While harmful algal blooms (HABs) are a significant issue in the Great Lakes that have been monitored using real-time data buoys, there is still no automated method for measuring the toxins produced by these blooms.

Todd Miller, an affiliate and associate professor at the University of Wisconsin-Milwaukee and principal investor at the Laboratory for Aquatic Microbiology and Chemistry, is currently working toward a solution that allows for the prediction of toxin concentrations through the use of real-time lake data.

The toxins produced by HABs—microcystins, cylindrospermopsin, and anatoxins—can be exposed to humans and animals through a variety of methods. While swimming exposure is common, Miller explains that these toxins can make their way into drinking water, food, beverages, dietary supplements, and other goods that use water for manufacturing.

“The best we can say right now is, ‘If it’s green, then stay out of the water.’ But that’s not always the best answer, because sometimes the toxins are there, even though the water may not look very green,” states Miller. And so, improving the way scientists measure HAB toxins is a key component of public health and safety.

A Panther Buoy deployed in Nicolet Bay (Green Bay) near Peninsula State Park in Wisconsin. (Credit: Todd Miller / University of Wisconsin-Milwaukee)

Predicting Harmful Algal Bloom Toxins

The first goal of the research is to characterize the frequency and timing of algal toxin production in aquatic environments, followed by determining potential risks to human health from algal toxins in aquatic environments.

The successful design and implementation of the toxin-measuring sensors on the buoys will improve existing monitoring methods and technologies for modeling and statistical analysis of toxins.

“While we are interested in building sensors for the toxins that are in the water, those currently do not exist or are still under development. In the meantime we are exploring use of real-time buoy data and other data sources to drive statistical models that predict toxin concentration,” states Miller.

Research technician Jeffrey Neuse loads the Mawikwe Panther Buoy onto the Apostle Islands work boat. (Credit: Todd Miller / University of Wisconsin-Milwaukee)

Miller builds his system out from a fleet of ten NexSens CB-150 data buoys with some newer systems on the smaller CB-75. He tests various sensors on the buoys and installs custom-made data loggers for real-time data collection and remote viewing.

Sensors must provide variables that are indicative of blooms and toxin concentration, so he chooses what sensors to use carefully.

He explains, “I have spent a considerable amount of time comparing sensor technologies for algal bloom detection. So, I know the limitations of most of the sensors on the market that can do this. We have taken them all apart and examined their individual components, and we have even built our own sensors for algal bloom detection.”

Miller continues, “So we really have an intimate understanding of how all of these optical sensors work at the circuit board level.” He uses this knowledge to design his own automated sensors that are trialed at the monitoring sites.

Technician Jeffrey Nuese holds the Panther Buoy dry well electronics, including the Panther Logger with cellular modem. (Credit: Todd Miller / University of Wisconsin-Milwaukee)

He also gathers additional water quality and weather data that is then shared publicly online. While a few of the systems are equipped with YSI EXO2 sondes, the majority use individual sensors for phycocyanin, dissolved oxygen, chlorophyll, and turbidity.

The Turner Designs sensors are ideal for Miller’s applications as they are easily replaced and upgraded, and being a shorter instrument overall helps them detect conditions closer to the surface.

Some of the buoys are also equipped with a temperature string to measure stratification and temperature throughout the water column, which contributes to HAB prediction and modeling.

On top of the buoys, Miller has a Young anemometer and a wave sensor, measuring wind speed and direction as well as wave height and period. Both of these environmental conditions can play a role in HAB formation and duration. Additionally, this data is used by the public to inform recreation on the water.

Todd Miller preparing to assemble a Panther Buoy for deployment in Lake Winnebago near Neenah, Wisconsin. (Credit: Jeffrey Nuese / University of Wisconsin-Milwaukee)

All of this data is compiled and transferred to servers at UW-Milwaukee, where custom-made PHP scripts sort out any outliers and then post it on the public dashboard at lakestat.com.

Miller analyzes the data and shares it with a number of organizations for modeling, reporting, and predictions, including the Great Lakes Observing System.

Each system has its own established maintenance schedule and is maintained regularly by Miller and his lab, who handle deployment and maintenance of the systems.

During the off-season, the buoys are retrieved from the water and sensors are stored in designated totes for each system. If necessary, the sensors are sent to Fondriest for repair.

Conclusion

According to Miller, this process of working on the instrumentation and trying to solve a problem impacting human and ecological health is rewarding and challenging in its own way.

“There is not a day I hate going to work. I love the multi-disciplinary aspect of our work studying harmful algal blooms. It involves biology, chemistry, physics, electrical engineering and getting out in the field,” explains Miller.

Research technician Jeffrey Neuse looks over the Mawikwe Panther Buoy prior to deployment in Lake Superior. (Credit: Todd Miller / University of Wisconsin-Milwaukee)