More Data Not Always Better For Hypoxia Models
Increasing use of ecological models for management and policy requires robust evaluation of model precision, accuracy, and sensitivity to ecosystem change. For coastal hypoxia, models are used to explore the underlining causes and to make specific management recommendations on nutrient management.
Therefore, optimizing their calibration becomes an important management, time and cost issue. Scientists at the University of Michigan supported under the National Centers for Coastal Ocean Science’s Coastal Hypoxia Research Program (CHRP) conducted an evaluation of hypoxia models for the northern Gulf of Mexico and Chesapeake Bay using historical data and by comparing model results based on 3-yr, 5-yr, and 7-yr datasets for model calibration.
For both systems the model sensitivity, precision and accuracy were optimized by calibrating the model to relatively short, recent 3-year datasets. Thus, hypoxia models may be as accurate using shorter time series datasets rather than longer time series datasets. These results were published by University of Michigan scientists Mary Anne Evans and Don Scavia in the journal Environmental Research Letters.