Interferences in the analysis of nanomolar concentrations of nitrate and phosphate in oceanic waters

This paper reports on investigations into interferences with the measurements of nanomolar nitrate+nitrite and soluble reactive phosphate (SRP) in oceanic surface seawater using a segmented continuous flow autoanalyser (SCFA) interfaced with a liquid-waveguide capillary flow-cell (LWCC). The interferences of silicate and arsenate with the analysis of SRP, the effect of sample filtration on the measurement of nanomolar nitrate+nitrite and SRP concentrations, and the stability of samples during storage are described. The investigation into the effect of arsenate (concentrations up to 100 nM) on phosphate analysis (concentrations up to 50 nM) indicated that the arsenate interference scaled linearly with phosphate concentrations, resulting in an overestimation of SRP concentrations of 4.6+/-1.4% for an assumed arsenate concentration of 20 nM. The effect of added Si(OH)(4) was to increase SRP signals by up to 36+/-19 nM (at 100 microM Si(OH)(4)). However, at silicate concentrations below 1.5 microM , which are typically observed in oligotrophic surface ocean waters, the effect of silicate on the phosphate analysis was much smaller (< or = 0.78+/-0.15 nM change in SRP). Since arsenate and silicate interferences vary between analytical approaches used for nanomolar SRP analysis, it is important that the interferences are systematically assessed in any newly developed analytical system. Filtration of surface seawater samples resulted in a decrease in concentration of 1.7-2.7 nM (+/-0.5 nM) SRP, and a small decrease in nitrate concentrations which was within the precision of the method (+/-0.6 nM). A stability study indicated that storage of very low concentration nutrient samples in the dark at 4 degrees C for less than 24 h resulted in no statistically significant changes in nutrient concentrations. Freezing unfiltered surface seawater samples from an oligotrophic ocean region resulted in a small but significant increase in the SRP concentration from 12.0+/-1.3 nM (n=3) to 14.7+/-0.6 nM (n=3) (Student's t-test; p=0.021). The corresponding change in nitrate concentration was not significant (Student's t-test; p>0.05).