ISWS - Hypoxia: Satellite Imagery of the Influence of Mississippi River Water...

This website is related to the publication, “Coastal Change and Hypoxia in the Northern Gulf of Mexico” by Krug (2007) and its allied document, “Marine Modification of Terrestrial Influences of Gulf Hypoxia” by Krug and Merrifield (2007). The national effort on Gulf hypoxia considers hypoxia to be the result of nutrients, particularly nitrogen (N), in river water from the Mississippi/Atchafalaya River Basin (MARB) to the Gulf. MARB N is supposed to fuel the growth of algae in the Gulf which, due to isolation of bottom water by summertime stratification, die and decay in bottom waters depleting dissolved oxygen (D.O.) values to < 2 mg L-1. This has been occurring within that part of the Gulf of Mexico bounded by the curve of the Louisiana-Texas coast (Figures 1 and 2). Such hypoxic D.O. values are lethal to many species of desirable aquatic and marine organisms. However, nutrient-enriched riverine discharges can also enhance fishery production on adjacent continental shelves, including the northern Gulf of Mexico.

Krug (2007) noted that “the observed recurrence of Gulf hypoxia starting with the 1973 flood and the more than doubling of its average annual starting with the 1993 flood is not in harmony with the observations that MARB nutrients, N and phosphorus (P), and freshwater delivery show negligible overall increase since comprehensive nutrient (N and P) measurements began in the early 1970s and the stabilization of the proportion of MARB water and nutrients flowing down the Atchafalaya and the Mississippi — with all Atchafalaya MARB water and nutrients and 53 percent of the Mississippi’s reported to contribute to hypoxia.” Furthermore, “the hypoxic zone also is experiencing massive physical, hydrological, chemical, and biological changes associated with an immense river-switching, delta-building event of the type that occurs here about once a millennium. Coastal change induced hypoxia in the northern Gulf of Mexico prior to European settlement” (Krug, 2007). Basically the area around the Mississippi River was and continues to lose wetlands at a great rate— 80 percent of all coastal wetland loss in the United States occurs here — leaking 100s to 1000s of years of stored nutrients and organic matter to the eastern half of the hypoxic zone. For example, the oxygen demand imposed by outflow from Barataria and Terrebonne estuaries (Figure 2) may equal or exceed that of the algae produced in the hypoxic zone itself (Krug, 2007).

The observed start of Gulf hypoxia coincides with the great flood of 1973. This is when the Atchafalaya’s 200 million m3 yr-1 sediment load — the amount of land moved to create the Panama Canal flow — achieved breakthrough to the Atchafalaya Bay with 50 percent sediment transfer efficiency to the Gulf (Krug, 2007). Whereas prior to 1973, the coastline of the western half of the hypoxic zone, Atchafalaya Bay itself, and estuaries immediately to the east of the bay were losing land, after 1973 there began a gaining of land even though most mud was lost offshore due to the erosive forces of wave, wind, current, tide, and storm (Wright, 1977; Roberts et al., 1980; Van Heerden et al., 1981; Wells and Kemp, 1981; Madden et al., 1988; Roberts, 1997). With the 1993 flood came increased efficiency of sediment transfer through the Atchafalaya and Atchafalaya Bay to the hypoxic zone (Adams and Baumann, 1980; Roberts et al., 1980; Donnell and Letter, 1992; Roberts, 1998; Anonymous, 1999) and the doubling of the extent of Gulf hypoxia by extension westward toward Texas and fuller coverage of the continental shelf west of the Mississippi Trough (Rabalais et al., 1999; Krug, 2007; Fig. 3). Basically, Gulf hypoxia formed and then expanded with the expansion of the mud from the Atchafalaya River (Krug, 2007). We have seen increased flux of nutrients and fresh water from the Atchafalaya into the Gulf, as well as mud. All are conducive to hypoxia formation.

Krug and Merrifeld went on to assess the relative efficiency with which Atchafalaya, Mississippi, and estuarine inputs promote Gulf hypoxia. They found significant marine modification of these terrestrial influences on Gulf hypoxia. Review and synthesis of the scientific literature and data led Krug and Merrifield to conclude that the hydrology of the northern Gulf of Mexico and of the hypoxic zone itself is such that water, nutrients and sediments delivered by the Mississippi River are relatively inefficient in promoting hypoxia compared to inputs from the Atchafalaya River. Inputs from the Barataria and Terrebonne estuaries are intermediate in their ability to promote hypoxia. Analysis of 1,000s of satellite images support the findings of Krug and Merrifield and are presented below.

The Mississippi River is the largest river in North America. The Mississippi and Atchafalaya Rivers drain a 1,270,000 mi2 basin (MARB) — 41 percent of the conterminous United States, including Illinois — into the northern Gulf of Mexico (Figure 1). The best viewing of the interaction of the Gulf with the Atchafalaya and Mississippi Rivers is during the winter and spring when these rivers deliver most of the water (Figure 3), sediment, and nutrients as can be seen in the 1-kilometer (km) resolution true color satellite image for January 8, 2002 (Figure 2). The largest hypoxia (22,000 km2 or 8,500 mi2) ever recorded was in summer 2002, even though the yearly flow from the MARB was just average with some months being higher and others lower than average flow (Figure 3).

In the January 8, 2002 true color image (Figure 2), deep blue indicates clear Gulf water and milky casts are consistent with the presence of coccolith algae (whose carbonate shells form chalk). The line of milky blue waters across the bottom of the image represents the strong eastward movement of the Gulf’s Boundary Current. The Boundary Current interacts with inshore Gulf water including Atchafalaya and Mississippi River water — as can be seen by the distribution of blue water against the distributions of yellow and brown sediment-laden water and greenish algae-laden water. Much water from the Atchafalaya River and its nearby shelf area is moved offshore and east toward the Mississippi River Delta. Water from the Mississippi’s southwest outlet becomes blocked from further westward movement; water from other Mississippi River outlets moves in various directions. Outflow is seen from the Barataria and Terrebonne estuaries, which lie between the Mississippi and Atchafalaya Rivers. These estuaries also lie at the head of the Mississippi Trough — an extension of deep water that runs across the continental shelf. The Atchafalaya empties into a bay from whose westernmost part muddy outflow moves west along the coast. This Atchafalaya “mud stream” reaches rivers near the Texas coast that exhibit offshore flow. This true color image shows that the Atchafalaya and Mississippi Rivers influence the area of the hypoxic zone and that the currents of the northern Gulf of Mexico mediate this influence.

Over the Mississippi Trough there is a four-way convergence of surface water — outflows of the Atchafalaya and Mississippi Rivers, outflows of the Barataria and Terrebonne estuaries, and inflow of clear, warm Gulf water. Convergence over the Mississippi Trough is the most common circulation pattern seen in the 2002 true color images. Convergence physically forces water downward. This physical forcing of water downward is further enhanced by a physicochemical processs known as cabbeling — the mixing of masses of water of different temperatures and salinities to produce water denser than the components of its parts. Cabbeling produces vertical velocities may be 1,000s of times greater than typical open ocean values. The Mississippi River plume forms frontal zones with Gulf waters creating zones of intense vertical motion: “Sinking due to this mixing process is known as cabbeling or cabballing; we may note that in those areas where it occurs, it will help maintain a discontinuity between the two different water types and hence maintain a sharp front between them” (Beer, 1997, pp. 123).

Apparently, this convergence and outflow through the Mississippi Trough maintains the trough. The Mississippi River delivers an enormous sediment load, nearly 20 times the sediment load of all of the rivers on the entire East Coast of the United States. Even though the Mississippi River deposits 70 to 75 percent of its sediment load westward into the vicinity of the Mississippi Trough (Scruton, 1956; Curtis et al., 1973; Corbett et al., 2004; McKee et al., 2004) bathymetry dating back to the 1700s and modern sediment sampling indicate that these dynamic currents have maintained the Mississippi Trough essentially as is, maintaining its features while accumulating just enough sediment to compensate for land subsidence (United States Coast Service, 1861; 1863; Corbett et al., 2004; McKee et al., 2004; Figure 2).

Thus the role of the Mississippi River’s great water and sediment loads appear to have a limited role in promoting hypoxia further west as the area around the Mississippi is a locus for transport of water and sediment offshore and away from the hypoxic zone (e.g., Dodge and Lang, 1983; Brooks and Legeckis, 1982; Lugo-Fernandez, 1998; Conkright et al., 1999; Hunter, 2001; Ohlmann et al., 2001; Weatherly et al., 2003; Wawrik and Paul, 2004; Krug and Merrifield, 2007). As can be seen from our 2002 animation, the Mississippi River itself discharges its waters beyond the shelf break, outside the hypoxic zone where much water is lost from being available to the hypoxic zone. Even if the water that moves west toward the hypoxic zone, its availability to the hypoxia zone is impeded.

This is not so with water and sediment and nutrients discharged by the Atchafalaya which discharges on the innermost continental shelf at the coastal center of the hypoxic zone. In recent decades, the Atchafalaya River has diverted 30 percent of the Mississippi's water and dissolved nutrients and 40 percent of its sediment. Here currents effectively distribute the Atchafalaya’s inputs throughout most of the area making the Atchafalaya highly influential in the formation of hypoxia. This finding is consistent with the geologic record showing that Atchafalaya-like distributaries of the recent geologic past were related to oxygen depletion of these coastal waters (Krug, 2007). As noted by Krug and Merrifield, “The expanding area of fluid mud of the Atchafalaya mud stream and its loose bottom mud on the continental shelf act as fluidized reactor beds where carbon and nutrients are heavily recycled (e.g., Trefry et al., 1994; Aller, 1998; Abril et al., 1999; 2004; Rowe et al., 2002; Gordon and Goni, 2003; Aller et al., 2004; Aller and Blair, 2004; Corbett et al., 2004; McKee et al., 2004; Sutula et al., 2004):
“‘Fluid muds and mobile surface material cause the seafloor and continental boundary to act as a massive, suboxic, fluidized bed reactor... Reoxidation, repetitive redox successions, metabolite exchange, and continual mixing-in of fresh planktonic debris with refractory terrestrial components, results in an effective decomposition system largely decoupled from net accumulation’” (Aller, 1998, p. 143)”.
Thus with the great flood of 1973, the Atchafalaya mud stream achieved breakthrough to the coast and permanently altered the coastal dynamics of the hypoxic zone (Roberts et al. 1980; Wells 1980; Wells and Kemp, 1981; 1982; Roberts, 1998; Huh et al., 2001; Draut et al., 2005) creating a large and expanding area of oxygen-consuming fluidized mud reactor to deplete the oxygen out of the low volumes of water inherent to these shallow water depths. The widespread effects of the Atchafalaya are readily seen in many of the true-color satellite images.

The oceanographic controls on the influence of Barataria and Terrebonne estuaries on Gulf hypoxia appear to make them intermediate between that of the Mississippi and Atchafalaya Rivers. Whereas these estuaries discharge at the innermost part of the continental shelf, they do so at the eastern edge of where hypoxia forms and into the mouth of the Mississippi Trough which lies just offshore of their outlets.

The 1-km resolution sea surface temperature (SST) satellite images supplement their visible counterparts. The spectrum of sea surface temperatures ranging from red through green illustrate the flow of the Gulf of Mexico’s Boundary Current and its Loop Current on 23 February 2003 (Figure 4). Fluid lines defining colored areas depict sea surface temperatures of the Gulf of Mexico. Sharp, speckled, and streaked colored (temperature) features and areas of white are clouds. The area of deepest red represents warm Caribbean seawater passing into the Gulf by the Yucatan, through and out the Gulf between Florida and Cuba with the Loop Current. The Boundary Current along the margins of the Gulf is highlighted by cold upwelled water flanked inshore and offshore by warmer waters. The westward flow of Mississippi River water stops and then falls back on itself much as a fountain does in trying to overcome the downward pull of gravity. In this case, the force operating against the Mississippi’s westward outflow is the eastward motion of the Gulf’s Boundary Current. The eastward flow of cold water out of the third bay to the west of the Mississippi is the outflow from the Atchafalaya River. Between the outflows from these two rivers there is the inflow of clear, warm water from further out in the Gulf. As with its true color counterpart, convergence is seen over the Mississippi Trough and this is the most common circulation pattern seen — as also is shown in the 18-image short 2002 SST animation, the SST images of the 365 day 2002 animation, 2003-2005 SST animation, and other imagery displayed by this website.

Higher resolution SST and turbidity satellite images of the Mississippi River Delta and the area of the Atchafalaya River for this day are accessed on the Louisiana State University’s Earth Scan Laboratory website: http://www.esl.lsu.edu/research/NOAA_AVHRR/archive_baywatch.php?day. Turbidity of water is measured by reflectance, the percent of light reflected by particles suspended in water, as is seen in this reflectance image for May 14, 1993 — the year of the 1993 flood and the start of doubling the average size of Gulf hypoxia (Figure 5)

The chlorophyll satellite images supplement their visible and SST counterparts albeit at a poorer, 9-km, resolution, as does this January 8, 2002 (Figure 6) image. This is a satellite image of the plant pigment chlorophyll a found in algae. Accordingly, this form of chlorophyll is used as a surrogate for algae. Even at this poorer 9-km resolution, the inflow area of chlorophyll-poor, clear, warm Gulf water can be seen to be bounded on the west, north, and east by chlorophyll-rich Atchafalaya, Mississippi, Barataria, and Terrebonne estuary waters, respectively. The entire year is depicted in 8-day composite chlorophyll images; unfortunately, time-averaging further diminishes their resolution. Additional daily 9-km resolution images are available in the 365-day image series. Also available are 4-km resolution chlorophyll images beginning in July 2002.

True Color - The deep blue of Gulf water is changed browns and yellows by sediment, green by algae, and lightened (dilute milky) are consistent with coccolith algae whose carbonate shells form chalk.

Sea Surface Temperature - The color scale indicates surface water temperature.

Chlorophyll - The color scale indicates the concentration of the chlorophyll a pigment and, therefore, the amount of algae present

Reflectance - The percentage of light reflected indicates how turbid water is.

The national effort on Gulf hypoxia bases its calculated effect of the Mississippi River discharge on Gulf hypoxia on idealized conditions. It is assumed that the algae produced in the Gulf by N in water from the MARB depletes oxygen in bottom waters and that all of the water which flows to the Gulf and from the Atchafalaya River and from the South Pass, Southwest Pass, and Grand and Tiger Passes of the Mississippi River — 53 percent of the Mississippi’s total flow — moves west to support algal production in 106,866 km2 of continental shelf water west of the Mississippi River to the Mexico/Texas border (Turner and Rabalais, 1991; Rabalais et al., 1999, p. 35). However, even in 2002 — the year of greatest recorded hypoxia — we see that these assumptions are not true. Most Mississippi water and sediment does not pass past the Mississippi trough to get at where most of the algae is reported to be produced and where most of the hypoxia occurs. And appreciable offshore/onshore exchange of water indicates that marine N naturally present in Gulf waters can also act to produced algae (Krug and Merrifield, 2007).

In addition to the above 2002 satellite imagery, multiyear web-based satellite imagery is also available to the public. These images, like the 2002 images, show the prevalence of impeded drainage through the hypoxic zone. For example, the NOAA Coastal Services Center posted about 2500 paired reflectance and SST for 1985-1999. Its main menu http://www.csc.noaa.gov/products/gulfmex/startup.htm uses a satellite image typical of the set showing impeded circulation, largely of convergence Mississippi and Atchafalaya area waters in the area of the Mississippi Trough: as does the satellite image used in the Project Description http://www.csc.noaa.gov/products/gulfmex/html/projdesc.htm and the satellite image used in product description http://www.csc.noaa.gov/products/gulfmex/html/data.htm and ecosystem description http://www.csc.noaa.gov/products/gulfmex/html/ecosys.htm. The following table of our analysis of 10 years — 1,768 pairs — of these posted reflectance and SST images of the northern Gulf of Mexico show that impedance of the flow of Mississippi River water through the hypoxic zone is prevalent. Of the 1,768 days of reflectance images posted for 1990-1999, 1,476 days are clear enough for reflectance (turbidity) to show directional flow in the area of the Mississippi River Delta and the hypoxic zone (Mississippi River to the Texas border). Of the 4 Mississippi distributaries whose waters are assumed to flow west through the hypoxic zone, Flow is visibly impeded in 1,222 flow through the hypoxic zone; the rest are of undetermined flow through the hypoxic zone because of cloud cover and/or incomplete satellite and feature coverage.

Each of these 1,476 reflectance images has a corresponding sea surface temperature (SST) image. SST can leave a bigger “footprint” — river water can retain a temperature signature after losing its visible turbidity and the flow of ocean currents may be measured beyond its shaping of visibly-turbid water. Of these SST images, 1,178 have sufficient detail to use for validation. This caused 76 changes resulting in a net increase of 74 days falling into the impeded flow category. Beginning 1 September 1997, 9-km resolution chlorophyll images are available on the web for comparison: http://oceancolor.gsfc.nasa.gov/cgi/level3.pl?DAY=&PER=&TYP=swchl&RRW=16. Despite their poorer, 9-km resolution, these caused 6 more days to fall into the impeded flow category. What this analysis shows is that the more information is available the more impeded flow is seen. The general flow characteristics seen in 2002 are typical of the 1990s.

The images from 2002 and the 1990s can be compared to the approximately 1700 documented NIMBUS-7 satellite passes of 18-km resolution chlorophyll and quasi-true color images posted for 1978-1986. We see from the posted 1978-1986 monthly composite chlorophyll images (Figure 7) circulation consistent with that seen for 19990-1999 and 2002. Individual NIMBUS 7 images for the Gulf of Mexico can be accessed by the public at NASA/Goddard Space Center’s Ocean Color website: http://oceancolor.gsfc.nasa.gov/cgi/browse.pl?sen=am.

Similarly, the Short Series of 2002 SST Images (Figure 4) can be compared with the Short Series of 2003-2005 SST Images (Figure 8).

"It is recommended that a broadened approach for better understanding the causes and controls of Gulf hypoxia be adopted including, but not limited to, MARB inputs and coastal change and marine processing of terrestrial and Gulf influences” (Krug and Merrifield, 2007).