Building a harmful algal bloom early warning system in western Lake Erie.
In Lake Erie, harmful algal blooms (HABs) typically begin as nutrient-rich water from the Maumee River drains into the warm, shallow western part of the lake.
These conditions enable the growth of a type of algae that creates a toxin called microcystin. In 2014, such a bloom caused the microcystin levels in Toledo tap water to exceed what is recommended by the World Health Organization, triggering a two-day “Do Not Drink” advisory.
Thanks to a grant through the IOOS Ocean Technology Transition project, GLOS partnered with seven other organizations to build an early warning system that will help to address this pressing regional health and safety concern.
Tim Kearns, CIO of GLOS, talks about this network of sensors feeding into the automatic alert system prototype from this project.
The project aimed to create a system that will keep people informed when portions of the lake develop harmful algal blooms (HABs). This was possible by bringing together live data from forecasts and a network of in-water sensors and other monitoring equipment, processing that data, and sending actionable text message alerts when conditions worsen.
By making high-quality, live information readily available, decision-makers in the western basin can better anticipate HABs and react more effectively.
In order to pinpoint the information needs of those who would be receiving an early warning and to inform a future business model to sustain the system, researchers at The Ohio State University surveyed hundreds of stakeholders and compiled a stakeholder assessment report.
At the core of the system is a network of sensors or “sondes” distributed through the at-risk areas. This network has expanded as more water treatment facilities deploy sensors, many purchased with project funding. While helpful in detecting algae that could produce toxins, this sensor network cannot answer the question, “Is the water toxic right now?”
Environmental sample processors (ESPs) from NOAA can help to answer this. An ESP, or “lab in a can,” automatically tests water for toxins, a process that, until recently, had to be performed by hand. This means that samples can be taken more frequently which increases the accuracy of harmful algal bloom predictions. When added to the network, this data will help enrich information sent via alerts.
Another important part of the system are HAB forecasts from NOAA. These forecasts are issued after sampling confirms presence of the HAB species microcystis during bloom season (typically July to October) when conditions are optimal for growth. The Lake Erie HAB Forecast is updated daily and provides information on current extent and trajectory of HABs as they form.
By combining these disparate data sets, alerts can become a high-value, unified decision support tool that actively communicates with users when they need to pay close attention to changing conditions, freeing them from continually monitoring multiple information sources.
The team then worked to develop the backend technology necessary to receive and process sources of data and send timely text alerts. A prototype user interface was developed and received favorable reviews from testers in 2019.
The GLOS team plans to incorporate more data sources, including real-time data from the ESP and build the alert functionality into the Seagull platform. In addition, the team will publish a series of recommendations that can be used to inform further buildout of the system.
A demo of the prototype web application shows a parameter-triggered text message alert with a link to a dashboard that visualizes the latest conditions.
More on Funding
Funding for this project has been provided by the U.S. IOOS Ocean Technology Transition Project. This national project sponsors the transition of emerging marine observing technologies, for which there is an existing operational requirement and a demonstrated commitment to integration and use by the ocean and Great Lakes observing community, to operational mode.