A Geostatistical Model of Hypoxia Formation in the Northern Gulf of Mexico

EPA Grant Number: FP917105
Title: A Geostatistical Model of Hypoxia Formation in the Northern Gulf of Mexico
Investigators: Obenour, Daniel Redd
Institution: University of Michigan
EPA Project Officer: Packard, Benjamin H
Project Period: September 1, 2010 through August 31, 2013
Project Amount: $111,000
RFA: STAR Graduate Fellowships (2010) RFA Text |  Recipients Lists
Research Category: Academic Fellowships , Fellowship - Water Quality: Coastal and Estuarine Processes


Hypoxia is an environmental problem common to many coastal waters, and that is particularly severe in the northern Gulf of Mexico. The purpose of this project is to enhance our understanding of Gulf hypoxia formation through the development of a geostatistical model capable of integrating data from a variety of relevant sources. The model will be used to estimate the extent and severity of hypoxia through time, and to evaluate factors related to its formation.


Many coastal ecosystems are subject to an environmental condition known as hypoxia, which is defined by oxygen levels too low for the support of aquatic life. A particularly severe example of this problem is the hypoxic zone in the northern Gulf of Mexico. This project will focus on the development of an innovative geostatistical model for assessing Gulf hypoxia. The model will be used to better understand the causes of hypoxia, and to estimate the size of the hypoxic zone throughout time.


This research will focus on development of a geostatistical model that will incorporate dissolved oxygen measurements along with a wide array of auxiliary information, such as bathymetry, remote sensing, hydrodynamics, wind stress, and river loading data. In general, the geostatistical model will take the form of an advanced system of kriging equations that accounts for spatial correlations and relationships among the auxiliary information and dissolved oxygen. Another component of this work will be development of a simple biophysical model, focused primarily on the temporal dynamics of hypoxia, and which will be linked with the geostatistical model to enhance its predictive capabilities.

Expected Results:

This project will improve our understanding of hypoxia formation in several ways. First, the geostatistical model will allow for estimation of the hypoxic extent (with confidence intervals) at regular time intervals throughout the year. Second, this research will determine which factors (and associated auxiliary data sources) are highly correlated with hypoxia formation and which are not. The time scales on which these factors influence hypoxia will also be evaluated, and the potential for the model to predict future hypoxia will be explored. The model will also be used to evaluate hypothetical scenarios, such as changes in the nutrient load from the Mississippi River. Finally, although this project will focus on hypoxia in the northern Gulf of Mexico, the tools and methodologies developed by this research should be applicable to the study of other environmental phenomena as well.

Potential to Further Environmental/Human Health Protection:

The results of this research project will be useful in the management and protection of the Gulf coast ecosystem and its associated fisheries. First, the project will improve our knowledge of how the extent and intensity of hypoxia vary throughout the year, so that aquatic scientists can make better assessments of how hypoxia is limiting available habitat. Second, the project will enhance our knowledge of the primary factors causing hypoxia, potentially allowing for more effective pollution mitigation strategies.

Supplemental Keywords:

water quality modeling, hypoxia, geostatistics, kriging, eutrophication, Gulf of Mexico,

Progress and Final Reports:

  • 2011
  • 2012
  • Final