From https://doi.org/10.1029/EO063i022p00522-01
Measuring the rates of biological processes in the open sea remains a challenging problem to experimentalists. The problems are especially acute in the central gyres of the subtropical oceans where the concentrations of plankton and nutrients take minimum values (i.e., in oligotrophic waters). Historically, this led to the use of radioisotopic tracers to measure fundamental rate processes of the plankton such as photosynthetic carbon assimilation. Such technology requires incubating samples of seawater in bottles. This has long been worrisome since enclosing the plankton for several hours may well lead to artifacts. There ought to be better ways of making the measurements; finding some is one of the PRPOOS tasks.
Another task is to sort out the flows of carbon and nutrient elements through the food web. Temperate coastal food webs have been viewed in the past as involving diatoms (primary producers) that are fed on by copepods (herbivores) and these by larger zooplankton and fish. Observations over the past decade in the central gyres and other nutrient impoverished waters indicate that most of the phytoplankton is tiny (<10 fxm) and is not diatoms, that most of the herbivores are protozoans, and that bacteria are much more important in biological and chemical transformations than we knew 10 years ago. Copepods are of course present and in fact, remain the predominant macrozooplankton just as they are in temperate coastal waters. Similarly, the tiny forms are ubiquitous, but they seem to be relatively more important in the oligotrophic, oceanic waters.
Understanding the relevant spatial scales of processes that influence plankton production and mortality is another difficulty. Some of the evidence that bottle experiments underestimate plankton production requires physical measurements related to mixing processes for their understanding, especially with respect to nutrient transport into the euphotic zone. Such physical studies should be considered once the biology is sorted out.
Optical data on light penetration into the central oceans, and that absorbed by phytoplankton, along with the expected quantum yield of photosynthesis, set upper limits on the expected rate of photosynthesis of the order of 500-1000 mg C m^-2 d^-1. Present 14C tracer measurements in these waters give lower values. In the past, this has been interpreted as suggesting sub-optimal growth rates of the phytoplankton resulting from nutrient, especially nitrogen, deficiency. Recent laboratory results suggest this may not be the case and that the low 14C values may be artifacts of the bottle incubation. Such artifacts could derive from several causes including trace metal poisoning with contaminated bottles, sampling gear or 14C solutions; mortality of phytoplankton due to herbivore grazing or sensitivity to surfaces; and microscale patchiness in nutrient fields which are destroyed in bottle incubations, etc.