HPLC analysis of algal pigments: A comparison exercise among laboratories and recommendations for improved analytical performance

Pure individual and mixed pigment standards were distributed among Joint Global Ocean Flux Study (JGOFS) pigment analysts to estimate the variability of their spectrophotometric and chromatographic systems. To monitor the integrity of the pigments during the comparison exercise, chlorophyll and carotenoid standards were archived and periodically analyzed by high-performance liquid chromatography (HPLC). Pigment standards stored in the dark under nitrogen at - 20 degrees C were found to be stable for periods of at least one year. Results from three separate intercalibration exercises document a better agreement for spectrophotometric analyses than for HPLC. For the spectrophotometric comparisons, 90\% of the pigments analyzed by participant laboratories were within +/- 6\% of the mean ''consensus'' values. By contrast, 65 and 85\% of the laboratories agreed to within +/- 10 and +/- 20\%, respectively, when chromatographic analyses were compared. Chlorophyll absorption measurements obtained with a diode array-type spectrophotometer were 6-9\% lower than those obtained with monochromator-type spectrophotometers. These underestimates probably result from chlorophyll fluorescence contamination associated with the optical configuration of the diode array spectrophotometer. It was also determined that HPLC methods which are not capable of separating monovinyl chlorophyll a from divinyl chlorophyll a can produce 15-25\% overestimates of total chlorophyll a concentration in Prochlorococcus-dominated oceanic waters. A simple dichromatic approach is described for eliminating this variable source of error caused by co-elution of these structurally-related pigments. The use of internal standards and periodic calibration checks with external standards is highly recommended for improving analytical performance.