Ecological dynamics

Preface: Different classes of tidal wetlands due to changes in salinity and tidal position

Salinity.
Situated between the land and the sea, tidal wetlands are defined in terms of salinity levels and by the vertical extent of the tides. Tidal wetlands include all wetlands subject to the influence of the tides yet vary in salinity, ranging from undiluted seawater to freshwater derived from inland sources such as rivers, runoff, and groundwater seeps. These broad variations in salinity create an intergrading sequence of tidal wetlands ranging from: (1) salt tidal marshes, to (2) brackish tidal wetlands, to (3) freshwater tidal wetlands. The relationship between salinity and wetland class can be approximated by Cowardin et al. (Cowardin et al., 1979; Theve, 2013). (In practice, the term “halinity” is sometimes used in place of salinity to denote ocean salts dominated by halide, NaCl. The units may be expressed in parts-per-thousand [ppt] or virtually equivalent Practical Salinity Units [PSU] (UNESCO, 1981) or commonly expressed as deciSiemans-per-meter [dSm-1]. An approximate conservative conversion is 1 dSm-1 = 0.640 ppt at 25 °C.) Salt marshes typically exhibit salinity levels greater than 18 ppt (28.1 dSm-1), considered polyhaline, 18 - 30 ppt (28.1- 46.9 dSm-1) and levels greater than 30 ppt (46.9 dSm-1), especially in situations where surface evaporation can concentrate salts. The salinity levels of brackish wetlands are more dilute ranging from 0.5 - 18 ppt (0.8 – 28.1 dSm-1), considered oligohaline, 0.5 – 5 ppt (0.8 – 7.8 dSm-1), to mesohaline, 5 – 18 ppt (7.8 – 28.1 dSm-1). Freshwater tidal wetlands exhibit salinity levels of 0.5 ppt (0.8 dSm-1) or less. Salt and brackish tidal marshes are considered estuarine habitats, and are directly influenced by tidal flooding. Salt marshes are found in sheltered, depositional areas located in coves, embayments, and behind barrier beaches. Brackish marshes exist either along landward seeps or drainageways of rivers and streams where fresh and saline waters mix. Freshwater tidal wetlands, while still physically affected by tidal forces, are beyond the reach of the salt front. Lacking salinity, freshwater tidal habitats are sometimes considered riverine habitats (Cowardin et al., 1979), yet other sources consider all tidal wetlands as estuarine (Odum et al., 1984; Odum, 1988). Freshwater tidal and brackish tidal wetlands are optimally developed along the lower reaches of large river systems with low gradients near the confluence with the sea. Furthermore, the floristic dissimilarities between these three classes of tidal wetlands provide a suitable justification for separate Ecological Sites.

Tidal position.
In the simplest terms, each class of tidal wetlands can be separated into two broad zones in reference to the vertical extent of the tides, 1) the “regularly” flooded zone (flooded daily on average); and 2) the “irregularly” flooded zone (flooded less than daily, i.e., only with spring tides). When projected onto the marsh profile these tidally influenced zones are commonly referred to as the “low” marsh (below Mean High Water, MHW) and the “high” marsh (above MHW). However, for the purposes of these ecological site descriptions, only in the case of the salt marsh, are the high marsh and the low marsh treated separately as two different ecological sites. While, brackish and freshwater tidal sites also exhibit similar shifts in plant distribution along the intertidal profile, the geographic extent is very limited, hence the ecological sites of brackish tidal wetlands and freshwater tidal wetlands are considered each in their entirety, aggregating a complex of vegetation-forms over the entire tidal profile.

Tidal salt high marshes of the northeastern United States
The ecological dynamics of the northeastern salt marshes are complex. Most plants have a very limited ability to grow in saline habitats except a small number of specially adapted plants, called “halophytes” (Flowers et al., 1986; Aslam et al., 2011). Furthermore, these same plants are also subjected to increasing physical harshness (decreasing redox potential and increasing salinity) associated with tidal flooding (Nichols, 1920; Miller and Egler, 1950; Niering and Warren, 1980; Bertness and Ellison, 1987). The mosaic pattern of vegetation across salt marshes of the northeast appears roughly in zones due to the limiting influence of physical factors, such as tidal flooding and salinity, and additional effects of both negative and positive biotic interactions among species, like competition and facilitation, respectively (Bertness, 1991a; b; Bertness and Shumway, 1993; Bertness and Hacker, 1994; Bertness and Leonard, 1997; Zhang and Shao, 2013). In addition, superimposed on this complex gradient are several disturbances to the high marsh including: relative sea-level-rise, waterlogging, salt stress, wrack, major storm events, ice-scouring/rafting, and herbivory. Furthermore, additional agents of salt marsh vegetation change are attributed directly or indirectly to human actions, including: accelerated sea-level rise, temperature warming, filling, ditching and draining, open-water-marsh-management (OMWM) and integrated-marsh-management (IMM), alteration of tidal exchange by road crossings, impoundment dikes, culverts, tide gates and other structures, fires, eutrophication, sudden vegetation “dieback”, and consumer control.

Biotic interactions: stress-tolerance / competition trade-off and disturbance-driven facilitation
Besides coping with the changing physical limitations of the intertidal gradient by way of stress and disturbance, plants also experience biotic interactions: competition or facilitation (Bertness and Shumway, 1993; Bertness and Callaway, 1994; Zhang and Shao, 2013).

Competition.
Ecological trade-offs between competition and stress-tolerance are opposing species strategies (Grime, 2006) that explain much of the salt marsh zonation. Since competition is often greatest where stress is minimal, salt marsh species are differentially limited landward by competition and seaward by physical harshness (Bertness, 1991a; b). This accounts for the competitive hierarchy of species observed in the typical northeastern salt marsh (reference state): the woody shrub, Iva frutescens, dominates the terrestrial border and competitively displaces Juncus gerardii to the upper reaches, which displaces Spartina patens to the meadow, which in turn displaces Spartina alterniflora to the low marsh. The same species are ranked in reverse for stress tolerance.

Facilitation.
Furthermore, the vegetation pattern is also the result of secondary succession following disturbance-driven facilitation (Bertness and Shumway, 1993). Facilitation is a positive interaction that occurs when a species ameliorates stress, making the environment more favorable for other, successive species (Connell and Slatyer, 1977; Bruno et al., 2003). Disturbance-generated bare patches and slight depressions known as “pannes” on the high marsh become hypersaline and or waterlogged. The initial colonizers (referred to as foundation species (Angelini et al. 2011)) are stress-tolerant forbs, such as Salicornia spp., or Spartina alterniflora or sometimes Distichlis spicata. The shade cast by these plants that limit salt accumulation, and the accretion of living peat that aerate the rhizosphere, are plant-induced processes that ameliorate the physical harshness of the sites, which can then lead to the succession of the less stress-tolerant competitive dominants, Spartina patens or Juncus gerardii (Bertness and Shumway, 1993; Bertness and Hacker, 1994; Bertness and Leonard, 1997).
In summary, facilitation is common in secondary succession under increased environmental stressors, but competition dominates under more benign physical conditions (Bertness and Shumway, 1993).

Relative Sea-level Rise
Under natural conditions, salt marsh elevations are generally near or approaching a dynamic equilibrium relative to rising sea level (Friedrichs and Perry, 2001; Morris et al., 2002). Historically, marsh-building has occurred both landward and seaward in southern New England during times of slow [historic] rates of sea-level rise at approximately 1.0 mm•yr-1 (Nixon, 1982; Donnelly et al., 2004) and marshes may appear to continue building during moderate rates of sea-level rise at approximately 2.5 mm yr-1 (Bricker-Urso et al., 1989; Donnelly et al., 2004; NOAA/NOS/CO-OPS, 2014). However, there is a concern that the accelerated rates of sea-level rise may eventually impose an upper limit or tipping point beyond the ability of coastal marshes to build and keep pace with sea-level rise (Bricker-Urso et al., 1989; Kirwan et al., 2010). (More details follow in the section on accelerated sea-level rise).

Waterlogging
Waterlogging within the marsh profile results from impediments to soil drainage due to changes in surface micro-relief (Miller and Egler, 1950; Niering and Warren, 1975; Nixon, 1982; Ewanchuk and Bertness, 2004). The surface of the high marsh is periodically interrupted with low depressions consisting of inadequately drained pannes, secondary pannes, permanently flooded pools and excavated pools. Waterlogging on the salt marsh generates a physically stressful habitat with anaerobic or anoxic (oxygen-deficient) conditions that challenge plant establishment, growth, and survival (Ewanchuk and Bertness, 2004; Tiner, 2013). Besides waterlogging, cycles of flooding and evaporative exposure create extremes in salinity, further contributing to the harshness of these habitats. Hence, extremely stressful pannes may be devoid of vegetation or covered with a mat of cyanobacteria or filamentous green alga, Cladophora gracilis. Intolerant to physical stresses, the main turf-forming graminoids (i.e., grass or grass-like plants), Spartina patens and Juncus gerardii are physically excluded. However, stress-tolerant forbs and short-form Spartina alterniflora are able to withstand some waterlogging and act as pioneering plants which persist for a while without competitive displacement (Bertness, 1991a, 1992; Ewanchuk and Bertness, 2004). These “safe-sites” for these species (Harper, 1977) persist as long as cycles of flooding and exposure maintain or renew the waterlogged nature of the site. However, these safe-sites may only serve as a temporary refuge. Often the mere occupation of a site by plants will have an ameliorating effect. Aeration of the rhizosphere and elevational accretion are plant-induced site modifications that can facilitate re-invasion by turf-forming graminoids (Bertness and Shumway, 1993; Bertness and Hacker, 1994; Tiner, 2013; Zhang and Shao, 2013).
Cyclical changes in Metonic high tides (18.6 year cycle) can affect plant life with increased tidal flooding frequency and duration (Tiner, 2013). The resulting prolonged waterlogging may instigate detrimental feedback, where limits to plant growth further limit vertical accretion which exacerbates waterlogging, etc. (Friedrichs and Perry, 2001). Waterlogging can be produced by relative sea-level rise, marsh subsidence, occluded creeks, micro-levees, and or reduced sediment load (Niering and Warren, 1975; Friedrichs and Perry, 2001; Ewanchuk and Bertness, 2004; Adamowicz and Roman, 2005). Waterlogging has also been implicated as a common stress associated with dieback (Smith and Carullo, 2007; Alber et al., 2008). The prevalence of severe waterlogging, i.e., drowning, may be symptomatic of accelerated sea-level rise (Warren and Niering, 1993; Hartig et al., 2002) and crossover to a different ecological site entirely (e.g., the regularly flooded, “tidal salt low marsh”).

Salt stress (salinity extremes)
Salinity in salt marshes can be quite variable (Tiner, 2013). Dilution can occur during freshwater inputs from high precipitation events producing upland runoff and groundwater seepage, and concentration can occur during cycles of tidal flooding and evaporation. In general, salinity gradients produced by these processes contribute to the physical stress gradient intensifying across the marsh from upland to open water. In addition, hyper-saline conditions can occur anywhere in disturbed bare patches or pannes (Nichols, 1920; Miller and Egler, 1950; Niering and Warren, 1980). Most plants have a very limited ability to grow in saline habitats except a small number of specially adapted plants, called “halophytes” (Flowers et al., 1986; Aslam et al., 2011). Accordingly, most salt marsh plants tolerate moderate salinity, but even fewer plants can withstand the elevated salinity of bare patches and pannes ranging from 35 to 50 ppt (54.7 – 78.1 dSm-1) and above (Niering and Warren, 1980). The ability to withstand conditions such as salt stress (and waterlogging), enables stress-tolerant plants such as Salicornia spp., stunted Spartina alterniflora, and other forbs a chance to escape immediate competition from the less stress-tolerant turf-forming graminoids, Spartina patens and Juncus gerardii. However, these refugia are often temporary. As these plants fill the site, their shade will limit evaporation and salt accumulation, thereby reducing physical stress, and ultimately making the site suitable for less stress-tolerant but more competitive plants (Bertness and Shumway, 1993; Zhang and Shao, 2013).
Predicted climate warming (IPCC, 2013b) may create hyper-saline conditions, ranging in effects from salt pannes to greater high marsh facilitation and dominance by turf building graminoids (Gedan and Bertness, 2009).

Wrack
The most common natural disturbance in Northeastern US salt marshes is the stranding of wrack. Deposited by extreme high and storm tides, the smothering action of wrack kills vegetation and renders the patch bare. In Northeastern US tidal marshes, wrack is made up mostly from decaying litter of Spartina alterniflora, and to lesser extent of Zostera marina (eelgrass). Wrack disturbance covers 6 to 15 percent of the high marsh surface annually (Hartman, 1984; Bertness and Ellison, 1987), albeit, the areas actually damaged by wrack are usually less if the wrack is not persistent (Valiela and Rietsma, 1995). The extent and distribution varies considerably over the marsh, but persistent wrack does accumulate more in the upper border (Miller and Egler, 1950; Bertness and Yeh, 1994), providing opportunities for Iva frutescens spp. oraria (maritime marsh elder) establishment. The damage facilitated by the smothering action of wrack is also a mechanism for secondary panne formation, and provides colonization opportunities for Distichlis spicata (saltgrass) and competitively subordinate plants (Brewer et al., 1998). Forbs frequently establish by seeds, whereas Distichlis spicata colonizes vegetatively (Bertness and Ellison, 1987).

Ice scouring and ice rafting
With increasing latitude, more northern salt marshes are progressively exposed to the effects of two different types of winter ice disturbances, ice scouring and ice rafting, which can trigger vegetation change. Winter ice damages low marsh, but ice also can impact the high marsh to some degree. Ice scouring is not as frequent as wrack disturbance, but generally more severe, leaving behind ice scars and hollows that persist for years where the marsh peat was removed or plucked (Ewanchuk and Bertness, 2003). Ice rafted debris was noted as a major source of sediment in Maine tidal marshes (Wood et al., 1989). Localized erosion and deposition from ice disturbances, enhance the surface microtopography and create bare areas that develop secondarily into pannes and pools.

Major storm events
Coastal storms have been associated with both marsh erosion and marsh accretion. In addition, subsurface effects such as subsidence can result (Cahoon, 2006). Changes in marsh topography and position have marked influence on the distribution of marsh plants. Direct wave erosion (scouring) of the marsh surface happens during periods of extreme high tide, but impacts are minimized due to the damping of flow through marsh vegetation. More commonly, direct erosion occurs along the creeks crossing the high marsh where strong tidal currents can undermine banks or along seaward bayfronts where there is heavy wave action (Friedrichs and Perry, 2001). Instead, storm related erosion is most often associated with events of ice-scouring and ice rafting in northern New England (Ewanchuk and Bertness, 2003) and the marsh-smothering aftermath of wrack deposition (Niering and Warren, 1980; Bertness, 1992) that leads to the creation of waterlogged pannes and pools which may degrade the interior of the marsh. Conversely, most sediment deposition in tidal marshes are attributed to coastal storm events (Harrison and Bloom, 1977; Stumpf, 1983; Gedan et al., 2009). During extreme high tides and storm surges that overtop the marsh are called “overwash”, and loads of sediment deposited are called “washovers”. The washovers often form progressive bands parallel to the marsh edge, potentially changing the microtopography of the marsh profile. If severe enough, washovers of sufficient material and magnitude, can convert the tidal marsh to “dunelands”, an entirely different ecological site. Otherwise, if the storm surge is more like a spring tide, deposition are controlled by the creeks, which scour creek and ditch bottoms and build levees (Friedrichs and Perry, 2001). The sedimentary record in a Rhode Island marsh revealed at least 7 hurricane-magnitude storms within the past 700 years (Donnelly et al., 2001). Climate change scenarios predict that the intensity of major storm activity will increase in the North Atlantic (IPCC, 2013a).

Herbivory
Herbivory is not usually considered a significant factor in the structuring of saline high marsh plant communities in the northeast. Although, pronounced insect herbivory has been reported in patches of Salicornia sp. (Ellison, 1987) and seed set reduction of 50 percent upon common marsh perennials (Bertness and Ellison, 1987). Yet, herbivory is more important in the low marsh, dominated by Spartina alterniflora, where triggered by indirect effects of human disturbance, significant grazing impacts have been noted by geese, snails, and crabs and insects (see later section about effects of Consumer Control).

Accelerated Sea Level Rise – vertical accretion and lateral migration
Stratigraphic evidence and tide gage information has indicated that sea-level rise in New England accelerated during the last century from about 1.0 mm yr-1 (Nixon, 1982; Donnelly et al., 2004) to a rate of approximately 2.6 mm yr-1 (1938 – 2006, USGS New London tide gage no. 8461490). The rate may be increased to 4.5 mm yr-1 if adjusted for the latest gage data trend segment from 1987-2013 (Boon, 2012; Sallenger et al., 2012). AR5, the Fifth Assessment Report, (IPCC, 2013b) predicts that the rate of global mean sea level (GMSL) rise may conservatively increase to 4.5 mm yr-1 by mid-century and continue to accelerate to 11 mm yr-1 or even as high as 16 mm yr-1 by the end of the 21st century in response to a decadal warming trend ranging from 0.08 to 0.14 °C. These recent trends and projections in accelerated sea level rise due to climate change (global warming ) (Varekamp and Thomas, 1998; Boon, 2012; IPCC, 2013b) have caused concern about (1) whether vertical salt marsh accretion can keep pace (Reed, 1995; Kirwan et al., 2010), (2) whether the horizontal development of marshes has sufficient neighboring space upon which to laterally migrate. Any significant changes in the vertical or lateral extent of the tidal marsh can lead to a state change in the vegetation.

Vertical accretion.
Vertical stability of the tidal marsh’s surface elevation is achieved when wetland surface elevation gains are approximately the same rate of sea level rise. Elevation gains are driven by the accumulation of organic matter (the persistent buildup of belowground vegetation growth and litter, or soil O- horizons) and the accumulation of inorganic sediment, minus any losses attributed to erosion, subsidence (i.e., decomposition, and compaction) (Morris et al., 2002; Cahoon et al., 2006; Fagherazzi et al., 2012). To precisely document vertical processes, a network of “Surface-Elevation Table and Marker Horizon” (SET-MH) set-ups exist throughout New England and elsewhere (Cahoon et al., 1999, 2006). The SET-MH results are used as a measure of marsh resiliency or vulnerabilityassociated with rising sea level and to document changes to vegetation states and to ecological sites. Depending on the location and geomorphic setting, the contribution of vertical processes of marsh accretion varies. In New England, peaty organic matter accumulation dominates vertical accretion (Bricker-Urso et al., 1989; Turner, 2011). Organic matter contributes more significantly to vertical marsh building on a volume basis than sediment accumulation, building “elevation capital” to buffer increase in sea level (Cahoon and Guntenspergen, 2010). Predictions of warmer temperatures and longer growing season may also enhance plant productivity and decomposition with the net effect of adding gains to offset marsh losses to rising sea levels (Charles and Dukes, 2009; Kirwan and Megonigal, 2013). However, as sea level continues to rise, organic matter accrual is eventually limited by “drowning’ when increasing tidal flood levels no longer allow plant growth (Morris et al., 2002; Kirwan et al., 2010). Yet increasing flood levels relative to lowered marsh elevation favor an increase in the rate of inorganic sediments (Friedrichs and Perry, 2001; Morris et al., 2002) which now become more important than organic matter accrual. (Fagherazzi et al., 2012; Kirwan and Megonigal, 2013). Due to low sediment concentrations, the peaty organic matter-based marshes of the northeastern United States may be less resilient to sea level increases than more sediment laden salt marshes in the southeastern and gulf regions of the United States (Gedan and Bertness, 2010; Weston, 2014). For example, models predict a threshold sea level rise rate of 5 mm yr-1 for the peat marshes in the Plum Island Estuary, MA assuming suspended sediment concentrations remain low (Kirwan et al., 2010). In New England, evidence of submerging marshes is revealed by shifts and disappearance of vegetation currently observed in southweatern Long Island Sound (Tiner et al., 2006), Jamaica Bay (Hartig et al., 2002), and Cape Cod (Smith, 2009). These predictions suggest that once the elevation capital is spent, these high marshes will convert to the “tidal salt low marsh ecological site”.

Lateral migration.
Another aspect of sea level rise is lateral migration, where salt marshes expand into adjacent low lying wetlands, up rivers, and onto uplands. This process, known as “marine transgression”, has continued at varying rates since the last glaciation (Redfield, 1972; Niering and Warren, 1980; van de Plassche, 1991). With accelerated sea level rise, the tidal marshes may appear to be retreating in areas where the low marsh boundaries and vegetation zones are shifting landward (Niering and Warren, 1980; Donnelly and Bertness, 2001) along with the encroaching-landward transgressive edge of the tidal marsh. However, any sea level induced losses can only be offset by transgression if suitable lowlands exist and human barriers for marsh transgression are minimal (Kirwan and Megonigal, 2013). Spatially explicit models showing future marsh migration locales are being used for insight into the into coastal planning and resource management (Fagherazzi et al., 2012). Sea Level Affecting Marshes Model (SLAMM) is a spatial GIS model that simulates the dominant processes involved in wetland conversions and shoreline modifications during long-term sea level rise (Mcleod et al., 2010). Distributions of wetlands are predicted under conditions of accelerated sea level rise, and results are summarized in tabular and graphical form; http://warrenpinnacle.com/prof/SLAMM/index.html. The Nature Conservancy’s Coastal Resilience Project includes a GIS tool that identifies coastal areas suitable for migration, as well as potential barriers to migration, and the relative importance of different marshes based on their potential to provide ecosystem services. In New England, the migration potential has been mapped for marshes in Long Island and Connecticut http://coastalresilience.org/geographies/new-york-and-connecticut ; http://maps.coastalresilience.org/nyct/#, (Hoover et al., 2010).

Temperature warming
In addition to driving accelerated sea level rise (IPCC, 2013b), temperature warming more subtly affects the composition and distribution of salt marsh vegetation directly and by extending growing season. In spite of waterlogging due to rising sea levels (Warren and Niering, 1993), higher temperatures, can affect greater evaporation, cause hypersalinity, and create drier conditions, as demonstrated by latitudinal and experimental studies (Pennings and Bertness, 1999; Bertness and Ewanchuk, 2002; Charles and Dukes, 2009; Gedan and Bertness, 2009). Depending on the rate of sea level rise and the marsh elevation, two scenarios are likely: 1) greater cycles of flooding and evaporation generating more forb or salt pannes at the expense of high marsh meadow (Warren and Niering, 1993; Pennings and Bertness, 1999; Kirwan and Megonigal, 2013), or 2) generating marsh meadow , at least temporarily, as the evapotranspiraton and productivity of forbs, lessen the stress of waterlogging and salinity, only to be replaced by facilitating turf-building graminoids such as Spartina patens (Gedan and Bertness, 2009, 2010).

Filling
Filling coastal marshes for land reclamation projects ranks among the most radical and greatest impact because the process almost always results in permanent habitat loss and conversion to a variety of “non-tidal and upland ecological sites”. Losses due to dredge and fill operations were common to improve navigation and expand seaports and marinas (Rozsa, 1995; Tiner, 2013). Other more subtle impacts include incidental fill and indirect hydrologic impacts resulting from road crossings and impoundment dikes that interfere with normal tidal flow causing a decrease in sediment supply to the marsh surface and a decrease in vertical accretion (Kennish, 2001). Anecdotal accounts often state coastal wetland losses at 50 percent (Gosselink and Baumann, 1980). Some tidal marsh assessments have estimated 30 percent loss for the Long Island Sound from 1880 to 1970 (Dreyer et al., 1995) and 37 percent for parts of Maine, New Hampshire, Massachusetts, and Rhode Island (from roughly 150 to 200 years ago until present) (Bromberg and Bertness, 2005). Coastal marsh losses in the greater Boston region is estimated at 81 percent (Bromberg and Bertness, 2005). It may be questionable whether those estimates for coastal salt marsh losses fully account the widespread filling that resulted from expansion of the maritime economy of New England in the mid 1700’s, the industrial revolution in the early 1800’s, and the full extent of coastal development over the last 250 years (Nixon, 1982).

Ditching
Historically, New England salt marshes have been extensively ditched and drained for salt hay farming and mosquito control (Miller and Egler, 1950; Redfield, 1972; Niering and Warren, 1980; Nixon, 1982). Conventional salt marsh haying appeared to have had little effect upon the ecology except promoting Spartina patens over Spartina alterniflora. However, many hayers also “worked” the marshes, meaning ditching and draining the marsh surface to further favor Spartina patens, (Buchsbaum et al., 2009). The most intensive ditching occurred in the 1930s, by the Civil Works Administration and Works Progress Administration, for the purposes of mosquito control and employment (Rozsa, 1995; Gedan et al., 2009). The type of ditch network most commonly excavated in New England was a pattern of successive straight parallel ditches running from upland to open water (sometimes called grid ditches even if cross-ditching was not evident). The intent of ditching was to decrease mosquito populations at the source by removing standing water where mosquito larvae develop; a practice that has since been judged of questionable or minimal benefit (Nixon, 1982; Tonjes, 2013). Currently, 94 percent of New England marshes have been ditched (Crain et al., 2009). Historically, pannes and pools were more common features on the high marsh and thus the scarcity of salt marsh pools today is a legacy of intensive ditching for mosquito control and salt hay operations (Ewanchuk and Bertness, 2004; Adamowicz and Roman, 2005).
Ditching is a major hydrological alteration to the salt high marsh, but less so to the low marsh. Ditches circumvent the natural drainage system with a network of shortcuts that increase the marsh surface area enhanced by tidal flooding and drainage that would not ordinarily be affected by tidal flooding except for extreme highs (Tonjes, 2013). The effects of disrupting the normal tidal fluctuations are dependent on the site tide range and density of the ditch network (Friedrichs and Perry, 2001). Yet there are some generalities, for example, newly cut ditches lessen waterlogging with peat drainage up to five meters away from ditches (Crain et al., 2009). By altering hydrologic conditions, ditching has pronounced effect on the vegetation pattern. Ditching impacts the surface drainage and microtopography plus drains the marsh pools, eliminating Ruppia maritima (Miller and Egler, 1950; Adamowicz and Roman, 2005). New ditches will ameliorate waterlogging favoring more competitive turf species such as Spartina patens and Juncus gerardii over forbs and Spartina alterniflora (Bertness, 1991a, 1992; Ewanchuk and Bertness, 2004). Just as creeks do, ditches also direct sedimentation to form levees that impede and convey surface drainage (Miller and Egler, 1950; Warren and Niering, 1993). Once large expanses of high marsh dominated by Spartina patens now appear as fragmented wetlands growing Iva frutescens and potentially invasive Phragmites australis on the levees that have built naturally by deposition or on levees artificially constructed of persistent spoils deposited ditchside or creekside. Ironically, the legacy of parallel ditching has often created large very pannes of stunted Spartina alterniflora and Salicornia spp. sometimes filling the inter-ditch area (Miller and Egler, 1950; Niering and Warren, 1980; Barrett and Niering, 1993). Ditches also create low marsh habitats supporting Spartina alterniflora along ditch banks and in aggraded ditches (Miller and Egler, 1950). Hence, ditches have expanded and accelerated sudden vegetation die-backs by increasing the area of susceptible low marsh habitat (Alber, Swenson, Adamowicz, & Mendelssohn, 2008; Coverdale, Bertness, & Alteiri, 2013). Enhanced ditch circulation translates to potential nutrient enrichment and eutrophication (Koch and Gobler, 2009; Tonjes, 2013) favoring either Spartina alterniflora or the invasive Phragmites australis in less waterlogged, less saline situations (Amsberry et al., 2000; Tiner, 2013).

Fire
Fire may play an important role in some coastal marsh ecosystems (Nyman and Chabreck, 1995), but the significance of fire is uncertain in high marshes of the northeastern US. Given the importance of early native American fires, speculation suggests that high marshes may have been burned (Miller and Egler, 1950). Today, accidental fires are more common to Phragmites australis brackish tidal wetlands (Roman et al., 1984).

Eutrophication (nutrient loading)
Eutrophication, particularly nitrogen inputs originating from shoreline development and removal of adjacent vegetation buffers is contributing to shifting changes in the salt marsh vegetation (Silliman et al., 2009; Gedan et al., 2011). Since most salt marsh plants are nitrogen limited, eutrophication disrupts the competitive hierarchy/stress and disturbance gradients associated with the typical zonation of the Northeastern US salt marsh (Emery et al., 2001). This process affects the expansion of Spartina alterniflora into the high marsh displacing Spartina patens and the proliferation of Phragmites australis at the upper boarder displacing Juncus gerardii (Levine et al., 1998; Bertness et al., 2002). Eutrophication is also implicated in reduced marsh accretion and subsidence leading to greater panne formation. Although nitrogen additions can increase aboveground biomass, eutrophication reduces the nutrient seeking belowground root and rhizomatous biomass volume and strength, ultimately leading to plant collapse, and potentially drowning (Valiela et al., 1976; Turner et al., 2009; Gedan et al., 2011). However some nitrogen enriched salt marshes already effected may not exhibit reduced biomass and elevation loss (Anisfeld and Hill, 2012). Eutrophication can trigger consumer control by herbivorous insects which can suppress primary production by as much as 50 percent (Bertness and Silliman, 2008).

Open Marsh Water Management and Integrated Marsh Management
Open water marsh management (OMWM) was a response of managers to address the long-term failures of ditching to more effectively control mosquitoes. The concept of OMWM is to re-create irregularly flooded saltmarsh pools that were eliminated by ditching to provide habitat for larvivorous mosquito-eating fish, notably the mummichog, Fundulus heteroclitus (Rochlin, James-Pirri, Adamowicz, Dempsey, et al., 2012). OMWM semi-tidal runnels (shallow ditches), pools, and larger ponds are created using rotary excavators and the selective plugging of existing ditches, an altogether uncertain methodology (Vincent et al., 2013). OMWM may have less impact on the physical, hydrological, and biological character of the tidal marsh than grid ditching. However, pools created from plugged ditches are not the same as natural pools. OMWM produces greater surface water, lower redox potential, low soil bulk density, and greater subsidence. Species associated with waterlogging are Spartina alterniflora and forb panne species, Triglochin maritimum, Plantago maritima, Atriplex patula, Gerardia maritima, Limonium carolinianum, and Sueda spp. (Vincent et al., 2013).
Integrated marsh management (IMM) is a multi-purposeful approach to salt marsh ecosystem management that uses best management practices of salt marsh restoration while also incorporating various mosquito control methods (Rochlin, James-Pirri, Adamowicz, Wolfe, et al., 2012). The main goal of IMM is to restore or emulate native salt marsh conditions using mostly geomorphic alterations to achieve tidal flow restoration, vegetation management, and aspects of OMWM. However, since IMM an OMWM projects tend to be highly physically disruptive, a systematic program is strongly recommended to first evaluate and always monitor the impacts and benefits to the vegetation (Potente, 2007).

Tidal restriction and exclusion due to causeways, impoundment dikes, culverts, and tide gates
Tidal restriction limits coastal marshes from essential marine tidal influences, which alters the physical and biotic conditions on the marsh leading to change in the vegetation. For example, tidal restrictions like dikes and impoundments are constructed across coastal wetlands and tidal waterways to manipulate coastal marshes for many reasons, e.g., salt haying, tide mill ponds, wildlife management, flood control, or to build transportation causeways, like road crossings (Kennish, 2001). Tidal restrictions may involve adjustable water control structures such as weirs, flap gates, and self-regulating gates. Additionally, tidal restrictions may simply be the result of undersized or inappropriately positioned road culverts (Roman et al., 1984). Berms and dikes that prohibit sheet flow of flood tides across the marsh surface coupled with one-way “flapper” gates are some of the most restrictive structures. Marsh habitats behind constructed dikes are effectively isolated from marine influence and degrade due to reduced accretion, lowered water tables, and reduced salinity. Reduced salinity results in in the invasion of brackish marsh species (Roman et al., 1984) such as cattails, (Typha latifolia and Typha x glauca) and bulrushes (Schoenoplectus spp.) as well as the common reed (Phragmites australis or possibly the native genotype, Phragmites australis ssp. americanus). In extreme situations, if drained, the aerated organic matter will decompose, oxidize pyrite (iron sulfides), and acidify the soil (Portnoy and Giblin, 1997). These acid sulfate soils, also called “cat-clays”, are severely acid enough to prohibit plants and may appear barren. Furthermore, interrupting tidal flow and increasing surface water impoundment not only collapses the marsh but reduces the “elevation capital” which diminishes the ability of the marsh to keep pace with sea level rise and contributes to salt marsh habitat loss over time (Gedan et al., 2009). Tidal restriction is an acute problem in the industrialized northeastern United States and a primary focus of restoration (Burdick and Roman, 2012).

Marsh sudden vegetation “dieback”
Sudden vegetation dieback, SVD, also known as simply “dieback”, is a condition that has been described as the loss or death of emergent vegetation in salt marsh systems. It was originally characterized by standing dead vegetation only, rather than sites that were over-grazed, wracked, ice-sheared, etc. (Alber et al., 2008). While the nature of marsh dieback is clear, the causes of dieback is poorly understood and controversial (Linthurst, 1979; Alber et al., 2008; Holdredge et al., 2009; Smith, 2009; Elmer and Marra, 2011; Anisfeld and Hill, 2012). Marsh dieback is not usually associated with the salt high marsh although exceptions exist on the tidal wetlands of Cape Cod, MA (Smith, 2009). Marsh dieback is a significant issue in the salt low marsh sites impacting Spartina alterniflora.

Consumer Control (subsidized or unchecked herbivory)
Among the impacts to salt marshes attributed to human activity, vegetation losses due to indirect and subtle factors, such as consumer control are less understood. Triggered by human disturbance, herbivory by plant consumers either goes unchecked or gets subsidized, resulting in significant vegetation loss in areas where plant consumers were not historically important (Power, 1992; Pace et al., 1999; Bertness and Silliman, 2008). Examples of consumer control triggered by human disturbance, include significant grazing impacts by geese, snails, and crabs and insects, mostly upon Spartina alterniflora, especially in the low marsh and only occasionally in the high marsh.
Salt marsh eutrophication from enriched discharges and runoff, enhance insect herbivory by grasshoppers, Conocephalus spartinae, and leafhoppers, Prokelisia marginata, reducing the vegetation by nearly 60 percent (Bertness et al., 2008; Sala et al., 2008).
Fueled by fertilizer subsidies to agricultural fields and golf courses, populations of gregarious and migratory geese inflated to nuisance levels. In such situations, snow geese, Chen (Anser) caerulescens, have denuded expansive marshes in the Subarctic (Jefferies et al., 2006) and to a lesser extent southeastern US (Smith and Odum, 1981), and Canada geese, Branta Canadensis, overgraze saltmarshes throughout New England (Buchsbaum et al., 1981; Teal, 1986).
On Cape Cod and perhaps elsewhere, recreational fishing has depleted predatory fish, allowing increases in the populations of the native, herbivorous purple marsh crab, Sesarma reticulatum, that completely denude the high marsh ditches and low marsh of Spartina alterniflora (Holdredge et al., 2009; Altieri et al., 2012; Coverdale, Herrmann, et al., 2013).
The waterlogged condition anticipated with accelerated sea-level rise is expected to heighten the grazing pressure of the long-introduced common periwinkle, Littorina littorea, (Bequaert, 1943) on Spartina alternifora along high marsh creekbanks and the low marsh (Bertness, 1984; Tyrrell et al., 2008).

State Transitions and Ecological Site Transitions
Several transitions occur that transform the typical salt high marsh reference condition. Two are state transitions that result directly from hydrological alteration and tidal restriction. Both states are subject to restoration actions. These state transitions are featured by appropriate numbers at the core of the state-and-transition model and within appropriate sections of the corresponding narrative. Four additional transitions are warranted that result in crossovers to entirely different ecological sites. These are external transitions that lead to, or, lead from, other ecological sites. (Unfortunately, current database limitations do not easily account for levels beyond a state transition. Therefore, an alternative solution is to simply include external transitions in the diagram without numbers or boxes and add a short description in the narrative.) External transitions to other ecological sites are nonetheless important to the ecological dynamics of the salt marsh ecosystem. These external ecological site transitions include highly significant yet commonplace processes or disturbances, and are listed as follows:

Transition to irreversibly Filled ecological sits of various types
The deposition of fill, human-transported material, to the extent of excluding the tides, creates non-tidal and upland ecological sites of various types.

Transition to Coastal Dunelands ecological site
Infrequent storms of severe magnitude create huge surges that deposit large volumes of sand in overwash. The resulting coastal dunelands ecological sites may be un-vegetated or sparsely vegetated with Ammophila breviligulata (American beachgrass), or only occasionally vegetated at lower elevations with Spartina patens (saltmeadow hay).

Transition from Tidal Brackish Wetland ecological site to Tidal Salt High Marsh ecological site.
Increasing flood stress associated with accelerated sea level rise results in the displacement of the brackish tidal wetland, dominated by Panicum virgatum (switchgrass), Typha spp. (cattails), and or Phragmites australis (common reed), by the tidal salt high marsh ecological site. Iva frutenscens ssp. oraria (maritime marsh elder) may also pioneer here.

Transition from Tidal Salt High Marsh ecological site to Tidal Salt Low Marsh ecological site
A shift up and landward in the regularly flooded (flooded daily) intertidal zone associated with accelerated sea level rise results in greater flood stress leading to the displacement of the Tidal Salt High Marsh ecological site to the Tidal Salt Low Marsh ecological site, dominated by Spartina alterniflora (smooth cordgrass).

The ecology of North Eastern tidal wetlands is well studied. Accounts of the vegetation and dynamics reported are derived largely from the literature and augmented by field sampling.

The information contained in this Ecological Site Description (ESD) and State and Transition Model (STM) were developed using historical data, professional experience, and scientific studies. The information presented is representative of a very complex set of plant communities. Not all scenarios or plants may be included. Key indicator plants, animals and ecological processes are described to inform land management decisions.

Additional information for Tidal Salt High Marsh R144AR002CT is based on original field observations and relevé plots sampled. The field work was conducted by the staff of State Soil Office 12-TOL, and NRCS Ecologist. Additional relevé plots were provided courtesy of VegBank, the CT Natural Diversity Database (Natural Heritage Program), and CT Department of Energy and Environmental Protection, Wildlife Division.