Posted under local birds & science
During the oil spill response in San Francisco last year, news organizations (and others, like bloggers) kept updating the public on the number of birds that died. NBC11 was neither better nor worse than the others, but I’ll use it as an example:
More than 2,000 birds have died as a result of the Cosco Busan oil spill in San Francisco Bay, according to updated numbers released Monday.
According to the Coast Guard, 1,591 birds have been discovered dead since the Nov. 7 spill, and another 534 died at cleaning and treatment facilities.
I hope you see the problem: the “more than 2000″ number isn’t a count of birds that have died, it’s birds that are known to have died. A bird that dies out at sea and isn’t discovered, or that is eaten by predators, or that dies on a remote beach where nobody finds it, won’t be included in this count. In fact, the standard guesstimate is that about ten times as many dead birds remain undiscovered than are found (though it could be 5 times or 25 times, depending on the number of searchers, the weather, the types of birds, and so on). So, yeah, it’s true to say that “more than 2000″ birds died, but it’s also true to say “gas costs more than 50 cents a gallon.” It’s true but badly misleading. They ought to say “more than 20,000 birds are thought to have died”, or “more than 10,000 birds have died, and perhaps as many as 50,000″, or something like that. By the way, it’s not a conspiracy to cover up the damage: even the International Bird Rescue Research Center makes the same mistake on their front page!
CWilson on 10 Feb 2008 at 8:17 pm #
Speaking as someone who has a fair amount of oil spill response experience (ex-USCG spill response team member, private spill response specialist), you are correct in stating that “more than 2000″ is a count of known bird deaths, not an estimated total bird deaths. But it’s probably pretty close. The 10X guesstimate was originally developed in the late ’60s before any actual research was done. There has been subsequent research to discredit that guesstimate.
You need to look where the numbers come from and what is being counted to come up with a reasonable estimate of bird deaths from the spill.
First off, there has been research by various folks (NOAA, CEDRE, a few others) into surveys and statistical sampling regarding non-recovered birds that died from oiling outside the impacted spill zone. The general conclusion was that the vast majority of oiled birds stay within the oiled area. Once they become contaminated, they generally don’t fly off or wander very far. Those that do generally aren’t oiled to the extent that it impacts their ability to live and thrive (i.e., they may have a small spot of oil, but they preen themselves and recover with little or no ill effects).
Second, the oiled area is pretty thoroughly gone over by multiple teams, and generally all oiled and dead birds are collected. So there are few or no missed birds in the oiled area.
Finally, the count of dead birds is a count of all dead birds, no matter what they died of. During the T/V Exxon Valdez spill, posting of dead birds indicated that something around 10% of the collected dead birds died from something other than the oil spill or its effects (causes included red tide poisoning, old age, attacks by predators, winter kill, etc.).
So, the 2,000+ number probably includes quite a few non-oil spill deaths, and there are probably few unrecovered dead birds.
On the other side of the coin, it has been well-documented that bird recovery efforts are minimally successful at best. Research from the TB Morris Berman spill indicates that less than 15% of the treated birds survive more than a few weeks in the wild. The hypothesis is that handling wild birds in the way necessary to clean them stresses them so much that they have trouble surviving in the wild.
Bottom line is that oil and birds don’t mix well, and an oiled bird is generally a dead bird. All the stronger case for spill prevention and prompt response.
admin on 11 Feb 2008 at 12:57 am #
CWilson, thanks for your comments. Very interesting and thought-provoking.
It seems to me that (if it hasn’t been done already) someone should do a formal study of the rate at which dead birds are found in the recovery effort. For example, someone could do a variation of what is called “mark-recapture”: a researcher could go out and find a bunch of dead birds, tag or mark them and leave them in place, and see how many are discovered in the recovery effort. If, say, 10% of them are missed, then you know that you are underestimating bird deaths by at least 10%. Unfortunately this only sets a crude lower limit, because birds that are washed out to sea, or are in really inaccessible places, won’t even be found the first time.
That said, I am pretty skeptical of the claim that the number of birds found dead is actually close to the number of birds that die. For example, I found an oiled bird at Lake Merritt (near a bird carcass, but I don’t know how that bird died and can’t attribute it to oil) and another in a nearby channel, both several miles from any oil, so I know oiled birds can and do travel quite a difference sometimes. Several hundred “lightly” oiled birds turned up at the Farallones, and no attempt was made to recover them (it would be interesting to hear if there is a follow-up on this story, especially an estimate of how many of these birds eventually died). I can certainly believe that the standard multiple of 10 is way too high and isn’t based on much data, but I find it hard to believe there aren’t lots of oiled birds that go uncounted. You can walk past a small dead bird like a plover and not see it even if you’re looking — it can be rolled in sand or covered with sea foam. I could believe the multiplier is 2 or 3 rather than 10, I guess.
As for whether oiled bird recovery efforts are worth the time and expense, this is something I’d like to see a widespread discussion about. I’m of two minds about it myself. First, let me say that I consider endangered species a special case and would encourage substantial expense to protect these birds and their habitat. But for common birds, there is a dilemma. My wife and I spent several hours to find and recover a single oiled surf scoter, and that effort was just the beginning, since it passed from us into the care of skilled volunteers who, I’m sure, gave it hours of special care. And that’s just for one bird. All told, thousands or even tens of thousands of hours of volunteer effort went into bird rescue. If that time (and money) had gone into, say, political lobbying or habitat restoration, we could have had benefits that would have carried forward for years, and possibly done much more good in the long run than the bird recovery did. But here’s the thing: you won’t get those thousands of volunteer hours if you are trying to get people to speak at hearings, write letters, or get out to pull weeds and plant reeds. A pathetic, frantic oiled bird cries out for action in a direct way that people are willing to respond to. I think the impulse to help these birds is entirely laudable and I commend the people who do it; I just wish that it would also inspire more people to take other, longer-lasting actions.
Thanks again for your comments, CW.
CWilson on 11 Feb 2008 at 8:38 pm #
A multiplier of 2 or 3 may be reasonable… I can’t seem to locate the research papers that discuss it in the archaeological dig that is my desk. A “mark-recapture” study as you propose would provide some interesting data.
Anecdotally, I’ve served on several SCATs (shoreline cleanup assessment teams) and by the time we survey the shoreline there isn’t much that hasn’t been collected and/or cleaned. The thoroughness probably varies from spill to spill, and shoreline type can have a big impact.
As to whether oiled bird recovery efforts are worth the effort, I tend to think that they are of very limited value, and in some cases may be detrimental to the species and the cleanup. It sounds crass and uncaring, but once oil hits the water we’ve basically lost the battle and all our efforts are really just damage control. Well-meaning people bringing in oiled wildlife for rehab divert scarce resources which could be better employed in the response (not folks like the IBRC, I’m talking about resources like security personnel, transport for oiled wildlife, keeping track of who is in/out of the contaminated area, etc.).
There was an interesting study done on sea otters after the Exxon Valdez (”Catastrophes and Conservation: Lessons from Sea Otters and the Exxon Valdez”, James Estes, Science, Vol. 254). Basically, since 70% of the sea otters brought in for rehab were either unoiled or only lightly oiled, and sea otters have a 5-10% stress-induced mortality from capture when they healthy, Mr. Estes drew the conclusion that deaths from capture stress were sufficient to crash the population of sea otters in Prince William Sound. Interesting reading, and I suspect a similar mechanism may exist for seabirds.
birds in texas on 01 Aug 2008 at 1:03 am #
birds in texas…
We put a feeder out a couple of months ago and were surprised to see that they flocked to it the very next day. Being that we had a ton of trees in the backyard it shouldn’t have been so surprising. We just lowered our feeder at our new place and hav…