The seasonal Gulf of Mexico hypoxic zone remains one of the largest in the world, and several scientists have developed models to help forecast the size of the zone each year. Measurements of the areal extent of the zone have occurred since 1985, and those data typically become available between late July and early August.
Trying to anticipate these measurements provides a “battle of the models” opportunity, as modelers can create their predictions and test their estimates against the official numbers. So this past June, Gene Turner, a professor of Oceanography and Coastal Studies at Louisiana State University and one of the original researchers who started studying the Gulf hypoxic zone, challenged SEFS Professor Sergey Rabotyagov to a friendly scientific wager to test their alternative models.
Using 2011 and 2012 numbers, Rabotyagov’s model had been more accurate than Turner’s and the other existing model for predicting the hypoxic zone, which was created by Professor Donald Scavia of the University of Michigan. When the hypoxic measurements became available for 2013, though, all three researchers were surprised to find their models over-predicted the size of the hypoxic zone. Professor Scavia’s estimates were actually closest this year, but Rabotyagov’s model did outperform Professor Turner’s model—and since the wager was only between the two of them, Rabotyagov claimed the prize: A frilly Detroit Red Wings beanie.
Still, two out of three for Rabotyagov is pretty impressive, and he also stresses other positive takeaways from these models and data. “Although the comparison of observed and predicted hypoxia clearly suggests that our ability to predict any given year’s hypoxic zone with a high level of accuracy is limited,” he says, “all of the existing models emphasize the role that upland areas have in creating and also mitigating this important environmental issue.”
Turner and Rabotyagov are working together as a part of a National Science Foundation-funded study to figure out how best to reduce the hypoxic zone, and they will likely tweak and test their models again next year!
Photo of Professor Rabotyagov © SEFS; graph below © Professor Rabotyagov.
- Comparison of the three models, with Professor Rabotyagov’s in green. His estimates were the closest to actual numbers in 2011 and 2012, and in 2013 his model performed second of the three (but bested Professor Turner’s to win the wager)!
