Southeastern Section - 63rd Annual Meeting (10–11 April 2014)

Paper No. 7
Presentation Time: 8:00 AM-12:00 PM

BUILDING RATING CURVES FOR LOW-ORDER STREAMS DRAINING SMALL WATERSHEDS OF THE ATLANTIC COASTAL PLAIN


PICKUS, Benjamin A., HERMAN, Janet S. and MILLS, Aaron L., Dept. of Environmental Sciences, University of Virginia, PO Box 400123, Charlottesville, VA 22904-4123, bap4db@virginia.edu

Small watersheds along the Virginia Eastern Shore extend from the drainage divide in the center of the Delmarva Peninsula eastward toward coastal lagoons and, eventually, the Atlantic Ocean. Instantaneous discharge at numerous streams was measured with hand-held flow meters occasionally over 5 years under varying flow conditions. These measurements were related to the stream stage, or water-surface elevation, measured at the same time for 4 of the streams to create rating curves. This study investigated the rating curves for inaccuracies in the relationship between discharge and stage, identifying the conditions giving greatest error and relating the errors to stream-channel geometry and to physical aspects of infrastructure installation. Continuous measurement (10-min intervals) of water-surface elevation was accomplished using submerged pressure transducers in stilling wells installed on the streambank downstream of a culvert that we assumed acted as a control structure ensuring a consistent channel cross section. Discharge measurements were rarely made during high flow, and rating curves had to be extrapolated to bracket the transducer record. The resulting continuous record of water-surface elevation was converted to discharge using the rating curves, and discharge was summed to give annual discharge for each stream. Aligning the precipitation record revealed that storm-event stream discharge was estimated to be unreasonably large in some of the small, flashy streams relative to the large streams. Discharge calculations considering the purported stage values and using the culvert and stream-channel geometry indicate the predicted discharge from the rating curve may be impossibly large, and the greatest error in annual discharge comes from a few high-stage events. We evaluated the quantitative effect on annual discharge of either truncating storm peaks at the highest stage measured by hand or using the calculated condition of pipe-full flow for the given culvert dimensions as an upper limit on stream discharge.