Cyanobacteria or blue-green algae blooms have become more common as ocean temperatures rise and rainfalls become more intense. Exacerbated by stormwater runoff that carries fertilizer, sediment, detergents and animal waste, these blooms block sunlight, deplete oxygen and produce toxins that cause a host of health issues, particularly for pets and children.
On April 2-4, the inaugural Brown University Health Data Fest gave undergraduate students a chance to tackle cyanobacteria blooms using local water-quality data.
Over the course of three days, students worked in teams to analyze data for trends and insights into cyanobacteria growth in local bodies of water and developed a model that predicted the start and end of cyanobacteria bloom events, with degrees of confidence. Students also looked at how the frequency and duration of blooms are related to the weather and how these might continue to increase in the future.
They leaned on guidance from Sara Horvet and Molly Welsh from the Stormwater Innovation Center, whose multimedia presentation showed how environmental and public health data are collected, managed and analyzed, and how these data can be used to better understand risks, evaluate interventions and inform policies that protect water quality and public health statewide.
“The judges and I were so impressed with how much progress was made in such a short amount of time—just 36 hours!” said Peter Lipman, associate professor of the practice of biostatistics at Brown. “Beyond exploratory analysis, students built predictive models and created dashboards to monitor cyanobacteria levels. And many teams moved beyond the data provided, incorporating other datasets and academic research to further explain how local geographic features play a strong role in determining the underlying trends. Students really showed how complex processes are at play in health datasets.”
