Characterising the phenomic response of plants by means of meta-analyses
There has been a wide effort to characterise the growth, physiology and morphology of plants as dependent on their environment. Most experimental investigations treated plants with 2 or 3 levels of one or two environmental factors. Although this research is very useful, it does not allow for an easy generalisation of responses of plants to their environment. For such an approach, it would be more fruitful to obtain full dose-response curves of plants.
Dose-response curves have been constructed long ago for the yield of agricultural species in relation to nutrient supply (e.g. Mitscherlich 1909). However, for most plant traits and most environmental factors, information is too scarce to enable construction of such curves. It is, nonetheless, possible to derive dose-response curves from a larger amount of fragmentary data.
A huge amount of data is available in the scientific literature of the last 60-100 years. However, this information is not accessible in a structured way. We are currently filling this gap, building a database that is using published experiments from such different fields as plant physiology, plant ecology, environmental sciences, agronomy, forestry and horticulture. This database is subsequently used to construct dose response curves for a wide range of traits relevant for plant growth and performance.