Loops, ecosystems and indicators

What is the difference between ecosystems and other systems? The main difference is that in ecosystems feedbacks are present. Ecology deals with a disparate array of things: flow of matter and energy, bioclimate, niches, population biology, competition, evolution, food webs, macroecology, but I am convinced that the unifying concept is that of feedback. Other systems can be very complex, but they present no loops. An exception is perhaps economy, where a few feedbacks are present.

It is often not realised that when you have a causal loop, like in the figure, you can do no statistics and even any quantification is difficult

In fact you cannot have differet fixed levels of A in order to study the relationship AB; when you increase A f a quantity g, it will increase a larger (or a lesser) quantity g + h due to indirect relationship via C.

But there is a way to turn around this problem. If you have a causal branch in the network that ends in D, measuring D you can asses the state of the loop ABC. The element D is an indicator. Ecologysts often frown at indicators they are reductionist, and they believe that elements indirectly related to a system are unimportant.  And yet indicator are the only way to quantify the elements of the system when loops – i.e. feedbacks – are present. The usual method is to build a reasonable model and to compare the prediction with the patterns observed. The problem is that the number of models that are compatible with observation is usually immense. Moreover, history of science teach us that often scientific laws are not reasonable –  it is neither intuitive nor reasonable that the sub turns around the earth, it seems exactly the opposite.

But I am afraid that it will take much untile ecologists will recognize that they look at ecosystems the worng way. Think for a moment to the successful theories in ecology, like island biogeography, they always concern systems without loops. The problems begin when you have loops.