Spatial variation in tidal channel cross-sectional geometry
Project Abstract
Tidal channels provide rearing habitat for threatened salmon and other fish and wildlife, support marsh resilience to sea-level rise by distributing suspended sediment, and support other ecosystem services. Because restoration site soils are oxidized they have lost organic content and become compacted, so that channel erosion from restored tidal processes is typically inhibited by thick clay layers. Consequently, channel excavation is critical to marsh restoration, but engineers are often unsure of how deep or wide excavated channels should be. This project analyzes previously surveyed channel cross-section data for 147 tidal channels in 8 river deltas (Nooksack, Skagit, Stillaguamish, Snohomish, Nisqually, Dosewallips, Duckabush, Hamma-hamma), and relate cross-section geometry to channel length, drainage basin area, marsh elevation, tide range, storm fetch, and salinity regime to provide design guidance to engineers.
Problem Statement
Tidal channel excavation is critical to marsh restoration to provide rearing habitat for juvenile salmon, but engineers are often unsure of how deep or wide excavated channels should be. We have predictive allometric models of tidal channel network planform geometry that can be used to guide restoration design (Hood 2007, 2015, 2020), but we do not have similar guidance for channel cross-sectional geometry. Such guidance is critical, as we cannot rely on passive channel development through tidal erosion, because many restoration sites are underlain by erosion-resistant clays or agricultural hardpans (e.g., Hood 2019a, b). To maximize timely benefit to juvenile salmon, tidal channels must often be excavated rather than waiting decades for passive channel development (Hood 2018). Accurately estimating channel cross-sectional geometry is necessary, because excavation and disposal of the soils entails a considerable cost, which can be managed most effectively if there are design tools that allow cut/fill planning.
In addition to engineering design, reference standards are necessary to evaluate restoration monitoring results. If a restoration site channel is not very deep, is it because it is constrained by a clay layer or is it because it is located in a low-elevation (subsided) marsh? A better understanding of the factors controlling channel cross-section geometry will allow better evaluation and adaptive management of post-restoration conditions.
Finally, results might be used to anticipate and plan for sea-level rise impacts, because sea-level rise uncompensated by marsh accretion is analogous to lowering marsh elevation within a tidal frame, which is hypothesized to affect cross-section geometry.
Hypothesis Statement
Tidal channel cross-sectional geometry is determined by channel size (e.g., length), tidal channel drainage basin size, tide range, marsh surface elevation, storm fetch, and salinity regime. thumb|439x439px|Figure 1. Comparison of six Skagit Delta tidal channel outlet cross-section depths for channels 13-16 m wide (left panels) and approximately 6 m wide (right panels) depending on marsh surface elevation (high for top four panels, low for bottom two panels). Note: Channels in low-elevation marsh tend to be wide and shallow, while those in high-elevation marsh tend to be narrow and deep. Fluvial channel cross-sectional geometry has long been known to scale with stream discharge (hydraulic geometry) (Leopold and Maddock 1953). The same principle has been applied to tidal channels by substituting tidal prism for stream basin discharge (Myrick and Leopold 1963). However, tidal prism is often difficult to calculate for individual tidal channels, because their drainage basins are often difficult to delineate due to the extremely flat topography characteristic of tidal marshes. Furthermore, several studies have shown that tidal exchange can occur between adjacent tidal channel basins, such that there is net tidal influx into one channel and outflux from another (e.g., French and Stoddart 1992). Thus, rather than attempting to measure tidal prism directly, the proposed work will measure correlates of tidal prism, including tidal channel size (as indicted by total length), tidal channel drainage basin size (measured in GIS using the Euclidean allocation tool), tide range, and marsh surface elevation. Marsh surface elevation has a complicated relationship with tidal channel prism, because while a lower marsh elevation might suggest greater tidal prism, much of the tidal exchange over low elevation marsh surfaces occurs as sheet flow and only becomes channelized flow as the falling water surface approaches the elevation of the marsh surface (Temmerman et al. 2005). Indeed, preliminary analysis of channel cross-section data shows an interesting relationship between tidal channel cross-sectional form and marsh surface elevation, where low-elevation channels appear to be much shallower and wider thanhigh elevation channels (Fig. 1). Storm fetch has been shown to affect planform tidal channel geometry (Hood 2015), probably by affect sediment balance in the affected tidal channels, so it seems at least possible that storm fetch will also affect channel cross-sectional geometry. Finally, salinity regime may indirectly affect tidal channel cross-sectional geometry via its influence on the lower limits of vegetation distribution and on vegetation height. Vegetation affects partitioning of tidal prism between sheet flow and channelized flow (Temmerman et al. 2005), and tall vegetation has a greater effect than short vegetation on tidal prism partitioning. Vegetation tends to be taller in low-salinity marshes (typically dominated by waist-high Carex lyngbyei) than in low salinity marshes (typically dominated by ankle-high Sarcocornia pacifica and Distichlis spicata), so we should expect tidal channels to be deeper in low-salinity marshes (at a given elevation and tide range) than in high salinity marshes.
Methods
Sampling approach
Tidal channels (n = 147) have been previously surveyed in a semi-random stratified design in eight Puget Sound river delta marshes. Randomness was constrained by accessibility; channels difficult to access at low tide by boat or on foot, usually in the largest and most complicated deltas, had to be omitted from consideration. In the smallest deltas, all or nearly all tidal channels were surveyed and so consisted of a census rather than a random sample. Nevertheless, the surveyed channels appear to be representative of the available environmental variation within selected strata. Strata consisted of river deltas, or portions of river deltas, with characteristic tide range, salinities, storm fetch, and marsh surface elevation ranges. Channels have been surveyed at their outlets, midpoints, and at points ¼ and ¾ of their mainstem lengths from the outlets, to provide standardized points of comparison for channels of different lengths. Channel lengths and cross-section survey locations were determined by GIS analysis of high-resolution aerial photos (<30 cm pixel resolution) prior to field surveys.
Sampling method
Channel cross-sections were surveyed with an RTK-GPS (3-cm horizontal and vertical resolution). Channel cross-sections included, at a minimum, points at the top of each bank, at the toe of each bank, at the channel thalweg, and at three or more points spanning at least 15 m of the marsh surface from each bank. Typically, point densities within the channel cross-sections were much greater than these minimal requirements.
Statistical analysis
RTK-GPS data will be plotted in GIS and channel cross-sections will be calculated and graphed in an Excel spreadsheet. Channel widths, mean depths, maximum depths, width/depth ratios, minimum elevations, cross-sectional areas, and mean elevations of adjacent marsh surfaces will be calculated from the spreadsheet data. The likely methodological choice for statistical analysis will be to use Generalized Linear Models (GLM) in the R programming environment, following guidance in Faraway (2016), to relate cross-sectional metrics to the independent explanatory variables postulated in the hypothesis statement.
References
Faraway JJ. 2016. Extending the Linear Model with R: Generalized Linear, Mixed Effects, and Non-parametric Regression Models, 2nd edition. CRC Press, Boca Raton, FL.
French, JR and DR Stoddart. 1992. Hydrodynamics of salt marsh creek systems: Implications for marsh morphological development and material exchange. Earth Surface Processes and Landforms 17: 235– 252.
Hood WG. 2020. Applying tidal landform scaling to habitat restoration planning, design, and monitoring. Estuarine, Coastal and Shelf Science.244: https://doi.org/10.1111/j.ecss.2018.12.017.
Hood WG. 2019a. Fir Island Farms 3rd-year Post-Restoration Tidal Marsh Monitoring Report. Prepared for Washington Department of Fish and Wildlife. Skagit River System Cooperative, LaConner, WA.
Hood WG. 2019b. zis a ba 2nd-year Post-Restoration Tidal Marsh Monitoring Report. Prepared for Stillaguamish Tribe Natural Resources Department. Skagit River System Cooperative, LaConner, WA.
Hood WG. 2018. Tidal Channel Erosion Rates Depend on Marsh Restoration Site Size. Draft manuscript prepared for the Estuary and Salmon Restoration Program, Washington Department of Fish and Wildlife. Skagit River System Cooperative, LaConner, WA.
Hood WG. 2015. Geographic variation in Puget Sound tidal channel planform geometry. Geomorph. 230:98-108.
Hood WG. 2007. Scaling tidal channel geometry with marsh island area: a tool for habitat restoration, linked to channel formation process. Water Resources Research. 43, W03409, doi:10.1029/2006WR005083.
Leopold LB and T Maddock. 1953. The hydraulic geometry of stream channels and some physiographic implications. USGS Professional Paper, No. 252. U.S. Government Printing Office, Washington, DC.
Myrick RM and LB Leopold. 1963. Hydraulic geometry of a small tidal estuary. USGS Professional Paper, No. 422-B. U.S. Government Printing Office, Washington, DC.
Temmerman S, TJ Bouma, G Govers, and D Lauwaet. 2005. Flow paths of water and sediment in a tidal marsh: Relations with marsh developmental stage and tidal inundation height. Estuaries 28: 338-352.
Relationships
- broader: channel structure
- involves organization: tribal
Source: Spatial variation in tidal channel cross-sectional geometry on Salish Sea Wiki