Linking the hydrologic and biogeochemical controls of nitrogen transport in near-stream zones of temperate-forested catchments: a review

We review the status of research concerning the links between hydrologic towpaths and the biogeochemicai environment controlling Nitrogen cycling and transport in near-stream saturated zones, centering on stream environments of the northern, teml~rate-forested zone. N retention, transformation and mobilization occur in streamside wetlands, floodplains, riparian zones, seepage faces, and the hyporheic zone. These areas are the focal point in non-point source loading of N to stream channels. They also represent areas where rapid changes in water-table and hydrologic towpaths occur during rainfall-runoff events. It is the combination of an abrupt change in biogeachemical environment, encountering a hydrologic boundary (the terrestrial/aquatic interface or ecotone), that make the near-stream/saturated zone critical for elucidating controls of N transport and transformation. We review published studies concerning the hydrologic controls of N transport in near-stream zones, and subsequently present several geomorphic and hydrodynamic scenarios relating N biogeochemistry and its response to hydrologic events (of both varying magnitude and seasons). It is at the critical junction between temporal and spatial conditions affecting N cycling in the near-stream zone, that research priorities must now be focused. 1. I n t r o d u c t i o n 1.1. Excess nitrogen in the environment Recent concern over increased human-induced atmospheric N-deposition, in addition to * Corresponding author. 0022-1694/97/517.00 © 1997Elsevier Science B.V. All rights reserved Pll S0022-1694(96)03286-6 C.P. Cirmo. J.J. McDonnell~Journal of Hydrology 199 (1997) 88-120 89 diffuse non-point source N-inputs to surface waters from agriculture and forestry practices, has stimulated investigations of controls on the biogeochemistry and transport of N. Much of this effort has been centered on forested and agricultural catchments within the humid-temperate zone of North America and Europe. Increased N concentrations have been observed in streams in the Adirondack and Catskill Mountains in the USA (Stoddard and Murdoch, 1991; Murdoch and Stoddard, 1992; Driscoll and van Dreason, 1993; Stoddard, 1994), in mid-Atlantic Appalachian watersheds of the USA (Kramer et al., 1986; Smith et al., 1987), in the Great Smoky Mountains of Tennessee, USA (Elwood et al., 1991), as well as in Germany (Hauhs et al., 1989) and the United Kingdom (Burt and Haycock, 1992). Attempts to link increased N concentrations in streams with atmospheric N-deposition have led to the development of "nitrogen saturation" hypotheses (Aber et al., 1989, 1991). These hypotheses consider a combination of increasing atmospheric N-deposition, widespread forest maturation, decreased forest cutting, and stressed forest health from acidic deposition, as possible contributing factors. A categorization of watersheds by their potential N-input sensitivity has also been developed (Malanchuk and Nilsson, 1989; Abet et al., 1991; Tamm, 1991; Stoddard, 1994). Many eastern USA forests and most forests in Europe are thought to be approaching a period of steady-state in terms of biological aggradation and N demand (Sullivan, 1993). The environmental consequences of additional N-inputs to surface waters include increased surface water acidification and possible N-based downstream eutrophication (Wright, 1991; Stoddard, 1994). In addition, an increase in N-loading from agricultural areas and artificial drainage systems has been widely documented in the United States (Kohl et al., 1971; Baker and Johnson, 1976; Chichester, 1976; Burton et al., 1977; Duda, 1982; Owens et al., 1991; Jordan et al., 1993), and in Europe (Iserman, 1990; Burt and Haycock, 1992; Heathwaite et al., 1993; Armstrong and Burt, 1993). There is abundant evidence of increased N-concentration in runoff from agricultural fields and from other land-disturbance activities (Likens et al., 1977; Meybeck, 1982; Lowrance et al., 1984a, b; Burr and Arkell, 1987). Increasing NO~ concentrations in public water supplies have also been observed, with subsequent concerns over drinking water potability (Dillon et al., 1991; Dourson et al., 1991). The sensitivity of regional groundwater recharge areas to N inputs is being studied in the headwater catchments of several major metropolitan water supplies, including the New York City water supply catchments in the Catskill Mountains of New York, USA (Stoddard, 1992; Murdoch and Stoddard, 1992), the Chesapeake Bay (USA) watershed (Glibert et al., 1991), and in the United Kingdom (Burt and Haycock, 1992). Galloway et al. (1995) have estimated that human activity has led to the fixation of an additional 140 Tg N per year, over and above natural processes like biotic N-fixation and lightning. Preliminary studies indicate that the fate of much of this additional N-input may involve continental processes, including storage in deep groundwater. Indeed, increasing concentrations of N in groundwater have recently been noted (Burt and Trudgill, 1993; Spalding and Exner, 1993; Sweeney, 1993; B6hlke and Denver, 1995). The correlation of increased N in stream water, with increased N in groundwater, is still somewhat unstudied. Attempts to identify sources of increased N in groundwater, along with the fate of N in near-stream environments, will be critical to a full understanding of the significance of the "nitrogen saturation" hypothesis to different catchments. 90 C.P. Cirmo, J.J. McDonnell~Journal of Hydrology 199 (1997) 88-120 1.2. Nitrogen-transformation zones within the watershed The identification of key landscape environments controlling N transport within catchments have been assisted by studies of biotic and abiotic controls of N transformation, and by the calculation of watershed N budgets. Several critical factors have been identified in both forested and agricultural catchments, including 1) the state of ecosystem maturation (age), 2) the in-situ decomposition rate (microbial status, soil fertility, moisture regime, etc.), 3) Cand N-limitation status, 4) the physical and chemical soil characteristics, and 5) the availability of moisture. Numerous investigations have shown that near-stream saturated zones and riparian wetlands are active sites of N-biogeochemical dynamics (Peterjohn and Correll, 1984; Lowrance et al., 1985; Cooke and Cooper, 1988; Cooper, 1990; Hill, 1990; Haycock, 1991; Mulholland, 1992). Furthermore, since near-stream/ saturated zones are interfaces between hillslope and stream channel dynamics, they should play a critical role in determining the amount and speciation of N entering the stream channel. The juxtaposition of the near-stream/saturated zone, the stream channel, the hyporheic zone, and the catchment hillslope, are shown diagrammatically in Fig. 1. The hydrologic routing of N from the upland hillslope through the near-stream zone, is relatively unstudied, as is the interaction between "hydrologic towpath" and "biogeochemical pathway" (Hill, 1990; Eshleman et al., 1994; O'Brien et al., 1994; BOhlke and Denver, 1995). N transformation and retention should occur where hydraulic residence time is increased and where saturated conditions prevail. Surface water-groundwater interface zones might include the hillslope-lowland interface, riparian wetlands and streambanks,