HYDROLOGICAL PROCESSES Hydrol. Process. 23, 3319–3333 (2009) Published online 15 September 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/hyp.7427 A comparison of pre- and post-remediation water quality, Mineral Creek, Colorado † Robert L. Runkel, 1 * Kenneth E. Bencala, 2 Briant A. Kimball, 3 Katherine Walton-Day 1 and Philip L. Verplanck 4 1 U.S. Geological Survey, Mail Stop 415, Denver Federal Center, Denver, CO 80225, USA 2 U.S. Geological Survey, Mail Stop 439, 345 Middlefield Road, Menlo Park, CA 94025 USA 3 U.S. Geological Survey, 2329 W Orton Circle, WestValley City, UT 84119, USA 4 U.S. Geological Survey, Mail Stop 973, Denver Federal Center, Denver, CO 80225, USA Abstract: Pre- and post-remediation data sets are used herein to assess the effectiveness of remedial measures implemented in the headwaters of the Mineral Creek watershed, where contamination from hard rock mining has led to elevated metal concentrations and acidic pH. Collection of pre- and post-remediation data sets generally followed the synoptic mass balance approach, in which numerous stream and inflow locations are sampled for the constituents of interest and estimates of streamflow are determined by tracer dilution. The comparison of pre- and post-remediation data sets is confounded by hydrologic effects and the effects of temporal variation. Hydrologic effects arise due to the relatively wet conditions that preceded the collection of pre-remediation data, and the relatively dry conditions associated with the post-remediation data set. This difference leads to a dilution effect in the upper part of the study reach, where pre-remediation concentrations were diluted by rainfall, and a source area effect in the lower part of the study reach, where a smaller portion of the watershed may have been contributing constituent mass during the drier post-remediation period. A second confounding factor, temporal variability, violates the steady-state assumption that underlies the synoptic mass balance approach, leading to false identification of constituent sources and sinks. Despite these complications, remedial actions completed in the Mineral Creek headwaters appear to have led to improvements in stream water quality, as post-remediation profiles of instream load are consistently lower than the pre-remediation profiles over the entire study reach for six of the eight constituents considered (aluminium, arsenic, cadmium, copper, iron, and zinc). Concentrations of aluminium, cadmium, copper, lead, and zinc remain above chronic aquatic-life standards, however, and additional remedial actions may be needed. Future implementations of the synoptic mass balance approach should be preceded by an assessment of temporal variability, and modifications to the synoptic sampling protocol should be made if necessary. Published in 2009 by John Wiley & Sons, Ltd. KEY WORDS acid mine drainage; synoptic sampling; temporal variation; mass balance; geochemistry; tracer injection Received 3 March 2009; Accepted 1 July 2009 INTRODUCTION A variety of techniques have been used to quantify and characterize the effects of mining on instream water quality (Whyte and Kirchner, 2000; Brake et al., 2001; Kimball et al., 2001; McKnight et al., 2001; Bird 2003; Schemel et al., 2006). These techniques range from relatively simple mass balance approaches (Foos, 1997; Yu, 1998) to detailed models of reactive transport (Runkel et al., 2007; Caruso et al., 2008). Included within this range is the synoptic mass balance approach of Kimball et al. (2002, 2007), in which detailed spatial profiles of constituent load are developed. Under the synoptic approach, numerous stream and inflow locations are sampled for the constituents of interest, and estimates of streamflow are obtained by tracer dilution. Estimates of mass load at each stream location are then determined * Correspondence to: Robert L. Runkel, U.S. Geological Survey, Mail Stop 415, Denver Federal Center, Denver, CO 80225, USA. E-mail: runkel@usgs.gov † This article is a US Government work and is in the public domain in the USA. as the product of concentration and streamflow. The resultant spatial profiles of mass load may be used to identify and rank sources of metal contamination (e.g. Kimball et al., 2001; Runkel et al., 2005). The synoptic mass balance approach described above is used herein to evaluate the effectiveness of remedial measures undertaken for Mineral Creek, an acid mine drainage stream in southwestern Colorado. Pre- (1999) and post-remediation (2005) data sets provide a basis for this evaluation. Quantifying the effects of remedi- ation is theoretically straightforward: spatial profiles of mass load from 1999 and 2005 may be directly compared for individual constituents; decreases in mass load over time, if present, suggest improved conditions that may be attributable to remediation. Despite this apparent simplic- ity, two confounding factors complicate the evaluation of remedial effectiveness. First, although both data sets were collected under nominally steady, low-flow conditions, both data sets appear to be affected by substantial tem- poral variation in constituent concentrations that occurred during synoptic sampling. The steady-state assumption that underlies the synoptic mass balance approach is Published in 2009 by John Wiley & Sons, Ltd.