Multiple regression models for hindcasting and forecasting midsummer hypoxia in the Gulf of Mexico

Abstract

A new suite of multiple regression models was developed that describes relationships between the area of bottom water hypoxia along the northern Gulf of Mexico and Mississippi-Atchafalaya River nitrate concentration, total phosphorus (TP) concentration, and discharge. Model input variables were derived from two load estimation methods, the adjusted maximum likelihood estimation (AMLE) and the composite (COMP) method, developed by the U.S. Geological Survey. Variability in midsummer hypoxic area was described by models that incorporated May discharge, May nitrate, and February TP concentrations or their spring (discharge and nitrate) and winter (TP) averages. The regression models predicted the observed hypoxic area within +/-30%, yet model residuals showed an increasing trend with time. An additional model variable, Epoch, which allowed post-1993 observations to have a different intercept than earlier observations, suggested that hypoxic area has been 6450 km2 greater per unit discharge and nutrients since 1993. Model forecasts predicted that a dual 45% reduction in nitrate and TP concentration would likely reduce hypoxic area to approximately 5000 km2, the coastal goal established by the Mississippi River/Gulf of Mexico Watershed Nutrient Task Force. However, the COMP load estimation method, which is more accurate than the AMLE method, resulted in a smaller predicted hypoxia response to any given nutrient reduction than models based on the AMLE method. Monte Carlo simulations predicted that five years after an instantaneous 50% nitrate reduction or dual 45% nitrate and TP reduction it would be possible to resolve a significant reduction in hypoxic area. However, if nutrient reduction targets were achieved gradually (e.g., over 10 years), much more than a decade would be required before a significant downward trend in both nutrient concentrations and hypoxic area could be resolved against the large background of interannual variability. The multiple regression models and statistical approaches applied provide improved capabilities for evaluating dual nutrient management strategies to address Gulf hypoxia and a clearer perspective on the strengths and limitations of approaching the problem using regression models.

Ensemble modeling informs hypoxia management in the northern Gulf of Mexico

“…We estimate that the probability that hypoxia would be <5,000 km 2 under an 80% load reduction is 87%. The 59% reduction is larger than the 45% reduction called for in the most recent Action Plan (17), but within the range of previous individual models (5,30,31,(33)(34)(35)(36). This percentage is also higher than recommendations made for other eutrophic systems.…”

Relative role of wind forcing and riverine nutrient input on the extent of hypoxia in the northern Gulf of Mexico

“…The shorter period begins with 1993, i.e., the year in which the relationship between hypoxic area and riverine N-loading was reported to have changed [Turner et al, 2008;Greene et al, 2009;Liu et al, 2010;Forrest et al, 2011]. [7] The predictors in our single and multiple regression analyses are: the new variable t Uwind (defined below), the 11-month averaged MR flow ( F 11 ) [Wiseman et al, 1997], the averaged spring flow ( F spring ; April, May and June), the averaged May flow ( F May ) [Greene et al, 2009], the May NO 3+2 loading [Turner et al, 2006] and the combined MayJune NO 3+2 loading [Donner and Scavia, 2007]. The regional wind data are from the North America Regional Reanalysis (NARR), which is a high resolution weather data set covering our study period at 3-hr intervals and 0.3 spatial resolution.…”

“…Greene et al [2009] used a multiple linear regression model to predict the area using nitrate and phosphate concentrations and river discharge. An important conclusion of both Turner et al [2008] and Greene et al [2009] was that the relationship between the hypoxic area and N-loading changed in the early 1990's. Scavia et al [2003] reproduced the hypoxic area by using a one-dimensional model driven by the May-June total N-loading from the MAR.…”

Sources, factors, mechanisms and possible solutions to pollutants in marine ecosystems

“…This can lead to further eutrophication of DOM-rich waters. Indeed, more extensive eutrophication and hypoxia have been observed in river-dominated ocean margins because of climate and land use changes (Bianchi et al, 2009;Greene et al, 2009;Howarth et al, 2011).…”