Spatial and temporal variation in the hydrodynamic landscape in intertidal salt marsh systems

Abstract

Lay Abstract The spatial and temporal variability of water flow at the scale of ecological processes in aquatic systems such as estuaries is not well understood. Current sampling schemes used to assess the impact of water flow on ecological processes are limited in their ability to encompass adequate spatial and temporal variability, confound spatial and temporal patterning, and do not provide clear connections to values of turbulent flow parameters. We examined spatial and temporal variation of water flow within and among four sites in Wassaw Sound, Georgia, USA, corresponding to different tidal types (i.e., neap, mean, and spring tides). Variation in water flow within the sites was similar and highly correlated at locations up to 20 m but was highly variable among distant sites. Some flow parameters also showed strong dependence on the site. Furthermore, flow parameters did not correlate with each other, indicating that specific parameters need to be identified to correctly assess the effect of water flow on ecological interactions. We provide several suggestions to improve current sampling schema, which include taking flow measurements concurrently with ecological experimentation, minimizing sequential sampling designs, and sampling water flow at similar spatial and temporal scales as ecological experimentation. Consideration of these issues should provide a better way in which to assess the interaction between water flow and ecological processes and help generate testable hypotheses that incorporate how these interactions may vary spatially and temporally.

Stratified turbulence in the nearshore coastal ocean: Dynamics and evolution in the presence of internal bores

“…Thus, to separate waves and turbulence, we employ the spectral “phase” decomposition method of Bricker and Monismith []. This method has been successfully used in previous field studies in a variety of coastal and estuarine locations [e.g., Hansen and Reidenbach , ; Wilson et al ., ; MacVean and Lacy , ]. Briefly, this method utilizes the phase lag between the velocity components (e.g., u and w ) of surface waves to interpolate the magnitude of the turbulence under the wave peak.…”

Assessment of single‐instrument techniques for removing wave bias from Reynolds stress estimates

“…A variety of techniques have been proposed to reduce the wave bias terms. This study presents a new single‐instrument method and compares it to two previously proposed single‐instrument techniques (Bricker and Monismith ‐ BM07; Wilson et al ‐ WWW13) as well as the most recent of the two‐instrument velocity differencing methods (Feddersen and Williams ‐ FW07). The background and details of implementation for these three methods are described in Supporting Information Appendix A.…”

“…While these two‐instrument techniques have proven effective in the offshore (Shaw and Trowbridge ) and nearshore (Feddersen and Williams ) regions, it is desirable from a financial and logistical standpoint to use a single‐instrument technique. Single‐instrument techniques have been proposed that rely on frequency domain analysis to separate the wave signal from the turbulence (e.g., Bricker and Monismith ; Wilson et al ), but suffer in effectiveness if the wave motions occupy the same frequency range as the turbulent fluctuations contributing to the Reynolds stress (Trowbridge ). Unfortunately, this is a common occurrence in the field.…”

“…For synthetic test data, a two degree instrument coordinate axis tilt and the presence of a wave field with significant wave height can result in an error in the estimated value (assuming the waves are propagating in the ‐direction). Since the observed Reynolds stresses near the bed in coastal and estuarine environments are typically on the order of (Shaw and Trowbridge ; Bricker and Monismith ; Feddersen and Williams ; Wilson et al ), this represents a significant error. In sheltered estuarine channels, the significant wave height may be on the order of centimeters, but must nonetheless be accounted for.…”

“…The proposed method also has broad similarities to the Benilov and Filyushkin () method, at least for the Reynolds normal stresses, that employs the coherence function between the water surface level and the velocity signal. The proposed method is compared to two frequency‐based single‐instrument techniques (Bricker and Monismith ; Wilson et al ), as well as the Feddersen and Williams () two‐instrument extraction technique for a series of synthetic data sets, 17 deployments at four sites in comparatively shallow estuarine channels near Wassaw Sound, Georgia, and 2 d of continuous monitoring at a single site in deeper water of Monterey Bay, California.…”