Research ArticleResearch Article
Open Access

Lessons Learned and Value of Early Post-Construction Monitoring of a Large Tidal Wetland Restoration Project

Henry M. Page, Stephen C. Schroeter, Andres Deza, Russell Johnston, Rachel S. Smith, Kathryn M. Beheshti, Daniel C. Reed and Justin C. Hoesterey
Ecological Restoration, December 2025, 43 (4) 260-275; DOI: https://doi.org/10.3368/er.43.4.260

Abstract

Coastal development has negatively impacted estuarine ecosystems over the past two centuries. In southern California, USA, salt marshes are valued for their high animal and plant diversity and for providing foraging and nesting habitat for species of conservation concern. Given the ecological importance of this regionally rare habitat, coastal managers have prioritized salt marsh restoration, despite its many challenges. We review lessons learned on the effect of design and adjoining landscape on early vegetation development in a 60.7 ha tidal wetland restoration project in a southern California estuary where nearly one-third of 37.6 ha of planned salt marsh was excavated to high marsh elevations. Initial planting to facilitate vegetation development in large areas of the high marsh (> 1.7 m MLLW) was unsuccessful, likely due to hypersaline soils. In some areas, subsurface freshwater intrusion likely enhanced planted vegetation, which outperformed vegetation planted at comparable high elevations in other areas. In the mid-marsh (< 1.7 m MLLW), vegetation development from natural recruitment exceeded that of the planted high marsh. After five years, vegetation cover in the high marsh remained low. To address this issue, the marsh plain was re-graded lower in some areas and a planting program was established to hasten vegetation development. Lessons learned pertain to the critical importance of initial wetland design, which in this project led to sparse vegetation cover and the necessity of re-grading and large-scale planting to facilitate vegetation establishment.

Keywords

Restoration Recap

  • We review the effect of topographic design on early vegetation development in a 60.7 ha tidal marsh restoration project in San Diego County, CA.

  • A large proportion of the site (11.4 ha, 31%) was planned high salt marsh. Initial planting to facilitate vegetation development was unsuccessful (< 10% cover) in large areas after five years due to hypersaline soils, whereas cover exceeded 30–50% at high marsh elevations in other areas in proximity to upland groundwater intrusion or the river channel.

  • In the mid-marsh, vegetation development from natural recruitment (no planting) exceeded that of planted areas in the high marsh. Poor vegetation development at high elevations triggered re-grading to lower elevations and implementation of a planting program.

  • A design that incorporates lessons learned from this study on the effect of graded topography of the marsh plain should alleviate the need for regrading to achieve timely vegetation cover.

Coastal development has negatively affected estuarine ecosystems over the past two centuries, leading to the degradation and loss of the critical ecosystem services and socioeconomic values that these systems provide (Kennish 2002, Zedler and Kercher 2005, Lotze et al. 2006, Valiela et al. 2009). Within the Southern California Bight, 48% of historical estuarine wetland habitat types have been lost since the mid 1800s, with estuarine vegetated wetlands experiencing a 75% loss through large-and small-scale habitat destruction and modification, including the installation of structures restricting tidal flow, dredging, and the filling of intertidal salt marsh to create upland and support development (Stein et al. 2014, Beller et al. 2014).

Salt marshes often form extensive habitat within estuarine ecosystems, providing shoreline stabilization, the preservation of native biodiversity, food chain support for invertebrates, fish and birds, and opportunities for research, teaching, and recreation (Onuf et al. 1979, Zedler and Kercher 2005, Beller et al. 2014). In the tidally influenced salt marshes of semi-arid southern California, USA, the higher elevation marsh is particularly valued as the location of high plant and insect diversity, and foraging and nesting habitat for endangered Passerculus sandwichensis (Belding’s Savannah Sparrow) and other species of conservation concern (Zedler et al. 1992, Powell and Collier 1998, Rosencranz et al. 2018). Salt marsh loss and degradation has caused population declines of obligate plant and animal species, and in some cases local extirpation of these species from sites (Ferren 1985, Sullivan 2001, Noe et al. 2019). Consequently, the preservation and restoration of salt marsh is a priority of government agencies and non-government organizations charged with the conservation of coastal ecosystems in California and elsewhere (SCWRP 2018, CWMW 2023).

Tidal wetland restoration projects are often considered successful if prescribed goals or success criteria are satisfied. For example, a success criterion could be that a restored wetland reaches specific habitat acreage targets or provides functions and values expected of a reference system(s) in the region (Mitsch and Wilson 1996, Zedler and Callaway 2000, Vivian-Smith 2001, Raposa et al. 2018). However, successfully satisfying these criteria can be difficult (Zedler and Callaway 2000, Thom et al. 2005, Callaway 2005). Challenges to achieving success begin during the initial design phase, and continue through construction where excavation and grading to appropriate tidal elevations is necessary to achieve the topography and hydro-periods required of the planned habitats (Callaway 2001, Wolters et al. 2008, Day et al. 2021, Wang et al. 2021).

For salt marsh restorations, project success generally includes the timely development of native vegetation, which provides habitat structure and food for marsh-associated invertebrate and vertebrate species. In mature marshes, there is a general association between tidal elevation and the distribution of salt marsh plant species and restoration project designs require an understanding of these relationships. These relationships are driven by species-specific tolerances to abiotic stressors that co-vary with elevation, and biological interactions (Zedler et al. 1999, Sullivan 2001, Janousek et al. 2019). However, planning for timely vegetation development using relationships between plant distributions and elevation within the wetland prior to restoration, or in mature reference wetlands, may be unsuccessful, particularly at higher tidal elevations, if abiotic conditions are stressful to new recruits or plantings (Zedler et al. 2003).

The vegetation of tidal marshes in southern California consists of low lying forbs and grasses that include Spartina foliosa (California cordgrass), typical at low marsh elevations, Salicornia pacifica (pickleweed), a plant typical of middle marsh tidal elevations, and a mixed assemblage of S. pacifica, Arthrocnemum subterminale (Parish’s glasswort), Limonium californicum (marsh rosemary), Frankenia salina (alkali seaheath), and Distichlis spicata (saltgrass), among the dominant species at higher elevations (Purer 1942, Onuf et al. 1979, Ferren 1985, Zedler et al. 1992).

Restoration projects may rely on natural recruitment to vegetate newly restored habitat (Sullivan 2001, Day et al. 2021). This approach is most feasible in areas of high seed supply, and where topographic and edaphic conditions are conducive to the germination, survival and growth of new recruits. However, the development of vegetation cover through natural recruitment may take years to decades to achieve equivalency with natural wetlands (Craft et al. 1999, Zedler and Callaway 1999, Williams and Orr 2002). Vegetation development may be especially problematic at higher tidal elevations (e.g., > 1.77 m Mean Lower Low Water, MLLW) that experience infrequent tidal inundation and hypersaline soils (Zedler et al. 2003, O'Brien and Zedler 2006, Beheshti et al. 2023).

More commonly, vegetation development at restored sites is achieved through a combination of natural recruitment and the planting of nursery grown plants (Sullivan 2001, Day et al. 2021). Planting not only reduces the reliance on seed supply that may be spatially and temporally variable (Rand 2000, Lindig-Cisneros and Zedler 2002, Morzaria-Luna and Zedler 2007, Xie et al. 2019, Liu et al. 2024), but also bypasses the vagaries of germination and early growth and survival. Planting can also accelerate the development of feedbacks between vegetation and soil conditions (e.g., soil shading, moisture retention) that may ameliorate high soil salinity and enhance vegetation development. The negative effects of some biotic factors (e.g., competition) on plant survival and growth that may be important in mature marshes (Bertness et al. 1987, Pennings and Callaway 1992, Emery et al. 2001) are expected to be less important in newly planted restoration sites since plant density is initially low. In contrast, edaphic factors that include soil salinity, soil moisture, and nutrient limitation are expected to be more important to achieving successful plant establishment in newly restored sites (Zedler et al. 2003, O’Brien and Zedler 2006).

Under favorable conditions, vegetation cover can develop quickly at planted sites. For example, at the Friendship (Model) Marsh Restoration site in Tijuana Estuary, planted vegetation achieved 100% cover within two years, although canopy height remained low compared with reference sites for four years (Keer and Zedler 2002). However, planted vegetation can perform poorly at restoration sites for various reasons, including high soil salinities, low soil nutrients and moisture content, and coarse soil texture atypical of salt marsh habitat (Haltiner et al. 1996, Zedler et al. 2003, Liu et al. 2024). For example, stem density and canopy architecture of planted S. foliosa in a constructed marsh in San Diego Bay were not comparable to those in a reference site even 10 years post-construction (Zedler et al. 1999). Topographic and hydrologic design could ameliorate conditions that impede successful vegetation development (e.g., poor natural recruitment, planting mortality), but the specifics of a particular design, which ideally include attention to elevation, slope and drainage, surface heterogeneity, soil properties, adjacent upland, and the planting program, may not be fully considered during the restoration planning process.

Vegetation developed unevenly within and among construction modules in the San Dieguito Wetlands Restoration Project in southern California despite planting following grading in the planned high marsh. Here, we review factors associated with restoration design and the adjacent upland that impeded, or facilitated, the early post-construction development of salt marsh vegetation within San Dieguito Lagoon, one of the largest extant estuarine systems in San Diego County, California. We used aerial imagery to compare early vegetation development between planned high and middle marsh among construction modules to evaluate the supposition that vegetation would develop through natural recruitment at planned middle marsh elevations, but that planting would be required at higher elevations. We used on-the-ground sampling of vegetation cover and soil properties to identify factors likely leading to the poor performance of planted vegetation at higher elevations in some areas, but not other areas. The goal of this article is to provide lessons that will inform the design and restoration of salt marsh habitat in California and elsewhere.

Material and Methods

Physical Setting

San Dieguito Lagoon is located approximately 32 km north of San Diego, California (32.970°N, 117.261°W), adjacent to the city of Del Mar (Figure 1). The lagoon experiences a semi-arid climate and strongly seasonal rainfall that occurs primarily from December through March. Annual rainfall measured at the San Diego International Airport, approximately 32 km from San Dieguito Lagoon, ranged from 8.4 cm to 57.4 cm with an annual average of 25.4 cm from 1965 to 2024 (San Diego County Water Authority, unpublished data). The wetland receives freshwater inputs primarily from the San Dieguito River with flows highest following rainstorms and declining during the dry summer and fall. Tidal exchange in the wetland occurs through an inlet to the Pacific Ocean. Tides are mixed semi-diurnal with planned high, mid and low marsh habitats expected to be exposed 99.6 to 99.9%, 81.7 to 99.6% and 61.8 to 81.7% of the time, respectively, following restoration (SCE 2005).

none
Figure 1.

Map showing the location of San Dieguito Lagoon, the construction modules, and habitat components within the San Dieguito Wetlands restoration site. A) Location of the project in San Diego County in coastal southern California, USA. B) Map of the wetland restoration construction modules alongside the San Dieguito River. This study was conducted in modules W2/3, W5, W10, and W16.

San Dieguito Lagoon and the San Dieguito Wetlands Restoration Project

Prior to the late 1800s, San Dieguito Lagoon was the largest of the six San Diego County estuaries, providing approximately 243 ha of estuarine open water and wetland habitat (MEC 1993, Stein et al. 2014, Beller et al. 2014). Much of the wetland was filled in during urban and agricultural development from the 1880s through the 1970s, which reduced the extent of estuarine wetlands to about 81 ha. The goal of the 60.7 ha San Dieguito Wetlands Restoration Project (SDW) is to restore primarily weedy, non-native ruderal uplands that were historical wetlands within the San Dieguito Lagoon system to fully tidal wetland habitats. Southern California Edison (SCE) designed and constructed this project as partial mitigation for the impacts of the San Onofre Nuclear Generating Station (SONGS) on the marine environment (CCC 1997, SCE 2005). The project is required to meet specific mitigation requirements, as performance standards, provided in the Coastal Development Permit for the operation of SONGS. Habitat targets were selected after consideration of several alternatives in the project EIR/EIS process that could satisfy the mitigation requirement required by the California Coastal Commission (CCC 1997). This article’s authors were not involved in project design, final decision on the selected project alternative, or in construction activities. The restoration was expected to provide ecosystem services like those of relatively undisturbed wetlands in the region (CCC 1997, SCE 2005).

Restoration was conducted in five construction modules (Table 1, Figure 1) that differed in size, topography and planned mix of habitats. Planned high marsh, graded to tidal elevations ranging from 1.86 m to 2.07 m (Mean Lower Low Water, MLLW, Table 1), comprised nearly one-third of the 37.5 ha of planned vegetated salt marsh. In two modules (W2/3, W10) high marsh comprised the majority (> 50%) of the planned salt marsh (Table 1). Nearly one-half of the remaining planned salt marsh was graded to middle marsh elevations ranging from 1.37 m to 1.86 m MLLW, whereas 23% was graded to low marsh elevations, ranging from 1.1 m to 1.37 m MLLW (Table 1).

Tidal creeks were not included in the Restoration Plan per se (SCE 2005), but three 50–60 m long linear precursor creeks were constructed in module W2/3 with the expectation that they would evolve over time into longer tidal creek networks. Seven creek networks were also constructed in module W4/16 (Litzba et al. 2011, Figure 1). Further details of restoration project design and construction are provided in SCE (2005) and Supplementary Material, Table S1.

View this table:
Table 1.

Planned areas (and percentage contribution) of low, middle, and high salt marsh habitat in the study restoration modules in San Dieguito Wetlands. Elevations of planned habitats: low marsh 1.09 m to 1.37 m, middle marsh 1.37 m to 1.86 m, and high marsh 1.86 m to 2.07 m MLLW (SCE 2005). This study was conducted in modules W2/3, W16, W5, and W10. See Figure 1 for module locations.

One goal of the restoration project was to achieve the high cover (i.e., > 85%) of salt marsh vegetation typical of relatively undisturbed, natural tidal wetlands in the region. Several plant species were expected to naturally colonize the planned high and middle salt marsh, including S. pacifica, A. subterminale, M. littoralis, D. spicata, and F. salina, while S. foliosa was expected to become established at low marsh elevations.

Planting supplemented natural recruitment in the high marsh. SCE contractors planted over 350,000 plants throughout the restoration at high marsh elevations (1.77 to 2.07 m MLLW) between December 2008 and March 2009 (Supplementary Material, Table S2). Most plants (90%) were S. pacifica. Plants were commercially sourced and grown from cuttings in 5 cm containers prior to planting (MCC 2009). During commercial grow-out, plants were initially watered with freshwater but acclimated to increasingly higher salinities prior to planting. Plants were planted at densities of between 2.0 and 4.7 plants m−2 (Table 2).

View this table:
Table 2.

Number of container plants and spacing between plants planted at high marsh elevations in the San Dieguito Wetlands project in December 2008–March 2009. Spartina foliosa not included (MCC 2009). Plant species provided in Table S2.

Remote Sensing of Vegetation Cover in High and Middle Marsh

We used multispectral aerial imagery to compare vegetation development in the planted high marsh (above the 1.77 m MLLW elevation contour) across modules and to the unplanted mid-marsh (Figure 1). The 1.77 m MLLW elevation contour was used to delineate the lower elevational boundary of high marsh in our analysis rather than the design elevation of 1.87 m MLLW because vegetation was planted above this elevation. Aerial imagery was acquired using fixed-wing aircraft flown at 1,829–3,048 m altitude in summer and fall (October 2008, 2009, August 2010, June 2012, August 2013), overlapping the period of highest plant productivity (Onuf 1987, Zedler et al. 1992). The 4-banded red-green-blue-near infrared (RGB-NIR) imagery was captured with a 1,024 × 1,024 pixel, 12-bit digital multispectral DMSC camera (Spec-Terra, LTD, Australia) (2008–2010) or a 14,430 × 9,420 pixel, 16-bit multispectral UltraCam camera (Microsoft, Austria) (2012–2013) flown at a ground sampling distance of 25 cm for the 2008–2010 imagery and 20 cm for the 2012 and 2013 imagery. Imagery was stored as ortho-mosaics. Imagery acquired in 2011 was excluded from the analysis due to the use of a 3-band camera.

We assessed vegetation cover from the aerial imagery using a pixel-based, supervised classification based on spectral characteristics, and from ground-based observations of the vegetation cover type to create a thematic raster dataset. Classes were grouped as vegetated or non-vegetated. Vegetation percent cover was calculated within module boundaries by first summing the area of pixel classes scored as vegetated within the contours delineating planned high and middle salt marsh and then dividing by the total pixel number. Training samples were created by drawing polygons around pixels that were representative of each vegetation cover class and used with the Support Vector Machine classifier in ArcGIS Pro 3.3.2. We compared percent cover estimates from aerial imagery with on the ground point contact estimates (see Methods in next section) in July 2009 and July 2012.

Performance of Planted Salt Marsh Plants

To compare changes in vegetation cover in the high marsh among modules and over time, we also sampled vegetation on the ground within 42 two-meter-wide belt transects extending from 1.77 m to 2.07 m MLLW. This sampling included module W5, the high marsh that was not well sampled using aerial imagery (see Results). Sampling was conducted along the same transects in April 2009, July 2009, and July 2012. Cover estimates were recorded at 0.5 m intervals along the transect. At each interval, contacts were recorded at five points (0.5 m apart) along a bar extending 2 m perpendicular to the transect. Only first contacts were recorded (vegetation layering was not measured). Point contact data were categorized by plant species, as bare space or other non-vegetated cover (e.g., wood, macroalgal wrack) for analysis. Percent cover of each category for each transect was calculated from the summed contacts by category divided by the total number of contacts scored.

Marsh Slope and Soil Properties in High Marsh

The slope of the constructed marsh plain can influence the drainage of marsh soils and thus soil properties that could affect vegetation development (Callaway 2001). We estimated the slope of the marsh plain between 1.77 m and 2.07 m MLLW using ArcGIS Pro 3.3.2 to measure the linear distance between these elevations along each transect line. Slope was expressed as “rise to run” ratio (1: linear distance between 1.77 m and 2.07 m) and as percent change in elevation between 1.77 m and 2.07 m.

Soil properties can affect plant growth and survival (Callaway et al. 2001). We characterized selected soil properties in the laboratory using two replicate soil samples taken each at the 1.77 m and 2.07 m MLLW endpoints of a subset (n = 20 of 42) of transects. Samples were collected using a 4 cm diameter core taken to a depth of 6 cm in September 2009. Percent cover of vegetation was estimated visually in two replicate 0.5 m × 0.5 m quadrats that included the soil sampling locations within each quadrat. Vegetation layering was not recorded.

Individual soil samples were wrapped in aluminum foil, returned to the laboratory in airtight plastic bags and refrigerated until analysis for soil salinity, soil moisture, soil (bulk) density, nitrogen content and percent organic matter. Estimates of soil salinity were made using a subsample from one core and the 1:1 (dry soil:water) soil dilution method (Callaway et al. 2001). Estimates of soil moisture were made using the second core from the change in weight of the soil sample before and after drying at 60°C for 48 hours. Soil bulk density was estimated using the weight of the dry core expressed per known core volume (modified from Callaway et al. 2001). Subsamples (50–90 mg sample) of the second core were ground with a ceramic mortar and pestle and analyzed for percent nitrogen using a CE440 CHN Elemental Analyzer (Exeter Analytical, Inc., Chelmsford, MA). Percent organic matter was measured as loss on ignition after combustion of a 7–14 g wet subsample in a muffie furnace at 400°C for 4 hours (Mook and Hoskin 1982, Callaway et al. 2001). Grain size analysis (% sand, silt, clay) was conducted in two construction modules by SCE contractors (Supplementary Material, Table S3).

Data Analysis

To explore the effect of module and time on vegetation cover in the planted high marsh, we used a beta regression model in the betareg package in R with transect data collected in modules W2/3, W5, W10, W16 and vegetation cover as the response variable, and module and dates as fixed factors. Beta regression models are appropriate with data bounded by 0 and 1 (Cribari-Neto and Zeileis 2010, Geissinger et al. 2022). We used posthoc Tukey pairwise tests to compare vegetation cover between modules.

To examine relationships between vegetation cover and soil properties at high marsh elevations across all modules, we fit a beta-regression model using the vegetation cover data taken in quadrats as the response variable with soil salinity, soil percent water, bulk density, percent organic matter, and percent nitrogen as the independent variables. We excluded two transects from these analyses dominated by the brackish-affinity reed Typha latifolia (broadleaf cattail). We examined correlation plots to assess collinearity prior to analysis and examined residual plots to ensure that model assumptions were met. All analyses were conducted using R version 4.4.0 (R Core Team 2024).

Results

Constructed Topography and Natural Recruitment of Vegetation

Aerial imagery revealed little vegetation development (< 5% cover) in module W2/3, above or below the 1.77 m MLLW contour in 2008 eight months following grading (Figures 2A, 3A). Vegetation cover below 1.77 m MLLW remained low in 2009 but increased through natural recruitment and growth to 9.7% in 2012 and 23.9% in 2013 (Figures 2A, 3B–E). In contrast, vegetation cover above 1.77 m MLLW remained sparse over the five-year period (1.8% in 2008 vs. 2.3% in 2013) despite the planting of nearly 184,000 plants across more than 5.7 ha in December 2008–January 2009 (Table 2, Figures 2A, 3, 4).

none
Figure 2.

Change in the percent cover of salt marsh vegetation over time estimated from aerial imagery in modules A) W2/3, B) W16, C) W5, and D) W10 from 2008 to 2013. Data unavailable from above the 1.77 m contour in W5 (but see Figure 7) and below this contour in W10.

In an attempt to increase flooding frequency of the marsh plain and enhance plant establishment, the short linear precursor creeks originally constructed in W2/3 were re-graded into longer creeks in November 2010 (Figure 2D). However, only ~11% of the module was re-graded to < 1.77 m. Consequently, inundation of the broader marsh plain remained relatively unchanged. Vegetation naturally colonized the edges (~1.3 m MLLW, Project Design Consultants, unpublished data) of the newly constructed creeks (Figure 4C), but cover remained < 5% on the high elevation marsh plain from 2012 through 2013 (Figures 2A, 3, 4).

Aerial imagery also revealed little vegetation development (< 1%) in module W16 above and below 1.77 m MLLW in 2008 eight months following grading (Figures 2B, 5A). Below 1.77 m MLLW, cover remained < 1% in 2009; however, cover increased through natural recruitment to 5.0% in 2010 and reached 20.6% in 2013 (Figures 2B, 5A). Over 123,000 plants were planted above the 1.77 m MLLW contour in W16 (Table 2, Figure 6). In contrast to module W2/3, vegetation cover in W16 above 1.77 m MLLW increased abruptly from < 1% in 2008 to 43.7% in 2009 following planting, but cover declined to 34.9% in 2010, and 8.5% in 2013 (Figures 2B, 5, 6).

none
Figure 3.

A–E. Change in the cover of salt marsh vegetation visualized using aerial imagery in module W2/3 from 2008 to 2013 (year indicated to the right of each figure). White dashed line indicates 1.77 m MLLW contour. Green = vegetation, brown = unvegetated. Note the constructed tidal creeks in 2012 and 2013 and change in location of the 1.77 m MLLW contour. The broader marsh plain was not re-graded.

none
Figure 4.

A) View of module W2/3 in 2009 showing berm in foreground and sparse vegetation cover on the relatively flat marsh plain, B) view of W2/3 in 2010 showing continuing sparse vegetation cover at higher tidal elevations (foreground) and higher vegetation cover at lower elevations to the left, C) view of Salicornia pacifica (white arrow) colonization along the banks of constructed creeks in W2/3, and sparse cover on the surrounding marsh plain, and D) naturally recruited vegetation below the 1.77 m contour (white arrow) in W2/3, and sparse vegetation cover above this elevation (black arrow) prior to re-grading the marsh plain lower in 2014.

The proportion of habitat above 1.77 m MLLW in module W5 consisted of a narrow strip ~3 m wide and sloped, with most of the module graded lower than 1.77 m (Tables 1, 3). The area between 1.77 m and 2.07 m MLLW was planted (Table 2), and sampled on the ground, but the resolution of the aerial imagery to estimate cover within this ~3 m wide strip was insufficient for this purpose and only cover below 1.77 m MLLW is reported here. Vegetation cover below 1.77 m MLLW increased from 2.4% in 2008 to 32.6% in 2013 (Figures 2C, 7). Conversely, W10 was planned entirely as high marsh. Thus, only cover above 1.77 m MLLW is reported here. Vegetation cover above 1.77 m in W10 increased from 14.4% in October 2008 (three months prior to planting) to 32.6% in 2013 (Figures 2D, 7).

Performance of Plantings in High Marsh, Topography, and Soil Properties

Beta regression analysis revealed no interaction between module and date (χ2 =5.292, df = 3, p = 0.15) on vegetation cover along transects in the planted high marsh (1.77–2.07 m); therefore, the analyses were re-run without the interaction term. There was a significant effect of both module (χ2 =41.730, df = 3, p < 0.001) and date (χ2 =5.241, df = 1, p = 0.022) on vegetation cover (Figure 8). Post-hoc Tukey pairwise tests found that all module combination pairs differed significantly (p < 0.05) except for W10 versus W16 (p = 0.782).

Vegetation growth along the transects was initially rapid in module W16, with mean cover increasing approximately ten-fold, 4.8% to 49.5%, from April to July 2009, then decreasing to 39.7% cover by July 2012 (Figure 8). In contrast, transect sampling revealed no significant increase in vegetation cover in module W2/3 from 2008 through 2012. This result was due to an observed nearly 100% mortality of planted plants within the first year (MCC 2009, Page and Schroeter, pers. obs.). Some natural recruitment was observed along the transects, in shallow depressions or at the toe of the earthen berm bordering the wetland-upland boundary, but this contributed little to overall cover that remained < 10% in July 2012 (Figure 8). Vegetation cover along the high marsh transects in W5 showed little increase, from 0% in April 2009 to 15% in July 2012 (Figure 8). Cover along transects in W10 increased by a similar amount from 26.8% in April 2009 to 44.3% in July 2012 (Figure 8). The 26.8% cover in W10 developed between the time it was graded (December 2008), planted (January 2009) and measured in April 2009.

none
Figure 5.

A–E. Percent cover of salt marsh vegetation visualized using aerial imagery in module W16 from 2008 to 2013. White dashed line indicates 1.77 m MLLW contour. Green = vegetation, brown = unvegetated.

Slope ratio of the constructed high salt marsh plain, measured along the transects from 1.77 m to 2.07 m MLLW, ranged from shallow slope values of 1:289 in W2/3 and 1:344 in W10, to a steeper slope of 1:10 in W5 (Table 3). Average width of the marsh plain between 1.77 m to 2.07 m MLLW ranged from 88 m in W2/3 and 106 m in W10 to 3 m in W5 (Table 3).

none
Figure 6.

View of planted vegetation primarily, Salicornia pacifica, looking east to west (A) May 2009, 3–4 months following planting and (B) June 2009, above the 1.77 m MLLW contour in module W16. Sparsely vegetated habitat below this elevation evident on the left side of (A). Same area in (C) February 2010 showing the boundary between naturally recruited (NR) and planted (P) vegetation.

Soil salinity within the high marsh was highly variable, ranging from < 10‰ to > 100‰ (Figure 9). Soil salinities exceeded 76‰ at several stations in July 2009, with the highest values measured in module W2/3. Low salinity values (< 32‰) along the northern edge of W16, where plant growth was greatest, reflected shallow groundwater seepage from adjacent developed upland. In W10 and W5, soil salinity was likely ameliorated in places by the proximity of sampling sites to the San Dieguito River and percolation of less saline riverine water into the soils of this module. Other than a brackish marsh with cattail and bulrush in the western portion of W16 that was excluded from our analysis, groundwater inputs were not evident elsewhere in our study modules (although there are a couple of small freshwater seeps elsewhere on the project site).

View this table:
Table 3.

Comparison of width and slope of the marsh plain among modules as measured along transects between 1.77 m and 2.07 m MLLW. Mean values ±1SE.

none
Figure 7.

A–E. Change in the cover of salt marsh vegetation visualized using aerial imagery in modules W5 and W 10 from 2008 to 2013. White dashed line indicates 1.77 m MLLW contour. Green = vegetation, brown = unvegetated. Black lines indicate module boundaries.

none
Figure 8.

Percent cover of salt marsh vegetation measured along transects in the high marsh (1.77–2.07 m MLLW) in San Dieguito Wetlands from April 2009 through July 2012. Mean values ± 1 SE. Number of transects = 8 (W2/3), 4 (W5), 4 (W10), and 11 (W16).

View this table:
Table 4.

Results of beta regression analysis evaluating relationships between percent cover of native marsh vegetation and measured soil properties. Data from 1.77 m and 2.07 m elevations combined.

Of the soil properties measured, soil salinity best explained variation in percent cover of vegetation among sampling stations (p = 0.008, Table 4). Other factors that could influence planting success, such as percent organic matter (p = 0.913, Table 4) and bulk density (p = 0.695, Table 4), were not significant explanatory variables, although a near significant p-value for soil percent water (p = 0.082) and percent nitrogen (p = 0.117, Table 4) suggested the potential importance of these properties to plant establishment at our stations.

none
Figure 9.

Pattern of soil salinity at high marsh elevations in A) module W2/3 and B) modules W5, W10, and W16 in September 2009.

Discussion

Marsh Topography and Adjoining Landscape Influence the Performance of Planted Vegetation

Typical of many wetland restoration projects, planting in San Dieguito Wetlands aimed to expedite vegetation development following construction (SCE 2005). Planting soon after construction focused on high marsh elevations where the natural recruitment of plants was expected to be slow. Planners projected that natural recruitment would be sufficient to enable timely vegetation development at middle marsh elevations. Expectations of successful plant establishment either through planting or natural recruitment were based solely on the relationships between plant distributions and tidal elevation observed at the site prior to restoration and elsewhere in the region (SCE 2005). Our results reveal that these expectations did not adequately consider the effect of constructed topography (width, slope and elevation of the marsh plain) on edaphic factors that could inhibit early plant establishment.

We observed that the increase in vegetation cover at high marsh elevations varied dramatically among modules, with soil salinity identified as the most important edaphic variable driving this difference. In the natural, semi-arid salt marshes of southern California, soil salinity typically increases with tidal elevation, and soils can become extremely hypersaline at higher elevations (Callaway et al. 1990, Pennings and Callaway 1992, Zedler et al. 2003, Beheshti et al. 2023). Typical seawater salinity of the eastern Pacific Ocean at the latitude of southern California is 34‰ (Sverdrup et al. 1962), which contrasts with the salinity of hypersaline soils of southern California salt marshes that can exceed 80‰ during the summer and fall. At San Dieguito Wetlands, we recorded salinities as high as 136‰ on the graded high marsh of restoration module W2/3. Salinities higher than ~70‰ are considered stressful to plant establishment in newly restored wetlands (Haltiner et al. 1996, Zedler et al. 2003). Planting success and the distribution of marsh plant species in restored and natural California salt marshes is strongly influenced by plant tolerance of high soil salinity (Mahall and Park 1976, Callaway et al. 1990, Pennings and Callaway 1992, Zedler et al. 2003, Beheshti et al. 2023), and stressful soil salinity is a likely explanation for poor vegetation development in areas planned as high marsh in module W2/3.

Hypersaline soil of module W2/3 was a product of the initially designed high tidal elevation (56% of habitat graded > 1.86 m MLLW, Table 1) and, therefore, the infrequent tidal inundation, broad width X̅, and shallow slope (1:289) of the planned high marsh. As a result, tidal waters that drained slowly or ponded on the marsh surface following flooding during the highest spring tides evaporated, contributing to a visible white salt crust. Because the marsh plain was not substantially re-graded lower when the short linear precursor creeks were lengthened in 2010, the marsh plain remained exposed except during the highest tides. This pattern was analogous to that seen in a San Diego Bay restoration where surface soil salinities following grading exceeded 100‰ at some locations (Zedler et al. 2003). The exposure of plants to salinity stress was likely compounded by lower moisture content of the soil due to infrequent tidal inundation, and elevated surface soil temperatures (not measured).

Freshwater inputs originating from adjoining upland ameliorated stressful edaphic conditions associated with high tidal elevation in portions of module W16. This subsurface flow traveled underneath the adjacent berm to penetrate the high marsh (MCC 2009) and supported the rapid growth of plants (primarily S. pacifica) planted in the high marsh. These inputs allowed for an unplanned “natural” experiment on the effects of reduced soil salinity and elevated moisture on the performance of planted plants. Ultimately, SCE contractors considered the subsurface seepage to be undermining the earthen berm and a drainage system was installed to divert this flow westward to the area of W16 where a brackish marsh developed in 2009. By 2019, surface soil salinities had increased in the absence of freshwater inputs to exceed 80‰ and the cover of healthy plants detectable using aerial imagery had declined to < 10% in places in the high marsh (Page et al. 2020, Beheshti et al., in preparation).

In the case of module W10, vegetation developed most rapidly along the northwestern edge of the module adjacent to the river channel. This module was graded entirely to high marsh elevations with a relatively flat marsh plain. Soil salinities were moderate (< 50‰), probably due to seepage of river channel waters. Module W5 was graded almost entirely (~90%) as middle and low marsh, where vegetation cover increased over time through natural recruitment and growth. These findings suggest that plant recruitment, and the growth of planted plants in the high marsh, might be most successful in areas bordering lower salinity waters (W10), and where high marsh is narrow and sloping (W5), enhancing the seepage and drainage of riverine and tidal waters that ameliorate high soil salinity.

Soil properties other than salinity could influence vegetation development. Inorganic nitrogen inputs that might originate in the seepage of groundwater (e.g., Page 1995) or river water that could stimulate plant growth were not reflected in soil N measurements. Although we did not measure soil particle size at the time, an analysis of data collected by SCE contractors in 2008 did not find a significant difference in the percentage of sand, silt, and clay between modules W2/3 and W16, although there was a trend of higher clay content in the soil samples of W2/3 (Supplementary Material, Table S3). We did not observe other stressors in San Dieguito Wetlands that could have explained the poor performance of planted vegetation such as sedimentation, the burrowing of invertebrates, herbivory, or unusually sandy soil reported at other restored sites (Zedler et al. 2003).

Constructed Topography and Vegetation Development from Natural Recruitment

Vegetation cover increased below ~1.77 m MLLW only through natural recruitment, and a distinct boundary was visible between this naturally recruited vegetation and vegetation planted above the 1.77 m elevation contour in module W16 (Figures 6A, C). Our topographic and vegetation surveys revealed that ~1.77 m MLLW could be a potential “critical tide level” (sensu Doty 1946) in those modules where inundation time decreased abruptly above this elevation. This elevation exceeded the estimated Mean Higher High Water (MHHW) of 1.55 m MLLW in San Dieguito Wetlands (Elwany, pers. com.). MacDonald (1988) analyzed the relationship between tidal elevation and the maximum duration of both continuous submersion and exposure of the marsh plain using data from Mission Bay marsh (San Diego County). This analysis noted a marked increase in the maximum duration of continuous exposure above MHHW: short tidal submergences (< 6 h) on several consecutive days were followed by several weeks or months of continuous exposure. Periods of continuous exposure would increase at higher elevations that are inundated even less frequently, and elevations of the graded high marsh, 1.77–2.07 m MLLW, are rarely hit by the highest tides. Plants may respond more to changes in the maximum duration of continuous submersion or exposure than to average annual total submergence times (MacDonald 1988).

The low successful recruitment of vegetation above ~1.77 m MLLW was clearly visible in module W2/3 (Figures 2A, 4) and in the strip of bare space in W16 (Figure 8C). It is likely that an abrupt reduction in inundation frequency above this elevation could reduce seed delivery and expose seeds and juvenile plants to prolonged edaphic stresses of higher soil salinity, temperature, and low moisture that could affect the successful germination and postgermination survival of seedlings (Shumway and Bertness 1992, Zedler et al. 2003). Our data support the supposition that natural recruitment would be higher below than above 1.86 m MLLW, which was initially planned as the upper elevational boundary of middle marsh, and that habitat higher than this would require a planting program to achieve vegetation cover goals (SCE 2005).

Adaptive Management of the San Dieguito Wetlands Restoration Project

Lessons learned following construction of the San Dieguito Wetlands confirm the importance of topographic design for timely vegetation development (Zedler and Callaway 2001, Smith and Warren 2012, Brooks et al. 2015, Craft 2022, Twomey et al. 2024). The consequences of the shallow constructed slope in module W2/3 (1:289, 0.3%) over a high and wide marsh plain were dramatically evident in the persistently very low overall vegetation cover post-construction. A steeper slope is typically recommended to facilitate drainage of tidal waters. For example, Sheehan (2013) recommended a slope for the marsh plain ranging from 1:200 (0.5%) to 1:100 (1%) while Craft (2022) recommended a slope of 1:100 (1%) to 1:33 (3%). Without monitoring and adaptive re-grading to address the underperformance of vegetation, it is likely that overall vegetation cover would have remained sparse over a large proportion of the W2/3 module, and the desired overall comparability of the San Dieguito Wetlands Restoration Project to natural wetlands in the region that typically have > 85% cover (Smith et al. 2023) would be impeded.

Following the recommendations of geomorphologists, the elevation of the marsh plain in module W2/3 was regraded lower by approximately 15–30 cm in March 2014 to increase inundation and encourage vegetation development (Sheehan 2013). The re-grading shifted the 1.77 m MLLW contour further inland from the river channel, increasing the proportion of module below 1.77 m exposed to regular tidal flooding from 40 to 90%. The marsh plain was also re-graded from a relatively flat slope to a slope of 1: < 100 (Sheets 1 and 2, SCE 2014) towards the constructed creeks to facilitate drainage and increase the range of elevations flooded during most high tides. More recently, the implementation of measures to address factors limiting planting success at elevations higher than 1.77 m that included soil decompaction, irrigation with freshwater, and planting (Beheshti et al. 2023), has increased vegetation cover such that areas sparsely vegetated in 2012 had at least 30% cover in 2023 (Smith et al. 2023, Beheshti et al. 2024). Overall, vegetation cover wetland-wide is on a promising trajectory of increase towards values typical of other wetlands in the region (Beheshti et al. 2024). The consequence of reducing stressful salinities demonstrated by the initial rapid growth of planted plants in module W16, followed by their decline when freshwater seepage stopped, further supports the importance of ameliorating high soil salinity to achieve planting success.

Design considerations should include not only the elevation, width, and slope of the marsh plain, but surface heterogeneity that includes tidal creeks (Callaway 2005, O'Brien and Zedler 2006, Craft 2022). Other than the 30–50 m long linear precursor creeks that were expected to evolve into tidal creeks over time, there was no planned incorporation of tidal creek networks or other topographic heterogeneity into module W2/3. Tidal creeks can enhance flooding and drainage of the marsh plain and provide a variety of other functions including enhancing seed dispersal (Shi et al. 2020), and plant species richness, biomass, and production (Zedler et al. 1999, Sanderson et al. 2000, Schile et al. 2011), as well as providing habitat and food resources for invertebrates, fish, and birds (Zedler et al. 1992, Desmond et al. 2000, Mallin and Lewitus 2004).

We learned from the San Dieguito Wetlands Restoration Project that re-grading (in 2010) to extend the short preursor creeks in module W2/3 did not lead to an increase vegetation cover in the module. Without more extensive lowering and contouring of the adjacent high marsh plain, vegetation remained sparse above the 1.77 m contour through 2013 (Figure 3A–E). We observed that vegetation did recruit to the lower elevation edges of the tidal creeks prior to re-contouring in 2014, but vegetation development occurred primarily along the edges of creeks with little recruitment on the adjacent ungraded high marsh (Figures 2, 4C).

In conclusion, the large-scale San Dieguito Wetlands Restoration Project initially exhibited shortcomings in project design (Zedler and Callaway 2001, Craft 2022). This has extended the timeline of development of vegetation cover across the entire planned salt marsh beyond the 3–5 years noted as typical for salt marsh restoration projects, including some sites in southern California (Keer and Zedler 2002). Others have found variable rates of vegetation development (Ambrose et al. 2006) but rarely has vegetation cover remained < 10% for multiple years over large areas as was the case in the high marsh of module W2/3. The low natural recruitment and poor performance of planted vegetation at higher elevations in W2/3 necessitated major intervention in the form of re-grading in 2014 to lower the elevation and re-contour of the marsh plain, and a follow-up large-scale planting program to facilitate salt marsh development. Such interventions come with logistical constraints that include the ability to access areas without damaging portions of the restoration that are progressing as planned, as well as additional costs and ecological impacts. The need for adaptive management can be greatly reduced if the planned design incorporates the lessons learned outlined here pertaining to the effects of graded elevation on vegetation development. If a goal of the restoration is to achieve high vegetation cover within a few years, then grading to elevations below 1.77 m MLLW initially will facilitate achievement of this goal. If a goal is to have ~30% of the salt marsh as high marsh above 1.77 m, as originally planned, then vigorous, extensive intervention in the form of soil treatments and irrigation to ameliorate high soil salinity should occur in the early phases of the project (Beheshti et al. 2023). In the absence of adaptive management, vegetation cover would likely have remained sparse throughout the restoration project for many years, impeding the development of cover similar to that in natural wetlands in the region.

none

Northern Harrier (Circus hudsonius), a winter denizen of southern California’s coastal marshes. Ed. Tudor Jenks. 1911. The World of Nature. Chicago, IL: The Educational Society. The Florida Center for Instructional Technology, fcit.usf.edu.

Acknowledgements

We thank K. Patonai, E. Blair, K. Hill, J. Tay, and J. Wolfe (UCSB) for field and laboratory assistance, D. Huang (UCSB) for data assistance, M. Hess (Ocean Imaging) for aerial imagery and initial vegetation classification, N. Garrity and L. Sheehan (ESA-PWA) for advice on tidal creek design and the SONGS Mitigation Monitoring Program Science Advisory Panel (R. Ambrose, P. Raimondi, R. Schmitt) for advice. SCE conducted the grading and planting as required by the California Coastal Commission under SCE’s coastal development permit (No. 6-81-330-A, formerly 183–73) for Units 2 and 3 of the San Onofre Nuclear Generating Station (SONGS). Data are publicly available through the Environmental Data Initiative (EDI).

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/) permitting copying and distributing the material in any medium or format in unadapted form only, for noncommercial purposes only, provided the original work is properly cited

References

    1. Ambrose R.F.
    , Vance R.R., Diaz N. 2006. Wetland restoration monitoring report for Naval Base Ventura County, Mugu Lagoon: July 2001 to September 2005. Unpublished report.
    1. Beheshti K.
    , Smith R., Page M., Schroeter S., Reed D. 2024. Annual Report of the Status of Condition A: Wetland mitigation. San Onofre Nuclear Generating Station (SONGS) Mitigation Program. Submitted to the California Coastal Commission, July 2024. https://marinemitigation.msi.ucsb.edu/sites/default/files/documents/wetland/2023_annual_report-SONGS_wetland_mitigation.pdf
  3. ↵
    1. Beheshti K.M.
    , Schroeter S.C., Deza A.A., Reed D.C., Smith R.S., Page H.M. 2023. Large-scale field studies inform adaptive management of California wetland restoration. Restoration Ecology 31(5):e13936.
    OpenUrl
    1. Beller E.E.
    , Baumgarten S.A., Grossinger R.M., Longcore T.R., Stein E.D., Dark S.J., Dusterhoff S.R. 2014. Northern San Diego County Lagoons Historical Ecology Investigation: Regional Patterns, Local Diversity, and Landscape Trajectories. Richmond, CA: San Francisco Estuary Institute.
  5. ↵
    1. Bertness M.D.
    , Wise C., Ellison A.M. 1987. Consumer pressure and seed set in a salt marsh perennial plant community. Oecologia 71:190–200.
    OpenUrlPubMed
  6. ↵
    1. Brooks K.L.
    , Mossman H.L., Chitty J.L., Grant A. 2015. Limited vegetation development on a created salt marsh associated with over-consolidated sediments and lack of topographic heterogeneity. Estuaries and Coasts 38:325–336.
    OpenUrl
  7. California Coastal Commission (CCC) 1997. Adopted findings and conditions. Permit amendment and condition compliance. San Onofre Nuclear Generating Station (SONGS) coastal development permit 6-81-330-A (formerly 183-73) for Units 2 and 3.
  8. California Wetland Monitoring Workgroup (CWMW) 2023. California Wetland Program Plan 2023–2028.
    1. Callaway J.C.
    2001. Hydrology and substrate. Pages 89–117 in J.B. Zedler (ed.), Handbook for Restoring Tidal Wetlands. New York, NY: CRC Press.
    1. Callaway J.C.
    , Sullivan G., Desmond J.S., Williams G.D., Zedler J.B. 2001. Assessment and monitoring. Pages 271–335 in J.B. Zedler (ed.), Handbook for Restoring Tidal Wetlands. New York, NY: CRC Press.
  11. ↵
    1. Callaway J.C.
    2005. The challenge of restoring functioning salt marsh ecosystems. Journal of Coastal Research SI40:24–36.
    OpenUrl
  12. ↵
    1. Callaway R.M.
    , Jones S., Ferren Jr. W.R., Parikh A. 1990. Ecology of a Mediterranean-climate estuarine wetland at Carpinteria, California: Plant distributions and soil salinity in the upper marsh. Canadian Journal of Botany 68:1139–1146.
    OpenUrl
  13. ↵
    1. Craft C.
    , Reader J., Sacco J.N., Broome S.W. 1999. Twenty-five years of ecosystem development of constructed Spartina alterniflora (Loisel) marshes. Ecological Applications 9:1405–1419.
    OpenUrl
    1. Craft C.B.
    2022. Creating and Restoring Wetlands: From Theory to Practice, 2nd edition. Amsterdam, The Netherlands: Elsevier.
  15. ↵
    1. Cribari-Neto F.
    , Zeileis A. 2010. Beta regression in R. Journal of Statistical Software 34(2):1–24.
    OpenUrl
    1. Day J.
    , Burdick D.M., Ibanez C., Mitsch W.J., Elsey-Quirk T., Rivaes S. 2021. Restoration of tidal marshes. Pages 443–475 in D.M. FitzGerald and Z.J. Hughes (eds.), Salt Marshes. Cambridge, UK: Cambridge University Press.
  17. ↵
    1. Desmond J.S.
    , Zedler J.B., Williams G.D. 2000. Fish use of tidal creek habitats in two southern California salt marshes. Ecological Engineering 14:233–252.
    OpenUrl
  18. ↵
    1. Doty M.S.
    1946. Critical tide factors that are correlated with the verical distribution of marine algae and other organisms along the Pacific coast. Ecology 27:315–328.
    OpenUrlCrossRef
  19. ↵
    1. Emery N.C.
    , Ewanchuk P.J., Bertness M.D. 2001. Competition and salt-marsh plant zonation: Stress tolerators may be dominant competitors. Ecology 82:2471–2485.
    OpenUrlCrossRef
    1. Ferren Jr. W.R.
    1985. Carpinteria Salt Marsh: Environment, History, and Botanical Resources of a Southern California Estuary. Santa Barbara, CA: The Herbarium, Department of Biological Sciences, University of California, Santa Barbara.
  21. ↵
    1. Geissinger E.A.
    , Khoo C.L.L., Richmond I.C., Faulkner S.J.M., Schneider D.C. 2022. A case for beta regression in the natural sciences. Ecosphere 13(2):e3940.
    OpenUrl
  22. ↵
    1. Haltiner J.
    , Zedler J.B., Boyer K.E., Williams G.D., Callaway J.C. 1996. Influence of physical processes on the design, functioning and evolution of restored tidal wetlands in California (USA). Wetlands Ecology and Management 43:73–91.
    OpenUrl
  23. ↵
    1. Janousek C.N.
    , Thorne K.M., Takekawa J.Y. 2019. Vertical zonation and niche breadth of tidal marsh plants along the northeast Pacific Coast. Estuaries and Coasts 42:85–98.
    OpenUrl
  24. ↵
    1. Keer G.H.
    , Zedler J.B. 2002. Salt marsh canopy architecture differs with the number and composition of species. Ecological Applications 12:456–473.
    OpenUrl
  25. ↵
    1. Kennish M.J.
    2002. Environmental threats and environmental future of estuaries. Environmental Conservation 29:78–107.
    OpenUrlCrossRef
  26. ↵
    1. Lindig-Cisneros R.
    , Zedler J.B. 2002. Halophyte recruitment in a salt marsh restoration site. Estuaries 25:1174–1183.
    OpenUrl
  27. ↵
    1. Litzba U.
    , Schroeter S., Page H.M., Elwany M.H.S. 2011. Biological importance of small tidal channels in wetland restoration projects. Shore and Beach 79:17–27.
    OpenUrl
  28. ↵
    1. Liu Z.
    , Fagherazzi S., He Q., Gourgue O., Bai J., Liu X. et al. 2024. A global meta-analysis on the drivers of salt marsh planting success and implications for ecosystem services. Nature Communications 15:3643.
    OpenUrlPubMed
  29. ↵
    1. Lotze H.K.
    , Lenihan H.S., Bourque B.J., Bradbury R.H., Cooke R.G., Kay M.C. et al. 2006. Depletion, degradation, and recovery potential of estuaries and coastal seas. Science 312:1806–1809.
    OpenUrlAbstract/FREE Full Text
    1. MacDonald K.B.
    1988. Coastal salt marsh. Pages 263–294 in M.G. Barbour and J. Major (eds.), Terrestrial Vegetation of California. Special Publication No. 9. Davis, CA: California Native Plant Society.
  31. ↵
    1. Mahall B.E.
    , Park R.B. 1976. The ecotone between Spartina foliosa Trin. and Salicornia virginica L. in salt marshes of northern San Francisco Bay: II. Soil water and salinity. Journal of Ecology 64:793–809.
    OpenUrl
  32. ↵
    1. Mallin M.A.
    , Lewitus A.J. 2004. The importance of tidal creek ecosystems. Journal of Experimental Marine Biology and Ecology 298:145–149.
    OpenUrl
  33. Marathon Construction Corporation (MCC) 2009. Letter from Marathon Construction Corporation, 1008 Riverford Road, Lakeside CA 92040 to Coastal Environments, 2149 Avenida de la Playa, La Jolla CA 92037. Subject: Plant Performance Issues at the W2A, W2B and W3 modules. (March 30, 2009).
  34. MEC Analytical Systems, Inc. (MEC) 1993. San Dieguito Lagoon Restoration Project Biological Baseline Study, March 1992–May 1993. Volumes I and II. Draft.
  35. ↵
    1. Mitsch W.J.
    , Wilson R.F. 1996. Improving the success of wetland creation and restoration with know-how, time, and self-design. Ecological Applications 6(1):77–83.
    OpenUrlCrossRef
  36. ↵
    1. Mook D.H.
    , Hoskin C.M. 1982. Organic determinations by ignition: Caution advised. Estuarine, Coastal and Shelf Science 15:697–699.
    OpenUrl
  37. ↵
    1. Morzaria-Luna H.N.
    , Zedler J.B. 2007. Does seed availability limit plant establishment during salt marsh restoration?. Estuaries and Coasts 30:12–25.
    OpenUrl
  38. ↵
    1. Noe G.B.
    , Fellows M.Q.N., Parsons L., West J., Callaway J., Trnka S. et al. 2019. Adaptive management assists reintroduction as higher tides threaten an endangered salt marsh plant. Restoration Ecology 27:750–757.
    OpenUrl
  39. ↵
    1. O'Brien E.L.
    , Zedler J.B. 2006. Accelerating the restoration of vegetation in a southern California salt marsh. Wetlands Ecology and Management 14:269–286.
    OpenUrl
    1. Onuf C.P.
    , Quammen M.L., Shaffer G.P., Peterson C.H., Chapman J.W., Cermak J., Holmes R.W. 1979. An analysis of the values of central and southern California coastal wetlands. Pages 186–199 in P.E. Greeson, J.R. Clark and J.E. Clark (eds.), Wetland Functions and Values: The State of Our Understanding. Proceedings of the National Symposium on Wetlands Held in Disneyworld Village, Lake Buena Vista, Florida, November 7–10, 1978. Minneapolis, MN: American Water Resources Association.
    1. Onuf C.P.
    1987. The Ecology of Mugu Lagoon, California: An Estuarine Profile. U.S. Fish and Wildlife Service Biological Report 85(7.15).
  42. ↵
    1. Page H.M.
    1995. Variation in the natural abundance of 15N in the halophyte, Salicornia virginica, associated with groundwater subsidies of nitrogen in a southern California salt-marsh. Oecologia 104:181–188.
    OpenUrlCrossRefPubMed
    1. Page M.
    , Schroeter S., Reed D. 2020. Annual Report of the Status of Condition A: Wetland Mitigation. San Onofre Nuclear Generating Station (SONGS) Mitigation Program. Submitted to the California Coastal Commission, October 2020. https://marinemitigation.msi.ucsb.edu/sites/default/files/documents/wetland/2019_annualreport-SONGS_wetland_mitigation.pdf
  44. ↵
    1. Pennings S.C.
    , Callaway R.M. 1992. Salt marsh plant zonation: The relative importance of competition and physical factors. Ecology 73:681–690.
    OpenUrlCrossRef
  45. ↵
    1. Powell A.N.
    , Collier C.L. 1998. Reproductive success of Belding’s Savannah Sparrows in a highly fragmented landscape. The Auk 115:508–513.
    OpenUrlCrossRef
  46. ↵
    1. Purer E.A.
    1942. Plant ecology of the coastal salt marshlands of San Diego County, California. Ecological Monographs 12:81–111.
    OpenUrlCrossRef
  47. R Core Team 2024. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
  48. ↵
    1. Rand T.A.
    2000. Seed dispersal, habitat suitability and the distribution of halophytes across a salt marsh tidal gradient. Journal of Ecology 88:608–621.
    OpenUrlCrossRef
  49. ↵
    1. Raposa K.B.
    , Lerberg S., Cornu C., Fear J., Garfield N., Peter C. et al. 2018. Evaluating tidal wetland restoration performance using National Estuarine Research Reserve System reference sites and the restoration performance index (RPI). Estuaries and Coasts 41:36–51.
    OpenUrl
  50. ↵
    1. Rosencranz J.A.
    , Thorne K.M., Buffington K.J., Takekawa J.Y., Hechinger R.F., Stewart T.E. et al. 2018. Sea-level rise, habitat loss, and potential extirpation of a salt marsh specialist bird in urbanized landscapes. Ecology and Evolution 8:8115–8125.
    OpenUrl
  51. ↵
    1. Sanderson E.W.
    , Ustin S.L., Foin T.C. 2000. The influence of tidal channels on the distribution of salt marsh plant species in Petaluma Marsh, CA, USA. Plant Ecology 146:29–41.
    OpenUrl
    1. Sheehan L.
    2013. San Dieguito Wetlands Restoration Plan—Third Review. Memorandum to S. Schroeter and M. Page. Environmental Science Associates, 9191 Towne Center Drive, Suite 340, San Diego, CA 92122.
  53. ↵
    1. Schile L.M.
    , Callaway J.C., Parker V.T., Vasey M.C. 2011. Salinity and inundation influence productivity of the halophytic plant Sarcocornia pacifica. Wetlands 31:1165–1174.
    OpenUrl
  54. ↵
    1. Shi W.
    , Shao., Gualtieri C., Purnama A., Cui B. 2020. Modelling long-distance floating seed dispersal in salt marsh tidal channels. Ecohydrology 13(1):e2157.
    OpenUrl
  55. ↵
    1. Shumway S.W.
    , Bertness M.D. 1992. Salt stress limitation of seedling recruitment in a salt marsh plant community. Oecologia 92:490–497.
    OpenUrlCrossRefPubMed
    1. Smith R.
    , Beheshti K., Page M., Schroeter S., Reed D. 2023. Annual Report of the Status of Condition A: Wetland mitigation. San Onofre Nuclear Generating Station (SONGS) Mitigation Program. 73 pp. https://marinemitigation.msi.ucsb.edu/sites/default/files/documents/wetland/2022_annual_report-SONGS_wetland_mitigation.pdf
    1. Smith S.M.
    , Warren R.S. 2012. Vegetation responses to tidal restoration. Pages 59–80 in C.T. Roman and D.M. Burdick (eds.), Restoring Tidal Wetlands: A Synthesis of Science and Management. Washington, DC: Island Press.
  58. Southern California Edison (SCE) 2005. San Dieguito Wetlands Restoration Project. Final Restoration Plan. Submitted to California Coastal Commission.
  59. Southern California Edison (SCE) 2014. Letter and Figure 1 (Sheets 1 and 2) to Ms. Toni Ross, California Coastal Commission, San Diego District, San Diego, California 92108. San Dieguito Wetlands Restoration Project (Project)—W2/W3, Completion and Response to CCC January 21, 2014, Letter on Grading of W2/W3.
  60. Southern California Wetlands Recovery Project (SCWRP) 2018. Wetlands on the Edge: The Future of Southern California’s Wetlands: Regional Strategy 2018. Prepared by the California State Coastal Conservancy, Oakland, CA. 142 pp.
    1. Stein E.D.
    , Cayce K., Salomon M., Bram D.L., De Mello D., Grossinger R., Dark S. et al. 2014. Wetlands of the Southern California Coast: Historical Extent and Change Over Time. Southern California Coastal Water Research Project (SCCWRP) Technical Report 826, SFEI Report 720.
    1. Sullivan G.
    2001. Establishing vegetation in restored and created coastal wetlands. Pages 119–155 in J. Zedler (ed.), Handbook for Restoring Tidal Wetlands. New York, NY: CRC Press.
    1. Sverdrup H.U.
    , Johnson M.W., Fleming R.H. 1962. The Oceans: Their Physics, Chemistry, and General Biology. Englewood Cliffs, NJ: Prentice-Hall, Inc.
  64. ↵
    1. Thom R.M.
    , Williams G., Borde A., Southard H.J., Sargeant S., Woodruff D. et al. 2005. Coastal restoration: Where have we been, where are we now, and where should we be going. Journal of Coastal Research SI 40:94–108.
    OpenUrl
  65. ↵
    1. Twomey A.J.
    , Nunez K., Carr J.A., Crooks S., Friess D.A., Glamore W. et al. 2024. Planning hydrological restoration of coastal wetlands: Key model considerations and solutions. Science of the Total Environment 915:169881.
    OpenUrlPubMed
  66. ↵
    1. Valiela I.
    , Kinney E., Culbertson J., Peacock E., Smith S. 2009. Global losses of mangroves and salt marshes. Pages 107–138 in C.M. Duarte (ed.), Global Loss of Coastal Habitats: Rates, Causes and Consequences. Bilbao, Spain: Fundación BBVA.
  67. ↵
    1. Vivian-Smith G.
    2001. Developing a framework for restoration. Pages 39–88 in J. Zedler (ed.), Handbook for Restoring Tidal Wetlands. New York, NY: CRC Press.
  68. ↵
    1. Wang Q.
    , Xie T., Luo M., Bai J., Chen C., Ning Z., Cui B. 2021. How hydrological connectivity regulates the plant recovery process in salt marshes. Journal of Applied Ecology 58:1314–1324.
    OpenUrl
  69. ↵
    1. Williams P.B.
    , Orr M.K. 2002. Physical evolution of restored breached levee salt marshes in the San Francisco Bay Estuary. Restoration Ecology 10: 527–542.
    OpenUrl
  70. ↵
    1. Wolters M.
    , Garbutt A., Bekker R.M., Bakker J.P., Carey P.D. 2008. Restoration of salt‐marsh vegetation in relation to site suitability, species pool and dispersal traits. Journal of Applied Ecology 45: 904–912.
    OpenUrl
  71. ↵
    1. Xie T.
    , Cui B., Li S., Bai J. 2019. Topography regulates edaphic suitability for seedling establishment associated with tidal elevation in coastal salt marshes. Geoderma 337: 1258–1266.
    OpenUrl
  72. ↵
    1. Zedler J.B.
    , Callaway J.C. 1999. Tracking wetland restoration: Do mitigation sites follow desired trajectories?. Restoration Ecology 7: 69–73.
    OpenUrl
  73. ↵
    1. Zedler J.B.
    , Callaway J.C. 2000. Evaluating the progress of engineered tidal wetlands. Ecological Engineering 15: 211–225.
    OpenUrlCrossRef
  74. ↵
    1. Zedler J.B.
    , Callaway J.C. 2001. Tidal wetland functioning. Journal of Coastal Research SI27: 38–64.
    OpenUrl
  75. ↵
    1. Zedler J.B.
    , Kercher S. 2005. Wetland resources: Status, trends, ecosystem services, and restorability. Annual Review of Environment and Resources 30: 39–74.
    OpenUrlCrossRef
  76. ↵
    1. Zedler J.B.
    , Nordby C.S., Kus B.E. 1992. The Ecology of Tijuana Estuary: A National Estuarine Research Reserve. Washington, DC: NOAA Office of Coastal Resource Management, Sanctuaries and Reserves Division.
  77. ↵
    1. Zedler J.B.
    , Morzaria-Luna H., Ward K. 2003. The challenge of restoring vegetation on tidal, hypersaline substrates. Plant and Soil 253: 259–273.
    OpenUrlCrossRef
  78. ↵
    1. Zedler J.B.
    , Callaway J.C., Desmond J.S., Vivian-Smith G., Williams G.D., Sullivan G. 1999. Californian salt-marsh vegetation: An improved model of spatial pattern. Ecosystems 2: 19–35.
    OpenUrlCrossRef
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Lessons Learned and Value of Early Post-Construction Monitoring of a Large Tidal Wetland Restoration Project
Henry M. Page, Stephen C. Schroeter, Andres Deza, Russell Johnston, Rachel S. Smith, Kathryn M. Beheshti, Daniel C. Reed, Justin C. Hoesterey
Ecological Restoration Dec 2025, 43 (4) 260-275; DOI: 10.3368/er.43.4.260
Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords