Measuring and Modeling Flow Rates in Tidal Creeks: A Case Study from the Central Coast of South Carolina

Abstract

Listen

Translate article

The purpose of this study was to collect site- and condition-specific hydrology data to better understand the water flow dynamics of tidal creeks and terrestrial runoff from surrounding watersheds. In this paper, we developed mathematical models of tidal creek flow (discharge) in relation to time during a tidal cycle and also estimated terrestrial runoff volume from design storms to compare to tidal creek volumes. Currently, limited data are available about how discharge in tidal creeks behaves as a function of stage or the time of tide (i.e., rising or falling tide) for estuaries in the southeastern United States, so this information fills an existing knowledge gap. Ultimately, findings from this study will be used to inform managers about numeric nutrient criteria (nitrogen-N and phosphorus-P) when it is combined with biological response (e.g., phytoplankton assemblages) data from a concurrent study. We studied four tidal creek sites, two in the Ashepoo-Combahee-Edisto (ACE) Basin and two in the Charleston Harbor system. We used ArcGIS to delineate two different watersheds for each study site, to classify the surrounding land cover using the NOAA Coastal Change Analysis Program (C-CAP) data, and to analyze the soils using the NRCS Soil Survey Geographic database (SSURGO). The size of the U.S. Geological Survey’s Elevation Derivatives for National Application (EDNA) watersheds varied from 778 to 2,582 ha; smaller geographic watersheds were delineated for all sites (except Wimbee) for stormwater modeling purposes. The two sites in Charleston Harbor were within the first-order Horlbeck Creek and the second-order Bulls Creek areas. The ACE Basin sites were within the third-order Big Bay Creek and the fourth-order Wimbee Creek areas. We measured the stage and discharge in each creek with an acoustic Doppler current profiler (ADCP) unit for multiple tide conditions over a 2-year period (2015–2016) with the goal of encompassing as large of a range of tide stage and discharge data measurements as possible. The Stormwater Runoff Modeling System (SWARM) was also used to estimate the potential water entering the creeks from the land surface; this volume was very small relative to the tide water volume except for the more-developed Bulls Creek watershed. The results show that the peak discharge occurred on the ebb tide and that the duration of the flood tide spanned a longer period of time; both of these observations are consistent with traits associated with an ebb-dominated tidal creek system. The tidal inflow and outflow (flood and ebb tides, respectively) showed an asymmetrical pattern with respect to stage and discharge; peak discharge during the flood (rising) tide occurred at a higher stage than for the peak discharge during the ebb (falling) tide. This is not an unexpected result, as the water on an ebb tide is moving down gradient funneled through the creek channel toward the coast. Furthermore, water moving with the rising flood tide must overcome frictional losses due to the marsh bank and vegetation; i.e., the peak discharge can only happen when the water has risen above these impediments. We infer from the flow dynamics data that faster water velocities during ebb tide imply that more erosive energy could transport a larger mass of suspended solids and associated nutrients (e.g., orthophosphate) from the estuary to the coastal ocean. However, the discharge and runoff modeling indicate that land-based flux was important in the developed Bulls Creek watershed, but not at the larger and less-developed Big Bay Creek watershed. At Big Bay Creek, the relatively large tidal discharge volume compared to the smaller potential runoff generated within the watershed indicates that the creek could potentially dilute terrestrial runoff contaminants. Smaller, more-urbanized tidal wetland systems may not benefit from such dilution effects and thus are vulnerable to increased runoff from adjacent developed landscapes.