No abstract

No abstract

A 500‐m2 watershed on a deposit of coal refuse was instrumented for measurements of runoff and sediment yield and monitored for 15 months. The range of 42 storm period sediment yields was greater for the watershed as a whole (16 to 3,480 g m−2) than for a 4.65‐m2 portion of interfluve surface (15 to 1,596 g m−2). The ratio of sediment yield values (watershed:interfluve) was not constant. Ratios of about 7:1 occurred in late spring, when gully floors were flushed of colluvium that had accumulated during winter. The ratio remained above 2:1 in summer, when intense runoff caused gully incision. Estimates of annual sediment yield from three reaches of gully (82 to 160 kg m−2 yr−1) are about an order of magnitude greater than an estimate for interfluves (12 kg m−2 yr−1). A correlation analysis indicated that 78% of the observed variance in watershed sediment yield values is associated with characteristics of storm rainfall and runoff.

Two watersheds in northwestern Indiana were selected for detailed monitoring of bacterially contaminated discharges (Escherichia coli) into Lake Michigan. A large watershed that drains an urbanized area with treatment plants that release raw sewage during storms discharges into Lake Michigan at the outlet of Burns Ditch. A small watershed drains part of the Great Marsh, a wetland complex that has been disrupted by ditching and limited residential development, at the outlet of Derby Ditch. Monitoring at the outlet of Burns Ditch in 1999 and 2000 indicated that E. coli concentrations vary over two orders of magnitude during storms. During one storm, sewage overflows caused concentrations to increase to more than 10,000 cfu/100 mL for several hours. Monitoring at Derby Ditch from 1997 to 2000 also indicated that E. coli concentrations increase during storms with the highest concentrations generally occurring during rising streamflow. Multiple regression analysis indicated that 60% of the variability in measured outflows of E. coli from Derby Ditch (n = 88) could be accounted for by a model that utilizes continuously measured rainfall, stream discharge, soil temperature and depth to water table in the Great Marsh. A similar analysis indicated that 90% of the variability in measured E. coli concentrations at the outlet of Burns Ditch (n = 43) during storms could be accounted for by a combination of continuously measured water-quality variables including nitrate and ammonium. These models, which utilize data that can be collected on a real-time basis, could form part of an Early Warning System for predicting beach closures.