natural model that a restoration project is designed to
achieve. They serve three primary functions: (1) they can
be used as models for developing restoration actions;
(2) they provide a target from which performance goals
can be derived and against which progress toward these
goals can be compared; and (3) they provide
a comparison system by which environmental fluctuations
unrelated to the restoration action can be assessed. Alternatively, degraded control sites can be used to show the
progress of the restored system away from the degraded
condition (NRC, 1992).
Horner and Radaeke (1989) identified the following
features that should be assessed for degree of similarity
between a reference site and the potential conditions at
a restoration site:
• Functional similarity
• Climatological and hydrologic similarity
• Similarity in influences of human access, habitation,
and economic activities and in the quantity and quality
of water runoff from these activities to the wetland
• Similarity in the history of and potential for such
activities as grazing, mowing, and burning
• Similarity in size, morphology, water depth, wetland
zones and their proportions, and general vegetation
types
• Similarity in soils and nonsoil substrates
• Similarity in access by fish and wildlife
A coast-wide reference monitoring system implemented
to evaluate wetland restoration trajectories in Louisiana
addresses the problem of identifying paired reference and
restoration areas by providing an array of reference sites
(Steyer et al., 2003; Cretini et al., 2012).
Enhancing predictive capability
Bradshaw (1987) proposed that the ability to predictably
and successfully restore an ecosystem is the ultimate
test – the “acid test” – of ecological understanding. The
theory and practice of restoration is currently constrained
by a lack of robust predictive capability (e.g., Cairns,
1995; NRC, 1992). Large-scale estuarine habitat restoration projects often require large financial investments
and have low assurances of successfully meeting their
goals. Research to increase predictive capability is occurring and has improved the predictions related to some
selected habitat types (e.g., restoration of Spartina
alterniflora marshes). The field of restoration ecology
uses ecological experiments designed to predict the effect
of a particular restoration effort (Jordan et al., 1987). This
synthetic approach to restoration has found that the most
powerful way of studying a habitat is to attempt to restore,
repair, and adjust it so that it works properly. The very act
of restoration provides new insight into the functioning of
ecosystems. In turn, greater understanding of how ecosystems function enhances our ability to predictably restore
them. The ability to predict what will be achieved by
restoration, and when, requires long-term (10 years or
longer) comparisons of restored and natural ecosystems.
Such long-term research goes beyond systematic
sampling by following promising leads and discarding
hypotheses that do not stand up to testing (NRC, 1992).
Most projects attempt to set up initial conditions at
a site that will likely lead to natural development of the site
into a desirable state. Thus, a restoration action is the catalyst that sets natural succession in motion. Recognizing
the dynamic nature of estuaries, the goal of estuarine
restoration is to assist the “self-healing” capacity of the
estuary, rather than to achieve some endpoint absolutely.
Landscape considerations in estuarine habitat
restoration
To restore estuarine habitats, the ecosystem needs to be
restored (NRC, 1992). Thus, for a habitat to be restored,
the processes that contribute to forming and maintaining
the habitat have to be active and intact in the landscape
within which the restoration site occurs. For example,
sediment from the watershed must be delivered to
a newly connected former wetland site that has subsided
in elevation while behind a levee.
Using the above example for seagrasses, to enhance
seagrass distribution and abundance, it is necessary to
reduce the input of inorganic nutrients from the watershed
and from wastewater outfalls. Hence, strategies in the
Chesapeake Bay have involved nutrient discharge abatement through enhanced treatment of wastewater and creation and protection of wetlands in the watershed (http://
www.chesapeakebay.net/). These wetlands removed
inorganic nitrogen that contributed to eutrophication of
the estuary, which, through increased turbidity, caused
loss of seagrass. Further, dense reefs of oysters that had
been substantially reduced by overharvesting once were
responsible for reducing turbidity in the Chesapeake Bay
through their filter feeding. Currently, oyster reef restoration is part of the suite of actions taken to reduce turbidity
and enhance seagrass recovery.
A principal landscape ecology concept is that most
elements (habitats) within a landscape (e.g., watershed)
function best when integrated with all other elements of
the landscape. For example, habitat patch size, shape,
location, and accessibility are critical to the formation of
structure and realization of ecological functions. Humancaused fragmentation of natural landscapes into fewer
and fewer smaller pieces leads to habitat shrinkage and
less exchange of materials and species among habitats.
Increasing fragmentation and decreasing habitat size can
result in local extinction of some populations.
A landscape is a heterogeneous matrix of smaller
habitats, and the arrangement, size, productivity, and resilience of these habitats within the matrix affect the flow of
energy, animals, and materials through the landscape. Loss
or degradation of one or more elements may lead to dysfunctional performance of the remaining elements. In
deciding on restoration strategies and sites, it is useful to
identify and consider the dysfunctional or absent elements.
278
ESTUARINE HABITAT RESTORATION
achieve. They serve three primary functions: (1) they can
be used as models for developing restoration actions;
(2) they provide a target from which performance goals
can be derived and against which progress toward these
goals can be compared; and (3) they provide
a comparison system by which environmental fluctuations
unrelated to the restoration action can be assessed. Alternatively, degraded control sites can be used to show the
progress of the restored system away from the degraded
condition (NRC, 1992).
Horner and Radaeke (1989) identified the following
features that should be assessed for degree of similarity
between a reference site and the potential conditions at
a restoration site:
• Functional similarity
• Climatological and hydrologic similarity
• Similarity in influences of human access, habitation,
and economic activities and in the quantity and quality
of water runoff from these activities to the wetland
• Similarity in the history of and potential for such
activities as grazing, mowing, and burning
• Similarity in size, morphology, water depth, wetland
zones and their proportions, and general vegetation
types
• Similarity in soils and nonsoil substrates
• Similarity in access by fish and wildlife
A coast-wide reference monitoring system implemented
to evaluate wetland restoration trajectories in Louisiana
addresses the problem of identifying paired reference and
restoration areas by providing an array of reference sites
(Steyer et al., 2003; Cretini et al., 2012).
Enhancing predictive capability
Bradshaw (1987) proposed that the ability to predictably
and successfully restore an ecosystem is the ultimate
test – the “acid test” – of ecological understanding. The
theory and practice of restoration is currently constrained
by a lack of robust predictive capability (e.g., Cairns,
1995; NRC, 1992). Large-scale estuarine habitat restoration projects often require large financial investments
and have low assurances of successfully meeting their
goals. Research to increase predictive capability is occurring and has improved the predictions related to some
selected habitat types (e.g., restoration of Spartina
alterniflora marshes). The field of restoration ecology
uses ecological experiments designed to predict the effect
of a particular restoration effort (Jordan et al., 1987). This
synthetic approach to restoration has found that the most
powerful way of studying a habitat is to attempt to restore,
repair, and adjust it so that it works properly. The very act
of restoration provides new insight into the functioning of
ecosystems. In turn, greater understanding of how ecosystems function enhances our ability to predictably restore
them. The ability to predict what will be achieved by
restoration, and when, requires long-term (10 years or
longer) comparisons of restored and natural ecosystems.
Such long-term research goes beyond systematic
sampling by following promising leads and discarding
hypotheses that do not stand up to testing (NRC, 1992).
Most projects attempt to set up initial conditions at
a site that will likely lead to natural development of the site
into a desirable state. Thus, a restoration action is the catalyst that sets natural succession in motion. Recognizing
the dynamic nature of estuaries, the goal of estuarine
restoration is to assist the “self-healing” capacity of the
estuary, rather than to achieve some endpoint absolutely.
Landscape considerations in estuarine habitat
restoration
To restore estuarine habitats, the ecosystem needs to be
restored (NRC, 1992). Thus, for a habitat to be restored,
the processes that contribute to forming and maintaining
the habitat have to be active and intact in the landscape
within which the restoration site occurs. For example,
sediment from the watershed must be delivered to
a newly connected former wetland site that has subsided
in elevation while behind a levee.
Using the above example for seagrasses, to enhance
seagrass distribution and abundance, it is necessary to
reduce the input of inorganic nutrients from the watershed
and from wastewater outfalls. Hence, strategies in the
Chesapeake Bay have involved nutrient discharge abatement through enhanced treatment of wastewater and creation and protection of wetlands in the watershed (http://
www.chesapeakebay.net/). These wetlands removed
inorganic nitrogen that contributed to eutrophication of
the estuary, which, through increased turbidity, caused
loss of seagrass. Further, dense reefs of oysters that had
been substantially reduced by overharvesting once were
responsible for reducing turbidity in the Chesapeake Bay
through their filter feeding. Currently, oyster reef restoration is part of the suite of actions taken to reduce turbidity
and enhance seagrass recovery.
A principal landscape ecology concept is that most
elements (habitats) within a landscape (e.g., watershed)
function best when integrated with all other elements of
the landscape. For example, habitat patch size, shape,
location, and accessibility are critical to the formation of
structure and realization of ecological functions. Humancaused fragmentation of natural landscapes into fewer
and fewer smaller pieces leads to habitat shrinkage and
less exchange of materials and species among habitats.
Increasing fragmentation and decreasing habitat size can
result in local extinction of some populations.
A landscape is a heterogeneous matrix of smaller
habitats, and the arrangement, size, productivity, and resilience of these habitats within the matrix affect the flow of
energy, animals, and materials through the landscape. Loss
or degradation of one or more elements may lead to dysfunctional performance of the remaining elements. In
deciding on restoration strategies and sites, it is useful to
identify and consider the dysfunctional or absent elements.
278
ESTUARINE HABITAT RESTORATION