PROOF COPY [EE/2002/023416] 005405QEE PROOF COPY [EE/2002/023416] 005405QEE Sorption Behavior and Long-term Retention of Reactive Solutes in the Hyporheic Zone of Streams Karin Jonsson 1 ; Ha ˚ kan Johansson 2 ; and Anders Wo ¨ rman 3 Abstract: This paper analyzes the transport of sorbing solutes by extending the advective storage path model developed for longitudinal transport of inert solutes in streams coupled with flow-induced uptake in the hyporheic zone. Independent observations of a conservative ( 3 H) and a reactive ( 51 Cr) tracer in both the stream water and the hyporheic zone were used to differentiate between hydraulic and sorption processes. The method of temporal moments was found to be inadequate for parameter determination, whereas fitting versus the entire tracer breakthrough curves with special emphasis on the tail indicates that the proposed model could be used to represent both conservative and reactive transport. Information on the tracer inventory of the conservative tracer in the hyporheic zone was found to be of vital importance to the evaluation of the hydraulic exchange. A model evaluation based on stream water data alone can yield predictions of a wash-out in the hyporheic zone that deviates markedly from the observed wash-out. This prohibits long-term predictions of the wash-out from the hyporheic zone as well as the evaluation of sorption properties. The sorption in the hyporheic zone was found to follow a two-step model, where the first step is instantaneous and the second kinetic. A model with a single-step sorption process could not reproduce the observed breakthrough curves. An evaluation of the relative importance of including sorption kinetics in solute stream transport models is elucidated by means of the analytical expressions for the temporal moments. The omission of the kinetics in the second sorption step in the hyporheic zone will result in relative errors in the moments of second order or higher. The error will increase with decreasing residence time in the hyporheic zone. Especially, long-term predictions of the wash-out from the hyporheic zone require consideration of the rate-limited sorption. DOI: 10.1061/ASCE0733-93722004130:51 CE Database subject headings: Sorption; Streams; Hydrologic models. Introduction Understanding the transport mechanisms of solutes in rivers and streams is of vital importance to the predictions of the effects of contaminants in river systems. Different physical, chemical, and biological processes affect a solute transported in the river. The relative influence of the different processes on the residence time of the solute in the system depends on the reactivity of the sub- stance and the prevalent environmental conditions. Hence, to be able to model the retention of solutes in streams, we have to parameterize the different processes. A number of model concepts and different evaluation methods have previously been reported in the literature Bencala and Walters 1983; Jackman et al. 1984; Bencala et al. 1990; Wagner and Harvey 1997; Wo ¨ rman et al. 1998; Fernald et al. 2001; Jo- hansson et al. 2001; Laenen and Bencala 2001. However, im- provements in model formulations combined with fieldwork per- formance and evaluation techniques still give new insights into the transport mechanisms of solutes. Previously proposed solute stream transport models, such as the commonly used transient storage model Bencala and Walters 1983; Harvey et al. 1996; Schmid 1997; Choi et al. 2000 or the concept in which the exchange with the hyporheic zone is de- scribed as diffusive Jackman et al. 1984; Elliott and Brooks 1997b; Wo ¨ rman 2000; Jonsson et al. 2003 have been success- fully used to represent solute breakthrough curves resulting from stream tracer tests. However, attempts have recently been made to formulate models with a more physically based description of the hydraulic exchange with the hyporheic zone Elliott 1990; Elliott and Brooks 1997a; Packman et al. 2000; Wo ¨ rman et al. 2002. The main purpose of formulating physically based expressions for exchange processes is that interpretation of the data provides a physical understanding and also facilitates future generalization of the models. If the expressions for exchange processes reflect the underlying geochemistry and physics, we can expect to be able to identify model parameters that vary less with environmen- tal conditions than will otherwise be the case. Wo ¨ rman et al. 2002 presented a residence time modeling framework for conservative solute transport in streams that couples the longitudinal transport with flow-induced uptake in the hyporheic zone the advective storage path model referred to as the ASP-model. The model agrees with several investigations of the exchange with the hyporheic zone, which indicate that the exchange mainly occurs through advection in and out of the 1 PhD student, Dept. of Earth Sciences, Uppsala Univ., Villava ¨gen 16, SE-752 36 Uppsala, Sweden corresponding author. E-mail: Karin.Jonsson@geo.uu.se 2 PhD, Dept. of Biometry and Informatics, Swedish Univ. of Agricultural Sciences, Johan Brauners va ¨g 3, P.O. Box 7013, SE- 750 07 Uppsala, Sweden. 3 Associate Professor, Dept. of Biometry and Informatics, Swedish Univ. of Agricultural Sciences, Johan Brauners va ¨g 3, P.O. Box 7013, SE-750 07 Uppsala, Sweden. Note. Associate Editor: Mark J. Rood. Discussion open until October 1, 2004. Separate discussions must be submitted for individual papers. To extend the closing date by one month, a written request must be filed with the ASCE Managing Editor. The manuscript for this paper was submitted for review and possible publication on November 26, 2002; approved on March 24, 2003. This paper is part of the Journal of Environmental Engineering, Vol. 130, No. 5, May 1, 2004. ©ASCE, ISSN 0733-9372/ 2004/5-1–12/$18.00. JOURNAL OF ENVIRONMENTAL ENGINEERING © ASCE / MAY 2004 / 1 PROOF COPY [EE/2002/023416] 005405QEE