Elsevier

Advances in Water Resources

Volume 26, Issue 9, September 2003, Pages 951-964
Advances in Water Resources

Comparison of transient storage in vegetated and unvegetated reaches of a small agricultural stream in Sweden: seasonal variation and anthropogenic manipulation

https://doi.org/10.1016/S0309-1708(03)00084-8Get rights and content

Abstract

In this work, we compare solute transport and hyporheic exchange in vegetated and unvegetated reaches of Säva Brook, an agricultural stream in Sweden subject to extreme variations in channel vegetation and morphology due to both natural seasonal effects and anthropogenic manipulation. A solute injection experiment was conducted in September, 2001 (late summer), at which time there was extensive in-channel vegetation in upstream reaches but none in downstream reaches due to channel excavation by farmers. Experimental results are interpreted using both the advective storage path model and the transient storage model. Results from the vegetated and excavated reaches of the stream are compared both with each other and with the results of a previous experiment conducted in April, 1998 (early spring), when the stream was not excavated but there was only minimal vegetation present in the area due to natural seasonal effects. Results from the two injection experiments are compared by using scaled parameters that appropriately include the effects of stream velocity and depth on hyporheic exchange. This analysis indicates that the variation of solute storage time in all non-excavated agricultural reaches is attributable to differences in stream flow depth, velocity, and the hydraulic conductivity of the streambed sediments. Mixing in vegetated reaches is characterized by rapid exchange and considerable lag of the mean solute peak relative to the mean channel velocity. In addition, excavation altered the stream channel geometry so as to increase the storage time of solutes and reduce the effective exchange rate. This work indicates the need to consider the effect of specific processes when analyzing hyporheic exchange using tracer-injection methods, and supports the use of model frameworks with the potential to explicitly include different formulations for various hyporheic and dead zone transport processes.

Introduction

There has been long-standing uncertainty as to the exact processes represented by the empirical hyporheic exchange parameters obtained from stream tracer studies. The seminal work of Bencala and Walters on the transient storage model (TSM) noted that the derived exchange parameters represent the effect of multiple processes including both in-stream mixing and stream–subsurface exchange [1]. Indeed, earlier studies of solute transport had attributed all solute detention to in-stream processes [7], [22], [23], and there is still ongoing work in hydraulics to analyze mixing with off-channel dead zones [21].

In recent years, there has been some tendency to ascribe all observed transient storage to subsurface mixing, for example by considering the so-called ‘hyporheic area’, (As), obtained from the TSM to be some physically-measurable ‘extent’ of the hyporheic zone. This concept is misleading for several reasons. First, all exchange parameters derived from fitting breakthrough curves of injected solutes reflect some sort of average between in-stream mixing processes and various subsurface exchange processes. At this point, there should be no question that such averaging occurs, and the only outstanding question is the exact magnitude of fluxes resulting from each of the underlying processes. Even when considering only hyporheic exchange, the exchange rate and depth of penetration are fundamentally coupled and not independent as normally assumed in analyzing the results of solute injection experiments. This effect is due to the fact that both the rate and the extent of exchange are controlled by the same underlying hydrodynamic processes; namely the interaction of stream and subsurface flows across the interface of the channel boundary [15]. Theoretical analysis indicates that this coupling between TSM parameters requires them to vary consistently when representing a particular exchange process [26], and this theory has been confirmed by the results of controlled experiments [27].

This interdependence of idealized solute transport model parameters makes it extremely difficult to evaluate the effects of changing stream conditions on hyporheic exchange. At the current time, the only practical method for probing hyporheic exchange over long stream reaches is to analyze transient storage behavior from the results of solute injection experiments [10]. However, almost any change in stream conditions will simultaneously modify both in-stream transport and hyporheic exchange. If in-stream transport and hyporheic exchange modify solute transport behavior in exactly the same way then it would not be possible to distinguish their effects from in-stream solute transport measurements, and the magnitude of hyporheic flows could then only be discerned by means of direct subsurface measurements. But when in-stream mixing and hyporheic exchange have different functionality with stream conditions, which is generally expected due to differences in the characteristic timescales for in-stream and subsurface transport processes, then the results of solute injection experiments can be used to assess the effects of changing stream conditions on hyporheic exchange.

In this work, we apply these concepts to Säva Brook, a small agricultural stream in Sweden that is subject to both natural seasonal variations and anthropogenic modification of the stream channel (morphology and vegetation). Säva Brook is a good test case for this analysis because it is much more geomorphologically homogenous than the highly variable mountain streams where hyporheic exchange has most frequently been studied. This means, for example, that the stream slope and energy grade line show little or no sharp discontinuities over long stretches of the stream. In addition, analysis of solute transport in the test section is simplified by the fact that the stream has no significant tributaries in this reach.

While the stream tends to be geomorphologically similar through the study area, it is subject to extreme seasonal variations in vegetation and also to anthropogenic modification. As the stream sits at 59°(45–55) north latitude, low light levels and temperatures inhibit macrophyte growth in the winter. Conversely, temperate summers with long days cause extensive proliferation of vegetation, sometimes throughout the entire active stream channel. Effectively, this produces a significant natural variation in the extent of in-channel dead zones. The stream has only modest seasonal variations in flow rate, which allows us to directly relate changes in solute transport during different seasons to the extent of in-stream mixing due to vegetation. Anthropogenic modification also affects solute transport in the stream. It is a common practice in this area (and throughout Northern Europe) for farmers to excavate channels to improve drainage of agricultural fields. Excavation removes all in-channel vegetation (indeed, this is the primary goal of excavation), so that the stream has some regions without vegetative cover even during the summer months.

The main goal of the work described here is to consider the effects of in-stream vegetation and anthropogenic modification of channel morphology on solute transport. A tracer injection was performed in Säva Brook in September, 2001 (late summer) at a time when there was extensive in-channel vegetation in upstream reaches, little in-channel vegetation along most of the study reach, and no vegetation in a long downstream reach that had been excavated just before the experiment. Both the TSM and our new advective storage path (ASP) model are used to analyze solute breakthrough curves at several downstream locations. The results from different reaches are compared with each other and with the results of a previous injection experiment that had been conducted in April, 1998 (early spring) at a time when there was much less vegetation throughout the study reach [12], [25]. The extreme difference in vegetative cover between natural and excavated reaches emphasizes the effect of in-channel mixing. The comparison of the September, 2001 and April, 1998 results shows the effect of natural seasonal variation on solute transport in the upstream reaches, and provides another basis for evaluating the effect of excavation on transport in the downstream reach. Physically based scaling arguments will be used to facilitate this comparison.

Section snippets

Site description

Säva Brook is a typical small stream in central Sweden, and is similar to many streams that drain low-lying agricultural lands throughout Northern Europe. The stream is about 34 km long and drains a watershed area of about 197 km2. It originates in forested lands to the Northwest of Uppsala, continues through around 25 km of agricultural fields through Uppland County to the West of Uppsala, and eventually discharges into Lake Mälaren, which is a large and important lake that connects with the

Theory

Currently available models for solute and contaminant transport in streams are limited in that they use empirical mass-transfer parameters to represent the effect of stream–subsurface exchange and they cannot distinguish actual stream–subsurface exchange from similar mixing into low-flow regions within the stream channel [1], [15]. As a result, mixing in vegetation patches and similar in-stream features can obscure measurements of hyporheic exchange. Herein we will apply the ASP model and the

Analysis of solute breakthrough curves from 2001 injection

The iodide breakthrough curves observed in 2001 are shown in Fig. 4 along with the optimized solutions from the ASP model. The ASP model simulations shown in Fig. 4 are fit to the solute breakthrough curves using an exponential residence time probability density function (T-PDF). The measured physical parameters and the optimized model parameters used to simulate the solute breakthrough curves are presented in Table 2. The TSM model simulation is shown in Fig. 5, and the best-fit TSM model

Seasonal variation and effect of vegetation

Natural seasonal variations obviously alter many aspects of stream systems. Normal hydrologic variations in stream discharge cause simultaneous variations in the stream velocity and cross-sectional geometry that influence both in-stream transport and hyporheic exchange. In addition, seasonal climatic variations also influence biological processes that have an effect on stream flow and solute transport. In the case of Säva Brook, located at almost 60° north latitude, there is an extreme

Conclusions

A tracer injection experiment was performed in Säva Brook, Uppland County, Sweden to assess stream–subsurface exchange and in-channel mixing in an agricultural stream. The advective-storage path (ASP) model and the TSM were used to represent solute transport results obtained at five sampling stations over a distance of 10.4 km. Both the ASP model and the TSM simulated the observed breakthrough curves well, and the mean storage zone residence times calculated by the two models were comparable.

Acknowledgements

We gratefully acknowledge the financial support for this research provided by the US National Science Foundation under Grant #0196368, travel funds provided by NSF’s Office of International Programs, and the financial support of the Swedish National Energy Administration, the Swedish Research Council and Elforsk AB. We would also like to thank Shulan Xu, Anna Lindahl and Mirja Törnquist of Uppsala University for their help in executing the tracer injection experiment.

References (28)

  • H. Johansson et al.

    Retention of conservative and sorptive solutes in rivers––simultaneous tracer experiments

    Sci. Total Environ.

    (2001)
  • K.E. Bencala et al.

    Simulation of solute transport in a mountain pool-and-riffle stream: a transient storage model

    Water Resour. Res.

    (1983)
  • Y. Bichsel et al.

    Oxidation of iodide and hypoiodous acid in the disinfection of natural waters

    Environ. Sci. Technol.

    (1999)
  • Y. Bichsel et al.

    Formation of iodo-trihalomethanes during disinfection and oxidation of iodide containing waters

    Environ. Sci. Technol.

    (2000)
  • Carter RW, Davidian J. General procedures for gaging streams, US Geological Survey, Techniques of Water Resources...
  • A.H. Elliott et al.

    Transfer of nonsorbing solutes to a streambed with bed forms: theory

    Water Resour. Res.

    (1997)
  • A.H. Elliott et al.

    Transfer of nonsorbing solutes to a streambed with bed forms: Laboratory experiments

    Water Resour. Res.

    (1997)
  • H.B. Fischer et al.

    Mixing in Inland and Coastal Waters

    (1979)
  • Harvey JW. Nationwide assessment of channel physical characteristics that influence solute storage and downstream...
  • J.W. Harvey et al.

    Evaluating the reliability of the stream tracer approach to characterize stream–subsurface waterexchange

    Water Resour. Res.

    (1996)
  • J.W. Harvey et al.
  • P.A. Hutchinson et al.

    Solute uptake in aquatic sediments due to current–obstacle interactions

    J. Environ. Eng., ASCE

    (1998)
  • Kilpatrick FA, Cobb ED. Measurement of discharge using tracers, US Geological Survey, Techniques of Water Resources...
  • H.M. Nepf

    Drag, turbulence and diffusion through emergent vegetation

    Water Resour. Res.

    (1999)
  • Cited by (66)

    • Hyporheic Zone and Processes

      2022, Encyclopedia of Inland Waters, Second Edition
    View all citing articles on Scopus
    View full text