Journal of the American Water Resources Association
"Nutrient Load Characterization from Integrated Source Data for the Lower Mississippi River," by J. E. Bollinger et al. (1)
Derek Winstanley (2)
Bollinger at al. (2000) have taken on the important and difficult task to provide a summary of nutrient data that may be useful in evaluation of the potential effects of Mississippi River water on the Gulf of Mexico. The authors conclude that loads of phosphorus and nitrogen increased from the 1960s to the1980s. Some questions arise from their data and analysis.
1. Why have nitrogen data been presented only from 1960, although the authors acknowledge that nutrient concentration data have been collected on a continuous basis since the mid-1950s? CENR (2000a) finds that hypoxia has increased only since the 1950s, so it is important to evaluate nutrient loads from the 1950s using all available data.
If 1950's nitrogen data had been presented, it would have been apparent that the concentration and load of total nitrogen were higher in the early 1950s than for any time since. Mayer et al. (1998) present data for the periods 1950-1952, 1953-1972, and 1973-1982. Goolsby et al. (1999) present nitrate data from1954 to 1996. Mayer et al. show that the average concentration of dissolved inorganic nitrogen+labile particulate nitrogen in the Lower Mississippi River was 164.9 uM (2.31 mg N/l) in 1950-1952. These data do not include dissolved organic nitrogen (DON). Using the concentration of 0.52 mg N/l DON in the Lower Mississippi River for the period 1955-1965 provided by CENR (2000b) gives an average concentration of total nitrogen in the Lower Mississippi River of ~2.83 mg N/l in the early 1950s. This is ~60 percent higher than the concentration of total nitrogen in 1953-1972 and ~25 percent higher than the concentration of total nitrogen of 2.26 mg N/l for the period 1980-1996 provided by Goolsby et al. (1999). The flow of the Mississippi River was also high in 1950-1952 (Goolsby and Battaglin, 2000), so the load of nitrogen in the Lower Mississippi River for the period 1950-1952 would have been very high. These data do not support the conclusion by Rabalais et al. (1999) that the concentration and flux of nitrogen has increased since the 1950s.
The delivery of high loads of bioavailable nitrogen to the Gulf in the early 1950s occurred at a time when the use of nitrogen fertilizer was minimal and the size of the hypoxic
zone in the Gulf is assumed to have been low (CENR, 2000a). The importance of addressing total nitrogen in surface waters is recognized by USEPA in setting nitrogen criteria (USEPA, 2000) and in the development of a Hypoxia Action Plan (MRGMWNTF, 2001).
2. What are the justification and implications of normalizing nitrate+nitrite data to USGS data where agency differences were found? It is stated simply that USGS data are generally more abundant.
3. It is stated that unfiltered nitrate from the St. Francisville segment was retained separately to characterize the nitrogen load summaries in 1960-1973. However, Figure 7 shows that some nitrate+nitrite date were also used in this period. Have FWPCA and FWQA nitrate+nitrite data prior to 1973 also been used, or do the St. Francisville data include some nitrate+nitrite data?
4. What are the implications of integrating and using in load calculations only data for river segments that were found to be not statistically different from the furthest segment downstream? Could this be a reason why segment-to-segment analyses revealed no statistically significant changes in nitrate+nitrite concentrations in the Mississippi River as it flows through the state of Louisiana?
5. What are the justification and implications of combining in the nitrogen load trend analysis (Figure 7) the mainly unfiltered nitrate concentrations prior to 1973 with unfiltered and filtered nitrate+nitrite post 1973? It is reported that significant differences were found between filtered and unfiltered nitrate, with the filtered water samples showing higher nitrate concentrations than the unfiltered by 125+/- 24 percent. Does the dominance of unfiltered nitrate concentration data in the 1960s produce an inhomogeneous time series biased towards low values in this period? The authors do acknowledge that the comparability of nitrogen nutrients was not established.
Goolsby et al. (1999) apparently regard the time series of nitrate concentration at St. Francisville from 1954 to the present as homogeneous, despite changes in sampling frequencies, filtering, storage, and analytical methods. Turner and Rabalais (1991) simply assume that all numbers reported in the literature are comparable.
In a recent comparison of the accuracy and precision of different methods of determination of NH4+ and NO3- (colorimetry; potentiometry; distillation; and diffusion), substantial differences often were observed for the same sample, and recovery of N was often incomplete by up to ~50 percent (Khan et al., 2000).
6. What are the justification and implications for normalizing Historic Segment F ADPCE and current Segment B LADEQ TKN data with USGS data, other than the greater abundance of USGS data (Segment F, ADPCE TKN> USGS TKN by 75 +/- 21 percent, and Segment B, LADEQ TKN >USGS TKN by 45 +/- 20 percent)?
7. Do the facts that the LADEQ historical values for total phosphorus were greater than those reported by USGS by 10 to 41 percent, but that no such discrepancies were apparent in the contemporary data, justify normalizing the historical LADEQ data to those of USGS?
8. How have FWPCA and FWQA total phosphorus data prior to 1973 been normalized to post-1973 USGS data, when it is stated that no USGS data exist prior to the 1970s? These FWPCA and FWQA data are critical in concluding that there has been an increase in total phosphorus load since the 1960s.
Bollinger et al. conclude that phosphorus and nitrogen loads overall were observed to have increased from the 1960s to the 1980s. However, the above questions and additional information challenge the trends shown in Figures 7 and 8. In addition, the authors present nitrate+nitrite and TKN data only since 1973, thus precluding any conclusion about total nitrogen loads prior to this date. In examining nutrient ratios and limiting nutrients, it would be more meaningful to compare total phosphorus with total nitrogen, rather than with nitrate.
Another challenge, but beyond the scope of the Bollinger et al. paper, is to identify the cause(s) of any increase in nitrogen loads. Nitrate concentrations in rivers in the Mississippi River basin are positively correlated with precipitation and river flow (e.g., Goolsby et al.,1999). Goolsby et al. report that an increase in precipitation in the Mississippi River basin from 1970 to 1983 contributed to increases in nitrate concentration, streamflow, and nitrate load. However, despite quantifying nitrogen inputs to the MRB, Goolsby et al. do not quantify the role of increased precipitation and streamflow in accounting for the increased concentration and load of nitrate.
Goolsby et al. (1999) conclude that there has been no statistically significant increase or decrease in the annual flux of phosphorous to the Gulf since records began in the early 1970s. Rabalais et al. (1999) report that some data suggest a two-fold increase in the concentration of total phosphorus between the early 1960s and the 1980s. Evidence for a four-fold increase in the load of total phosphorous from the 1960s to1980s identified by Bollinger et al. should necessitate an evaluation of the significance of this large increase in phosphorus load for hypoxia formation in the Gulf, especially since Rabalais et al. conclude that phosphorus limitation is much more likely than was originally expected. Ferber (2001) reports that in the Kattegat strait between Denmark and Sweden oxygen levels have increased over the last 14 years as phosphorus, not nitrogen, levels have fallen 80 percent. In central California, it is concentrations of iron in offshore waters that appear to control phytoplankton biomass, primary production and community structure (Johnson et al., 2001).
A four-fold increase in the load of total phosphorus from the 1960s to about 1980 could also be one reason for the more than doubling of the nitrate load in the Mississippi River over the same time period. It is well known that phosphorus is rapidly taken up by micro-organisms and can stimulate nitrogen fixation through the growth of blue-green algae (NRC, 1969; Vallentyne, 1974; Paerl, 1990).
Establishing the homogeneity and accuracy of historical nutrient data for use in trend analysis remains a challenge. I agree totally with the recommendation by Bollinger et al. for implementation of standard methods for collecting and analyzing water quality data.
Bollinger, J.E., L.J. Stenberg, A.J. Englande, J.P. Crews, J.M. Hughes, C. Velasco, K.H. Watanabe, C.M. Swalm, J.M. Mendler, L.E. White, and W.J. George, 2000. Journal of the American Water Resources Association (JAWRA) 36(6): 1375-1390.
CENR. 2000a. Integrated Assessment of Hypoxia in the Northern Gulf of Mexico. National Science and Technology Council Committee on Environment and Natural Resources, Washington, D.C.
CENR. 2000b. Hypoxia in the northern Gulf of Mexico - responses to public comments. National Science and Technology Council Committee on Environment and Natural Resources, Washington, D.C.
Ferber, D. 2001. Keeping the Stygian waters at bay. Science 291: 968-971, 973.
Goolsby, D.A., W.A. Battaglin, G.B. Lawrence, G.B., R.S. Artz, B.T. Aulenbach, R.P. Hooper, D.R. Keeney, and G.J. Stensland, 1999. Flux and sources of nutrients in the Mississippi-Atchafalaya River Basin. National Science and Technology Council Committee on Environment and Natural Resources Hypoxia Work Group, Topic 3 Report, Washington, D.C.
Goolsby, D.A. and W.A. Battaglin, 2000. Nitrogen in the Mississippi Basin - estimating sources and predicting flux to the Gulf of Mexico. USGS Fact Sheet 135-00, U.S. Geological Survey.
Johnson, K.S., F.P. Chavez, V.A. Elrod, S.E. Fitzwater, T. Pennington, K.R. Buck, and P.M. Walz, 2001. The annual cycle of iron and the biological responses in central California coastal waters. Geophys. Res. Ltrs., 28(7): 1247-1250.
Khan, S.A., R.L. Mulvaney, K. Strle, and B.P. Horgan, 2000. Evaluation of diffusion for inorganic-nitrogen analysis of natural water and wastewater. Journal of Environmental Quality 29(6): 1890-1895.
Mayer, LM, R.G. Kail, S.A. Macko, S.B. Joye, K.C. Rutenberg, and R.C. Aller, 1998. Importance of suspended particulate in riverine delivery of bioavailable nitrogen to coastal zones. Global Biogeochemical Cycles 12(4): 573-579.
MRGMWNTF. 2001. Action plan for reducing, mitigating, and controlling hypoxia in the northern Gulf of Mexico. Mississippi River/Gulf of Mexico Watershed Nutrient Task Force (www.epa.gov/msbasin).
NRC, 1969. Eutrophication: causes, consequences, correctives. Procs. of a Symp. National Research Council, National Academy of Sciences, Washington, D.C.
Paerl, H.W, 1990. Physiological ecology and regulation of N2 fixation in natural waters. Advances in Microbial Ecology: 11(8), 305-344.
Rabalais, N.N., R.E. Turner, T.D. Justic, Q. Dortch, and W.J. Wiseman, 1999. Characterization of hypoxia. National Science and Technology Council Committee on Environment and Natural Resources Hypoxia Work Group, Topic 1 Report, Washington, D.C.
Turner, R.E., and N.N. Rabalais, 1991. Changes in Mississippi River water quality this century. BioScience 41: 140-147.
USEPA. 2000. Nutrient criteria technical guidance manual: rivers and streams, United States Environmental protection Agency, Washington, D.C., EPA-822-B00-002.
Vallentyne, J.R., 1974. The algal bowl: lakes and man. Canada Dept. Environment Misc. Special Pub. 22, Ottawa, Canada.
KEYWORDS: Nitrogen; nitrate; nutrients; Mississippi; hypoxia; phosphorus.
| Home | Highlights | Staff | Data | Information | Centers | Site Map |
Email the Web Administrator with questions or comments.Last Modified: November 17, 2008