REVIEW OF AUGUST 2004 USEPA REGION 4 REPORT

"EVALUATION OF THE ROLE OF NITROGEN AND PHOSPHOROUS IN CAUSING OR CONTRIBUTING TO HYPOXIA IN THE NORTHERN GULF"

 

Derek Winstanley

Illinois State Water Survey

August 20, 2004

 

COMMENTS AND RECOMMENDATIONS

 

1.     The report presents some interesting and valuable data and analyses that are different from the data and analyses presented in the Committee on Environment and Natural Resources (CENR) assessment reports (Rabalais et al., 1999; Goolsby et al., 1999; CENR, 2000). However, some available data have not been incorporated in the Region 4 report and some of the data that are included are misleading and probably erroneous. An earlier version of the Region 4 report contains important, relevant data and information that has been omitted from the latest version of the report. Nevertheless, the contents and overall tenor of the report indicate that a new, comprehensive, INDEPENDENT assessment of the causes, consequences, mitigation, and control of midsummer hypoxia in the northern Gulf of Mexico needs to be conducted. Such an assessment, 5 years after the 1999 CENR assessment, would be consistent with the adaptive management strategy adopted by the Mississippi River/Gulf of Mexico Watershed Nutrient Task Force (the Task Force). The national research and monitoring agenda for nutrient prevention in coastal waters (Howarth et al., 2003), which, like the CENR reports, is almost exclusively focused on nitrogen, also needs to be revisited and revised.

 

2.    The Task Force developed its 2001 Action Plan to reduce the size of the midsummer hypoxic zone in the northern Gulf of Mexico on the assumed credibility of the CENR assessment reports. The Task Force's Action Plan calls for a 30% reduction in the average annual load of total nitrogen delivered by the Mississippi/Atchafalaya River to the Gulf in order to reduce the size of the midsummer hypoxic zone. This Task Force strategy is based on the flawed CENR conclusion that it is a system that is in approximate Redfield balance and the flawed sequel that controlling nitrogen loads to the Gulf, especially from agricultural fertilizer application in the Midwest, will reduce excessive algal growth and midsummer hypoxia formation. The Region 4 report demonstrates that the Task Force has been fed with selective, inappropriate, and misleading information flowing from the CENR assessment reports.  This information also has been adopted in the setting of a national research agenda for nutrient pollution in coastal waters (Howarth et al., 2003).

 

3.     The Region 4 report concludes correctly that nitrogen is an excess nutrient, and most importantly, that phosphorus is potentially the most important limiting nutrient in the lower Mississippi/Atchafalaya River in spring and early summer. The Report also concludes correctly that phosphorus is the nutrient that potentially controls primary productivity and hypoxia in the northern Gulf of Mexico in midsummer. These findings have major policy and resource management implications. In particular, the data suggest that the wrong pollutant - nitrogen - is being controlled to effectively and efficiently reduce the size of the midsummer hypoxic zone.

 

4.    Whereas both nitrogen and phosphorus occur in large amounts in the lower Mississippi/Atchafalaya River, the data indicate that a reduction in the load of the potentially limiting nutrient - phosphorus - is most likely to result in a reduction in the size of the midsummer hypoxic zone.  A 30% reduction in annual average orthophosphate load would be about 13,000 metric tons, compared with a 30% reduction in total nitrogen of about 470,000 metric tons. However, in this phosphorus-limited system, a 30% or 40% reduction in nitrogen load would not be expected to reduce the size of the midsummer hypoxic zone. A reduction of at least 80% in the load of dissolved inorganic nitrogen (~800,000 metric tons) would be needed to drive the phosphorus-limited system to nitrogen limitation and reduce midsummer hypoxia. Such a reduction is infeasible. This number has been calculated independently by scientists at the Illinois State Water Survey and USEPA Region 4. Data were included in the earlier Region 4 report, but have been omitted from the August report. Also, driving a system to nitrogen limitation can stimulate the fixation of nitrogen by blue-green algae and create another undesirable ecological situation. The justification presented in the Region 4 report for continuing to reduce the nitrogen load is weak and speculative and is appropriately presented under the banner of "it is not obvious what would happen under various nutrient load reduction scenarios."

 

5.    The CENR reports do not contain information on methods, feasibility, costs, and effectiveness of phosphorus control, demonstrating that in scoping out the assessment there was a preconceived bias to identifying excess nitrogen as the problem and nitrogen control as the solution. The CENR reports also are weak in identifying phosphorus sources. A new assessment is needed to identify and quantify phosphorus sources, to determine the bioavailability of different forms of phosphorus, to evaluate methods, costs, feasibility, and benefits of phosphorus control, and to evaluate the reductions in phosphorus loads needed to drive the system to phosphorus starvation and to reduce midsummer hypoxia. .

 

6.   The focus of the CENR assessment, annual hypoxia monitoring, and the Task Force Action Plan is midsummer hypoxia. Rabalais et al. (1999) report that  "persistent hypoxia and often anoxia [occur].for extended parts of the record from May-September" and that "Hypoxia has not been recorded later than the first week of October...".   Hence, discussions of nitrogen and phosphorus control should continue to focus on reducing and mitigating midsummer hypoxia. As the system appears to be strongly phosphorus limited for 11 months of the year and approaches Redfield balance only in October, any further discussion of nitrogen control to achieve a reduction in hypoxia should recognize the fact that it is primarily a phosphorus-limited system and hypoxia generally does not occur in October.

 

6. The Region 4 report states that increasing loads of phosphorus from fertilizer use and municipal and industrial discharges are the causes of the calculated 140% increase in reactive phosphorus concentration in the northern Gulf since 1960, even though their data for the Mississippi River are for total phosphorus and not reactive phosphorus, and their 1960's phosphorus data are "reconstructed" and not measured.  Despite much freshness and honesty in the Region 4 report, this section appears to be a rehash of the questionable science in the CENR assessment reports.

 

Goolsby et al. (1999) found no data on phosphorus concentrations prior to 1972, but reported "no apparent long-term trend in either ortho P or total P concentrations or in the ratio of ortho to total P in the Mississippi River at St. Francisville since the period of record began in the early 1970s." The CENR Integrated Assessment (CENR, 2000) concludes that "Phosphorous loads have not changed significantly since the early 1970s when records began,..." However, Rabalais et al. (1999) conclude that there has been a "two-fold increase in the total phosphorous concentration" in the lower Mississippi River from 1973 to 1987, that there has been " a trend for increase in orthophosphate", and estimated that "the total P concentration increased two fold between 1960-62 and 1981-1987."  Much of total P load is not biologically available and, as noted above, their 1960's phosphorus data are "reconstructed."

 

The Illinois State Water Survey has phosphorus data prior to 1972. The data show that the average concentration of orthophosphate (as P) at Peoria on the middle Illinois River decreased rapidly from 0.75 mg/L in 1966 to 0.32 mg/L in 1973. Goolsby et al. (1999) report the mean concentration of orthophosphate (as P) at Marseilles (Upper Illinois River) for the period 1980-1996 to be 0.250 mg/L and 0.131 mg/L at Valley City (Lower Illinois River). Assuming that the mean concentration at Peoria was about 0.23 mg/L, then these data suggest a decrease in orthophosphate concentration of about 0.5 mg/L, or about 70% since 1966. Since Goolsby et al. (1999) identify the Illinois River as the largest contributor of 42 interior basins to the orthophosphate load in the Lower Mississippi River and some orthophosphate data exist from 1966, then any reported (calculated or reconstructed) increasing trend in the concentration of orthophosphate in the Mississippi River and northern Gulf should be regarded as highly dubious.

 

8.     The Region 4 report overemphasizes the agricultural sources of phosphorus, especially the use of fertilizer. Goolsby et al. (1999) reported that "About 31% of the phosphorus flux to the Gulf is estimated to come from fertilizer" ... and "About 41% of the annual phosphorus flux comes from sources that are not quantified but are associated with basin runoff. The most important of these is believed to be phosphorus in sediment associated with soil erosion."  Goolsby et al (1999) do not identify or quantify specific sources of orthophosphate ("the only form of P that can be utilized by algae, bacteria and plants") that contributes only about 30% of the total phosphorus load at St. Francisville on the lower Mississippi River. They do quantify orthophosphate concentrations, loads, and yields in 42 interior basins and show the highest median concentration in the Mississippi River Basin is in the Upper Illinois River Basin (Chicago metropolitan area). Although the area of the Upper Illinois River Basin is less than one half of the Lower Illinois River Basin, the average annual flux of orthophosphate from the Upper Basin is double the flux from the Lower Basin, which is predominantly agricultural. The yield of orthophosphate per square mile is almost 5 times greater in the Upper Basin than the Lower Basin. This is quite different from nitrate yield, which is only 65% higher in the Upper Basin than the Lower Basin.

 

9.     The Region 4 report omits data from an earlier version of the report that shows that the concentration of orthophosphate in the lower Mississippi and Atchafalaya Rivers peaks in late summer and the fall, which is the low flow period. The concentration of nitrate is highest at high flow in spring. These characteristics are suggestive of mainly non-point sources of nitrate and point-sources of orthophosphate.

 

10.     The Region 4 report appropriately analyses nutrient data at Belle Chasse, which is some 100 miles closer to the Gulf than St. Francisville, whose data are analyzed by Goolsby et al. (1999) and assumed to represent the flux of nutrients to the Gulf. However, the Region 4 report omits reference to the significant differences between the Belle Chasse and St. Francisville data discussed in an earlier Region 4 report. The earlier report found that the concentration and load of orthophosphate at Belle Chasse were considerably higher than at St. Francisville due to large, unpermitted, point-source discharges from the cities and industrial plants in the Baton Rouge-New Orleans corridor.

 

11.    The authors of the Region 4 report are to be congratulated for incorporating Transect C data for 1994, 1995, and 1997. They state that Drs. Rabalais, Turner, and Wiseman compiled these data. These data show DIN/DIP elemental ratios of 140:1 in April and May - key months for generating hypoxic conditions in summer. The question should be asked why these vitally important data were not included in the CENR assessment reports? 

 

12.     The Region 4 report focuses on nutrient concentrations and loads from the Mississippi/Atchafalaya River, but there are major uncertainties regarding other factors that cause or contribute to hypoxia in the northern Gulf. Two examples follow:

  •  Rabalais et al. (1999) report that  "The relative contribution of offshore sources of nutrients from upwelled waters of the continental slope is unknown." As the upwelling of nutrients is the major cause of hypoxia in many parts of the world's coastal oceans, this source of nutrients needs to be quantified.
  • Louisiana is losing about 27 square miles of nutrient rich wetlands and coastal marshes each year and the trend is expected to continue. Although satellite photographs clearly show bleeding of water and nutrients from coastal waters, estuaries, marshes and wetlands, their contribution to eutrophication and hypoxia in the northern Gulf has not been quantified. This needs to be quantified.

13.     The Region 4 report suggests Chesapeake Bay as a model to be emulated in controlling excess phytoplankton production. Unfortunately, two reports in the Washington Post in July 2000 illuminated the lack of progress in reducing eutrophication in Chesapeake Bay and criticized the use of computer models for reporting progress that is not observed in the real world. As in the Gulf of Mexico, finding an effective strategy to control eutrophication and hypoxia in Chesapeake Bay apparently would benefit from the injection of some good science and fresh ideas.

 

14.    Reading the report raises many questions regarding the CENR assessment reports and the assessment process. For example, why was an inappropriate nutrient ratio used as a basis for evaluating the roles of nitrogen and phosphorous in causing hypoxia, and for setting the scientific basis for nitrogen control? Why was the limiting nutrient not clearly identified, as was recommended in the Illinois assessment review comments? Why were nutrient data for stations below St. Francisville not used in the CENR assessment? Why were the 1994, 1995, and 1997 nutrient concentration data for Transect C in the Gulf of Mexico not included in the CENR assessment? Why did Goolsby et al. (1999) not analyze elemental nutrient ratios? Why did the economics report not address phosphorus control? How could the flawed and selective analysis pass peer review and be approved by the CENR Editorial Board and embraced by the National Research Council? Answers to these questions should be sought.

 

 

REFERENCES

 

CENR. 2000. Integrated Assessment of Hypoxia in the Northern Gulf of Mexico. National science and technology Council Committee on Environment and Natural Resources, Washington, D.C.

 

Goolsby et al. 1999. Flux and Sources of Nutrients in the Mississippi-Atchafalaya River Basin." NOAA Coastal Ocean Program Decision Analysis Series No.17.

 

Howarth, R., R. Marino, and D. Scavia. 2003. "Priority Topics for Nutrient Pollution in Coastal Waters: an Integrated National Research Program for the United States". NOAA, National Centers for Coastal Ocean Science.

 

Mississippi River/Gulf of Mexico Watershed Nutrient Task Force. 2001. Action Plan for Reducing, Mitigating, and Controlling Hypoxia in the Northern Gulf of Mexcio. USEPA.

 

Rabalais et al. 1999. Characterization of Hypoxia. NOAA Coastal Ocean Program Decision Analysis Series No15.

 

| Hypoxia | Water Survey Home |