The structure of marine pelagic food webs determines the fate of organic carbon and productivity, but it is difficult to measure. We compared two common methods (stable isotope analyses, SIAs, and biovolume spectrum theories, BSTs) of estimating trophic positions (TPs) of mesozooplankton. Two sets of stations across the North Atlantic (Iceland Basin, Irminger Basin, Labrador Sea) were clearly separated. In the East, we observed a very early spring bloom, with mixed layer depths >500 m, chlorophyll a evenly distributed, and the Calanus population was dominated by CV/adults. Here, TPs based on both methods were comparable, with a TP of 2 for small zoo-plankton and 2-3 for larger species. In the West, a more advanced stage of the bloom was observed, with mixed layer depths <100 m, surface maxima of chlorophyll a, higher proportions of young stages of Calanus and more abundant microzooplankton. Here, significant differences in TPs were observed, with those based on BST being similar to 1 and 3 higher than those based on SIA, for small (TP similar to 3) and large (TP similar to 5) species, respectively. We conclude that BSTs capture energy flow through the microbial food web that is undetected by estimations using stable isotopes.