote at the bottom of a daily news update, and I like it:
The aim of a joke is not to degrade the human being but to remind him that he is already degraded. -- George Orwell
I was never made to read George Orwell at school, or collage, and so I came to appreciate what a good author he was without the negative memories of time in Engrish class. Which is good, this way I like him!



Here's some more science from the Conference, this one being talk number 54, titled "What's killing North America's mammals? A meta-analysis of cause specific mortality." It was authored by Christopher Collins and Roland Kays, out of New York State Museum.

As they write in their Abstract, everything eventually dies. What we'd like to know is across all mammals, what are the major drivers of mortality? We can quantify this by `Collering and follering` - putting on GPS or radio collars to track animals until they die, and investigating the cause of death.

Any given study can only look at so many animals at one time. GPS and radio collars are expensive, and so are putting them on critters. It's also very time intensive. So any given study only gets so much information, and that information is generally most applicable to a given species, or a group of species in a similar area. However, people can go in after studies are published and collect lots of peoples' different studies and bring them together. This is called a Meta-analysis. Meta-analyses frequently need lots of boring mathematical treatment, because people invariably collect their data in different ways, but well done meta-analyses can be very informative about large scale patterns that no one study would reveal.

Collins and Kays gathered in all the studies they could find with mortality data, and after working with the data, they found that surprisingly, 43.8% of across-species mortality was caused by humans. Most of this was hunting, though not all; Road-kill accounted for 7.9% of all mammals' mortality reported. Now, natural causes accounted for 40.9%, and unknown causes accounted for 15.3%, which is very large indeed. Natural mortality is probably much larger, because that 15.3% of unknown is likely a lot of disease related death, and predation. However, in those cases, the researchers got to the bodies late, or just couldn't figure out how the animal died - it's harder to figure out what disease killed an animal than you'd think!

Still, even if all of the unknown mortality was caused natural causes, that still leaves 43.8% in the hands of humans. Where other factors used to dominate, now humans apply the "Natural Selection" in North America. This has very profound implications for the direction that evolution will take in the long-haul, as species slowly adapt to human presence. Well, either that or they'll go extinct. One of the two.

I once took a course in Urban Ecology, which is the study of natural systems in human areas - farms, villages, dockyards, and big cities. What really stuck in terms of animal resilience was that species that thrived around humans either had a long history around human cities (usually Europe or Asian weeds and vermin), or animals that are so behaviourally complex, and so plastic and amenable to changing conditions, that humans really couldn't disrupt them. Racoons in the lower 48 are the perfect example of this.

With this in mind, and knowing what Collins and Kays said about human-cuased mortality, what Alaskan Mammals do I see with the best chance at persistence in 200 -400 years?
  • Arctic Fox
  • Red Fox
  • Coyote
  • Feral Domestic dogs
  • Racoon - invade and expand.
  • Ermine
  • White Tail Deer - invade and expand
  • Snowshoe Hare
This isn't a comprehensive list, but species that are behaviourally plastic, and stick out in my mind as things that can dodge human-induced mortality to a sufficient degree. I don't include moose because they're highly susceptible to diseases carried by White Tailed Deer, who are slowly invading northward.

There's some hope for other species, though. According to Collins and Kays, when humans designate an area as protected, it seems to work by-and-large. There's still human caused mortality, but it becomes considerably less. There just needs to be the political will to do it, and the public support for the designation.

Easier said than done.

Tuesday, 30 June 2009

Feeling hungover without having drank

My head is doing things that I lack words to explain, but luckily, Thursday and Friday are holidays for the University, and I'm getting out while I can. Add to that I'm owed a vacation day from three weeks ago that I got billed for, but never got to take! The conference is completely over, the guests are gone, the mess is all that remains and (luckily) I don't have to clean any of it. I'm back in my office (for now!). It appears I accidentally left lunch in here before things got busy(er).

Enough about me, though. I want to talk about some of the things I learned at the meeting. That's purpose number one of a meeting - communicate science between researchers. Purpose two is to get to know people, find people to work with, etc. I have people who are going to contact me about mink samples, for example. I need to email someone about how they did something so I can do it too. Anyhow. Here's some things:

Talk 262: Phylogeography of Matschie's Tree Kangaroo
My take home message was that first, expert hunters' contributions should be taken skeptically. The researchers found that something akin to 80% of one set of samples were mis-identified to their species, when they did precise analysis. 80% is worse than randomly putting a label on it! The other sets of samples were around 70% and 70% mis-assigned. The author couldn't find any "conservation units," which is a group that can be protected and encompass a representative chunk of diversity, but this might have been because they were looking at slowly evolving mtDNA, and not other, faster DNA markers.

Talk 23: Winter habitat selection by elk in Wyoming with respect to habitat improvement areas
Model stew. It wasn't the fault of the researcher, but the poor quality of data they inherited. They rarely found any signifigant findings (statistically signifigant, that is - which means unlikely to be caused by chance observation), and their best winter habitat models included more parameters than most scientists are comfortable with, espeically given such low support values. I talked to other management people, and they weren't convinced the findings were robust either. Someone is likely to replicate this experiment...

Talk 31: Hibernation and the Ark
This guy knew how to get a crowd, with that title! Half his talk was poking fun of creationists who used his hibernation research to explain how Noah got every animal + dinosaur onto the ark without drowning in anaq and pee (let me add: ew). The other half his talk was about how species that can enter topor (and apparently Australia has a lot of them - this guy was an Aussie) seem to go extinct less than other groups. This might be a reporting bias, but interesting if true.


And speaking of Aussies... There were a ton of them. I wanted to call them all Bruce.

No, I didn't do it.

Anyhow, I have to wonder what it is about Australia and Alaska that causes so many Aussies to go to AK and so many Alaskans to go to Australia. It's crazy.

Thursday, 25 June 2009

ASM Day 1, bullet point style.

Day one of the conference is over - 6 more to go. I made it a point, yesterday, to meet someone new at the ASM social. Here's some new things I learnt from new people after I had a few glasses of liquid courage:

  • Sin nombre hanta virus prevalence among deer mice in Colorado is currently low.
  • Sin nombre hanta virus prevalence in CO doesn't follow a elevational gradient, either(?!?!).
  • Our niche modelling for wolf lice in AK might be trained by the distribution and abundance of lice in the lower 48.
  • ... but it might also not.
  • The army doesn't like the delta herd bison 100%, because their calving grounds are protected.
  • Bison have calving grounds! And site fidelity!
  • I found someone who studied Sitka Black Tailed Deer dispersal, but one of my co-authors aready knew about him,.
  • The name Thor is pronounced `Tor.`
  • The Army keeps staff wildlife biologists.
  • Germany has a very different university system.
  • Germany's politics are radically controlled by American politics, and American political decisions.
  • The Germans don't know why so many Germans visit Alaska.
  • We've got high hopes for the GMU 20A controlled burn.
  • Sea Otters go straight to hypothermia if they can't groom their pelt for even a short period...
  • ... or even if they can't groom an area the size of a quarter.
That's just what I remember. Today will be... well, maybe better? Maybe not. But definitely longer.

Click for Fairbanks, Alaska Forecast