Elsevier

Ecological Modelling

Volume 410, 15 October 2019, 108722
Ecological Modelling

Development and implementation of an empirical habitat change model and decision support tool for estuarine ecosystems

https://doi.org/10.1016/j.ecolmodel.2019.108722Get rights and content

Abstract

Widespread land use change in coastal ecosystems has led to a decline in the amount of habitat available for fish and wildlife, lower production of ecosystem goods and services, and loss of recreational and aesthetic value. This has prompted global efforts to restore the natural hydrologic regimes of developed shorelines, especially resource-rich estuaries, but the resilience of these restored ecosystems in the face of accelerated sea-level rise (SLR) remains uncertain. We implemented a Monitoring-based Simulation of Accretion in Coastal Estuaries (MOSAICS) in R statistical software to address uncertainty in the resilience of modified estuarine habitats, using the Nisqually River Delta in the Pacific Northwest USA as a case study. MOSAICS is a spatially explicit model with a numerical foundation that uses empirical monitoring datasets to forecast habitat change in response to rising tidal levels. Because it accounts for the crucial ecomorphodynamic feedbacks between tidal inundation, vegetative growth, and sediment accretion, MOSAICS can be used to determine whether alternative management scenarios, such as enhanced sediment inputs, will bolster estuarine resilience to SLR. Under moderate SLR (0.62 m), the model predicted that a two-fold increase in mean daily suspended sediment during the rainy season was sufficient to maintain Nisqually’s emergent marshes through 2100, but under high SLR (1.35 m) MOSAICS indicated that greater sediment additions would be necessary to prevent submergence. A comparison between a restored marsh with subsided and high-elevation areas and a relict marsh demonstrated that the subsided restoration area was highly susceptible to SLR. Findings from the MOSAICS model highlight the importance of a site’s initial elevation, capacity for producing above and belowground biomass, and suspended sediment availability when considering management actions in estuaries and other coastal ecosystems.

Introduction

Estuaries and their associated tidal marshes connect freshwater riverine ecosystems to the open ocean, thereby playing a critical role in supporting marine ecosystem function (McLusky and Elliot, 2004; Beaumont et al., 2007). They supply valuable economic, cultural, and ecological goods and services, including habitat for fish, waterbirds, and mammals (Simenstad and Cordell, 2000; Greenberg et al., 2006; Barbier et al., 2011). Consequently, their preservation is a priority for coastal management. Over the coming century, global sea-level rise (SLR) is expected to diminish the extent of coastal wetlands by as much as 90%, converting productive tidal marshes into subtidal habitat (Nicholls et al., 1999; Galbraith et al., 2002; Craft et al., 2009; Crosby et al., 2016). Shoreline armoring, levees, and dams further compound the threat of SLR by removing many estuaries from tidal influence, preventing them from migrating inland, and reducing sediment accretion (Kirwan and Murray, 2008; Jackson, 2010). Major efforts are underway to eliminate unnecessary barriers to tidal exchange, but restored estuaries are often subject to degradation, which can reduce their resilience to SLR and other natural hazards (Callaway et al., 2007; Vandenbruwaene et al., 2011).

Historically, estuarine tidal marshes have been able to counter the effects of SLR via an ecomorphodynamic feedback cycle whereby tidal inundation, vegetative growth, and sediment accumulation allow the marsh plain to keep pace with gradually increasing tidal levels (Kirwan et al., 2010; French, 2006; Fitzgerald et al., 2008). Specifically, marsh vegetation traps suspended sediment and slows the flow of water across the marsh surface, thereby facilitating sediment deposition and accretion (D’Alpaos et al., 2007, 2011). Habitat degradation due to disturbance, development, or management actions may inhibit this feedback cycle by facilitating erosion, hindering sediment delivery, or limiting above and belowground productivity (Syvitski et al., 2009; Vandenbruwaene et al., 2011; Burdick and Roman, 2012). Practitioners can augment the capacity of at-risk tidal marshes to respond to disturbance (i.e., their resilience) through the physical modification of sedimentation and subsidence rates, but there are often ecological, physical, or demographic limitations to the continued stability of estuarine habitat (Kirwan and Megonigal, 2013). Given these limitations, practitioners may benefit from a decision support tool that uses empirical data to identify vulnerable areas in a restored or degraded estuary, predicts changes to the estuarine habitat mosaic, and simulates the effects of alternative management scenarios given projected estimates of SLR.

Marsh accretion models can be used as decision support tools to help managers determine a target estuary’s long-term resilience. There is considerable breadth in the type and utility of marsh accretion models (Rybczyk and Callaway, 2009; Mcleod et al., 2010), ranging from generalized numerical equations and simplified inundation simulations (Titus and Richman, 2001; Kirwan et al., 2010), to spatially explicit, process-based models (Schile et al., 2014; Morris et al., 2016; Thorne et al., 2018). At present, tradeoffs between accuracy and simplicity have opened a niche for habitat change models that combine a transparent numerical foundation with localized, empirical data such as that commonly collected in wetland monitoring programs. A similar approach has already been used to predict water temperature (Morrill et al., 2005; Sharma et al., 2008), phytoplankton productivity (Joint and Groom, 2000; Barnes et al., 2011), marsh vegetation (Sanderson et al., 2001), tidal channel formation (Hibma et al., 2004), and organic and inorganic sediment accretion (Swanson et al., 2014; Morris et al., 2016) in a variety of freshwater and marine ecosystems. Thus, there is reason to believe that the combined use of existing monitoring datasets with numerical models may also be suitable for predicting estuarine habitat response to SLR.

This study evaluates the effects of rising tidal levels and modified sediment management strategies on estuarine habitat by developing a Monitoring-based Simulation of Accretion in Coastal Estuaries (MOSAICS). We calibrated and tested MOSAICS in R statistical software (R Core Development Team, 2017) using seven years of post-restoration monitoring data from the Nisqually River Delta (NRD) in southern Puget Sound, Washington, USA (Ellings et al, 2016; Woo et al., 2018). The NRD is a macro-tidal system comprised of a diverse habitat mosaic of upland tidal forests, freshwater, brackish, and saline emergent marshes, and intertidal mudflats. These attributes make it an ideal case study to evaluate habitat resilience to SLR and resultant changes in habitat distribution among historically managed and unaltered (relict) areas. Restoration actions conducted between 1996 and 2009 were led by Billy Frank Jr. Nisqually National Wildlife Refuge (hereafter “Refuge”) and the Nisqually Indian Tribe (hereafter “Tribe”). The restored marshes have provided ample foraging and roosting sites for economically-valuable waterbirds and crucial nursery habitat for juvenile Chinook salmon (Oncorhynchus tshawytscha; Ellings et al., 2016; Davis et al., 2018; Woo et al., 2018) listed under the Endangered Species Act. As such, state, federal, and tribal managers have a vested interest in maintaining the current extent of wetland habitat over the coming century.

Here we describe our steps in constructing, validating, and evaluating MOSAICS, which we used to evaluate long-term habitat trajectories in a subsided restoration area, a higher elevation restoration area, and a relict tidal marsh. Our overarching objectives for this study were: 1) develop an empirical model to project long-term trajectories of elevation change and habitat distribution; 2) manipulate tidal level and sediment input parameters to determine how management scenarios may influence the estuarine habitat mosaic under impending SLR; and 3) compare estimated changes in the habitat mosaic for restored and relict sites. We anticipated that subsided restoration areas would exhibit less resilience to rising sea levels than the relict marsh due to lower initial elevations and a lack of vegetative biomass (Morris et al., 2002). Our approach demonstrates how monitoring data can play a crucial role in modeling efforts and can be used as part of a decision support tool to gauge the resilience of restored estuaries in the face of climate change.

Section snippets

Study site

The Nisqually River originates from glacial discharge at the southwestern base of Mount Rainier, Washington, USA. From there the river meanders 130 km northward, where it empties into Puget Sound between the cities of Olympia and Tacoma (47.08 °N 122.70 °W), forming a large river delta with an areal extent of approximately 20 km2. Between 1904 and 2009, roughly 5 km2 of the delta’s marshes were diked or leveed for agriculture, resulting in habitat loss for a variety of fish and wildlife

Model structure and validation

We successfully parameterized each hierarchical component of MOSAICS (I, S, V, ΔE) using the empirical datasets (Table 1, Table 2). Model RMSE was less than 16% of the range of observed values for each component (Table 3). For ΔE, the parameters rb in the sediment capture equation (Ecap) and va in the organic matter deposition equation (Eorg) converged on zero, so they were omitted from the final model. This meant that Eorg essentially had no effect on elevation change in the NRD (contrary to

Discussion

Our study effectively demonstrates how empirical monitoring data can be used as part of a decision support tool to forecast differences in the long-term resilience of restored and relict estuaries when confronted with SLR. We evaluated the relative effects of alternative sediment management scenarios on the persistence and composition of estuarine habitats using a hierarchical numerical model (MOSAICS). Model predictions indicated that post-restoration marsh accretion in the NRD is occurring at

Conclusion

Empirically-based decision support tools that allow for multiple climate change and management scenarios are important for restoration planning, especially given the threat of global SLR. Managers rarely have the opportunity to manipulate suspended sediment inputs on a whole-estuary scale (Thom, 2000; Teal and Weishar, 2005; Callaway et al., 2007); however, adaptive reservoir management options to increase sediment delivery to sediment-deficient river deltas may help ameliorate the effects of

Uncited-references

Rogers et al. (2012).

Funding

This research was funded by EPA Tribal Assistant Grant no. PA-00J15001, the Estuary and Salmon Restoration Program (Project #13-1583P), USGS/USFWS Science Support Program, National Fish and Wildlife Foundation, USGS Biologic Carbon Sequestration Program, USGS Ecosystem Mission Area, USFWS Coastal Program, and WERC program funds. Author M. Davis was partially supported by an American Dissertation Fellowship through the American Association of University Women.

Data accessibility

Water logger data are available at: https://doi.org/10.5066/F7SJ1HNC (Thorne, 2015). Vegetation biomass data can be found at https://doi.org/10.5066/F77943K8 (Byrd et al., 2017). R code and R Markdown documents outlining the model building and calibration process will be available on https://github.com/USGS-R.

Declaration of Competing Interest

None.

Acknowledgments

This research was made possible through cooperation between the U.S. Geological Survey Western Ecological Research Center (USGS WERC), Nisqually Indian Tribe, Billy Frank Jr. Nisqually National Wildlife Refuge, and the Nisqually River Foundation. Numerous USGS employees contributed to more than seven years of data collection, including J.Y. Takekawa, L. Shakeri, S. Blakely, A. Munguia, K. Turner, L. Belleveau, P. Markos, S. Kaviar, M. Holt, H. Minella, H. Allgood, A. Goodman, C. Freeman, G.

References (99)

  • M.L. Kirwan et al.

    Ecological and morphological response of brackish tidal marshland to the next century of sea level rise: Westham Island, British Columbia

    Global Planet. Change

    (2008)
  • E. Mcleod et al.

    Sea-level rise impact models and environmental conservation: a review of models and their applications

    Ocean Coast. Manage.

    (2010)
  • R.J. Nicholls et al.

    Increasing flood risk and wetland losses due to global sea-level rise: regional and global analyses

    Global Environ. Change

    (1999)
  • K. Rogers et al.

    Modelling wetland surface elevation dynamics and its application to forecasting the effects of sea-level rise on estuarine wetlands

    Ecol. Modell.

    (2012)
  • E.W. Sanderson et al.

    A simple empirical model of salt marsh plant spatial distributions with respect to a tidal channel network

    Ecol. Modell.

    (2001)
  • S. Silvestri et al.

    Tidal regime, salinity and salt marsh plant zonation

    Estuarine Coast. Shelf Sci.

    (2005)
  • C.A. Simenstad et al.

    Ecological assessment criteria for restoring anadromous salmonid habitat in Pacific Northwest estuaries

    Ecol. Eng.

    (2000)
  • C.A. Simenstad et al.

    When is restoration not?: incorporating landscape-scale processes to restore self-sustaining ecosystems in coastal wetland restoration

    Ecol. Eng.

    (2006)
  • R.P. Stumpf

    The process of sedimentation on the surface of a salt marsh

    Estuarine Coast. Shelf Sci.

    (1983)
  • J.M. Teal et al.

    Ecological engineering, adaptive management, and restoration management in Delaware Bay salt marsh restoration

    Ecol. Eng.

    (2005)
  • S. Temmerman et al.

    Modelling long-term tidal marsh growth under changing tidal conditions and suspended sediment concentrations, Scheldt estuary, Belgium

    Mar. Geol.

    (2003)
  • R.M. Thom

    Adaptive management of coastal ecosystem restoration projects

    Ecol. Eng.

    (2000)
  • W. Vandenbruwaene et al.

    Sedimentation and response to sea-level rise of a restored marsh with reduced tidal exchange: comparison with a natural tidal marsh

    Geomorphology

    (2011)
  • D. Verdonck

    Contemporary vertical crustal deformation in Cascadia

    Tectonophysics

    (2006)
  • J.A. Warrick et al.

    Large-scale dam removal on the Elwha River, Washington, USA: source-to-sink sediment budget and analysis

    Geomorphology

    (2015)
  • L. Ballanti et al.

    Remote sensing for wetland mapping and historical change detection at the Nisqually River Delta

    Sustainability

    (2017)
  • E.B. Barbier et al.

    The value of estuarine and coastal ecosystem services

    Ecol. Monogr.

    (2011)
  • C. Barnes et al.

    Predicting marine phytoplankton community size structure from empirical relationships with remotely sensed variables

    J. Plankton Res.

    (2011)
  • L.J. Belleveau et al.

    Vegetation community response to tidal marsh restoration of a large river estuary

    Northwest Sci.

    (2015)
  • R. Bivand et al.

    Package ‘rgdal’: Bindings for the ‘Geospatial’ Data Abstraction Library. R Package Version 1.4-4

    (2019)
  • B. Bolker

    Package ‘bbmle’: Tools for General Maximum Likelihood Estimation. R Package Version 1.0.20

    (2017)
  • P.D. Bromirski et al.

    Dynamical suppression of sea level rise along the Pacific coast of North America: indications for imminent acceleration

    J. Geophys. Res.

    (2011)
  • D.M. Burdick et al.

    Salt marsh responses to tidal restriction and restoration

  • K.P. Burnham et al.

    Model Selection and Multi-Model Inference: A Practical Information-Theoretic Approach

    (2002)
  • K.B. Byrd et al.

    Biomass/Remote Sensing Dataset: 30m Resolution Tidal Marsh Biomass Samples and Remote Sensing Data for Six Regions in the Conterminous United States

    (2017)
  • D.R. Cahoon et al.

    High-precision measurements of wetland sediment elevation: II. The rod surface elevation table

    J. Sediment. Res.

    (2002)
  • J.C. Callaway et al.

    Emerging issues for the restoration of tidal marsh ecosystems in the context of predicted climate change

    Madroño

    (2007)
  • C. Craft et al.

    Forecasting the effects of accelerated sea-level rise on tidal marsh ecosystem services

    Front. Ecol. Environ.

    (2009)
  • C.A. Curran et al.

    Suspended Sediment Delivery to Puget Sound from the Lower Nisqually River, Western Washington, July 2010–November 2011. Scientific Investigations Report 2016-5062

    (2016)
  • J.A. Czuba et al.

    Sediment Load from Major Rivers into Puget Sound and its Adjacent Waters. U.S. Geological Survey Fact Sheet 2011-3083

    (2011)
  • J.A. Czuba et al.

    Geomorphic Analysis of the River Response to Sedimentation Downstream of Mount Rainier, Washington. U.S. Geological Survey Open File Report 2012-1242

    (2012)
  • J.A. Czuba et al.

    Changes in Sediment Volume in Alder Lake, Nisqually River Basin, Washington, 1945-2011. U.S. Geological Survey Open File Report 2012-1068

    (2012)
  • A. D’Alpaos et al.

    Landscape evolution in tidal embayments: modeling the interplay of erosion, sedimentation, and vegetation dynamics

    J. Geophys. Res.

    (2007)
  • A. D’Alpaos et al.

    Dynamic response of marshes to perturbations in suspended sediment concentrations and rates of relative sea level rise

    J. Geophys. Res.

    (2011)
  • A.E. Davenport et al.

    Comparing automated classification and digitization approaches to detect change in eelgrass bed extent during restoration of a large river delta

    Northwest Sci.

    (2017)
  • M.J. Davis et al.

    Gauging resource exploitation by juvenile Chinook salmon (Oncorhynchus tshawytscha) in restoring estuarine habitat

    Restorat. Ecol.

    (2018)
  • R.M. DeConto et al.

    Contribution of Antarctica to past and future sea-level rise

    Nature

    (2016)
  • R.D. DeLaune et al.

    Some influences of sediment addition to a deteriorating salt marsh in the Mississippi River deltaic plain: a pilot study

    J. Coast. Res.

    (1990)
  • J.Z. Drexler et al.

    Carbon accumulation and vertical accretion in a restored versus historic salt marsh in southern Puget Sound, Washington, United States

    Restorat. Ecol.

    (2019)
  • Cited by (6)

    View full text