Complex systems often operate on multiple scales. Remember the boiling water example? Imagine the path of a single water molecule, followed for a short period of time; probably no pattern will be seen. But taking a step back, we see a general circular pattern when many molecules are watched for the same time period (or a single molecule is watched for a longer time). Thus, patterns may appear and disappear depending on the scale of the study. A process driving a pattern at one scale may be unimportant in the bigger picture (although these discontinuities result in heterogeneity).

...

Scale must be considered in any study. Experimental ecologists like to draw inferences about large-scale processes based on small-scale experiments. This assumes that either 1) the pattern and underlying process do not change with scale, or 2) the pattern changes in a predictable fashion, and there are no threshold effects (e.g. percolation theory). For organisms, "scaling deals with the structural and functional consequences of changes in size or scale among otherwise similar organisms" (Schmitt-Neilson,1984). The size of an organism is perhaps its most apparent characteristic. In ecology, many attributes of ecology, life history, and physiology can be described by power laws based on the organism's size (see Functional similarity section above). Understanding the origin of these biological scaling laws will provide critical insight into the mechanistic processes operating within ecological systems.