Nitrogen

Miracle Farm Blueprint

Organic Farming Manual

Get Instant Access

Nitrogen yields from undisturbed catchments are not well understood because recent human activities have more than doubled the global supply of reactive N, influencing most landscapes around the globe. Exports from undisturbed tropical rivers likely are minimally affected, however. Rivers of the Americas and the Gambia River in Africa had an average TN yield of 5.1 kg ha1 year1. Because the nitrate yield was about half of the total and other forms of DIN typically are minor, export of N in organic form was substantial (Lewis et al. 1999). This contrasts with human-impacted rivers where most N is exported as nitrate. Total N yields from pristine temperate rivers are estimated to have ranged between 0.76 and 2.3 kg ha1 year1 (Howarth et al. 1996), which is consistent with an average value of 0.86 kg ha1 year1 for 85 relatively undisturbed sites in the United States (Clark et al. 2000). The higher N yields of pristine tropical rivers, in comparison with temperate systems in their undisturbed state, is attributed to greater N-fixation and atmospheric deposition, and to the absence of N limitation in tropical terrestrial systems, which favors the export of N (Downing et al. 1999, Holland et al. 1999).

Where anthropogenic inputs are important, N values are elevated, sometimes greatly. In the Ohio River, for example, the total N yield is 9.4 kg ha1 year1, of which two thirds is nitrate (Howarth et al. 1996). An impressive 80% of the variation in the export of nitrate from 35 major world rivers was explained by fertilizer application, atmospheric deposition, and human sewage (Caraco and Cole 1999). Nitrogen yields of rivers draining northwestern Europe and the eastern United States are up to 19 times higher than under pristine conditions (Howarth et al. 1996). Indeed, most of the increase in total N is attributable to increases in nitrate, due mainly to increased fertilizer application. In a river draining an agricultural watershed in Illinois, nitrate concentrations were usually between 8 and 10 mg L 1 (David et al. 1997), far higher than recorded at undisturbed sites (Table 11.1). Long-term data from the Mississippi River establish that nitrate concentrations changed little from the turn of the century until the 1950s and then roughly doubled in the following 35 years, coincident with a steady rise in fertilizer application over the same time period (Turner and Rabalais 1991). In addition, atmospheric deposition of N in the northern hemisphere has increased by a factor of four, and some regions have experienced a 16-fold increase in comparison to preindustrial times (Holland et al. 1999).

The relative importance of various anthropogenic N inputs varies with the extent of human presence in catchments and especially with the intensity of agriculture. In a survey of 928 streams throughout the United States that were considered to be relatively uncontaminated but nonetheless affected to varying degrees by fertilizer runoff (Omernik 1977), concentrations of nitrate and TDN were proportional to percent of land in agriculture and inversely proportional to percent of land in forest (Figure 11.2). DIN increased from about 18% of total N in streams draining forested watersheds to nearly 80% in streams draining agricultural watersheds, presumably due to the use of N fertilizers. In rivers of the Chesapeake Bay drainage, percent of cropland correlated strongly with concentrations of nitrate and total N (Jordan et al. 1997). In tributaries of an agricultural catchment in southeastern Michigan, the ratio of agricultural to forested land explained 94% of the variation in annual average concentrations of nitrate (Castillo et al. 2000). In the mid-Atlantic region of the United States, half of the variation in total N concentrations was explained by the amount of agriculture in the catchment, and atmospheric deposition explained an additional 27% (Jones et al. 2001).

The Mississippi River Basin is an informative case study, important because N loading to the Gulf of Mexico appears to be responsible for a large anoxic region created by the decomposition of algal blooms resulting from the delivery of riverine nutrients (Section 13.2.3). Nitrate concentrations in the Mississippi River have increased markedly over the last 100 years, showing a particularly strong increase from 1970 to 1983, while fertilizer applications have increased sevenfold since 1960 (Goolsby et al. 1999). Nitrogen budgets reveal that almost 90% of the total N transported by tributaries of the Mississippi River derives from diffuse sources. Fertilizer and soil organic matter contribute 50%, atmospheric deposition, groundwater, and soil inputs supply 24%, and the application of animal manure provides 15% of total inputs. The remaining 11% is provided by discharges from waste water of urban and industrial origin. By region, the Ohio River is the source of one third of the nitrate discharged by the Mississippi into the Gulf of Mexico, but another third originates in the intensively agricultural lands of Illinois and Iowa despite the much smaller water discharge from that portion of the Mississippi Basin (David and Gentry 2000). The availability of such a detailed accounting is extremely useful because it helps to identify sources and thereby suggest what management practices should try to achieve.

Although agricultural activities are frequently the most important N input and atmospheric deposition also can be significant, the absolute and relative magnitude of anthropogenic sources varies greatly with human presence and activities. A comparison of 16 catchments in the northeastern United States, all with extensive forest cover (48-87%), found substantial differences in N inputs and exports related to variation in human population density and extent of agriculture (Boyer et al. 2002). Total inputs of N across the catchments was negatively correlated with the fraction of land area in forest, and positively correlated with the fraction of land area in agriculture and with the fraction of disturbed land (agriculture plus urban land) (Figure 11.13). On average, atmospheric deposition was the dominant N source (31%), followed by imports of N in food and feed (25%), fixation in agricultural lands (24%), fertilizer use (15%), and fixation in forests (5%). The lesser importance of fertilizer, relative to studies described above, is

FIGURE 11.13 Nitrogen inputs estimated from budgets constructed for 16 catchments of the northeastern United States and draining to the North Atlantic Ocean are strongly correlated with land use in the catchments: (a) a negative relationship with forested land, (b) a positive relationship with agricultural land, and (c) an even stronger positive relationship with the sum of urban and agricultural land. (Reproduced from Boyer et al. 2002.)

FIGURE 11.13 Nitrogen inputs estimated from budgets constructed for 16 catchments of the northeastern United States and draining to the North Atlantic Ocean are strongly correlated with land use in the catchments: (a) a negative relationship with forested land, (b) a positive relationship with agricultural land, and (c) an even stronger positive relationship with the sum of urban and agricultural land. (Reproduced from Boyer et al. 2002.)

unsurprising for catchments that are half or more forested. When all 16 catchments are compared, it is apparent that N inputs reflect various anthropogenic sources, which differ in both absolute and relative magnitude across forested, urban, and agricultural catchments.

Comparison of the export of N by rivers to all catchment inputs reveals that river export and human-derived inputs are highly correlated (Figure 11.12). On average rivers export only about 25% (Howarth et al. 1996, Boyer et al. 2002) to 40% (Goolsby et al. 1999) of these loadings. De-nitrification in wetlands and aquatic ecosystems likely is a major sink for the remainder (van Bree-men et al. 2002), but export of food and wood products also can be significant loss terms. There can be little doubt that high N yields, which in the rivers studied by Howarth et al. (1996) all exceed 10 kg ha"1 year"1, are the result of increased nonpoint inputs due to fertilizers and atmospheric deposition, with the former responsible for roughly two thirds of the total.

Was this article helpful?

0 0

Post a comment