ERDC/TN EMRRP-EM-14

January 2025

Mediated Model Development for

Coastal Marsh Management in the
Chesapeake Bay
by Emily R. Russ, Darixa D. Hernandez-Abrams, Brook D. Herman,
Jackie Specht, Raul Osorio, Adrianna Zito-Livingston, Kevin
Thompson, Andrea Landaverde, and Todd M. Swannack

PURPOSE: The purpose of this technical note is to develop a conceptual model that describes
the critical processes, stressors, and interactions that affect coastal marsh dynamics within the
Chesapeake Bay, as identified by subject matter experts, and then link those factors to specific
management actions. Managing coastal marshes within Chesapeake Bay involves multiple
stakeholders across federal, state, local, and nongovernmental agencies. Reaching consensus
among large stakeholder groups can be difficult, since each has their own perspective and
requirements for management. Mediated modeling is a technique that facilitates consensus
building among stakeholders and provides a transparent roadmap for decision-making. This
technical note describes how mediated modeling was applied to marsh management in Chesapeake
Bay. On 4–5 May 2022, The Nature Conservancy (TNC) and the US Army Engineer Research and
Development Center (ERDC) Integrated Ecological Modeling Team (EcoMod) partnered for a
multistakeholder mediated modeling workshop to (1) build a conceptual model that depicts the
relevant processes impacting marsh dynamics, and (2) identify indicators that are necessary for
tracking marsh conditions, which inform needed management strategies. This conceptual model
provides the foundation for the development of a marsh management decision framework that will
use indicators to identify marsh conditions that subsequently trigger management decisions.

INTRODUCTION AND BACKGROUND: Coastal marshes provide valuable ecosystem and

economic services, including flood mitigation and carbon sequestration (Barbier et al. 2011).
These ecosystems are globally threatened by rapid sea-level rise (SLR). The average rate
of relative SLR in the Chesapeake Bay was 4.1 mm y−1 between 1969 and 2014; * this is higher
than the current global average rate (3.2 mm y−1) and is a result of regional subsidence (Beckett et
al. 2016). In addition, Maryland has lost 70% of its original wetland acreage due primarily to land
use change (Dahl 1990). This loss elevated the urgency to preserve the remaining wetlands
and coastal habitats. Yet, this historic loss is exacerbated by an anticipated 50% decline in tidal
high marsh habitat along the lower Eastern Shore by 2050. Losses of this scale will have a dramatic
effect on water quality and marsh species within the Chesapeake Bay. Marshes respond to
SLR in three ways: (1) accreting sediment at or greater than the rate of SLR, (2) migrating upland,
or (3) drowning (Morris et al. 2002; Molino et al. 2022; Kearney et al. 2002). Salt marshes
in the Chesapeake Bay are especially vulnerable to drowning from a combination of limited
sediment input due to microtidal setting and accelerating SLR rates (Stevenson et al. 1985;

* For a full list of the spelled-out forms of the units of measure used in this document, please refer to US
Government Publishing Office Style Manual, 31st ed. (Washington, DC: US Government Publishing Office, 2016),
248–52, https://www.govinfo.gov/content/pkg/GPO-STYLEMANUAL-2016/pdf/GPO-STYLEMANUAL-2016.pdf.

Distribution Statement A. Approved for public release: distribution is unlimited.

ERDC/TN EMRRP-EM-14
January 2025

Kearney et al. 2002). Development can also indirectly contribute to salt marsh degradation and
loss through hydrological alterations, excess nutrient runoff, and increased invasive species
(Kutcher et al. 2022).

Marsh health is driven by system-wide processes—hydrology, water quality, and sedimentation—

that are often controlled by factors external to the marsh footprint. And yet, climatically focused
marsh management is inhibited by the lack of scalable (e.g., site-level versus system-wide) data
needed to describe marsh system processes. Site-level data are important for highlighting the need
for management action and are best paired with system-wide data to guide management actions.

Additionally, the interaction of these system-wide coastal processes, and their compounding
effects, must be considered to inform appropriate management actions. For instance, thin-layer
placement of dredged material may be an option for elevating marshes to prevent drowning
(hydrologic process). However, the marsh elevation effort will have a limited timescale of habitat
enhancement if further actions are not taken to reinvigorate the marsh’s ability to accrete sediment
(sedimentation process) at rates that outpace SLR. Further, if the thin-layer placement effort is not
managed correctly, invasive species may take advantage of the recently disturbed marsh complex,
degrading the biodiversity benefits of the restoration effort. It is essential to identify the conceptual
connections between local and system observations and important data gaps to improve targeted
conservation activities.

With current and projected rates of habitat loss, proactive management of the Chesapeake Bay
marshes is necessary. Management actions must comprehensively address the stressors of marsh
habitats across scales. A strategic decision framework can help meet goals of comprehensive
management, while also helping to streamline marsh management activities. To begin developing
a strategic decision framework for Chesapeake Bay marsh management, TNC and ERDC EcoMod
led a virtual, expert-driven community modeling workshop. The purpose of this workshop was to
convene Chesapeake Bay marsh management subject matter experts (SMEs) and incorporate
marsh system processes across multiple scales (i.e., site to system) to identify (1) potential
management strategies to mitigate habitat degradation, and (2) the marsh condition indicators that
would trigger management actions.

Conceptual models help depict tidal marsh system processes. These models also highlight key
indicators to consider when evaluating marsh health. Identification of key indicators will allow
managers to proceed with the management strategy best suited to the complex needs of their marsh.
Marsh health indicators conventionally include ecological functions (vegetation and soil),
connectivity, trophic diversity, and hydrology. Other aspects to consider in the face of marsh
migration and SLR are accretion/erosion rates (and factors that influence these [e.g., sediment
availability, tidal range, biomass, and edge type]), geomorphology (e.g., elevation and slope), and
stressors (e.g., anthropogenic development) that limit the area of potential marsh migration.

This conceptual model is the first step in the marsh management decision framework, which will
enable mangers to focus on the relevant indicators of marsh degradation and use a decision tool to
select the most appropriate management strategies (Figure 1). The framework includes steps that
will collectively address the management challenges specific to a marsh complex and guides
managers through the process.

2
 ERDC/TN EMRRP-EM-14
January 2025

Figure 1. Framework for developing marsh management strategies.

The conceptual model informs the next steps in the framework, which involve evaluating the
condition of the relevant indicators and using a decision framework to propose comprehensive,
financially feasible management strategies. The subsequent steps were not within the scope of this
workshop and are not discussed further.

WORKSHOP OVERVIEW: TNC and ERDC EcoMod partnered to host and facilitate the virtual
“Holistic Marsh Management Tool Workshop” on 4–5 May 2022. Thirty SMEs from academia,
federal and state agencies, nongovernmental organizations (NGOs), and the private sector in the
Chesapeake Bay and mid-Atlantic region attended the workshop. * On 4 May, TNC opened the
workshop with a presentation on the project background and desired outcomes for the workshop.
Then, EcoMod provided an overview of the foundation of conceptual modeling and how to
develop these models for applied management.

The workshop participants were split into four breakout groups (with roughly equal representation
from academia, federal and state agencies, NGOs, and the private sector) to develop their own
conceptual models on marsh processes. The first stage of conceptual model development focuses
on developing the narrative of marsh management. Participant were asked to tell their “story” of
marsh management in Chesapeake Bay by answering the following questions:

• What aspect(s) of the marsh do you want to restore?

• What should the conceptual model seek to do?
• Who is the target audience for the conceptual model?
• What variables or processes should appear in your conceptual model?
• What are the major components of your conceptual model?

* Participants included Chesapeake Bay National Estuarine Research Reserve Maryland, Lower Shore Land
Trust, Maryland Coastal Bays Program, Maryland Department of Natural Resources, Maryland Sea Grant, National
Audubon Society, National Oceanic and Atmospheric Administration, Partnership for the Delaware Estuary,
Sustainable Science LLC, The Nature Conservancy, University of Maryland Center for Environmental Science,
University of Maryland Environmental Finance Center, University of Virginia, US Army Corps of Engineers, US
Fish and Wildlife Service, and Virginia Institute of Marine Science.

3
ERDC/TN EMRRP-EM-14
January 2025

Each group then developed their own conceptual model with the marsh processes/variables they
identified. The various conceptual models are described in more detail in the section “Summary
of Breakout Group Discussion.”

Day 2 of the workshop opened with a brief recap and review of the conceptual models developed
by SMEs from Day 1 and was followed by a second presentation by EcoMod on identifying
conditions and stressors for management. The entire group then participated in an interactive
brainstorming session to identify marsh management strategies as well as management
priorities/goals. The management strategies in this master list were grouped into four
subcategories: (1) morphological and hydrological restoration (e.g., thin layer placement, tidal
channel restoration), (2) socioecological management (e.g., flood risk management, recreational
improvements), (3) land use management (e.g., removing barriers that impede migration, ditch
plug removal), and (4) ecological restoration (e.g., vegetation planting, endangered species
management). The breakout groups then linked a subset of management strategies to measurable
marsh system processes, or indicators. Each group chose the top 2–4 indicators relevant for making
management decisions and listed conditions that would trigger management actions.

SUMMARY OF BREAKOUT GROUP DISCUSSIONS

Conceptual Model Development: Although each group developed a unique conceptual model,
there was significant overlap among the processes/variables included. Two groups developed box
and arrow diagrams that showed the links between hydrologic (e.g., tides, waves), geomorphic
(e.g., sediment supply, relative elevation), ecological processes (marsh vegetation, plant density),
and depicted stressors (SLR, storms, and development) that will likely alter marsh processes in the
future. The remaining two groups developed a more high-level conceptual model that described
the drivers of change in marshes, including many of the same variables/processes as the other
groups, and indicators of degradation (e.g., marsh erosion, fragmentation, habitat transition). One
group highlighted potential management goals that may be applied to a degraded marsh, such as
maintaining marsh spatial extent, protecting species of interest, preventing invasive species spread,
and increasing public access. Another group listed various metrics that can be used to measure
marsh change, such as elevation, inundation duration and frequency, and primary productivity.
Additionally, this group noted socioecological drivers of change (e.g., urbanization,
population/demographic change, commodity crops/deforestation), indicators of degradation (e.g.,
damage to critical infrastructure, poor water quality, decreased recreational fishing), and metrics
linked to those indicators (e.g., economic losses, reduced recreation, and fecal coliform bacteria
levels). See the Appendix for all the conceptual models.

Synthesis of Breakout Group Conceptual Models: Following the workshop, EcoMod and TNC
synthesized these individual models into a single, comprehensive, box-and-arrow conceptual
model (Figure 2), which included links between hydrodynamics, geomorphology, ecology, and
natural or anthropogenic stressors.

4
 ERDC/TN EMRRP-EM-14
January 2025

Figure 2. Final conceptual model of relevant marsh processes and indicators that will be used for
decision tool development. Each box represents a variable or process with arrows linking to other
variables/processes. The colors indicate categories of variables/processes, in which red, yellow,
blue, and green reflect stressors, geomorphology, hydrology, and ecology, respectively.

The conceptual model highlights that there are several variables and processes that contribute to
marsh health, with each variable and process having either a direct or indirect effect on multiple
other marsh properties. For example, the final conceptual model shows development is a stressor
that can directly affect (1) fluvial input by altering flows, (2) sediment availability by increasing
sediment runoff during construction, and (3) vegetation density by converting habitat. Although
not directly linked to marsh morphology (i.e., relative elevation, marsh zonation/slope), this can
lead to increased accretion from enhanced sediment input, which could increase the elevation of
the marsh surface and subsequently alter the vegetation community. This conceptual model does
not specify all possible scenarios but does attempt to clarify the connections and feedbacks
between the different marsh variables and processes, which is a necessary step for identifying the
data needed to evaluate marsh condition.

Linking Indicators to Management Strategies: Table 1 summarizes the indicators and

conditions identified by SMEs at the workshop that would trigger management actions as well as
potential management strategies.

The overlap in conditions that trigger management actions across different indicators (i.e.,
extended inundation for relative elevation, flood risk, nuisance flooding) demonstrates that these
degraded conditions are not caused by a single stressor but are likely a combination of several.
Many management strategies are also appropriate for different conditions demonstrating the need
to approach marsh management holistically rather than attempting to remediate individual issues.

5
ERDC/TN EMRRP-EM-14
January 2025

Table 1. Indicators, conditions that trigger management actions, and management strategies
determined through group breakout session.
Indicator Conditions that Trigger Management Management Strategies
Relative • Extended inundation (increased • Beneficial use of dredged material
elevation frequency and duration) • Thin-layer placement
• Deviations from expected vegetation • Nearshore placement
zonation
• Unhealthy plants
Altered • Increase in flooded area • Tidal channel restoration
hydrology • Expanding ponds and/or ditches • Improved connectivity
• Loss of edge stability and peat • Runnels for drainage
• Sediment loss • Extend creeks to reduce ponding
Water quality • Increased nutrients • Stormwater management practices
• Eutrophication • Manage point sources
• Decreasing plant/animal population
• Increased bacteria levels
• Low dissolved oxygen
• Turbidity
Flood risk • Shoreline degradation/erosion • Regulation
• Increased wave action • Sediment placement
• Increased storm surge impacts • Oyster reef creation
• Increased development at low elevations • Incentives
• Increased subsidence • Breakwaters
• Extended inundation • Living shorelines
• Loss of submerged aquatic • Dune creation
vegetation (SAV) • Flood gates
• SAV restoration
Ecological value • Degraded habitat • SAV restoration
(cultural, • Mowing/destroying wetland • Invasive species removal
recreational, and • Increased development • Policy and regulation
aesthetic) • Increased invasive species • Outreach and education
• Ghost forests
• Decreasing plant/animal population
• Shifting plant communities (fresh to
saltwater marsh)
• Decreased diversity
Sediment • Marsh drowning • Culvert sizing for tidal passage
availability • Subsidence • Soil remediation
• Erosion • Ditch plug removal
• Expanding pools • Improve hydrologic connectivity
• Changes in marsh elevation
• Vegetation loss
• Loss of marsh extent
• Loss of nesting habitat
Fragmentation • Expanding pools • Remove barriers to migration
• Lack of vegetation • Create marsh migration buffer
• Increased area of mud flats zones
• Reduction of high marsh • Conservation leases
• Increased development • Long-term protective easements
• Mosquito ditching
• Expanding phragmites

6
 ERDC/TN EMRRP-EM-14
January 2025

Table 1 (cont.). Indicators, conditions that trigger management actions, and management
strategies determined through group breakout session.
Nuisance • Damage to infrastructure • Culvert sizing for tidal passage
flooding • Erosion • Removal of infrastructure
• Loss of sediment
• Habitat transition (e.g., high to low
marsh, low to open water, forest to ghost
forest)
• Vegetation die off
• Extended inundation
• Saltwater intrusion
• Root collapse/subsidence
Stormwater • Increased sediment supply • Improve hydrologic connectivity
runoff • Increased nutrient delivery • Removal of impervious surfaces
• Harmful algal blooms
• Increased fecal coliform bacteria
• Erosion
• Increased pooling/ponding
• Extended inundation
• Increased turbidity
Native • Loss of native species • Native plantings
vegetation • Low plant density • Debris removal
• Low productivity
• Low biomass
• Loss of coverage (increased
ponds/runnels)
• Loss of root mass
• Toxic nutrient levels
Sensitive • High predation • Land protection
species: • Lack of food/foraging • Invasive species removal
threatened and • High disturbance from recreation • Endangered species management
endangered, • Lack of habitat/nesting structure • Prescribed burns
candidate, • Non-native competition • Facilitated bird nesting
commercial, and • Pollution/contaminants
recreation • Fragmentation by development
• Lack of cover
Development/ • Heavy industrial area with • Land protection
land use uncontrolled runoff • Removal of impervious surfaces
• Increased urbanization
(impervious surfaces)
• Fragmentation by development
Edge type/ • Shoreline eroding/retreating • Living shorelines
shoreline • Hard infrastructure (seawall, ajax, dikes, • Breakwaters
position roads, revetments)

CONCLUSIONS AND FUTURE WORK: Conceptual model development was the first step in
the marsh management decision framework that will be used to address future management
challenges in the Chesapeake Bay. The next steps will involve identifying data that can be used to
assess the condition of marsh health, which will then be input into a decision tool to select
appropriate management strategies.

7
ERDC/TN EMRRP-EM-14
January 2025

POINT OF CONTACT: For additional information, contact Emily Russ (828-413-7589,

[email protected]). This technical note should be cited as follows:

Russ, E. R, D. D. Hernandez-Abrams, B. D. Herman, J. Specht, R. Osorio, A. Zito-

Livingston, K. Thompson, A. Landaverde, and T. M. Swannack. 2025. Mediated
Model Development for Coastal Marsh Management in the Chesapeake Bay.
ERDC/TN EMRRP-EM-14. Vicksburg, MS: US Army Engineer Research
and Development Center.

REFERENCES
Barbier, E. B., S. D. Hacker, C. Kennedy, E. W. Koch, A. C. Stier, and B. R. Silliman. 2011. “The Value of
Estuarine and Coastal Ecosystem Services.” Ecological monographs 81 (2): 169–93. https://doi.org/10.1890
/10-1510.1.
Beckett, L. H., A. H. Baldwin, and M. S. Kearney. 2016. “Tidal Marshes across a Chesapeake Bay Subestuary Are
Not Keeping up with Sea-Level Rise.” PLoS One 11 (7): e0159753. https://doi.org/10.1371
/journal.pone.0159753.
Dahl, T. E. 1990. “Wetlands Loss Since the Revolution.” National Wetlands Newsletter 12 (6): 16–17.
Kearney, M. S., A. S. Rogers, J. R. G. Townshend, E. Rizzo, D. Stutzer, J. C. Stevenson, and K. Sundborg. 2002.
“Landsat Imagery Shows Decline of Coastal Marshes in Chesapeake and Delaware Bays.” Eos, Transactions
American Geophysical Union 83 (16): 173–78. https://doi.org/10.1029/2002EO000112.
Kutcher, T. E., K. B. Raposa, and C. T. Roman. 2022. “A Rapid Method to Assess Salt Marsh Condition and Guide
Management Decisions.” Ecological Indicators 138: 108841. https://doi.org/10.1016/j.ecolind.2022.108841.
Molino, G. D., J. A. Carr, N. K. Ganju, and M. L. Kirwan. 2022. “Variability in Marsh Migration Potential
Determined by Topographic Rather Than Anthropogenic Constraints in the Chesapeake Bay Region.”
Limnology And Oceanography Letters 7 (4): 321–31. https://doi.org/10.1002/lol2.10262.
Morris, J. T., P. V. Sundareshwar, C. T. Nietch, B. Kjerfve, and D. R. Cahoon. 2002. “Responses of Coastal
Wetlands to Rising Sea Level.” Ecology 83 (10): 2869–877. https://doi.org/10.1890/0012-9658(2002)083
[2869:ROCWTR]2.0.CO;2.
Stevenson, J. C., M. S. Kearney, and E. C. Pendleton. 1985. “Sedimentation and Erosion in a Chesapeake Bay
Brackish Marsh System.” Marine Geology 67 (3–4): 213–35. https://doi.org/10.1016/0025-3227(85)90093-3.

NOTE: The contents of this technical note are not to be used for advertising, publication, or promotional
purposes. Citation of trade names does not constitute an official endorsement or approval of the use of
such products.

8
 ERDC/TN EMRRP-EM-14
January 2025

APPENDIX: BREAKOUT GROUP CONCEPTUAL MODELS

Breakout group marsh model conceptual diagrams.

Figure A-1. Group 1.

Figure A-2. Group 2. Note: P = phosphorus, N = nitrogen, and BOD = biochemical oxygen demand.

9
ERDC/TN EMRRP-EM-14
January 2025

Figure A-3. Group 3.

Figure A-4. Group 4.

10