Historically, salt marshes experienced a host of direct anthropogenic stressors ranging from modification of hydrology through mosquito ditching and ditch-plugging practices (Meredith et al. 1985, Wolfe 1996, Vincent et al. 2014, Burdick et al. 2020), fill-in for urbanization and infrastructure (Broomberg and Bertness 2005, Gedan et al. 2009), and development at the immediate upland edge (Bozek and Burdick 2005, Pontee 2013). Over the past thirty years, sea level rise (SLR) has been recognized to pose the largest threat to salt marsh systems on the East Coast in the United States. Recent estimates of rapid SLR rates, however, may outpace salt marsh elevation gain (Crosby et al. 2016), which can further disrupt marsh building by increasing plant mortality and decreasing belowground biomass production (Payne et al. 2019). Increased flooding can also be exacerbated by limited sediment supplies (Ganju et al. 2017) and legacy effects of prior agricultural alterations (Mora and Burdick 2013, Adamowicz et al. 2020, Smith et al. 2021). Large loss of coastal wetlands from SLR has been documented in Maryland (Schepers et al. 2020), Mississippi Delta (Day et al. 2011), New England (Watson et al. 2017), New York (Smith et al. 2021), and Virginia (Burns et al. 2021).
One of the main pathways for marsh loss to SLR is oversaturation, short-form Spartina alterniflora panne formation, and subsequent creation and expansion of pools on the interior marsh platform (Redfield 1972, Vincent et al. 2013, and Raposa et al. 2017). Compared to low marsh areas immediately along tidal creeks and shorelines, the higher elevation marsh platform is especially vulnerable to deterioration associated with reduced inorganic sediment inputs away from creeks (Temmerman et al. 2003, 2005, and Fagherazzi et al. 2013) and declines of belowground biomass inputs from dominant graminoids caused by increased flooding (Watson et al. 2016, Payne et al. 2019). Oversaturation of the interior marsh platform increases stressors on the vegetation community including anoxia, increased salinity, and build-up of hydrogen sulfide concentrations in the root zone (Van Huissteden and Van De Plassche 1998, Mendelssohn and Morris 2000, Berkowitz et al. 2018, Himmelstein et al. 2021). Over time high marsh graminoids (S. patens, Distichlis spicata, Juncus gerardii) die off and are replaced by more stress-tolerant short-form S. alterniflora and Salicornia spp. or open water (Warren and Niering 1993, Raposa et al. 2017) with less consolidated soils and shallow standing water (Berkowitz et al. 2018, Burdick et al. 2020). Conversion of high marsh meadows to short form S. alterniflora pannes and pools has been attributed to population declines of endemic species such as Ammospiza caudacauta (Saltmarsh Sparrow) as well as A. nelsoni (Nelson’s Sparrow) (Gjerdum et al. 2005, Shriver et al. 2016).
Small pools with stable, vertical banks are natural features on the marsh interior (Redfield 1972, Adamowicz and Roman 2005), however, coastal ecologists have recently been documenting large swaths of interior platform being lost to pool expansion, especially at lower elevations and flat slopes (Wilson et al. 2014, Ganju et al. 2020). Continued vegetation die-off and decrease in elevation (e.g., surface erosion, subsidence, and organic matter decomposition) may “push” the platform over a tipping point to vegetation collapse and pool formation (DeLaune et al. 1994, Day et al. 2011, Chambers et al. 2019). Runaway pool collapse occurs when the inorganic sediment supply is insufficient for the marsh to keep pace with SLR and stressful biogeochemical conditions at pool edges causes further vegetation die-off and bank collapse (Mariotti 2016, Mariotti et al. 2020, Himmelstein et al. 2021). The interior marsh can convert to large areas of standing water as smaller pools expand and merge over time (Himmelstein et al. 2021). Although pool recovery has been well-documented after reconnection with a tidal creek (Wilson et al. 2014, Smith and Pellew 2021), coastal ecologists have identified insufficient sediment inputs, microtidal ranges (< 1 m), and high tidal prisms as factors that further exacerbate pool expansion after tidal reconnection (Ganju et al. 2017, Schepers et al. 2017, 2020, Vinent et al. 2021).
Coastal ecologists have implemented a restoration strategy, termed runnels, to improve drainage of oversaturated soils without the impacts of over-aeration of the peat soil column (Wigand et al. 2017). Runnels are shallow vegetated swales (15–30 cm wide, 20–30 cm deep) constructed through areas of standing water or at the edge of pools which connect to the nearest hydrologic channel (ditch or creek). Improvements in drainage across the salt marsh platform is expected to decrease biogeochemical stressors and allow for eventual recolonization of high marsh graminoids. Besterman et al. (2022) documented a pattern of vegetation recovery over seven years after runnel construction in Rhode Island: (1) initial drainage of standing water and exposing bare ground, (2) colonization of S. alterniflora and Salicornia spp., and (3) replacement of low marsh vegetation with high marsh graminoids of S. patens, Distichlis spicata, and Juncus gerardii. Additionally, Perry et al. (2021) observed revegetation across salt marsh platforms as well as no net increase in peat oxidation after runnel construction. On longer time scales, revegetation and drainage enhancement of the salt marsh surface may lead to gains in elevation.
Runnelling is an additional tool for coastal ecologists and land managers to conserve and improve the resiliency of salt marsh systems. Adapted from Open Marsh Water Management methods to enhance tidal flow of mosquito breeding depressions (Wolfe et al. 2021), runnels have only been recently implemented and monitored for the purpose of rebuilding and improving high marsh vegetation communities. Questions remain about the efficacy across different tidal ranges, sediment inputs, and tidal prisms as well as maintenance requirements over time (Besterman et al. 2022). Recent applications of runnels have focused on recently converted low marsh pannes, where elevation and vegetation losses are less severe than marshes converting to pools (see Perry et al. 2021, Besterman et al. 2022). We constructed runnels in Spring 2015 at two large pools in the Parker River National Wildlife Refuge (Newbury, MA) to reconnect tidal exchange and in doing so, arrest pool expansion and enhance revegetation of the salt marsh surface. Both pools have been present on the landscape as early as 1965 based on historic aerial imagery (United States Geologic Survey - EarthExplorer), however evidence of recent expansion spurred restoration efforts. By 2014, one year before runnel construction, the north and south pools had expanded to sizes of 0.60 ha and 3.52 ha and perimeters of 327 m and 1536 m, respectively. We monitored hydrology, vegetation, and elevation over seven growing seasons to document the efficacy of the runnels and the trajectory of the pool and adjacent marsh platform recovery. Additionally, we used remote sensing analysis of publicly available aerial imagery to document system-wide vegetation recovery over time.