Читать книгу Practical Field Ecology - C. Philip Wheater - Страница 54

There are two main approaches to collecting field data: experiments and surveys. An experiment involves the manipulation of a system, whilst a survey depends on observations being taken without manipulation. For example, if we were interested in how many invertebrates could be found under logs of varying size, we could either survey a woodland floor finding as many logs as possible and recording both the number of invertebrates and the size of the log, or we could devise an experiment where we placed logs of differing sizes on a woodland floor and after a period of time examined the number of invertebrates underneath them. The advantage of the experimental approach would be that we could standardise all aspects of the logs except for size; for example, age, the degree of decay, the type of wood, and the distance between logs. All of these factors may influence the invertebrates found and confuse any relationship with log size. However, with a survey we would get an impression of what was happening in a real‐life situation (i.e. under logs that had been naturally deposited). Moreover, we may decide that the experimental approach is damaging to the environment; here, artificially placing logs in a natural system. In addition, for practical reasons, we might decide that the colonisation of newly introduced logs by invertebrates would take longer than the time available for the project to be completed. In most environmental research programmes, surveys are useful for generating ideas about important factors, but because of the additional complexity in real situations, surveys cannot identify cause and effect. Because experiments strip away the additional complexity, they are more useful in identifying cause and effect, but less likely to be applicable to real‐life situations.

When designing experiments, it is important that as many factors as possible are kept constant. So, for example, if we are interested in identifying whether an increase in insecticide concentration will lead to a decrease in aphid infestation of a crop, then the same amount of water (assuming this is the solvent or carrier for the pesticide) should be used for each application (irrespective of the concentration applied) so that we are testing the amount of pesticide added, rather than the amount of water added. In addition, it is important, where possible, to include a control treatment. In this example we would use a water only treatment to see if the addition of any water had an impact. If we did not do this and found a reduction in aphid numbers with any application of pesticide, we would be unable to tell whether this was due to the pesticide or the fluid added.