River N export

Increases in river N export over time often correlate relatively well with increases in N inputs to watersheds. However, the response of river N export to decreases in watershed N inputs is not necessarily as rapid or well understood. In at least one case, the Patuxent River (MD, USA), point source reductions in N resulted in rapid reductions in N export (Boynton et al., 1995 Jordan et al., 2003). However, N export is generally dominated by non-point sources, not point sources (Carpenter et al., 1998 Dumont et al., 2005 Seitzinger et al., 2005b others), so substantially reducing N export often requires a reduction in non-point N inputs. Reductions of non-point nutrient inputs have been attempted in a number of systems, but response of river N export to these reductions is often much slower and less predictable than responses to reductions in point sources (Stalnacke et al., 1999). 

Human modification of water discharge associated with dams and consumptive water use can greatly decrease river N export. Both dissolved and particulate N forms are affected. Dams trap particulates and also can increase water residence time, which can increase the proportion of N inputs that are denitrified or buried. Consumptive water use (e.g. irrigation) transfers N out of the river system to terrestrial soils where it can be assimilated by crops, fixed by soils, or denitrified, and therefore not transported downstream. Consumptive water use also can influence PN export by decreasing water flow, which increases suspended sediment trapping within the river system. 

For open ocean regions atmospheric deposition is calculated as a percent of biological N2-fixation plus atmospheric deposition, assuming river N inputs are removed within coastal and shelf sediments. For continental shelves the range includes uncertainties in river N inputs for open ocean estimates, the range is that calculated with and without including DON in rainwater. See Fig. 6.6 for the contribution of atmospheric deposition in watersheds to river N export. 

FIGURE 3 Relationships between annual runoff and watershed export of DOC in streams and rivers reported in the literature. The respective lines extend only over the range of runoff values included in the dataset. The slopes of each line are approximately equivalent to the mean annual DOC concentration for that group (in parentheses). Sources for each relationship are as follows streams with wetlands, temperate (Mulholland, 1997) streams with wetlands, N. Carolina (Mulholland and Kuenzler, 1979) large rivers, global (Spitzy and Leenheer, 1991) large rivers, N. America (Mulholland and Watts, 1992) streams, tropical (McDowell and Asbury, 1994) streams, N. America (Mulholland, 1997). 

During the last several decades there have been many studies quantifying and characterizing N export by individual rivers to the coastal zone through direct measurements of N concentrations and water discharge (e.g. Boynton et al., 1995 ... 

Relatively few large rivers account for a large proportion of the N exported to coastal zones globaUy for ah N forms, based on NEWS model predictions. For example, the 25 rivers with the greatest water discharge globaUy (<0.5% of the rivers with a watershed >2500 km ) account for approximately half of the exported DIN and DON, and about a quarter of the exported PN, globaUy (Table 9.1). Some of... 

The response of coastal ecosystems to river nutrient inputs depends, in part, on the seasonal pattern of the N dehvery, including seasonal changes in the form of N delivered (e.g., DIN, DON, and PN). Most studies of river nutrient export only report annual rates and often only report TN export. Where seasonal patterns of N export by individual river basins have been reported, a number of patterns have been observed, and the different forms of N can exhibit different patterns of export (e.g., Alexander et ah, 1996 Eyre and Pont, 2003 Lesack et ah, 1984) (Figs. 9.9 and 9.10). Seasonal patterns are observed in both N concentration and in N export (Fig. 9.10). 

Figure Global seasonal patterns of N export as predicted using hydrology alone. Hydrographs of five large, representative rivers are shown.

There is substantial variability in the magnitude of N inputs to coastal regions, as well as the contribution of different sources driving that export, that is starting to be addressed by spatially explicit models. There are now many watersheds throughout the world with high yields (i.e., kg N exported by rivers to the coastal... 

Often only a relatively small proportion of the total N inputs to a watershed are exported by rivers to the coast, although the range is considerable (e.g., <1% to >40%). Much of the N is denitrified within soils and freshwater ecosystems. Hydrology has a large impact on the proportion of N exported by rivers with the largest proportion exported from watersheds with high water runoff (>1 m yr ) and lowest from basins with extensive damming or very low annual precipitation and runoff. 

Figure 18.13 Scatter plot of percent N exported versus water residence time for a variety of estuarine ecosystems. Figure was re-drawn from Nixon et al. (1996). Patuxent and Choptank Rivers, tributaries of Chesapeake Bay, were not included in the original analysis. Data for those systems were from Boynton et al. (1995) and Fisher (unpublished data).

Figure 36.10 Comparison of anthropogenic nitrogen inputs and associated riverine N exports (kg per km of watershed area per year) in two contrasting water regions of the USA The northeast USA and the Mississippi River hasin. Modified from NRC (2000) based on data from Howarth et al. (1996).

---