Addressing the measurement of particulate organic and inorganic phosphorus in estuarine and coastal waters

• We evaluate methods for the speciation of particulate phosphorus in waters.
• Inorganic and organic compounds, algal cultures and estuarine samples were tested.
• Results are specifically discussed reviewing P speciation in phytoplankton.
• The choice of methods takes into account intracellular phosphate and polyphosphate.
• Our findings show clear evidence of the necessity to fully describe method procedure.

The speciation of particulate phosphorus (TPP) into inorganic (PIP) and organic phosphorus (POP), has been scarcely investigated in suspended particulate matter (SPM), contrary to numerous sediment studies. This work evaluates the actual capacity of methods for TPP, POP and PIP determination in SPM to measure efficiently and specifically each pool. Methods were tested on a range of plankton-associated organic and inorganic P compounds, on microalgal cultures and estuarine samples. Then results are specifically discussed reviewing P speciation in phytoplankton. For TPP analysis, the method of Solórzano and Sharp (1980) is recommended compared with that of Aspila et al. (1976), originally dedicated to sediment, as there is one step less, the protocol is more explicit (use of MgSO4) and it provides the best results. The application of the original protocol of the Aspila method for TPP (no MgSO4) results in underestimated and highly variable values. For POP, the CWO (Chemical Wet Oxidation) method tested here succeeds in providing sufficient oxidation of organic P, but, inorganic polyphosphate is yet too significantly hydrolysed, and intracellular phosphate, an underestimated PIP fraction, contributes to the POP pool with this method. Finally the Aspila method for PIP is the best compromise in estimating the contribution of PIP to TPP and POP by difference. Even if particulate polyphosphate is only partially hydrolysed (41–49%), intracellular phosphate is included and only a very low percentage of organic P, essentially nucleotidic P, is attacked. In any case, our findings show clear evidence of the necessity to fully describe the procedure when P speciation methods are used.