Sensors for Monitoring Harmful Algae, Cyanobacteria and Their Toxins

15 2) The complexity of ‘what to regulate’ was discussed. Specifically, what defines a ‘bloom’ varies widely between species and what defines a toxic event varies widely among compounds, so it’s very difficult for regulators to make decisions about what levels and parameters constitute regulatory concerns. While this remains a challenge for several HAB genera/species, in other instances there are guidelines in place (e.g. the microcystin 1.0 ug/L threshold for human consumption recommended by the World Health Organization; though this may not be a direct correlation to abundance of cells in the environment). 3) Discussion focused on how data types often do not translate to ecological processes. When managers make observations they want to know processes that drive events (e.g., specific nutrients, grazers, food web disruption, presence/absence of important conspecific species). The idea behind this is if a causative agent can be managed (e.g. N or P, salinity intrusion, etc.), then we can more effectively prevent blooms in the first place. Management should be pro-active more than reactive, saving both time and money in the process. 4) Sample frequency is often seen as inadequate. Loss of info and data results in less certainty in decision process: increased sampling frequency may allow for better answers. 5) Interpretation of data and technology needs to be simple and clear to understand. 6) Technology platforms are not addressing trophic transfer. This is especially key in dynamic systems where currents and mixing from upwelling can result in the rapid movement of cells and toxin (e.g. Monterey Bay). 7) Large scale mapping capabilities are often inadequate, thereby providing a limited scope of an actual event or conditions leading up to that event. 8) Mobility can hinder the use of technologies, especially for mapping purposes, however platform development is currently underway to address this (e.g. 3G ESP). This is an important feature to work towards for both ‘Tesla-level’ and ‘Honda Civic-level’ platforms. 9) Surveillance and forecasts need to include both cells and toxins. 10) There would be great benefit to having more generic platforms that have the ability to house a variety of HAB sensors. 11) More assessment of current research is needed, with guidelines and algorithms to help stakeholders. • Q4: Are there viable solutions in the R & D pipeline? 1) New platforms are in development that allow for mobile adaptive sampling (e.g. mobile ESP, mobile IFCB) 2) Better photo plankton class discrimination is being pursued on imaging platforms for better classification abilities (e.g. IFCB)