Sensors for Monitoring Harmful Algae, Cyanobacteria and Their Toxins

17 Breakout Session D – How can detection methods be integrated into existing systems? • Q1: What performance assessments (QA/QC) methods are in use for current technologies? The group discussed these methods in terms of three available platforms. ESP – analytical accuracy and precision are addressed in the lab, but in the field the same level is not expected given that an integrated time sample is collected. In the lab, arrays are “calibrated” with standards that are run pre- and post-deployment, and there is no in-field calibration performed. Standard curves are based on cell counts and array intensities. Getting a “same true sample” in the field is challenging due to the instrument pumping for hours to get the desired volume. As a result, cell and DA detection is a sequential process and not from the “same” water mass. IFCB – obtains results from a 5 ml sample approximately every 20 mins and data are typically reported as a mean of observation for 1 hour. This time series provides a much more realistic view than a traditional single bottle sample (daily or weekly) which offers a ‘snapshot in time’. Comparisons with manual microscopy counts are performed for validation, and no standard QA/QC protocols have been established. ELISA – extraction efficiency can be measured through spiked samples and blanks. Standard curves are produced to determine dynamic working range, and internal controls should fall within the expected limit. Overall, QA/QC considerations can be a barrier to developing new methodology – the goal is to identify limitations in technology in prototypes, and to use that information to develop the next prototype. Some of the current metrics used in QA/QC may not be applicable to new technology. For example, CFU/mL is replaced by cycle time (CT) in qPCR, and CT is unique to this assay type. On the other hand, in practice QA/QC may not be the important driver. It could be pressures in the market place. Instrument makers should establish the metrics and provide performance specifications. In addition, there may be extra steps in the QA/QC process depending on application (e.g. FDA regulations versus ELISA field screening for toxins in shellfish) and agencies trying to develop standard protocols for calibration and operation. • Q2: What ground-truth methods are in use for current technologies? Combining data from one or more sensors can be very helpful for ground-truthing each of those platforms. This can be explored on different spatial scales (or temporal scales, as in over a season), for example coupling an IFCB to a fluorometric profiler or using a coordinated network of in situ observations to support remote sensing models/forecasts. One challenge to these more complex networks is obtaining representative field samples. Boat-based CTD casts are often utilized for field sample comparisons matched to time of sampling by a deployed instrument. As for detection assays, there is a need for more standard protocols for new molecular qPCR approaches. These would include defining specificity of primers and quantitative