Predicting summer hypoxia in the northern Gulf of Mexico: Riverine N, P, and Si loading R.E. Turner a, * , N.N. Rabalais b , D. Justic a a Coastal Ecology Institute, Louisiana State University, Baton Rouge, LA 70803, USA b Louisiana Universities Marine Consortium, 8124 Highway 56, Chauvin, LA 70344, USA Abstract We conducted a statistical analysis to discern the relative strengths of the loading of various forms of nitrogen, phosphorus, dissolved silicate and their molar ratios on the variance in the size of the summertime low oxygen zone found off the Mississippi River, northern Gulf of Mexico. A stable statistical model that included Year and riverine nitrate + nitrite loading for the 2 months prior to the mea- surement of hypoxic zone size described 82% of its variation in size from 1978 to 2004. The usefulness of the term Year is consistent with the documented increase in carbon stored in sediments after the 1970s, which is when the hypoxic zone is predicted to have become a regular feature on the shelf and to have expanded westward. The increased carbon storage is anticipated to cause a sedimentary respi- ratory demand influencing the size of the zone, and whose temporal influence is cumulative and transcends the annual variations in nitro- gen loading. The variable Year is negatively correlated with the TN:TP ratio in a way that suggests N, not P, has become more important as a factor limiting phytoplankton growth in the last 20 years. Nitrogen, in particular nitrate + nitrite, and not phosphorus, dissolved silicate, or their molar ratios, appears to be the major driving factor influencing the size of the hypoxic zone on this shelf. This conclusion is consistent with cross-system analyses that conclude that the TN:TP ratio in the Mississippi River, currently fluctuating around 20:1, is indicative of nitrogen, not phosphorus, limitation of phytoplankton growth. Nutrient management that places stronger emphasis on reducing nitrogen loading as compared to phosphorus loading, is justified. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Hypoxia; Oxygen; Mississippi River; Eutrophication; Nitrogen; Phosphorus; Silicate; Nutrients; Monitoring; Water quality; Management 1. Introduction The largest zone of oxygen-depleted coastal waters in the United States and in the western Atlantic Ocean is in the northern Gulf of Mexico on the Louisiana/Texas con- tinental shelf at the terminus of the Mississippi River (Fig. 1). The size of the Gulf of Mexico hypoxic zone, oper- ationally defined as bottom water with a dissolved oxygen concentration less than 2 mg l À1 , reached 22,000 km 2 in mid-summer 2002, but was barely existent during the drought in 1988 (Rabalais and Turner, in press). The main purposes of this analysis are to predict variations in the size of the hypoxic zone and to more accurately define the pri- mary driving forces behind these variations. The shelfwide distribution of hypoxia has been deter- mined in a systematic way from midsummer cruises begin- ning in 1985 (Fig. 1). The hypoxic water mass during these cruises extends west from the Mississippi River birdfoot delta along the Louisiana shelf and onto the upper Texas coast, and from within 1 km of the barrier shoreface to as far as 125 km offshore. A linear regression analysis of its summertime area and length as it spreads along the coast from east to west yields an adjusted coefficient of determination (R 2 ) of 0.79 (Scavia et al., 2003). There are no shelfwide data on its distribution from before 1985. The size of the hypoxic zone has been measured for smaller areas and with lesser frequency as far back as 1972, and some of these data were used to estimate the size of the 0025-326X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2005.08.012 * Corresponding author. E-mail address: euturne@lsu.edu (R.E. Turner). www.elsevier.com/locate/marpolbul Marine Pollution Bulletin 52 (2006) 139–148