Wetland restoration begins and ends with water. The single most important step in any project is re-establishing the hydrology that once kept the site saturated, because nearly every other wetland function, from soil chemistry to plant communities to wildlife habitat, follows from getting the water regime right. Beyond hydrology, a successful restoration typically involves reshaping the land surface, managing invasive plants, reintroducing native vegetation, and committing to years of monitoring and adaptive management. The process is rarely quick, and outcomes rarely match pristine reference sites perfectly, but the techniques available today are grounded in decades of field research across a wide range of wetland types.
Fixing the Water First
Most wetlands that need restoring were drained on purpose. Agricultural ditches, road culverts, and levees pulled water off the land, lowered the water table, and dried out soils that had been saturated for centuries. Reversing that drainage is the foundation of every restoration project, regardless of whether you are working on a coastal salt marsh, an inland peat bog, or a prairie pothole.
The practical methods depend on how the site was drained. Where an existing canal or ditch network removes water, installing adjustable water control structures can slow drainage and raise water levels across the target area. In the Great Dismal Swamp in Virginia, for example, managers placed flashboard risers at strategic points in the canal system. These structures use removable boards to set canal water levels, which reduces the pull of groundwater from the surrounding swamp into the ditch and keeps the soil profile wetter for longer periods.1Ecological Engineering. Impact of control structures on hydrologic restoration within the Great Dismal Swamp The advantage of adjustable structures is flexibility: managers can raise or lower boards seasonally as conditions demand, which matters because most wetlands do not stay at a single water level year-round.
In other settings, restoration means filling ditches entirely, breaching levees, or removing berms so that floodwater can spread across the landscape again. Where streams have cut deeply into their channels, reconnecting the water to the floodplain takes more creative engineering. Whatever the method, the goal is the same: get water staying on the site long enough to support wetland soils and vegetation, without flooding the site so deeply that you create open water where marsh should be.
Shaping the Ground Surface
Natural wetlands are not flat. They have a complex surface of slight mounds and shallow depressions, sometimes only a few centimeters apart in elevation, and that microtopography controls which spots are flooded, which are merely damp, and which stay dry enough for upland species to persist. When a site has been farmed or graded, that subtle landscape is gone, replaced by a uniform surface that responds uniformly to water. The result is often a restored wetland that is either all flooded or all dry, with little of the habitat variety you would find in a natural system.
Re-creating microtopography means building hummocks (small mounds) and hollows (shallow depressions) on otherwise level terrain. Research at a restored site in North Carolina found that this kind of reestablishment produced distinct differences in hydrology, soils, and vegetation between hummock and hollow positions, creating habitat diversity on a site that would otherwise have been monotonous.2Restoration Ecology. Hydrologic, Edaphic, and Vegetative Responses to Microtopographic Reestablishment in a Restored Wetland In practice, this is often done during initial earthmoving, using excavators to vary the grading rather than producing a smooth, even surface. It adds relatively little cost to the site preparation phase and can substantially accelerate the development of diverse plant communities.
What Happens Underground
Once you raise the water table, the soil chemistry begins to shift. Saturated soils become oxygen-poor, which triggers a cascade of chemical changes collectively known as reduction. These changes are essential to how wetlands process nutrients and store carbon. In drained soils, oxygen allows organic matter to decompose quickly and nutrients like nitrogen and phosphorus to remain mobile. When water returns and oxygen disappears, decomposition slows, carbon accumulates as peat or organic muck, and certain nutrient pathways shut down.
A year-long field study at a restored lowland wet grassland on nutrient-rich peat in England tracked how water table height controlled soil chemistry throughout the profile. The researchers found a strong relationship between water table height and soil redox conditions at all depths measured, with the deeper soil layers remaining in a permanently reduced state associated with denitrification, even during drier periods.3Geoderma. Hydrological controls on soil redox dynamics in a peat-based, restored wetland This matters because denitrification is one of the key ecosystem services wetlands provide: converting excess nitrate into harmless nitrogen gas. For anyone restoring a wetland in an agricultural landscape, this is often the primary water quality benefit they are after.
Microbial communities also respond to rewetting. A study of former cranberry farmland found that flooded and restored sites had microbial communities most similar to natural reference wetlands, while active farms had the least similarity.4PLOS ONE. Flooding and ecological restoration promote wetland microbial communities and soil functions on former cranberry farmland The microbes that drive wetland nutrient cycling do not need to be imported; they recolonize on their own once conditions are right. The bottleneck is hydrology, not biology.
Managing Invasive Vegetation
Invasive plants are one of the biggest obstacles to a successful restoration, and in many North American wetlands, the single worst offender is common reed, Phragmites australis. The non-native lineage of this grass forms dense monocultures that crowd out native species, alter hydrology, and degrade habitat. Controlling it is often the most labor-intensive and expensive part of a restoration project.
The evidence is clear that herbicide application is the most effective tool for large-scale Phragmites control. Across Chesapeake Bay wetlands, researchers found that multiyear herbicide treatments greatly reduced Phragmites cover while increasing the diversity and quality of native plant communities.5Ecosphere. Landscape and site factors drive invasive Phragmites management and native plant recovery across Chesapeake Bay wetlands Combining herbicide with a follow-up burn can speed the process: at two freshwater wetland sites, both herbicide alone and herbicide-plus-burning dramatically reduced Phragmites and increased plant biodiversity, but the burned site saw quicker regrowth of native plants.6Restoration Ecology. Common Reed Phragmites australis: Control and Effects Upon Biodiversity in Freshwater Nontidal Wetlands A key finding from that work was that recovery depended on a pre-existing native seed bank in the soil. Where the seed bank has been depleted, additional seeding or planting is necessary.
Mechanical control, such as mowing or cutting, is less effective on its own. A review of 40 years of Phragmites management in the United States concluded that mowing alone does little beyond temporary containment unless it is combined with either plastic sheeting or herbicide. The most effective mechanical approach involves cutting in spring and then covering the stubble with plastic for a full growing season, which reduced shoot density from about 21 per square meter down to 0.1 per square meter in one experiment.7AoB PLANTS. Phragmites australis management in the United States: 40 years of methods and outcomes The catch is that this approach is extremely labor-intensive and practical only on small areas.
Establishing Native Plants
After invasives are knocked back, the question becomes how to get native wetland plants established. There are two broad strategies: passive recolonization (waiting for plants to arrive on their own from nearby seed sources) and active seeding or planting of nursery-grown stock.
Passive recolonization works best when natural wetlands are close by and a viable seed bank persists in the soil. When those conditions are not met, the results can be poor. A seed-based restoration study found that passive recolonization was insufficient to increase native plant cover and limit invasive species at restored emergent wetland sites, even when those sites were geographically close to each other. Extreme weather, including severe drought in the first year and prolonged flooding in the second, overwhelmed the effects of active seeding treatments as well.8PubMed Central. Bet Hedging to Aid Seed-Based Wetland Restoration Under Hydrologic Extremes The takeaway is that even well-planned seeding can fail if hydrology is not stable, and that planting a diversity of species with different flood tolerances (a “bet-hedging” approach) can improve the odds in unpredictable conditions.
For many projects, active planting of container-grown plugs is the most reliable method for establishing target species, though it is expensive. A common compromise is to plant key structural species (such as sedges or bulrushes that form the foundation of the plant community) while allowing other species to colonize naturally over time.
Beaver Dam Analogues and Process-Based Restoration
In the American West and other semi-arid landscapes, a different approach to wetland restoration has gained traction: building structures that mimic the dams beavers would naturally create. Beaver dam analogues, usually constructed from fence posts, woven willow branches, and sediment, are installed across incised stream channels to slow water down, raise it up, and push it out onto the surrounding floodplain.
Research in a semi-arid floodplain showed that beaver dam analogues increased both surface and groundwater levels, promoted flow reversals where stream water recharged the floodplain during high-flow events, and reduced the overall gradient pulling groundwater back toward the stream.9Hydrological Processes. Impact of beaver dam analogues on hydrology in a semi‐arid floodplain In deeply incised channels where beavers would not return on their own, these structures offer a way to restore floodplain wetland function at relatively low cost. They do require regular maintenance, since they are built from natural materials that degrade over time.
Even in storm-driven urban streams, beaver dam analogues show promise. A study of structures installed in a flashy, stormwater-impacted stream documented substantial sediment deposition upstream of the dams, with the median grain size declining from 27 to 14 millimeters and fine material increasing after installation.10Water Resources Research. Hydrologic Retention and Sediment Redistribution Following Beaver Dam Analog (BDA) Installation in a Flashy, Ephemeral, Stormwater‐Impacted Stream That shift from coarse gravel to finer sediment reflects reduced stream energy and the beginning of floodplain-building processes, which is exactly what these structures are designed to catalyze.
Special Challenges for Coastal Wetlands
Tidal marshes face a threat that inland wetlands largely do not: rising sea levels. When a coastal marsh cannot build elevation fast enough to keep pace with rising water, it gradually drowns and converts to open mud or shallow water. In urbanized settings, marshes often cannot migrate inland because roads, buildings, or seawalls block the way. This leaves restoration practitioners with a difficult problem.
One technique gaining acceptance is thin-layer sediment placement, where dredged material is spread across the marsh surface to raise its elevation. At Seal Beach National Wildlife Refuge in southern California, a thin-layer application of locally dredged sediment raised marsh elevation by about 25 centimeters across roughly 3 hectares.11Ecological Engineering. Thin-layer sediment addition to an existing salt marsh to combat sea-level rise and improve endangered species habitat in California, USA A similar project at Blackwater National Wildlife Refuge in Maryland applied thin layers of sediment dredged from the adjacent river onto 16 hectares of subsiding and fragmenting tidal marsh to increase resilience to sea-level rise.12Shore & Beach. Tidal marsh restoration at Blackwater National Wildlife Refuge, Maryland: A case study in thin-layer placement These projects are essentially giving the marsh a head start on the vertical race against rising seas. The approach requires a nearby source of clean sediment and careful application to avoid burying existing vegetation too deeply.
The Carbon Trade-Off in Peatland Rewetting
Peatlands, which include bogs, fens, and some types of swamp, are among the most carbon-dense ecosystems on Earth. When they are drained, the stored peat oxidizes and releases large amounts of carbon dioxide. Rewetting them stops that CO₂ loss, but it creates a new issue: methane. Flooded, oxygen-free peat produces methane, a far more potent greenhouse gas over short time scales.
This presents what researchers have called a biogeochemical compromise. Reflooded peatlands sequester carbon and rebuild peat soil, but the methane emissions make them strong greenhouse gas sources in the short term.13Geophysical Research Letters. A Biogeochemical Compromise: The High Methane Cost of Sequestering Carbon in Restored Wetlands However, the picture improves with time. Because methane has a much shorter atmospheric lifetime than CO₂, and because rewetted peatlands produce negligible CO₂ and nitrous oxide, the total warming effect of all three greenhouse gases combined reaches a plateau relatively quickly after rewetting.14PubMed Central. Prompt rewetting of drained peatlands reduces climate warming despite methane emissions Over decades, the net climate benefit is positive compared to leaving the peatland drained. This is an important consideration for projects seeking carbon credits or climate mitigation benefits, because the payoff is not immediate.
When the Animals Come Back
If the hydrology and vegetation recover, wildlife usually follows, but the speed and completeness of that recolonization depends heavily on the surrounding landscape. Amphibians are often used as indicator species because they are sensitive to water quality and depend on wetlands for breeding.
In Minnesota, monitoring of seven recently restored wetlands found that eight amphibian species rapidly colonized and established breeding populations. However, reference wetlands supported twelve species, including four that were absent from the restored sites. Wetland size and isolation were the strongest predictors of species richness; water chemistry and vegetation cover mattered less than how big the wetland was and how far it sat from other wetlands.15The American Midland Naturalist. Colonization of Restored Wetlands by Amphibians in Minnesota The practical lesson: a small, isolated restored wetland will attract some species but probably not the full complement, while a larger wetland connected to others through the landscape has a much better chance.
Studies in central Iowa reached a similar conclusion, finding few differences in amphibian species richness or survival between restored and reference wetlands, though the two types differed in water chemistry and disease prevalence. The researchers emphasized that maintaining a complex of fish-free wetlands with varied water levels across the landscape was likely the best strategy for amphibian persistence.16Wetlands. Restored Agricultural Wetlands in central Iowa: Habitat Quality and Amphibian Response In a European context, restored wetlands in Hungary showed significantly higher occupancy for fire-bellied toads and tree frogs compared to non-restored control sites, suggesting that restoration can provide habitat for species losing breeding sites elsewhere.17Global Ecology and Conservation. Effectiveness of wetland restoration on amphibians in a drying grassland habitat losing its breeding sites
Monitoring That Actually Tells You Something
A restored wetland does not take care of itself, at least not for a long time. Monitoring is how you know whether the project is on track and when to intervene. A review of 99 wetland restoration studies found a broad variety of post-restoration indicators in use, spanning vegetation structure, composition, function, and landscape configuration.18Ecological Indicators. Review Indicators of vegetation development in restored wetlands The sheer variety can be overwhelming, so choosing a manageable set of indicators that match your project goals is essential.
For tidal marshes along the Gulf of Mexico, researchers found that recovery of plant biomass and sediment characteristics, which are important but expensive to measure, could be predicted from simpler variables like time since restoration, percent plant cover, and sediment bulk density.19Journal of Applied Ecology. Evaluating restoration success using metric‐based indicators of ecosystem recovery in tidal marshes along the northern Gulf of Mexico This kind of practical shortcut is valuable because most restoration projects operate on tight budgets, and intensive laboratory analysis of soil carbon and root biomass at every monitoring visit is not realistic.
In the regulatory context, wetland mitigation banks in the United States are required to meet specific ecological performance standards. An analysis of 22 mitigation banks in the Chicago district found that banks typically met standards related to native species richness and dominance but struggled to meet standards limiting non-native species.20Wetlands. Compliance with Regulatory Performance Standards in Wetland Mitigation Banks Perennial plant species richness and floristic quality generally improved over the five-year monitoring period, but invasive control remained a persistent challenge. This finding echoes a theme running through the entire restoration literature: getting native plants established is usually achievable, but keeping invasives at bay requires ongoing effort.
Long-Term Management With Fire and Grazing
Once a wetland is restored and the initial planting phase is over, the work is not done. Without some form of ongoing disturbance, many wetlands become dominated by aggressive species like reed or woody shrubs, which reduces habitat quality and biodiversity. In natural systems, fire, grazing, and flooding all served as periodic disturbances that kept the plant community open and diverse. Restoration managers need to replicate at least some of those forces.
Long-term research on subtropical wetland plant communities found that moderate grazing levels maintained high plant diversity and prevented encroachment by trees and shrubs. Where grazing was removed, prescribed fire became essential to maintaining species evenness.21PubMed Central. Long-term response of wetland plant communities to management intensity, grazing abandonment, and prescribed fire Removing all disturbance, both grazing and fire, led to lower diversity and floristic quality within just three years.22Journal of Applied Ecology. Interactive effects of pasture management intensity, release from grazing and prescribed fire on forty subtropical wetland plant assemblages A separate field experiment showed that low-intensity cattle grazing combined with late-summer burning effectively reduced reed cover and increased both plant and amphibian diversity in marsh habitats.23Biological Conservation. Spatiotemporally variable management by grazing and burning increases marsh diversity and benefits amphibians: A field experiment
The specific management prescription varies by region and wetland type, but the broad principle is consistent: wetlands are disturbance-dependent systems. A restoration plan that does not include a long-term management strategy for maintaining open habitat is likely to degrade over time, even if the initial results look good.
Realistic Expectations for Recovery
Perhaps the most sobering finding in the wetland restoration literature comes from a global meta-analysis of 621 wetland sites. Even a century after restoration, biological structure (primarily plant communities) and biogeochemical function (primarily soil carbon storage) remained on average 26% and 23% lower, respectively, than in reference wetlands.24PubMed Central. Structural and Functional Loss in Restored Wetland Ecosystems Large wetlands (over 100 hectares), those in warmer climates, and those with more hydrologic exchange (riverine or tidal) recovered more quickly than small, cold, or depressional wetlands. An earlier study of a restored estuarine wetland found that only a few of 16 measured ecosystem attributes were on a trajectory toward equivalency with natural wetlands.25Ecological Applications. Functional Equivalency Trajectories of the Restored Gog‐Le‐Hi‐Te Estuarine Wetland
This does not mean restoration is pointless. A wetland functioning at 75% of its reference condition is vastly better than drained farmland or a parking lot. But it does mean that restoration should not be treated as a perfect replacement for preservation. If a healthy wetland can be protected rather than destroyed and rebuilt, that is always the better outcome. When restoration is the option available, going in with realistic expectations about timeline and end state helps set appropriate goals and avoid premature declarations of success.
Stormwater Wetlands in Urban Settings
Not all wetland restoration looks like returning a drained prairie pothole to its former glory. In urban areas, constructed wetlands are increasingly built to treat stormwater runoff before it reaches rivers and lakes. These systems combine many of the same principles as ecological restoration, including managing water levels, establishing vegetation, and promoting the soil conditions that support nutrient removal, but they are engineered from the start rather than returned to a historical condition.
A study of a stormwater constructed wetland in Strasbourg, France, that combined a sedimentation pond with a vertical-flow sand filter found that the system removed between 50% and 100% of various pollutants from stormwater during rain events. The filter captured the majority of both dissolved and particle-bound contaminants, while the pond’s efficiency was reduced when incoming flow stirred up previously deposited material.26PubMed. Micropollutants removal and storage efficiencies in urban stormwater constructed wetland During dry periods, several heavy metals accumulated at high concentrations in both the pond and filter soils, raising questions about long-term management of those accumulated contaminants. Urban stormwater wetlands trade some of the ecological complexity of a natural system for targeted pollutant removal, and they require their own form of maintenance, including periodic sediment removal and vegetation management to keep flow paths open.