The Ecological Forecasting Initiative (EFI), in collaboration with the National Ecological Observatory Network (NEON), has conducted a forecast challenge to grow the field of ecological forecasting by using data collected by NEON and weather forecasts made by the National Oceanic & Atmospheric Administration (NOAA). We contributed to the challenge by forecasting dissolved oxygen (DO) using atmospheric pressure in a state space framework at an aquatic NEON site, Lake Barco in Florida. Utilizing Bayes Theorem allows for the uncertainties to be quantified and partitioned into different sources of error and process variability, and then allows us to propagate them into the forecast. By utilizing the positive correlation between DO and atmospheric pressure, we hypothesized DO concentrations to be higher along with higher levels of atmospheric pressure because water is able to hold more DO at higher atmospheric pressures due to gas saturation decreasing. The results of this project contribute to the EFI-NEON goal of understanding aquatic systems by showing how a covariate of atmospheric pressure contributes to and affects ecological forecasts of concentration levels of DO.
Louis Stokes Alliance for Minority Participation (LSAMP) Scholarship
SACNAS, INRSEP Research Symposium