Thingstad TF, Cuevas LA
microbial food web, stoichiometry, Predator control, mesocosm, Ny Alesund, Svalbard, Norway, Denmark, 1 m3
Although explanatory and predictive powers are 2 closely interconnected aspects ofconceptual and mathematical models of complex systems, the two are not equivalent. The 2 aspectsare discussed here for the microbial part of photic zone food webs of the marine pelagic. We focus onthe specific question of how limiting nutrients are transferred from the dissolved form, through themicrobial food web, to mesozooplankton. For this purpose, 5 different nutrient addition experimentsare reviewed and compared to a ‘simplest possible’ conceptual food web model. The experimentsrange in scale from artificial food webs constructed in laboratory chemostats, via mesocosm experiments,to a Lagrangian open-ocean addition experiment and cover time scales from days to weeks.We conclude that main system responses in all cases can be explained within the framework of thesimple model, and that each experiment therefore also adds credibility to the basic concepts of thismodel. However, different system attributes profoundly affect the pathway and speed of nutrienttransfer in each experiment. A re-occurring theme seems to be how the interactions between flexiblestoichiometry and predatory processes modify experimental outcomes. Understanding the flexibilityin the behavior of the system has thus increased with each experiment, but the requirement for newad hoc assumptions to be added to the basic model structure in each case makes reliable predictionsof the experimental outcome appear only possible with further model elaboration