Relating nutrient uptake with transient storage in forested mountain streams

Streams control the timing and delivery of fluvial nutrient export from watersheds, and hydraulic processes such as transient storage may affect nutrient uptake and transformation. Although we expect that hydraulic processes that retain water will increase nutrient uptake, the relationship between transient storage and nutrient uptake is not clear. To examine this relationship, we injected a conservative tracer and nutrients (ammonium and phosphate) into 13 streams for a total of 37 injections at Hubbard Brook Experimental Forest (HBEF), New Hampshire. Transient storage was estimated by fitting conservative solute data to a one-dimensional advection, dispersion, transient storage model. To correct for variation in depth and velocity among streams, we considered nutrient uptake as a masstransfer coefficient (Vf), which estimates benthic demand for nutrients relative to supply. Transient storage decreased with increasing specific discharge (discharge per unit stream width). Transient storage explained only 14% of variation in ammonium Vf during the entire year and 35% of variation during summer months. Phosphate uptake was not related to transient storage, presumably because P uptake is predominantly by chemical sorption at HBEF. At HBEF, surface water pools can store water but were not modeled as such by use of the transient storage model. These pools were probably not important areas of nutrient uptake; further variation in the relationship between nutrient uptake and transient storage may be explained by biological demand. Streams are important landscape features because they provide an avenue for nutrient loss from the terrestrial landscape and subsequent delivery to downstream ecosystems. Streams are not simply conduits, however, because they alter the form and amounts of nutrients through uptake and transformation of dissolved and particulate forms (Burns 1998; Fisher et al. 1998; Alexander et al. 2000). The pattern of element loss observed from forests (e.g., Likens and Bormann 1995) may, in part, be a function of in-stream processes (Hall et al. 2001). One way to examine how in-stream 1 Corresponding author ([email protected]). Present address: Department of Zoology and Physiology, University of Wyoming, Laramie, Wyoming 82071. 2 Present address: Nicholas School of the Environment, P.O. Box 90329, Duke University, Durham, North Carolina 27708-0329. Acknowledgments Thanks to Stuart Levenbach, who assisted with field work and excelled at ammonium analysis, and to George Bernhardt III, who also helped with field work. Comments from Mike Paul, Mike Marshall, Brad Taylor, and two anonymous reviewers improved the manuscript. Discussions with Doug Smith and Steve Thomas helped with our interpretations of transient storage. This paper is a contribution to the Hubbard Brook Ecosystem Study and to the program of the Institute of Ecosystem Studies. The Northeastern Research Station, U.S.D.A. Forest Service, operates and maintains HBEF. Financial support was provided by the Andrew W. Mellon Foundation to G.E.L., NSF predoctoral fellowship to E.S.B., and NSF LTER and LTREB. processes affect nutrient transport is to measure nutrient uptake length, which is the average distance downstream traveled by a nutrient atom before being removed from the water. This measure indicates the degree of retentiveness for a given element (Newbold et al. 1981, 1983). Essentially, a small amount of nutrient or isotopic tracer is added to a stream to estimate uptake relative to a nonreactive hydrologic tracer. This approach has been used to describe stream nutrient dynamics and has effectively demonstrated the importance of streams in processing nutrients (Newbold et al. 1981; Munn and Meyer 1990; Mulholland et al. 1997). To understand the role of in-stream processes in determining watershed nutrient export, we need to estimate what factors control uptake length in streams. Uptake length varies among streams, and this variation may be caused by hydrologic, geomorphological, and biological processes. For example, streams with greater depth and velocity will have longer uptake lengths, and indeed most of the variation in uptake length can be a function of these two components of discharge (Valett et al. 1996; Butterini and Sabater 1998). Nonetheless, many studies use uptake length as a measure of nutrient uptake, but because uptake length is sensitive to stream discharge, it is difficult to compare lengths among different size streams. One way to correct for this effect is to calculate a mass-transfer coefficient of uptake (Stream Solute Workshop 1990; Davis and Minshall 1999), which represents demand for nutrients relative to supply in the water column.