Feng, Yang (2012-05). Statistical and Realistic Numerical Model Investigations of Anthropogenic and Climatic Factors that Influence Hypoxic Area Variability in the Gulf of Mexico. Doctoral Dissertation. Thesis uri icon

abstract

  • The hypoxic area in the Gulf of Mexico is the second largest in the world, which has received extensive scientific study and management interest. Previous modeling studies have concluded that the increased hypoxic area in the Gulf of Mexico was caused by the increased anthropogenic nitrogen loading of the Mississippi River; however, the nitrogen-area relationship is complicated by many other factors, such as wind, river discharge, and the ratio of Mississippi to Atchafalaya River flow. These factors are related to large-scale climate variability, and thus will not be affected by regional nitrogen reduction efforts. In the research presented here, both statistical (regression) and numerical models are used to study the influence of anthropogenic and climate factors on the hypoxic area variability in the Gulf of Mexico. The numerical model is a three-dimensional, coupled hydrological-biogeochemical model (ROMS-Fennel). Results include: (1) the west wind duration during the summer explain 55% of the hypoxic area variability since 1993. Combined wind duration and nitrogen loading explain over 70% of the variability, and combined wind duration and river discharge explain over 85% of the variability. (2) The numerical model captures the temporal variability, but overestimates the bottom oxygen concentrations. The model shows that the simulated hypoxic area is in agreement with the observations from the year 1991, as long as hypoxia is defined as oxygen concentrations below 3 mg/L rather than below 2 mg/L. (3) The first three modes from an Empirical Orthogonal Function (EOF) analysis of the numerical model output results explain 62%, 8.1% and 4.9% of the variability of the hypoxic area. The Principle Component time series is cross-correlated with wind, dissolved inorganic nitrogen concentration and river discharge. (4) Scenario experiments with the same nitrogen loading, but different duration of upwelling favorable wind, indicate that the upwelling favorable wind is important for hypoxic area development. However, a long duration of upwelling wind decreases the area. (5) Scenario experiments with the same nitrogen loading, but different discharges, indicate that increasing river discharge by 50% increases the area by 42%. Additionally, scenario experiments with the same river discharge, but different nitrogen concentrations, indicate that reducing the nitrogen concentration by 50% decreases the area by 75%. (6) Scenario experiments with the same nitrogen loading, but different flow diver- sions, indicate that if the Atchafalaya River discharges increased to 66.7%, the total hypoxic area increases the hypoxic area by 30%, and most of the hypoxic area moved from east to west Louisiana shelf. Additionally, if the Atchafalaya River discharge decreased to zero, the total hypoxic area increases by 13%. (7) Scenario experiments with the same nitrogen loading, but different nitrogen forms, indicate that if all the nitrogen was in the inorganic forms, the hypoxic area increases by 15%. These results have multiple implications for understanding the mechanisms that control the oxygen dynamics, reevaluating management strategies, and improving the observational methods.
  • The hypoxic area in the Gulf of Mexico is the second largest in the world, which has received extensive scientific study and management interest. Previous modeling studies have concluded that the increased hypoxic area in the Gulf of Mexico was caused by the increased anthropogenic nitrogen loading of the Mississippi River; however, the nitrogen-area relationship is complicated by many other factors, such as wind, river discharge, and the ratio of Mississippi to Atchafalaya River flow. These factors are related to large-scale climate variability, and thus will not be affected by regional nitrogen reduction efforts.

    In the research presented here, both statistical (regression) and numerical models are used to study the influence of anthropogenic and climate factors on the hypoxic area variability in the Gulf of Mexico. The numerical model is a three-dimensional, coupled hydrological-biogeochemical model (ROMS-Fennel). Results include: (1) the west wind duration during the summer explain 55% of the hypoxic area variability since 1993. Combined wind duration and nitrogen loading explain over 70% of the variability, and combined wind duration and river discharge explain over 85% of the variability. (2) The numerical model captures the temporal variability, but overestimates the bottom oxygen concentrations. The model shows that the simulated hypoxic area is in agreement with the observations from the year 1991, as long as hypoxia is defined as oxygen concentrations below 3 mg/L rather than below 2 mg/L. (3) The first three modes from an Empirical Orthogonal Function (EOF) analysis of the numerical model output results explain 62%, 8.1% and 4.9% of the variability of the hypoxic area. The Principle Component time series is cross-correlated with wind, dissolved inorganic nitrogen concentration and river discharge. (4) Scenario experiments with the same nitrogen loading, but different duration of upwelling favorable wind, indicate that the upwelling favorable wind is important for hypoxic area development. However, a long duration of upwelling wind decreases the area. (5) Scenario experiments with the same nitrogen loading, but different discharges, indicate that increasing river discharge by 50% increases the area by 42%. Additionally, scenario experiments with the same river discharge, but different nitrogen concentrations, indicate that reducing the nitrogen concentration by 50% decreases the area by 75%. (6) Scenario experiments with the same nitrogen loading, but different flow diver-
    sions, indicate that if the Atchafalaya River discharges increased to 66.7%, the total hypoxic area increases the hypoxic area by 30%, and most of the hypoxic area moved
    from east to west Louisiana shelf. Additionally, if the Atchafalaya River discharge decreased to zero, the total hypoxic area increases by 13%. (7) Scenario experiments with the same nitrogen loading, but different nitrogen forms, indicate that if all the nitrogen was in the inorganic forms, the hypoxic area increases by 15%. These results have multiple implications for understanding the mechanisms that control the oxygen dynamics, reevaluating management strategies, and improving the observational methods.

publication date

  • May 2012