Important Earth Observation System objectives are to quantify Chlorophyll a (Chla) in marine environments, know the measurement uncertainty and ensure data are valid over the long-term. Most satellite missions require on-orbit accuracy to within 35% and for remote sensing calibration, HPLC is the reference method. Assuming total uncertainty is split evenly between satellite and field components, the HPLC accuracy for Chla must be within 25%, but accuracy within 15% (presumably) offers opportunities to improve remote sensing algorithms. Intercomparisons among laboratories measuring marine pigments have demonstrated that average accuracy for Total Chla to within 6-7% is possible for methods that include a quality assurance (QA) process. For other pigments, average accuracy to within 25% is possible, but this can be improved to 15% if methods are quality-assured. An important result of the HPLC intercomparisons was the identification of performance criteria that allow a determination of whether a method is capable of achieving specific accuracy requirements. Inter-laboratory comparisons cannot, however, describe whether a laboratory�s results will remain accurate at all points in time; for this, a QA Plan is needed. A QA plan includes repetitive application of quality control measurements and routine assessment that such measurements fall within quantified limits, for in the absence of such rigor, it is not possible to know whether the accuracy requirements set for researchers will be achieved. Horn Point Laboratory proposes to provide HPLC pigment analyses for field samples of NASA investigators. Our qualifications include: attainment of consistently accurate results in inter-laboratory comparisons (averaging 4.3% for TChla and 11.1% for other pigments), implementation of a QA plan, sixteen years experience providing HPLC pigment analyses to multiple clients, and fourteen years experience researching method improvements. Results will be audited by a NASA representative for compliance with performance criteria and research will be conducted to refine QA procedures.
