Years ago I heard a conservation biologist give a talk about conserving a bird species using habitat around an artificial lake in Arizona. His job involved working with non-scientist government employees. I actually found his lunch-time conversation after the talk more interesting than the talk itself. He said the most difficult thing about dealing with the government managers was that they wanted everything to be cut and dried. They wanted to know EXACTLY how much land was necessary to preserve the population of this bird to prevent it from going locally extinct.
To a conservation biologist this is an impossible request. The best you could do, if you had really really good long term data, is come up with some sort of probability distribution of the population going extinct based on the area protected (e.g. if you preserved 5000 acres then there is a 60% chance they will be gone in 20 years, 10,000 acres there is a 20% chance, 15000 acres a 5% chance, etc.).
Why is there this level of uncertainty? The bird population is a cloud. All kinds of unpredictable factors affect its long term viability: climate (both major events such as droughts and the overall trend), infectious diseases, fires, declines or increases in the populations of food insect species, and so on. Even with a really big population there is no guarantee of survival.
At this point I want to stop and say that I am not attacking the legal and political communities for the kind of viewpoint described above. If you have a trial it is necessary to have a verdict of guilty or not guilty. The jury can't say that they think there is an 85% probability that the defendant is guilty. Similarly a law has to be either unconstitutional or constitutional. An orderly society requires that decisions be made.
The problem comes when extremely cloud-like natural phenomena are viewed through the legal/political world view. In other words a cloud is treated like a clock. The most obvious (and important) example of this today is climate change. The climate is a cloud. General features are predictable over long time periods but the more specific you get, the short your time frame of accuracy (I'm not a climatologist so my numbers may be off a bit here but the general idea is right). Based on what we know it is reasonable to predict that the average temperature on earth will be higher 100 years from now than it is now. However a prediction about a particular period of time (e.g. will August be cooler or warmer than a normal August) can only be made a few months in advance. The prediction of the weather on a specific day (as we know) is accurate for no more than a few days in advance. And the prediction of the weather at an exact point in space and time (e.g. will it be raining at my house at 3:30 PM) has an even shorter predictive range.
So what's the problem? Lets look at imaginary example from another cloud system - evolution in populations. Imagine a remote oceanic island with sparse vegetation. An insect species colonizes the island and, as is common in these situations, slowly evolves shorter wings. Longer winged insects that fly more are more likely to be blown out to sea by the wind.
Over the long term evolutionary biology can tell what will (probably) happen. The average wing length in the population will get shorter and eventually the species will become wingless. However the specifics of what will happen to any one insect are completely unpredictable. Some long-winged insects will live long and happy lives and die leaving dozens of grandchildren.
We could, after the fact, divide the fate of the insects into those that were blown off the island and those that died from other causes. And then we could say that one group died as a result of being on the island and the other group didn't. However this statement would be not useful and untrue. Natural selection is caused by differences in survival and reproduction - all the deaths get factored into determining the strength of selection on wing length, not just the ones caused by drowning offshore.
Also, all the insects died because they are on the island. They wouldn't have existed in the first place as those particular insects if their ancestors hadn't made it to the island. The island is an integral part of the system in which the population exists and evolves.
So hopefully you can see where I'm going with this. Claiming that any individual weather event or season is evidence for or against climate change is nonsensical. We all (hopefully) here at DK readily agree that a cold and snowy winter is not evidence against climate change. Similarly a particular hot summer (by itself) is not evidence for climate change. It would be if the climate was a clock and all the parts always operated in exactly the same way. As it is someone can point out that some summers in the 1930s were equally as hot as recent summers (although winters were colder).
Unfortunately we as a society seem to be obsessed with assigning particular events to climate change or to other factors. This is nonsensical. Weather patterns are always due to local factors. Those particular combinations of local factors wouldn't exist if we hadn't increased the concentration of CO2 in the atmosphere. Every particular detail is going to be different than it would have been if humans hadn't started pumping extra carbon into the air. We have no idea how those particular details would be different any more than we would know the fate of the insects that would have be born, lived, reproduced, and died on the mainland if their ancestors hadn't ended up on the island.
The important data are things like: nine out ten of the hottest years on record have occurred since 2000, or the general trend of warming since 1850 when accurate world wide measurement began. These are indications of a change to the overall cloud that is the global climate rather than individual storms, droughts, and heat waves which are like the molecules of the cloud.
The problem is that humans tend to think of problems as clocks. This makes sense when we deal with the simple issues our ancestors (and most of us) face frequently. Building things, fixing mechanical devices, etc. These are clock issues. You do things the right way and things will work out. Even problems that are cloud like we often treat as clocks. Imagine you have two different routes you could drive home. The best solution is find out what the traffic is like and choose based on that information. If that is not possible then the next best solution would be to randomly choose a route each day and then time the trip. After a large number of replications you would take the average and decide which one is best overall. However my experience is that most people try each route once and then decide. This works well if there is no random variation but not if there is.
I am very comfortable in a world of uncertainty but most non-scientists aren't. I think the most valuable contribution to science policy would be a voting public that could be coaxed from clock thinking to cloud thinking. But I have no idea how to do that. Any takers.
I'm cross posting this to backyard science as this is exactly the sort of thinking underlying the gathering of phenological data.
Comments are closed on this story.