Here is the better study.
http://www.int-res.com/articles/meps/8/m008p015.pdf
"The average atomic ratio for particulate matter from the stored system was 113:15:1 which is a good fit to the model (the Redfield ratio of C:N: P is 106:16:1). The concentration of particulate matter from the recirculating system was considerably lower than that from the stored system, but the average ratio of 98:14: 1 is insufficiently different from the Redfield ratio."
Now that I understand what evidence you're using I'll try to answer your first question: I dont believe you can use the CNP ratio of an organism to predict how to help it 'thrive or decline'.
I think varying nutrient levels in a given tank may be very helpful in limiting certain undesirable 'stuff' - just that the CNP ratios of the organism can not help you predict whether a given nutrient will help it thrive or decline (this was your thesis in your OP - I believe). Somehow we've gotten off the original topic so I'd like to bring this back there.
The Redfield ratio is used in the study of plankton. (not bacteria - not Cyanobacteria) so the study you quote (which focus on plankton) cannot be used to say the same the same things about 'non-harmful bacteria' or Cyanobacteria. So for a minute lets focus on phtytoplankton:
As I posted once before -
Redfield himself (and this is accepted currently in marine ecology) - decided that it was the phytoplankton themselves that were influencing N P ratios in Seawater - not the other way around (which also accounts for the results in your study which showed the 2 groups were statistically insignificant):
Redfield:
N P in seawater “must tend to approach that characteristic of protoplasm in general” (Redfield 1934). Furthermore, Redfield proposed thermostat like scenario in which the activities of nitrogen fixers and denitrifies keep the nitrate to phosphate ratio in the seawater near the requirements in the protoplasm. (Redfield, 1958). Redfield proposed that the ratio of Nitrogen to Phosphorus in plankton resulted in the global ocean having a remarkably similar ratio of dissolved
nitrate to
phosphate (16:1). He considered how the cycles of not just N and P but also C and O could interact to result in this match.
As to Bacteria and CNP (Cyano and heterotrophic). Here is 1 study abstract looking at CNP ratios in other types of 'bacteria' (i.e. not phytoplankton). "The baseline for studies of the elemental stoichiometry of marine microorganisms was established more than 60 years ago by Alfred Redfield, who estimated that marine plankton (and organic marine detritus) have a carbon/nitrogen/phosphorus (C/N/P) ratio of 106/16/1 (BNID
112423). The Redfield ratio remains the foundation for studies of the elemental composition of marine microorganisms and organic matter, particularly in the deep oceans [refs 4,15]. Indeed, the observed elemental ratios can vary greatly with the component of marine organic biomass that is under consideration: for example, estimates of
C/N/P for marine heterotrophic bacteria are reported to be around 69/16/1 (primary source 5), and measurements for individual cyanobacterial isolates include 46/10/1 (for Prochlorococcus sp. MED4 under phosphorus‑replete conditions) and 301/49/1 (for Synechococcus sp. WH8013 under phosphorus‑limited conditions) (primary source 12). Thus, the elemental stoichiometry of bacterial cells varies with taxa and growth conditions and often differs substantially from the Redfield ratio (reviewed in ref. 16)."
The numbers in that study are quite different than the numbers in your OP - (Non harmful marine bacteria with a typical N: P ratio of 50:1), Cyanobacteria has a N: P ratio of between 1:1 and 5:1) In their study - heterotrophic bacteria were 16:1 and Cyanobacteria 10:1 and 49:1 (depending on phosphate concentration). As you said - dinoflagellates are 'all over the map'.
Based on what I've read - certain NP ratios (in the water) can enhance or inhibit certain strains of microorganisms. However, the NP ratio of the organism itself cannot be used to predict the concentrations to use to inhibit or enhance it. Use the example of the cyanobacteria in the study above - in phosphorous depleted conditions - the NP ratio was 50:1. In phosphorous replete conditions the NP ratio was 10:1 (though they are different species as well). However - organisms also have mechanisms for storing elements for when conditions are 'lean' - so an organism may have an overabundance of an element simply because there is more of that element at that moment.