DYE TRACING IN THE JORDAN SANDSTONE NEAR THE CRYSTAL SPRINGS STATE FISH HATCHERY, WINONA COUNTY, MINNESOTA

Southeast Minnesota’s karst lands support many trout streams, which depend on the cool and clear water sourced by springs. These springs are under increasing pressure from human activities; greater withdrawals of groundwater may be decreasing spring flow, while domestic and agricultural pollutants are affecting water quality. This study aimed to initiate delineation of the springshed surrounding the Crystal Springs State Fish Hatchery (CSSFH) in Winona County, Minnesota. CSSFH springs emerge from near the St. Lawrence Formation/Jordan Sandstone contact. A 1980 dye trace in the Kieffer Valley (across the Whitewater River north of CSSFH) from a stream sink near the top of the St. Lawrence reached four different springs lower in St. Lawrence, up to 1.7 km away, in less than a week. This trace, which was replicated in 2008, demonstrated rapid conduit flow in the St. Lawrence.

Two dye traces were attempted in or through the Jordan Sandstone. One dye was injected in a sinking stream in the Kieffer Valley about 600 m northwest, upstream of the 1980/2008 sink point. This sinkpoint was at or just above the Jordan/St. Lawrence contact. A second dye was injected about 2 km southwest of CSSFH near the Oneota Formation/Jordan Sandstone contact in the terminal sink of a sinking stream. Charcoal detectors were installed in 2 springs at CSSFH, 4 other springs, 6 surface stream locations and one private well. The detectors were initially changed on weekly to biweekly intervals and then at monthly intervals for five months. The dye packets were eluted with basic isopropanol and analyzed on a scanning spectrophotometer. No dye was detected from either injection.

Negative dye traces are ambiguous. Four classes of explanations exist: 1) The dye has not yet reached the detection points. The flow is slower than hypothesized and we have not waited long enough. 2) The dye was diluted below the detection limit. We did not use enough dye. 3) The dye flowed to unmonitored locations. We did not look in the right places. 4) The dye may have been adsorbed or chemically changed. The third explanation seems to be the most reasonable in this case, suggesting the groundwater flow systems in this area are more complex than initially assumed.