Nitrate Dynamics in Riparian Forests: Microbial Studies

While riparian forests have a demonstrated ability to remove nitrate (NO3-) moving from uplands before it enters streams, there is considerable uncertainty as to the mechanisms of NOr removal in these areas. We characterized spatial and temporal variation in denitrification enzyme activity (DEA), microbial biomass C and N content, soil respiration and potential net N mineralization and nitrification in three riparian forest sites consisting of soil catenas containing mod. erately well, somewhat poorly, poorly, and very poorly drained soils (inceptisols and entisols). These measurements were made in conjunction with studies of NOfremoval from groundwater during growing and dormant seasons that are reported in a companion paper. Two f the sites were on stratified glacial drift, one with an undeveb oped upland and one with an upland with high density unsewered residential development that produced groundwater at the edge of the riparian zone with NOf-N concentrations between 8 and 12 mg/L The third site was on unstratified glacial drift with an undeveloped upland. Hydric surface (0-15 cm) soils (poorly and very poorly drained) sistently had higher DEA than upland-wetland transition zone (moderately well and somewhat poorly drained) surface soils. Spatial patterns of microbial biomass C and N content were more variable but showed the same general pattern as DEA. Levels of DEA and microbial biomass were consistently low or undetectable at and below the seasonal high water table. Surface soil DEA and microbial biomass were correlated with NOr emoval from groundwater during the growing season. Low levels of DEA and microbial biomass in the subsurface however, suggested that plant uptake was the dominant groundwater NOr sink during the growing season. During the dormant season, water table levels were higher and groundwater-borne NOfwas ble to interact with near surface soil and be removed by denitrification and/or microbial immobilization. Potential net N mineralization was quite variable both within and between sites, while potential net nitrification was very low at most sites. A notable exception was in the transition-zone soils at the stratified drift site with a densely developed upland, which ad relatively high rates of net NOr production. The high nitrification rates observed at this site may suggest that the longterm buffering potential of this site is limited. W RIPARIAN FORESTS have a demonstrated ability to remove NO~ moving from uplands before it enters streams (Lowrance et al., 1984; Peterjohn and Correll, 1984; Jacobs and GiIliam, 1985; Pinay and Decamps, 1988; Cooper, 1990), there is considerable uncertainty as to the mechanisms of NO~ removal in these areas. The strong affinity of plants and microorganisms for N creates a high potential for NO~ removal in riparian zones, but there is considerable spatial and temporal variability in plant and microbial activity and consequently, in riparian zone N dynamics. Understanding this variability is necessary for the development of broadly applicable models and management strategies for the use of riparian zones for water quality maintenance and improvement (Lowrance and Shiromhammadi, 1985). Dep. of Natural Resources Science, Univ. of Rhode Island, Kingston, RI 02881. Current address of P.M. Groffman is Inst. of Ecosystem Studies, Box AB, Millbrook, NY 12545. Received 4 Sept. 1991. *Corresponding author. Published in J. Environ. Qual. 21:666-671 (1992). Possible NO g fates in riparian zones include uptake by plants, denitrification (conversion of NO~ into gases by facultative anaerobic microorganisms), and microbial immobilization. Within riparian zones, marked changes in vegetation, soils and hydrology occur over small spatial scales (Warwick and Hill, 1988; Cooper, 1990). These changes cause marked spatial variation in NOg attenuation processes. In addition to spatial variation, seasonal variation, especially in vegetation uptake, is likely to be significant. Over longer time scales, the nature and extent of N input into riparian zones will likely affect attentuation processes. Riparian zones that have been subject to long-term NO3 inputs may have different attenuation capacities than non-NOg enriched areas. Understanding the relative magnitude of different NO~ attenuation mechanisms, or the partitioning of NO~ to different fates in riparian zones, is important for evaluating the long-term effectiveness of these zones as NO ~ sinks. While denitrification results in removal of NOg from the system as N gas, NO~ removed by plant uptake and microbial immobilization is subject to remineralization and release back to the soil solution. Over time, plant and microbial pools can become enriched, or "’saturated" with N, resulting in a decline in their N absorbing capacity (Aber et al., 1989). Longterm N inputs to riparian zones may lead to increases in N mineralization and nitrification induced by N enrichment of soil organic matter and microbial populations. Increases in mineralization and nitrification could increase within-site NO ~ production and export from the riparian zone (Hill and Shackleton, 1989). Sites where dentrification is the dominant NO3 sink should be more effective as long-term buffer zones than areas where plant uptake and immobilization are the dominant sinks. In this study, we characterized microbial processes in three forested riparian zones with catenas consisting of moderately well, somewhat poorly, poorly, and very poorly drained soils. The study presented here was carried out in conjunction with direct measurements of NO g removal from groundwater at these sites (see the companion paper by Simmons et al., 1992). We measured microbial biomass C and N content as indices of the size of the active pool of microorganisms (Smith and Paul, 1990), and several indices of microbial activity including DEA, soil respiration, and potential net N mineralization and nitrification. Our objectives were to: (i) characterize spatial variation microbial processes in riparian zones and (ii) elucidate the mechanisms responsible for the patterns of NO3 attenuation observed in the companion groundwater study. MATERIALS AND METHODS