Insight into the ecology of our world is difficult to obtain. Organisms, habitats, and the physical world are interconnected as ecosystems in such a way that studying one in the absence of the others can be misleading. It’s a bit like juggling with one ball: sure it’s possible, but it is really even juggling anymore? We also know that that our observations of the environment are fundamentally flawed. If you want to know how many birds are in the local park, you go out and count them. But, to determine how many actually use the park, you need to know how many you counted, how good you are at counting animals, when they are around, and how often they move in at out of the park.
This is why experimental design and statistical modeling are so important to ecology. Experimental design allows us to keep all the juggling balls in the air while we focus on the one that is important. Statistical modeling helps us understand where we err, why it happened, and how it influences our understanding of ecological processes. To make accurate predictions about ecosystems—that is, to describe how the world works and project what will happen in the future—is a daunting task. Through quantitative ecology, we can make these predictions better by assessing the accuracy of our knowledge and striving for improvement in our science.