I joke that when I go for a run on the trails around my home in San Geronimo Valley, I'm doing tick surveys. Which is essentially true--one method of catching ticks for study is to drag fabric through grass, and the ticks latch on. That is what I do when my legs brush past grass and twigs.
I don't keep formal track of the ticks I find--or more accurately, the ticks that find me. Which is unfortunate, because there is a nice dataset there. But from my memory, what has struck me is the variability. Just when I think I see a pattern, and understand what is going on, and formulate a hypothesis, I observe ticks in places that confound that pattern. Or I hear about others' observations that don't match my mental model.
Take this year for example. I had not seen a tick until last week, yet word on the street (or fire road?) was it was a bad tick season. So I began putting together a hypothesis about the cold December, dry winter, and very wet March: that prior to March, ticks were scarce (at least, scarce in my surveys) in grassy areas where I was running because of the low growth and short grass. Since March the heavy rain impacted the ticks in these areas in some way, possibly washing them out of the grass and drowning them.
But how does this mesh with others' observations of a bad tick season? Well, my tick surveys occur predominantly in grassy areas, and my hypothesis does not apply to wooded areas. People could be getting fewer ticks in the grass and more ticks in the woods.
But then last week I saw a tick on the tip of a blade of grass next to the trail on which I was running. Now, this single observation shouldn't necessarily blow my whole theory out of the water. But since my conceptual models of tick ecology are built on small sample sizes, and it was the first tick I've seen of the year, I had to extrapolate and assume this single observation meant there were a lot more ticks out there waiting for me on blades of grass.
This could just mean that it has been dry enough over the last week to allow ticks to not get knocked off grass by raindrops, and my mental model is a good one. But when I tried to make the tick move, it didn't. So does that mean that my early morning tick surveys during the cold part of the day are less likely to gather ticks because they are lethargic due to the cold temperature? Which calls into question the effectiveness of my sampling and my hypotheses?
Let's move to the Delta now. In understanding Delta smelt* salvage at the giant export pumps in the south Sacramento-San Joaquin Delta, one faces a similar problem. The pumps and salvage facilities (where fish are "salvaged," i.e. counted and released but likely killed) can be considered to be a sampling device, drawing in the water in the Delta. It is like running a fishing net through the Delta all day every day. They catch millions of fish each year.
Delta fish are entrained into the flow of water to the pumps, but before they get to the pumps, they are carried through the poor habitat of the southern Delta, where they are eaten by predators. So the salvage numbers are just a smaller indicator of the much larger number of fish killed by entrainment.
So you can see the parallels with tick surveys. You don't know you've got a tick until you see it on your clothes... but observing fish in salvage is like waiting a day and seeing a tick on yourself the next day. By then you might have to dig it out, so you want to catch it right away, brush it off, and avoid hiking in that area. With Delta fish, operators want to see fish getting entrained right away so they can slow the pumps and not suck the fish into the south Delta. So there are early detection surveys--of course, you don't want to depend on these because they might miss fish, operators are slow to respond to the data, and what do you do during a government shutdown?
In a complex ecosystem, it is hard to know what the fish (and ticks) are doing, so the conceptual models of when they (you) are more at risk are more valuable, so you can reduce pumping (or avoid ticky areas) at those times BEFORE you cause a problem. So you begin to develop rules based on observations. But what do you do when the observations don't seem to match your rules? You must adapt as you go and act conservatively.
And what does all this have to do with avalanches? Well, anytime a skier skis on a potentially dangerous slope, they are doing an avalanche survey. If they are trained in avalanche safety, they will hopefully never trigger an avalanche, because there are very good models and forecasts and the steep areas can be avoided when they are dangerous. But everyone always wants to push the limit as far as possible. If the 60-degree slope near the top of the ridge is dangerous, perhaps the 45-degree slope lower down is okay. That is where the steep and deep powder is, and skiing in those fun areas is the whole point of being out there. So a skier is always playing a game of risk, but the only time there is positive feedback is when it is too late and you trigger an avalanche. So you end up with self-fulfilling prophesies--nothing happened, so it must have been safe. But you don't know how close to the edge you were. The next time it might not turn out okay.
There are a lot of examples of this, where you don't get feedback until it is too late--eating sugar and getting diabetes is another. Or speeding and not getting into an accident.
By the time you know for sure, it is too late. You must learn as much as you can,** and act conservatively. The consequences of mistakes are too great.
*Note that I discuss Delta smelt here because it is the species that gets the most public recognition and attention, and often is the species with potential to control exports that has the least known about its distribution. However, salinity control, water quality, and endangered salmon runs are responsible for far more restrictions on exports than Delta smelt. In many years Delta smelt do not trigger export restrictions at all, contrary to the impression one would get from the complaints of agribusiness interests who would like to eliminate the Endangered Species Act. I have a paper submitted on this topic that I'll link to from here once it is published.
**I wrote this blog, and later that same day I saw this awesome study in the Journal of Medical Entomology that answers many of my questions about tick habitat suitability: Modeling Climate Suitability of the Western Blacklegged Tick in California
5/11/18 Addendum
The day after I wrote this, there was a very informative Forum radio show on ticks and Lyme disease in Northern California. Also, in the week since I wrote this, I've observed ticks on every Open Space trail in grassy areas on which I've been. So this year, it was very striking how certain areas went from no ticks in April to lots of ticks in early May.
No comments:
Post a Comment