Time filter

Source Type

Eaton B.C.,University of British Columbia | Andrews C.A.E.,University of British Columbia | Giles T.R.,British Columbia Ministry of forests | Phillips J.C.,Skagit River System Cooperative
Geomorphology | Year: 2010

In 2003, a wildfire burned nearly two-thirds of the Fishtrap Creek watershed, including the forested floodplain. Annually repeated surveys of the stream channel - combined with bed material tracer studies in 2006 and 2007 - indicate that the channel has become unstable and that the bed material transport dynamics have changed since the fire. Using these data, the morphologic changes have been quantified and event-scale bed material transport rates have been estimated for four post-fire freshets. The immediate response of the channel (during the 2005 and 2006 freshets) was a progressive shift in channel morphology from a relatively featureless plane-bed morphology devoid of significant sediment accumulations to a riffle-pool morphology in which bars cover a significant proportion of the bed. This transformation occurred in association with moderate bed material sediment transport rates and only minor bank erosion within the study reach. Extensive bank erosion occurred during the 2007 and 2008 freshets, producing a dramatic change in channel morphology and increasing the local bed material transport rates by an order of magnitude. Post-fire monitoring of streamflow and suspended sediment concentrations (a surrogate for upstream sediment supply) indicates that post-fire increases are not detectable (presented in Eaton et al., in press), and our analysis demonstrates that major morphologic changes can occur following wildfire even when the common external drivers of stream discharge and sediment supply remain unchanged. © 2010 Elsevier B.V.

Scaling relationships in landforms are a signature of locally stable, self-organized critical states, which in tidal marshes result from the interaction of hydrodynamics, sediment dynamics, and biota. Empirical scaling relationships for tidal channel planform were developed for reference tidal marshes in four of the largest river deltas in Puget Sound to explore the potential underlying generative process of the observed patterns and to provide design guidance for restoration of estuarine rearing habitat for juvenile salmon. The length, surface area, and drainage basin area of the largest, 2nd-largest, 3rd-largest, etc., up to 15th-largest tidal channels that drain a marsh island, as well as the lengths of the largest through 5th-largest tributaries to the largest and 2nd-largest channels scaled with marsh area. Additionally, regression of the scaling relationship y-intercepts against channel rank for each delta showed that the rate of channel size decrease from one rank to the next was well fit by a power function, with R2 values approaching 1. These relationships reveal predictable structure in many aspects of tidal channel planforms and allow engineers to design channel excavation in considerable detail. A simulation model of channel formation through recursive marsh island conglomeration in river deltas reproduced the scaling behavior of the empirically observed marsh channels, thereby linking observed patterns to the underlying generative process. Previous allometric modeling has provided predictions of the number of tidal channels a marsh restoration site should have; this study provides a method to predict the size distribution of those channels so that engineers, planners, and restoration scientists can better plan, design, and monitor marsh restoration. © 2016 Elsevier B.V.

Effective tidal marsh restoration requires predictive models that can serve as planning and design tools to answer basic questions such as which, if any, plant species will colonize a proposed restoration site. To develop such a tool, a predictive model of oligohaline tidal marsh vegetation was developed from reference marshes in the Skagit River Delta (Washington, USA) and applied to a 1.1-ha restoration treatment site. Probability curves for the elevational distributions of common marsh species were generated from RTK-GPS point samples of reference tidal marshes. The probability curves were applied to a LIDAR-derived digital elevation model to generate maps predicting the occurrence probability of each species within treatment and control sites. The treatment and control sites, located within a recently restored area that had been diked but never completely drained, were covered by a mono-culture of non-native Typha angustifolia L. (narrow-leaf cattail) growing 40-60 cm lower in elevation than in the reference marsh. The T. angustifolia was mowed repeatedly in the treatment site to allow colonization by predicted native marsh species. Four years after mowing, T. angustifolia was replaced on 60 % of the treatment site by native sedges (Carex lyngbyei, Eleocharis palustris), consistent with the predictive vegetation model; the control site remained covered by T. angustifolia. The mowing experiment confirmed that pre-emptive competition from T. angustifolia was preventing vegetation recovery in the restoration site following dike removal, and implied that some vegetation species may be refractory to environmental change, such as dike removal or sea-level rise, because of differences in recruitment and adult niches. © 2013 Springer Science+Business Media Dordrecht.

Kairis P.A.,Skagit River System Cooperative | Rybczyk J.M.,Western Washington University
Ecological Modelling | Year: 2010

The dynamics that govern the elevation of a coastal wetland relative to sea level are complex, involving non-linear feedbacks among opposing processes. Changes in the balance between these processes can result in significant alterations to vegetation communities that are adapted to a specific range of water levels. Given that current sedimentation rates in Padilla Bay, Washington are likely less than historical levels and that eustatic sea level rise is accelerating, the extensive Zostera marina (eelgrass) meadows in the bay may be at risk of eventual submergence. We developed a spatially explicit relative elevation model and used it to project changes in the productivity and distribution of eelgrass in Padilla Bay over the next century. The model is mechanistic and incorporates many of the processes and feedbacks that govern coastal wetland elevation change. Accretion estimates made using 210Pb dating of sediment cores, sediment characteristics measured within cores, and eelgrass productivity and decomposition data were used to initialize and calibrate the model. Validation was performed using an elevation change rate measured with a network of surface elevation tables. Both the field data and model simulations revealed a net accretion deficit for the bay. Simulations using current rates of sea level rise indicated an overall expansion of eelgrass within Padilla Bay over the next century as it migrates from the center of the bay shoreward. © 2009 Elsevier B.V. All rights reserved.

Channel meander dynamics in fluvial systems and many tidal systems result from erosion of concave banks coupled with sediment deposition on convex bars. However, geographic information system (GIS) analysis of historical aerial photographs of the Skagit Delta marshes provides examples of an alternative meander forming process in a rapidly prograding river delta: deposition-dominated tidal channel meander formation through a developmental sequence beginning with sandbar formation at the confluence of a blind tidal channel and delta distributary, proceeding to sandbar colonization and stabilization by marsh vegetation to form a marsh island opposite the blind tidal channel outlet, followed by narrowing of the gap between the island and mainland marsh, closure of one half of the gap to join the marsh island to the mainland, and formation of an approximately right-angle blind tidal channel meander bend in the remaining half of the gap. Topographic signatures analogous to fluvial meander scroll bars accompany these planform changes. Parallel sequences of marsh ridges and swales indicate locations of historical distributary shoreline levees adjacent to filled former island/mainland gaps. Additionally, the location of marsh islands within delta distributaries is not random; islands are disproportionately associated with blind tidal channel/distributary confluences. Furthermore, blind tidal channel outlet width is positively correlated with the size of the marsh island that forms at the outlet, and the time until island fusion with mainland marsh. These observations suggest confluence hydrodynamics favor sandbar/marsh island development. The transition from confluence sandbar to tidal channel meander can take as little as 10 years, but more typically occurs over several decades. This depositional blind tidal channel meander formation process is part of a larger scale systemic depositional process of delta progradation that includes distributary elongation, gradient reduction, flow-switching, shoaling, and narrowing. © 2010 John Wiley & Sons, Ltd.

Analysis of historical aerial photos shows that Skagit Delta (Washington, USA) distributary dynamics are consistent with the Slingerland and Smith model of avulsion dynamics where the ratio of the water surface slopes of the two branches of a bifurcation predicts avulsion stability. This model was extended to predict distributary inlet (upstream) width and bankfull cross-sectional area. The water surface gradient ratio for a bifurcation pair predicted distributary width well; the lowest R2 was 0.61 for the 1937 data points, but R2 ranged from 0.83 to 0.90 for other year-specific regression lines. Gradient ratios were not constant over the historical record; from 1937 to 1972 the mainstem river channel lengthened by 1250m in the course of marsh progradation, while distributary lengthening was comparatively negligible. Consequently, the gradient advantage of the distributaries increased and their channels widened. After the mainstem river terminus stabilized from 1972 to the present, the distributaries continued to lengthen with marsh progradation, so that distributary gradient advantage steadily declined and the distributaries narrowed. While distributary cross sections were not available for the historical period, they were surveyed in 2007 near the distributary inlets. Gradient ratio was more closely related to distributary inlet bankfull cross-sectional area (R2=0.95) than to minimum distributary width for any photo year examined. Applying this form of analysis to Skagit Delta distributaries that have been dammed in the course of agricultural development suggests that their restoration to stabilize eroding marshes at their outlets and recover salmon migration pathways would be feasible without significant risk of full river avulsion. © 2010 Elsevier B.V.

Tidal marsh restoration generally involves dike breaching rather than complete dike removal to restore tidal inundation. Dike removal more completely restores hydrodynamic processes, but dike breaching is more economical. Thus, without a clear demonstration of the ecological benefits of complete or extensive dike removal, economic considerations are likely to cause restoration planners and engineers to prefer dike breaching. To provide some insight into the relative benefit of dike breaching versus dike removal, tidal channel planform geometry was compared between historical dike breach sites (>15 years old) and reference tidal marsh. Tidal channel networks were examined because they mediate many hydrodynamic, sedimentary, and ecological processes. Dike breach sites were found to have fewer tidal channel outlets than reference sites, but greater total channel surface area and length. These differences were likely the result of remnant dikes constraining tidal prism entirely to channel networks rather than allowing a portion of the prism to transit site boundaries as sheet flow. Allometric analysis of GIS-calculated tidal channel drainage basin area relative to total marsh area indicated the proportion of tidal prism comprised of sheet flow was inversely related to total marsh area, with the smallest marsh islands having no tidal channels and all of their tidal prism consisting of sheet flow. This suggests dike removal to restore sheet flow is most important for small restoration projects. However, dike removal may still be important for large restoration sites depending on issues not examined in this paper, e.g., remnant dike effects on river flood hydrodynamics. © 2014 Elsevier B.V.

Hood W.G.,Skagit River System Cooperative
Geomorphology | Year: 2015

Tidal channels are central elements of salt marsh hydrodynamics, sediment dynamics, and habitat. To develop allometric models predicting the number and size of tidal channels that could develop following salt marsh restoration, channels were digitized from aerial photographs of Puget Sound river delta marshes. Salt marsh area was the independent variable for all dependent channel planform metrics. Tidal channel allometry showed similar scaling exponents for channel planform metrics throughout Puget Sound, simplifying comparisons between locations. Y-intercepts of allometric relationships showed geographic variation, which multiple-regression indicated was associated with tidal range and storm significant wave height. Channel size and complexity were positively related to tidal range and negatively related to wave height. Four case studies, each with paired regions of similar tidal range and contrasting wave environments, further indicated wave environment affected channel geometry. Wave-mediated sediment delivery may be the mechanism involved, with wave-sheltered areas experiencing relative sediment deficits, such that some salt marshes in Puget Sound are already suffering sea-level rise impacts that are reflected in their channel network geometry. © 2014 Elsevier B.V.

Hood W.G.,Skagit River System Cooperative
Wetlands | Year: 2012

Beaver (Castor spp.) are considered a riverine or lacustrine animal, but surveys of tidal channels in the Skagit Delta (Washington, USA) found beaver dams and lodges in the tidal shrub zone at densities equal or greater than in non-tidal rivers. Dams were typically flooded by a meter or more during high tide, but at low tide they impounded water, allowing beaver to swim freely while quadrupling pool habitat for fish compared to channels without dams. Seven fish species were caught in low-tide pools, including threatened juvenile Chinook salmon (Oncorhynchus tshawytscha), whose densities (by volume) averaged 3.2 times higher in low-tide pools than shallows. Accounting for the total contribution of pools and shallows to juvenile Chinook abundance, beaver pools tripled shrub zone channel capacity for juvenile Chinook salmon at low tide relative to herbaceous zone marsh without beaver pools. Current Chinook recovery efforts focus on restoring herbaceous zone tidal marsh for rearing juveniles, but this focus overlooks presently rare and poorly understood habitat, like tidal shrub marsh, that was historically common and likely important to beaver and small estuarine or anadromous fish. © Society of Wetland Scientists 2012.

Hood W.G.,Skagit River System Cooperative
Ecological Engineering | Year: 2015

Tidal channels are structurally and functionally prominent features in tidal marshes, so their restoration is central to tidal marsh habitat restoration. Consequently, an important question in tidal marsh restoration is how many tidal channels can a restoration site support, and thus, how many dike breaches should be made to restore tidal inundation and tidal channels, if the dike is not to be removed entirely. Allometric analysis of reference tidal marshes in Puget Sound river deltas and the lower Columbia River Estuary showed that channel outlet count scales with marsh area, and that completed and proposed tidal marsh restoration projects had 5-fold fewer channel outlets than reference marshes. This deficiency likely impacts fish access to the restoration sites. After addressing the question of tidal channel outlet count, or dike breach count, the next design questions are how should dike breaches be oriented in tidal marsh islands and how should they be spaced. GIS and statistical analysis of reference marsh islands indicated that outlets of the two largest tidal channels draining a marsh island are typically oriented downstream, in parallel with the nearest river channel. However, the outlets of smaller tidal channels are oriented randomly. Tidal channel outlet spacing is generally independent of site size and constant within a river delta. Geometric mean spacing ranged from 122 m in the Snohomish Delta to 280 m in the North Fork Skagit Delta. These results provide important guidance to improve tidal marsh restoration design, and illustrate a useful approach to restoration design evaluation. © 2015 Elsevier B.V.

Loading Skagit River System Cooperative collaborators
Loading Skagit River System Cooperative collaborators