Friday, 29 October 2010

Topic 1: Uncertainty in Ecology

‘A little knowledge is a dangerous thing’

My first karate instructor once told me this, as he looked at me straight in the eyes. In my mind he was massive, a towering figure that had survived feudal Japan and somehow manifested as an East Londoner. Once I had looked back at some old photos I realised he was 65 and probably around 5.4ft. Nevertheless, he left a very distinctive mark on my young mind. I took the advice at face value at the time; learn more martial arts before you start running around thinking you know it all. Yet I recently found myself applying this statement to my own scientific musings.

The study of ecosystems is a young field, with the synthesis of underlying ideas from geography and natural history occurring gradually (somewhat ecologically even!) at the end of the 19th century. For a long time the study of biology was restricted to the gathering of samples in order to document which species existed, to describe their physiology and record their behaviour. The new science of ecology sought to analyse the numbers, births, deaths, movements and all manner of changes in organism’s populations. This is all very exciting and ambitious stuff, and it relies on vast quantities of information. You need to go out and find animals and plants, record where they are and where they are not at regular intervals over extended periods of time, this is no mean feat.

I am often struck about how little we actually know about the world’s animals and plants. OK, so many species are hard to spot, present in very small numbers and in remote locations. The places where lots of humans live and work have been transformed and many animals and plants we historically shared habitats with have gone. Even so, the amount of knowledge we have today is pretty small, when compared to other fields.

The marine census being carried out worldwide at the moment is presented as a breakthrough, a global stocktaking of what we have in the world’s seas. I am surprised and shocked that no one has considered anything of such a nature before. Individual nations have records of fisheries abundance as this is something that has direct economic concerns but these records are weak and patchy outside of certain states and particular fish stocks. Censuses of human populations have been carried out for thousands of years, analysis of voting trends and consumer information has reached amazing levels, stock taking and yield analysis in agriculture is widespread, yet so little time and effort has been placed globally on the millions of other species that share the earth with us.

The fact that we know little is worrying, the fact that we are trying to predict things when we know so little is even more worrying. The acknowledgement that our knowledge of ecological systems and the data behind them is often uncertain, patchy or biased has dragged behind the science of modelling these systems.

Modelling methods developed in economics, statistics, physics and engineering are all being applied to ecological problems. Yet very few address the fundamental issue that we only know a tiny fraction of what is going on out there. Natural systems are very different to nearly all man made systems; if we build an energy grid or transport system we know all the links in it. The same cannot be said for a food web, there are so many uncertain links. Ecology is a big deal; other organisms determine what we eat, the air we breathe, our climate and our water supply. All these things are essential to human life. Essential, that is the important word here, the consumer price index, the nature of radioactive decay, advances in oil drilling efficiency, IPods, medicine, transport and money are all useful, all make life easier (In many cases a great deal easier) but are not essential every day for humans to live. Ecology is the science of food, air and water security. I am still surprised that there is little knowledge, because it is becoming an increasingly dangerous thing.


  1. No equations ?! That is BS and you know it.

    I think it is interesting that the more you look into a discipline the more you realise how little we actually know about it, and how that's kind of scary.

    I was talking to a Mathematician who had just come back from seven years as a Mathematical Biologist and he was talking about how difficult it was to do anything truly mathematical with biological systems because we have barely any of the mathematical tools we need to really have a decent go at the problems. It's actually quite shocking to think of how incapable we are at understanding even simple PDE's (take the Navier Stokes as an example) fluid flow seems like an obvious, and well understood, system and we can't even prove the equation has unique solutions.

    It could be hundreds of years before there is a mathematics that would be sufficient to base a 'theoretical biology' on (although you probably know way more about this than me) (& that is assuming a linear growth of mathematical understanding, it'll probably be twelve seconds after the singularity occurs and then we can move on).

    I think one of the reasons it is easy to believe that humanity is 'so knowledgeable' is because the research horizon in most fields is beyond people's sight and it's easy to extrapolate. "There are people who know way more than me about this subject, therefore they probably know most of what there is to know" is a simpler thing to understand than "there are people who know way more than me about this subject and they know basically nothing and the possibilities of knowledge stretch off to an infinite horizon".

    The latter has recently caused me to non-trivially freak out, it's nice in the middle where there are no edges, it's hard going on the open ocean.

    Cool idea for a blog, hope you keep it up, seems we have a similar Christianity leads to science and eastern religion deal going on so I'll probably post a lot on here.

    It'll be mostly flame warz and trolling.

    And thereby I will ownz you.

    - Jon

  2. Hi Jon,

    Good to hear from you, I thought this may be right up your alley. I agree with you that many people assume there is this nice tight association between what we think we know now and what there is to know, a fact that is so terribly wrong that it has blown up in our collective faces on more than one occaision.

    As far as coming to exact solutions to Biological problems, I basically dont see that ever happening (well maybe post singularity as you say!). Biological systems are noisy, chaotic and complex. Biological problems lend themselves to short term predictions made by apporiate but rough and ready models. I think we as scientists (and just about everyone else) need to move away from thinking about 'the' answer and towards things about 'an' answer that works for now.

    Hope to hear more from you,