Organisms need the right mixtures of chemical elements to build their bodies, acquire energy, and reproduce. The study of these relative demands is called ecological stoichiometry, and it helps us predict when an organism will release nutrients into the environment (i.e. mineralization). To make this prediction, we need to identify the element that is limiting organism growth and reproduction.
Measuring the nutrient limitation of organisms is difficult under natural conditions. Fertilization experiments are useful, but can have unintended consequences (e.g. acidifying the soil) and cannot separate direct effects of nutrients from indirect effects of organism interactions. Instead, Zoë Lindo, Paul Frost, and I are using an organism's biochemistry to assay nutrient limitation. The idea is that organisms are biochemically constrained, so limitation leads to an accumulation of the limiting nutrient in essential compounds. As a consequence, the metabolites present in an organisms body and the abundance of RNA coding for the relevant proteins should change. We use machine learning or discrimination statistics to identify patterns of limitation and test individuals from the field whose nutrient limitation is unknown.
Our research will make a significant contribution to our understanding of nutrient cycling. For example, we will be able to explain when and why co-limitation by multiple nutrients develops. Our research will also help test whether models with multiple nutrients are predicting the correct nutrient limitation.