Connecting the green and brown worlds: allometric and stoichiometric predictability of above- and below-ground networks

We examine the potential of trait-based parameters of taxa for linking above- and below-groundecologicalnetworks (hereafter ‘green’and‘brown’ worlds) tounderstandand predict community dynamics. This synthesis considers carbon, nitrogen and phosphorusrelated traits, the abundance of component species and their size distribution across trophic levels under different forms of management.We have analysed existing and novel databases on plants, microbes and invertebrates that combine physico-chemical and biological information from(agro)ecosystems spanning the globe.We found (1) evidence that traits from above- and below-ground systems may be integrated in the same model and (2) a much greater than expected stoichiometric plasticity of plants and microbes which has implications for the entire food-webmass–abundance scaling. Nitrogen and phosphorus are primary basal resources (hence, drivers) and more retranslocation of P than of N from leaves will lead to higher N:P in the litter and soil organic matter. Thus, under nutrient-rich conditions, higher foliar concentrations ofNand P are reflected by lowerN:P in the brown litter, suggesting less P retranslocated than N. This apparent stoichiometric dichotomy between green and brown could result in shifts in threshold elemental ratios critical for ecosystem functioning. It has important implications for a general food-web model, given that resource C:N:P ratios are generally assumed to reflect environmental C: N:P ratios. We also provide the first evidence for large-scale allometric changes according to the stoichiometry of agroecosystems. Finally,we discuss insights that can be gained from integrating carbon and nitrogen isotope data into trait-based approaches, and address the origin of changes in D13C and D15N fractionation values in relation to consumer–resource body-mass ratios