Interactions between plant traits and sediment characteristics influencing species establishment and scale-dependent feedbacks in salt marsh ecosystems

Geomorphology 250 (2015) 298–307

Contents lists available at ScienceDirect

Geomorphology journal homepage: www.elsevier.com/locate/geomorph

Interactions between plant traits and sediment characteristics influencing species establishment and scale-dependent feedbacks in salt marsh ecosystems C. Schwarz a,b,c,⁎, T.J. Bouma b, L.Q. Zhang a, S. Temmerman c, T. Ysebaert b,d, P.M.J. Herman b a

State Key Laboratory of Estuarine and Coastal Research, East China Normal University, Shanghai 200062, China Royal Netherlands Institute for Sea Research (NIOZ-Yerseke), 4400 AC Yerseke, The Netherlands c Ecosystem Management Research Group, Universiteit Antwerpen, Universiteitsplein 1, B-2610 Wilrijk, Belgium d Institute for Marine Resources and Ecosystem Studies, Yerseke, The Netherlands b

a r t i c l e

i n f o

Article history: Received 23 October 2014 Received in revised form 12 September 2015 Accepted 20 September 2015 Available online 25 September 2015 Keywords: Saltmarsh Biogeomorphology Habitat modification Scale dependent feedbacks Sediment

a b s t r a c t The importance of ecosystem engineering and biogeomorphic processes in shaping many aquatic and semiaquatic landscapes is increasingly acknowledged. Ecosystem engineering and biogeomorphic landscape formation involves two critical processes: (1) species establishment, and (2) scale-dependent feedbacks, meaning that organisms improve their living conditions on a local scale but at the same time worsen them at larger scales. However, the influence of organism traits in combination with physical factors (e.g. hydrodynamics, sediments) on early establishment and successive development due to scale-dependent feedbacks is still unclear. As a model system, this was tested for salt marsh pioneer plants by conducting flume experiments: i) on the influence of species-specific traits (such as stiffness) of two contrasting dominant pioneer species (Spartina alterniflora and Scirpus mariqueter) to withstand current-induced stress during establishment; and ii) to study the impact of species-specific traits (stiffness) and physical forcing (water level, current stress) on the large-scale negative feedback at established tussocks (induced scour at tussock edges) of the two model species. The results indicate that, not only do species-specific plant traits, such as stiffness, exert a major control on species establishment thresholds, but also potentially physiologically triggered plant properties, such as adapted root morphology due to sediment properties. Moreover, the results show a clear relation between speciesspecific plant traits, abiotics (i.e. sediment, currents) and the magnitude of the large-scale negative scaledependent feedback. These findings suggest that the ecosystem engineering ability, resulting from physical plant properties can be disadvantageous for plant survival through promoted dislodgement (stem stiffness increases the amount of drag experienced at the root system), underlying the importance of scale-dependent feedbacks on landscape development. © 2015 Elsevier B.V. All rights reserved.

1. Introduction In recent years the importance of biogeomorphic feedbacks in shaping landscape evolution has become increasingly clear (Hupp et al., 1995; Phillips, 1995; Stallins, 2006; Murray et al., 2008). Biogeomorphic feedbacks emerge if ‘ecosystem engineers’ influence geomorphologic (i.e. landscape-shaping) processes. Ecosystem engineers are organisms, able to actively (allogenically) or passively (autogenically) influence their immediate abiotic surroundings (Jones et al., 1994, 2010; Bouma et al., 2013), invoking feedback loops that can improve living conditions for themselves and potentially associated species (Jones et al., 1994; Mullan Crain and Bertness, 2006). If such feedbacks affect geomorphologic processes, they become biogeomorphic feedback loops, by which ⁎ Corresponding author at: University of Antwerp, Department of Biology, Ecosystem Management Research Group, Universiteitsplein 1, 2610 Wilrijk, Belgium. E-mail address: [email protected] (C. Schwarz).

http://dx.doi.org/10.1016/j.geomorph.2015.09.013 0169-555X/© 2015 Elsevier B.V. All rights reserved.

ecosystem engineers can shape various coastal and inland habitats, such as tidal wetlands (D'Alpaos et al., 2007; Temmerman et al., 2007), alluvial floodplain rivers (Murray and Paola, 2003; Tal and Paola, 2007) and fluvial hillslope systems (Istanbulluoglu and Bras, 2005). Since biogeomorphic feedbacks require the presence of ecosystem engineering organisms (i.e. plants) their occurrence is dependent primarily on the establishment probability of their initial units (e.g. plant seeds and seedlings). Further, following establishment, the occurrence of biogeomorphic feedbacks typically requires a critical biomass threshold to be surpassed (scale-dependent feedback), with no feedback present at lower densities (Bouma et al., 2009a,b; Friess et al., 2012). This raises the question (1) how biotic (i.e. plant traits) and abiotic (i.e. sediment, currents) factors influence the initial establishment in these biogeomorphic landforms, and (2) how the subsequent development from small, just established plants into larger-scale units (plant patches) depends on biotic-abiotic properties inducing biogeomorphic feedbacks.

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

1.1. Initial establishment of salt marsh seedlings Seedling establishment is a crucial step in disturbance-driven ecosystems (e.g. salt marshes, mangroves, alluvial floodplains and river macrophytes) to enable vegetation succession. Previous studies have focused on the impact of hydrodynamics (i.e. hydroperiod, currents and waves), salinity, temperature and soil moisture on seedling establishment (Scott et al., 1996; Patterson et al., 1997; Corenblit et al., 2007). Only recently, it was shown that initial seedling settlement is strongly dependent on interactions between sediment dynamics (i.e. erosion/accretion) and the magnitude of hydrodynamic stress (i.e. currents, waves). It was shown that periods of low hydrodynamic stress (i.e. windows of opportunity) can facilitate temporary establishment of seedlings on mudflats. Although temporarily established, these seedlings still stay vulnerable to changes in sediment dynamic and hydrodynamic properties (Bouma et al., 2009b; Balke et al., 2011, 2013; Infantes et al., 2011; Gurnell et al., 2012). The magnitude of this vulnerability was, for mangrove trees, shown to be strongly dependent on species specific plant traits (physiological plant properties such as growth velocity) (Balke et al., 2013), but data for non-woody pioneer species (as present in salt marshes) are still lacking. Moreover, sediment properties might influence this relationship, because both plant growth rate and plant-hydrodynamics interactions may depend on sediment type. To our knowledge, this aspect has not been tested to date. 1.2. Subsequent development from seedlings to plant patches Once established, plants tend to form round patchy vegetation units, further referred to as tussocks, scattered over the habitat (e.g. mudflat, channel reach) (Castellanos et al., 1994; Allen, 2000; Langlois et al., 2003; Schoelynck et al., 2012). At this development stage (i.e. intermediate scale) the interaction between plant traits and abiotic factors (e.g. currents and sediment properties) continues to be important (Vandenbruwaene et al., 2011). However, their impact shifts from influencing survival probability to controlling the plant's ability to grow, expand and ecosystem engineer its environment (van Wesenbeeck et al., 2008b; Schwarz et al., 2011). Tussocks exhibit a small-scale positive feedback (improving living conditions) through reducing hydrodynamics and hence promoting sediment accretion within their patches (Bouma et al., 2009b; Balke et al., 2012). This is in opposition to a larger-scale negative feedback, where flow-routing induced erosion (i.e. scouring) deteriorates living conditions at the tussock edges (van Wesenbeeck et al., 2008a; Bouma et al., 2009a). The occurrence and magnitude of these so-called scale-dependent feedbacks often depends on abiotic conditions, hence exhibiting abiotic context-dependency (Jones et al., 2010). Abiotic context-dependency describes whether the structural change generated by the ecosystem engineer leads to abiotic change. Previous studies have shown that the abiotic context dependency of scale-dependent feedbacks in a geomorphological context is influenced by estuarine-scale processes (e.g. tidal flow induced transport of sediment-rich estuarine water into and over the salt marsh), as well as local site-specific processes (e.g. sediment transport to and from the salt marsh depends on factors such as: local velocity of water flow; size, density of the sediment particles and bottom topography) (Allen, 2000; Wolanski et al., 2004). In this study we are focusing on scale-dependent feedbacks and their dependence on local (i.e. salt marsh) site-specific processes. As previously shown, factors as heterogeneity of bottom topography (Schwarz et al., 2014) and species properties (Langlois et al., 2003) influence local scale-dependent feedbacks. However knowledge on specific processes precipitating the dependency of local scale-dependent feedbacks on abiotic and biotic conditions, although already acknowledged in theoretical literature, is still lacking (Jones et al., 2010; Wolanski et al., 2004).The occurrence of scaledependent feedbacks (small-scale positive and large-scale negative feedback) is an important factor influencing the transition from small established plant units into larger landscapes and therefore are

299

important agents for landscape development (D'Alpaos et al., 2007; Temmerman et al., 2007; Vandenbruwaene et al., 2011; van de Koppel et al., 2012). Apart from a single hydrodynamic study on scale-dependent feedback strength on real plants without measuring actual sediment dynamics/erosion (Bouma et al., 2013), and flume studies on patches made of artificial rods (Zong and Nepf, 2010; Meire et al., 2014), we are not aware of studies addressing this issue, despite its importance for understanding the development of large-scale landscapes that are sculptured through biogeomorphic feedbacks. 1.3. Study approach In coastal tidal wetlands the biogeomorphic feedbacks between plants, water-flow and sediment, where the exerted friction of aboveground plant stems modifies the water-flow and its entrained sediment, have been determined to be the main factors driving large-scale landscape evolution (D'Alpaos et al., 2007; Temmerman et al., 2007). For this reason coastal wetlands are ideal model ecosystems to study impacts of biogeomorphic feedbacks on landscape evolution. The questions of initial establishment and subsequent development are assessed by analysing these processes for two different (i.e. with respect to their properties) but co-occurring salt marsh pioneer species, stiff Spartina alterniflora (further referred to as Spartina) and the flexible Scirpus mariqueter (further referred to as Scirpus). The two species differ in physical properties of the above-ground material (stiffness, maximum plant height, maximum stem diameter), the below-ground material (root network organization) and in physiological properties such as growth rate and stress tolerance (e.g. salinity) (Sun et al., 2001; He et al., 2012). We specifically investigated: (1) how initial seedling establishment thresholds (small scale) depend on the interaction between abiotic traits (sediment properties, i.e. cohesiveness) and species-specific plant traits (stiffness, root growth) in a stressed environment (current stress), and (2) how interactions between abiotic and species-specific traits (stiffness) influence the magnitude of the large scale negative feedback (induced scour at tussock edges, intermediate scale) in a stressed environment (current stress).

Finally, we discuss their implications on salt marsh-scale species establishment and expansion patterns. 2. Methods 2.1. Species and field characterization S. alterniflora and S. mariqueter are the most abundant pioneer halophytes present on salt marshes in the Yangtze estuary, China (Fig. 1). They differ in physical properties (e.g. Spartina: stiff, max. Shoot height 3 m; Scirpus: flexible, max. Shoot height 0.8 m), growth season (Spartina: May to October, Scirpus: April to November), stress tolerance (e.g. salinity and anoxia) and competitive strength (growth form and growth rate favours Spartina over Scirpus in direct competitive interactions) (Sun et al., 2001; Li et al., 2009; Schwarz et al., 2011; He et al., 2012). Scirpus is endemic to the Yangtze estuary and used to be the most abundant species observed in pioneer zones and low salt marshes. Spartina is an invasive North American species first found in the Yangtze estuary in 1990 and thereafter started to outcompete the endemic species (Huang and Zhang, 2007). On Chongming Island, the biggest Island in the Yangtze Estuary (31.6619°N, 121.4780°E), Spartina was planted in the northeast in 2001. Thereafter, it began to spread across existing Scirpus salt marshes accounting for 50% coverage of vegetated intertidal area in 2005 (Wang et al., 2006). The predominant spreading direction

300

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

Fig. 1. Field site, Chongming Island, Yangtze estuary, China; the hatched area shows the salt marshes present in its eastern parts. The right side shows a false colour aerial image from 2014 showing Spartina in the north depicted by lighter pixels and Scirpus in the south, depicted by slightly darker pixels.

of Spartina on Chongming Island was northwards (into areas characterized by muddy, cohesive sediment) whereas only limited spreading from the planted sites towards the south (into areas characterized by sandy, non-cohesive sediment) could be detected (Fig. 1) (Li et al., 2009; Wang et al., 2010). 2.2. Initial establishment of salt marsh seedlings 2.2.1. Seedling growth conditions Seeds from Spartina and Scirpus were collected from salt marshes on Chongming Island, Yangtze estuary, China and brought to The Netherlands. They were germinated and cultivated in a climate chamber, which was held at 25 °C and provided with an average of 12 h d−1 of 550 μmol m−2 s−1 photosynthetic active radiation (PAR) and a semi-diurnal tide mimicking field conditions. Seedlings were planted at 1 cm depth below the sediment surface in four different sediment types in individual PVC pipes (160 mm height and 120 mm diameter), with open bottoms and placed in large polyethylene bags (Fig. 2). The composition of the four different sediment types was chosen

following those present on the pioneer zones (i.e. mud flat — salt marsh transition) in the Yangtze estuary. Artificial silt, sand and clay (Sibelco benelux®) with a standardized grain size were combined with a nutrient mix (Osmocote®) to mimic 3 types of natural sediments (further referred to as mud, muddy sand, sandy mud). The properties of these mixed sediment types were; mud (sand:silt:clay — 0:60:40) with a D50 of 3.5 μm, muddy sand (sand:silt:clay — 37:52:11) with a D50 of 35.6 μm and sandy mud (sand:silt:clay — 49:43.5:7.5) with a D50 of 60.3 μm. The mineralogical composition consisted of a mixture between kaolinite, sericite/illite and quartz. The fourth sediment type (sand) was taken from a mudflat in the Oosterschelde estuary (The Netherlands) (sand:silt:clay — 96:3:1) with a D50 of 184.9 μm, also subsequently mixed with Osmocote® to ensure a surplus of nutrients. Fig. 3 shows the comparison of the chosen sediment mixtures to sediment samples taken in the pioneer zones in the Yangtze estuary. The sediments were all mixed with the same volume of water and kept waterlogged throughout the experiment. The duration of the growth phase was chosen to be 33 days, representing the average time available for seedlings before the onset of the disturbed period in the Yangtze estuary

Fig. 2. Experimental set-up to test initial establishment (right) and subsequent development (left). In the initial establishment experiment the critical vertical erosion threshold was assessed, by eroding sediment around seedlings of Spartina and Scirpus (adding discs at the bottom of the grown seedlings, and removing protruding sediment at the top with four different sediment types) and subsequently exposing them to current stress. The influence of plant traits in subsequent development was tested, by measuring patch-adjacent erosion of our two plant species (Spartina and Scripus) during current stress on a sand bed.

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

301

pairwise, comparisons were carried out to identify the pairs that significantly differ from each other using the Mann–Whitney U test with a 5% significance level (MATLAB software package). 2.3. Subsequent development from seedlings to plant patches

Fig. 3. Composition (sand:silt:clay) between our chosen sediment mixtures (large symbols) compared with sediment samples from salt marsh pioneer zones at various locations in the Yangtze estuary (small black dots), the black horizontal line indicates the empirical cohesive/non-cohesive boundary [van Ledden et al., 2004].

(Schwarz et al., 2011). During their growth phase all treatments experienced a semi-diurnal 1.5 h flooding, simulating inundation on a tidal flat. 2.2.2. Flume experiment on critical seedling erosion thresholds The flume at the NIOZ-Yerseke (Royal Netherlands Institute for Sea Research) consists of a 17.5 m-long and 0.6 m-wide oval channel that can produce currents and waves (Bouma et al., 2009b; Balke et al., 2011). At the double bottom test section, the cut PVC pipes filled with sediment were inserted flush with the flume bed, leaving only the seedling exposed (Balke et al., 2011, 2013; Han et al., 2012). No significant scouring was observed. Flume water depth was maintained at 32 cm. A constant current of 0.35 m s−1, representing the averaged field situation (Schwarz et al., 2011), was applied. For seedlings resisting the drag imposed by this current, the critical vertical erosion threshold was determined. The critical vertical erosion threshold was defined as the amount of sediment that needed to be removed from around the seedling, and after which the current caused the seedling to topple (Fig. 2). This definition was used as in a field situation toppling over would be followed by burial under sediment and therefore death of the seedling. The amount of critical erosion for seedling toppling was quantified by raising the sediment incrementally through placing PVC discs of 1 to 3 mm thickness underneath the pipe. After the addition of a disc, sediment at the top of the PVC pipe was gently removed using a water jet to keep the sediment level with the flume bottom. Subsequently, seedlings were put back into the flume and exposed to the same current stress (0.35 m s−1, see above) (Balke et al., 2013). This cycle was repeated until toppling occurred. For Scirpus, which also grew vegetative ramets, erosion cycles were continued until all ramets toppled over (Fig. 2). The differences in toppling between the two plant species were tested on significance using the Mann–Whitney U with a 5% significance level (MATLAB software package). After toppling, seedlings were carefully removed from the sediment and their biomass was measured. Above- and below-ground biomass was separated and the dry weight (drying until constant weight at 80 °C) determined. For below-ground biomass three subcategories were chosen: 0 to 2 cm sediment depth, 2 to 4 cm sediment depth and deeper than 4 cm. Differences in biomass allocation between the four different sediment classes at each plant species were compared using the nonparametric Kruskal–Wallis test at each of the above-mentioned subdepth (0–2 cm, 2–4 cm, 4 cm-end). In case of significant differences

Sediment and flow dynamics were investigated in and adjacent to vegetation patches in a flume experiment. In the set-up, half the width of the flume was filled by a patch of vegetation (Fig. 2). To assess sedimentation and erosion patterns in relation to physical plant properties, we cultivated Spartina (shoot density: 800 m−2, average stem diameter: 3.62 ± 0.75 mm) and Scirpus (shoot density: 1500 m−2, average stem diameter: 2.37 ± 0.48 mm) in their natural densities (as found in the Yangtze estuary) from seed to a height of 33 cm (Sun et al., 2002; Li and Yang, 2009). These densities also resulted in similar frontal plant areas facing the incoming current for Scirpus (0.018 m2) and Spartina (0.019 m2). To be able to investigate sediment dynamics also in front of and behind the vegetation patch, a 1 m-long and 30 cm-wide patch of vegetation was placed adjacent to the flume wall in the middle of the 2 m test section of the NIOZ flume (Fig. 2) (Bouma et al., 2005b, 2009a). For these experiments, we only used the above-mentioned sand category, since sand erosion thresholds are lower compared to cohesive sediment types, enhancing the potential occurrence of scour. As flow could move freely through and around the plants, this set-up allowed us to study how flow and subsequently local sediment transport are influenced by plant species traits (stiffness). A bare sediment control was also investigated. Current velocity was measured using an acoustic Doppler flow sensor (Nortek Vectrino) at different heights above the bottom to achieve a depth averaged flow velocity during the experiment of 0.35 m s−1. Two scenarios of water levels were used, one in which the plants were emergent (water level 0.22 m) and one in which the plants were fully submerged (0.37 m). In both scenarios flow velocity measurements at different longitudinal locations along the gap at 1 cm above the bottom (Figs. 2, 7c, diamond symbols) were carried out. The significance of velocity differences between these two scenarios was assessed using the Mann–Whitney U test with a 5% significance level (MATLAB software package). Erosion and deposition were measured after 30 min exposure to free flow conditions, via the distance scan mode of the acoustic Doppler sensor. The spatial grid had a resolution of 50 mm in the x- (longitudinal, parallel to the flow) and y- (transversal, cross-stream) directions, with a sub-mm vertical resolution. Erosion and deposition within the vegetation patch were not measured due to logistical constraints, and our focus was on the strength of the negative scale-dependent feedback (patch-adjacent erosion). The percent (%) of flow affected by vegetation was calculated by multiplying the average stem diameter by plant density and plant height and subsequently relating this area to the cross-sectional wet flume area. For Scirpus the height of the bent plant height was used, according to observations during the flume experiment. 3. Results 3.1. Initial establishment of salt marsh seedlings The erosion resistance of seedlings differed markedly between species and sediment types. In the mud sediment treatment Spartina could withstand significantly more erosion than Scirpus before toppling over (p = 0.039). As we reduced the degree of cohesiveness (i.e., clay content; mud to muddy sand to sandy mud treatment), Spartina and Scirpus exhibited similar critical erosion thresholds (Fig. 4). After further reduction in clay content (sand treatment), the situation reversed, with Scirpus becoming significantly more resilient against sediment erosion under hydrodynamic stress than Spartina (p = 0.016) (Fig. 4). The critical erosion threshold of Scirpus was linked to the biomass density in deeper sediment depths, showing significant differences in biomass

302

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

0.0016 m/submerged = 0.0013 m). When vegetation was present, sediment elevation changed and sediment was eroded adjacent to the vegetation patch. In the emergent scenario (water level did not exceed vegetation height), the eroded volume adjacent to the patch was in the same range for both species (0.0147 m3/m2 for Scirpus and 0.0139 m3/ m2 for Spartina; Scirpus/Spartina = 1.058). In the submerged scenario (water level exceeded vegetation height) the treatment with the stiff plant (Spartina) eroded about 40% more sediment adjacent to the patch, than the treatment with the flexible plant (Scirpus) (Scirpus/ Spartina = 0.589) (Fig. 6). It was further visible that the location of the erosion zone differed between the plant species. For both water levels, the flexible vegetation exhibited the onset of a scouring hole further downstream compared to the stiff vegetation (Fig. 6). Elevation within the vegetation patch was, due to logistical constraints, measured in presence of the plants. Although the measured values (in all treatments) within the patch suggested the occurrence of sediment accretion, we had to exclude this area from further calculations due to high measurement errors. Fig. 4. Critical vertical erosion threshold for Spartina alterniflora and Scirpus mariqueter seedlings grown on different bottom sediments; filled markers represent average values, whiskers represent the standard deviation.

allocation deeper than 4 cm between the muddy sand and sand (p = 0.011) treatment and the sandy mud and sand treatment (p = 0.013) (Fig. 5). For the cohesive sediment treatments (mud, muddy sand and sandy mud) Scirpus had a tendency for higher biomass density in the highest sediment layer (0 cm to 2 cm) compared to Spartina (Fig. 5). This consequently led to a lower biomass density in the deepest sediment layer (deeper than 4 cm). In all treatments Spartina was prone to higher biomass density in the deepest sediment layer than Scirpus. Nevertheless, the trend in critical erosion threshold with sediment type was not linked to below-ground biomass density for Spartina, which does not show significant differences between treatments (Fig. 5). A comparison of absolute biomass values of Spartina and Scirpus between treatments showed no significant differences in either aboveor belowground biomass, probably due to the high variability. 3.2. Subsequent development from seedlings to plant patches 3.2.1. Sediment erosion adjacent to the plant patch Bare sediment controls exhibited no significant change in sediment elevation (average elevation difference: emerged =

3.2.2. Flow around the plant patch Flow velocity measurements at different longitudinal locations along the gap at 1 cm above the bottom (Fig. 7a–c) showed flow acceleration in the downstream direction of the vegetation patch's leading edge for both species at all water levels. In the submerged scenario the flow velocity increased about 33 ± 1.8% (of the initial reference value) for Spartina and 29 ± 2.3% (of the initial value) for Scirpus. In the emerged scenario increases were 22 ± 4.1% for Spartina and 24 ± 4.8% for Scirpus. We did not observe significant differences in velocity at any position along the gap in the emerged scenario. In the submerged scenario, the gap flow velocity of the Spartina treatment appeared to be significantly different from the Scirpus treatment according to the performed Mann–Whitney U test (5% significance level; p-value/position: 0.036/0.55 m; 0.018/0.90 m; 0.030/1.25 m; 0.013/1.60 m; Fig. 7a–c). The flow area blocked by vegetation was calculated, using the projected stem density, stem diameter, plant height (stiff) and the height of the bent plant (flexible) (Fig. 7d). In the emerged scenario approximately the same area of the water column was blocked by the two species, whereas in the submerged scenario the Scirpus vegetation blocked about 40% less flow than Spartina (Fig. 7d). 4. Discussion Previous studies made it increasingly clear that biogeomorphic feedbacks play a key role in shaping many aquatic and semi-aquatic landscapes. These studies either focus on the entire landscape (e.g. Vandenbruwaene et al., 2012; Wang and Temmerman, 2013), individual vegetation patches (intermediate scale) (e.g. Balke et al., 2012; Zong and Nepf, 2012; Bouma et al., 2013) or initial establishment units (small scale) (e.g. Bouma et al., 2009b; Engels et al., 2011; Balke et al., 2013). This study is a first attempt of a comprehensive approach evaluating factors influencing the emergence of biogeomorphic feedbacks at two scales. Specifically we show that interactions between sediment properties and species traits are important for seedling survival (small scale) and the magnitude of scale-dependent feedbacks adjacent to established vegetation patches (intermediate scale). Both processes are essential for the occurrence of biogeomorphic feedbacks and their implications on landscape development as previously demonstrated through modelling by Temmerman et al. (2007) and Schwarz et al. (2014). 4.1. Initial establishment of plant seedlings — small scale

Fig. 5. Allocation of below-ground plant biomass of Spartina alterniflora and Scirpus mariqueter to different layers across the four different sediment treatments. Bar height denotes the average value of biomass allocation in percent, whiskers denote the standard deviation.

Previous studies on seedling establishment of pioneer salt marsh species mainly focused on the influence of abiotic stressors, such as hydrodynamics and sediment dynamics (accretion/erosion) on initial

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

303

Fig. 6. Deposition/erosion measurements resulting from a flume experiment with Spartina alterniflora and Scripus mariqueter; left: eroded sediment volume [m3/m2], middle: erosion map (the green square identifies the place of the vegetation patch), right: ratio of the eroded sediment volumes per treatment (emerged/submerged).

plant units (seedlings) (Peralta et al., 2008; Balke et al., 2013). The present study extends this understanding by including the influence of physical sediment properties (grain size/cohesiveness) in combination

with species-specific-traits on seedling establishment thresholds. This extension revealed surprising differences in response between Spartina and Scirpus, as elaborated below.

Fig. 7. Flow measurements: velocities at 1 cm above the bottom in the gap adjacent to the vegetation patch for the submerged (a) and emerged (b) scenario; (c) flume setup and current velocity measurement locations (◊); error bars signify the standard deviation over the Vectrino measurement interval, (d) flow blocked out by the vegetation stems [%] in the submerged and emerged scenario.

304

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

Spartina shows higher resilience to erosion than Scirpus in muddy sediment, but not in sandy sediments (Fig. 4). Surprisingly, Spartina did not significantly change its relative below-ground biomass allocation pattern throughout our different sediment treatments (Fig. 5). This was confirmed by visual comparison of the developed root networks (Fig. 9). Hence, for Spartina the reduction in critical vertical erosion threshold going from muddy to sandy sediment, was solely related to sediment cohesiveness. Sediment cohesiveness is linked to structural sediment properties. Non-cohesive sediments are granular with little interaction between sediment particles, with their erosion behaviour depending on factors such as grain size distribution, the shape, and the density of individual particles. In contrast, for cohesive sediments electrochemical forces between particles are important leading to the observed cohesion, increasing the force necessary for erosion (Partheniades, 1965; Mitchener and Torfs, 1996; Winterwerp and Van Kesteren, 2004; Grabowski et al., 2011). Spartina was able to grow equally well in all investigated sediment types, showing no significant changes in above or below-ground biomass between treatments. Scirpus follows a completely different growth strategy to Spartina (Fig. 8a,b). During the short time-frame of our experiment, the rhizomes of Spartina mainly grow vertically downwards. In contrast, the rhizomes of Scirpus mainly grow laterally ending in daughter ramets, which themselves grow roots. Our experiments show that the depth to which rhizomes of Scirpus were able to penetrate was strongly linked to sediment properties (Fig. 5), influencing its resilience against erosion. In the mud sediment type the root network did not penetrate into deeper sediment layers, causing a lower critical erosion threshold compared to Spartina (Figs. 4, 8c–f). In the sand sediment type, Scirpus also

invested biomass into deeper sediment layers by growing ramets in diagonal direction (Figs. 5, 8c–f), which had a strong influence on its ability to withstand erosion (Figs. 4, 8c–f). The observation that Spartina and Scirpus have similar erosion thresholds at muddy sand and sandy mud sediment types, although Spartina had a larger root-biomass density in deeper sediment layers, might be due to plant stiffness of the above-ground plant material. Shoot morphology (stiff or flexible) is an important factor for hydrodynamic drag exerted on the plant (Järvelä, 2002). In order to withstand strong drag forces (induced by current or waves), plants need to invest in tissue construction and shoot anchoring to prevent an enhanced risk of breaking or being detached (Bouma et al., 2005a). This could explain the investment into deeper sediment layers for the stiffer plant (Spartina) paired with the similar erosion behaviour of the more flexible plant (Scirpus) (Cotton et al., 2006; Bouma et al., 2010). It is not possible to deduce from our experiments whether Scirpus was unable to grow deeper roots in cohesive sediments due to mechanical or physiological limitations or whether it alternatively showed inherited adaptive behaviour where investment in roots is restricted to sediment types likely to undergo erosion in the field. The artificial sediments used were poor in organic matter, so that sulphide stress or oxygen limitation was unlikely (Koch et al., 2009). In fact, the sandy sediment was in comparison more organically enriched and thus more likely to yield physiological stress. The experimental cohesive sediments were not very consolidated, rendering mechanical limitation also unlikely. Interestingly, literature predicts that bottom sediment composition is mainly a result of the existing hydrodynamic conditions (Orton and

Fig. 8. Comparison in plant habitus between our two species at different sediment types; a) Spartina alterniflora; b) Scirpus mariqueter; c) Scirpus mariqueter grown on sand (non-cohesive) after several steps of sediment erosion, d) Scirpus mariqueter grown on mud (cohesive) after several steps of sediment erosion, it is visible that in the sand sediment type roots, tillers and rhizomes can penetrate deeper in the sediment than in the mud sediment type, this is schematized in e) growth on sand and f) growth on mud.

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

305

Fig. 9. Comparison of root biomass investment of Spartina alterniflora seedlings grown in different sediment types (mud, muddy sand, sandy mud, sand).

Reading, 1993). This opens the possibility that the observed response in morphology and growth strategy of Scirpus to sediment composition is the consequence of an evolutionary adaptation to sediment properties and the corresponding hydrodynamic climate. Thus, deeper rooting in sandy sediment may be advantageous because these sediments usually experience stronger hydrodynamic stress. Similar observations were already reported by van Hulzen et al. (2007), who linked Spartina anglica's phenotypic plasticity (changes in stem density) to changes in hydrodynamic climate. Our results further indicate that not only different species growth forms (phenotypic plasticity) but also their establishment probability are prone to be pre-selected by sediment properties and hydrodynamic conditions. This underlines the necessity to incorporate sediment properties in the evaluation of species distributions along hydrodynamic stress gradients (Bertness, 1991; van Wesenbeeck et al., 2007; Marani et al., 2013). Although the present study is not able to evaluate the influence of sediment properties on species distributions, it nevertheless opens an interesting direction for future research. 4.2. Subsequent development from seedling to plant patches — intermediate scale The flume experiment, investigating the influence of species-specific traits and physical forcing (biotic-abiotic context dependency) on scaledependent feedbacks at tussock scale, showed that in the emerged stage approximately equal volumes of sediment were eroded next to the vegetation patches of the two species. However in the submerged stage, 40% more sediment was eroded next to the Spartina patch compared to the Scirpus patch (Fig. 6). Flow velocity measurements in the gap showed that this increased erosion was due to increased flow adjacent to the Spartina patch (Fig. 7a–c). This increase in gap flow was explained by the differences in blocked flow by the vegetation (Fig. 7d). In the emerged scenario, the amount of blocked flow by vegetation stems was similar. During the submerged scenario Scirpus stems blocked less flow due to bending, thus permitting more flow over the submerged vegetation canopy than in the Spartina case, leading to less gap erosion. These findings are in agreement with previous studies investigating the flow routing behaviour around stiff and flexible vegetation patches. They show that due to the absence of bending, stiff vegetation is prone to re-route more flow towards their patch edges than flexible vegetation (Dijkstra, in preparation; Vandenbruwaene et al., 2011; Bouma et al., 2013). We extended on these insights by showing that the increased flow around stiff vegetation patches also leads to increased patch adjacent erosion (see results submerged case), which is the process inhibiting patch expansion (D'Alpaos et al., 2007; Temmerman et al., 2007; van Wesenbeeck et al., 2008a). The rationale for using the sand sediment type in our flume experiment was our focus in comparing species-specific traits, which is facilitated by its lower erosion threshold. Based on theoretical literature on sediment transport, we expect the observed differences between stiff and flexible vegetation to decline when sediment cohesiveness under constant

current forcing is increased (van Ledden et al., 2004; Winterwerp and Van Kesteren, 2004). Although we tested patch-adjacent erosion at constant current velocity with two different water levels (whereas in the field flow velocities vary with water level), we nevertheless consider this to be a good approximation for actual field conditions. Velocity-water level relationships are connected to factors such as tidal amplitude and shore geometry but in general high velocity is found at relatively intermediate water levels during large tides (Pestrong, 1965; Davidson-Arnott et al., 2002; Bouma et al., 2005b).

5. Synthesis and conclusions It is shown that species–specific interactions with abiotic forcing (1) are determining factors for the establishment of initial units and (2) are further able to determine the magnitude of scale-dependent feedbacks at established patches. Through the comparison between our two different scales it is established that traits advantageous at one scale can be disadvantageous at others. Albeit high ecosystem engineering ability can be advantageous in trapping sediment within stems, it can be disadvantageous in causing increased erosion at tussock edges, which might prevent permanent establishment. We propose sediment properties also exert a major impact on the ability of plants to sculpture geomorphologic features such as tidal channels (as shown in Temmerman et al., 2007). This study highlights the importance of incorporating sediment properties in the assessment of species establishment and showed its importance on the development of scale-dependent feedbacks at a small and intermediate scale. Although the small- and intermediate-scale interactions potentially influence landform development via altering scale-dependent feedbacks and survival probability of established units, further tests on this hypothesis still need to be provided. A deeper understanding of biogeomorphic feedback loops influencing seedling establishment and their ability to sculpt the landscape will have further consequences for wetland restoration and management.

Acknowledgements We thank the Programme Strategic Alliances between the People's Republic of China and The Netherlands for funding our research (PSA 04-PSA-E-01). We want to thank especially our Chinese collaboration partners of the SKLEC research institute at the East China Normal University in Shanghai for their energetic support during and after our field assessment, nameley Ma Zhigang, Zhu Zhenchang, Yuan Lin and the local farmers who supported us during field measurements and seed collection. We also want to thank Bas Koutstaal of the NIOZYerseke for his assistance during our seedling germination and growth experiments.

306

C. Schwarz et al. / Geomorphology 250 (2015) 298–307

References Allen, J., 2000. Morphodynamics of Holocene salt marshes: a review sketch from the Atlantic and Southern North Sea coasts of Europe. Quat. Sci. Rev. 19 (12), 1155–1231. Balke, T., Bouma, T.J., Horstman, E.M., Webb, E.L., Erftemeijer, P.L.A., Herman, P.M.J., 2011. Windows of opportunity: thresholds to mangrove seedling establishment on tidal flats. Mar. Ecol. Prog. Ser. 440, 1–9. Balke, T., Klaassen, P.C., Garbutt, A., van der Wal, D., Herman, P.M.J., Bouma, T.J., 2012. Conditional outcome of ecosystem engineering: a case study on tussocks of the salt marsh pioneer Spartina anglica. Geomorphology 153, 232–238. Balke, T., Webb, E.L., den Elzen, E., Galli, D., Herman, P.M.J., Bouma, T.J., 2013. Seedling establishment in a dynamic sedimentary environment: a conceptual framework using mangroves. J. Appl. Ecol. 50 (3), 740–747. Bertness, M.D., 1991. Zonation of Spartina patens and Spartina alterniflora in a New England salt marsh. Ecology 72 (1), 138–148. Bouma, T.J., De Vries, M.B., Low, E., Kusters, L., Herman, P.M.J., Tanczos, I.C., Temmerman, S., Hesselink, A., Meire, P., van Regenmortel, S., 2005a. Flow hydrodynamics on a mudflat and in salt marsh vegetation: identifying general relationships for habitat characterisations. Hydrobiologia 540, 259–274. Bouma, T.J., De Vries, M.B., Low, E., Peralta, G., Tanczos, C., Van de Koppel, J., Herman, P.M.J., 2005b. Trade-offs related to ecosystem engineering: a case study on stiffness of emerging macrophytes. Ecology 86 (8), 2187–2199. Bouma, T.J., Friedrichs, M., Klaassen, P.C., van Wesenbeeck, B.K., Brun, F.G., Temmerman, S., van Katwijk, M.M., Graf, G., Herman, P.M.J., 2009a. Effects of shoot stiffness, shoot size and current velocity on scouring sediment from around seedlings and propagules. Mar. Ecol. Prog. Ser. 388, 293–297. Bouma, T.J., Friedrichs, M., van Wesenbeeck, B.K., Temmerman, S., Graf, G., Herman, P.M.J., 2009b. Density-dependent linkage of scale-dependent feedbacks: a flume study on the intertidal macrophyte Spartina anglica. Oikos 118 (2), 260–268. Bouma, T.J., De Vries, M.B.D., Herman, P.M.J., 2010. Comparing ecosystem engineering efficiency of two plant species with contrasting growth strategies. Ecology 91 (9), 2696–2704. Bouma, T.J., Temmerman, S., van Duren, L.A., Martini, E., Vandenbruwaene, W., Callaghan, D.P., Balke, T., Biermans, G., Klaassen, P.C., van Steeg, P., 2013. Organism traits determine the strength of scale-dependent bio-geomorphic feedbacks: a flume study on three intertidal plant species. Geomorphology 180, 57–65. Castellanos, E., Figueroa, M., Davy, A., 1994. Nucleation and facilitation in saltmarsh succession: interactions between Spartina maritima and Arthrocnemum perenne. J. Ecol. 82 (2), 239–248. Corenblit, D., Tabacchi, E., Steiger, J., Gurnell, A.M., 2007. Reciprocal interactions and adjustments between fluvial landforms and vegetation dynamics in river corridors: a review of complementary approaches. Earth Sci. Rev. 84 (1), 56–86. Cotton, J., Wharton, G., Bass, J., Heppell, C., Wotton, R., 2006. The effects of seasonal changes to in-stream vegetation cover on patterns of flow and accumulation of sediment. Geomorphology 77 (3), 320–334. D'Alpaos, A., Lanzoni, S., Marani, M., Rinaldo, A., 2007. Landscape evolution in tidal embayments: modeling the interplay of erosion, sedimentation, and vegetation dynamics. Journal of Geophysical Research: Earth Surface 112 (F1) (2003–2012). Davidson-Arnott, R.G., van Proosdij, D., Ollerhead, J., Schostak, L., 2002. Hydrodynamics and sedimentation in salt marshes: examples from a macrotidal marsh, Bay of Fundy. Geomorphology 48 (1), 209–231. Dijkstra, J., 2009. Measurements and Modelling of the Effects of Seagrass Meadows on Flow and Sediment Transport in the Bay of l'Ecluse, Dinard, France (in preparation). Engels, J.G., Rink, F., Jensen, K., 2011. Stress tolerance and biotic interactions determine plant zonation patterns in estuarine marshes during seedling emergence and early establishment. J. Ecol. 99 (1), 277–287. Friess, D.A., Krauss, K.W., Horstman, E.M., Balke, T., Bouma, T.J., Galli, D., Webb, E.L., 2012. Are all intertidal wetlands naturally created equal? Bottlenecks, thresholds and knowledge gaps to mangrove and saltmarsh ecosystems. Biol. Rev. 87 (2), 346–366. Grabowski, R.C., Droppo, I.G., Wharton, G., 2011. Erodibility of cohesive sediment: the importance of sediment properties. Earth Sci. Rev. 105 (3), 101–120. Gurnell, A.M., Bertoldi, W., Corenblit, D., 2012. Changing river channels: the roles of hydrological processes, plants and pioneer fluvial landforms in humid temperate, mixed load, gravel bed rivers. Earth Sci. Rev. 111 (1), 129–141. Han, Q., Bouma, T.J., Brun, F.G., Suykerbuyk, W., Van Katwijk, M., 2012. Resilience of Zostera noltii to burial or erosion disturbances. Mar. Ecol. Prog. Ser. 449. He, Y., Li, X., Guo, W., Ma, Z., 2012. Division of labor in rhizomatous species: comparative performance of native and invasive species in the tidal marshes of the Yangtze river estuary, China. J. Exp. Mar. Biol. Ecol. 422, 122–128. Huang, H.M., Zhang, L.Q., 2007. A study of the population dynamics of Spartina alterniflora at Jiuduansha shoals, Shanghai, China. Ecol. Eng. 29 (2), 164–172. Hupp, C.R., Osterkamp, W., Howard, A.D., 1995. Biogeomorphology, Terrestrial and Freshwater Systems. Elsevier Science B.V. Infantes, E., Orfila, A., Bouma, T.J., Simarro, G., Terrados, J., 2011. Posidonia oceanica and Cymodocea nodosa seedling tolerance to wave exposure. Limnol. Oceanogr. 56 (6), 2223. Istanbulluoglu, E., Bras, R.L., 2005. Vegetation-modulated landscape evolution: effects of vegetation on landscape processes, drainage density, and topography. J. Geophys. Res. 110 (F2), F02012. Järvelä, J., 2002. Determination of flow resistance of vegetated channel banks and floodplains. River Flow 311–318. Jones, C., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers. Oikos 69 (3), 373–386. Jones, C., Gutiérrez, J.L., Byers, J.E., Crooks, J.A., Lambrinos, J.G., Talley, T.S., 2010. A framework for understanding physical ecosystem engineering by organisms. Oikos 119 (12), 1862–1869.

Koch, E.W., Barbier, E.B., Silliman, B.R., Reed, D.J., Perillo, G.M., Hacker, S.D., Granek, E.F., Primavera, J.H., Muthiga, N., Polasky, S., 2009. Non-linearity in ecosystem services: temporal and spatial variability in coastal protection. Front. Ecol. Environ. 7 (1), 29–37. Langlois, E., Bonis, A., Bouzille, J., 2003. Sediment and plant dynamics in saltmarshes pioneer zone: Puccinellia maritima as a key species? Estuar. Coast. Shelf Sci. 56 (2), 239–249. Li, H., Yang, S.L., 2009. Trapping effect of tidal marsh vegetation on suspended sediment, Yangtze delta. J. Coast. Res. 25 (4), 915–936. Li, B., Liao, C.H., Zhang, X.D., Chen, H.L., Wang, Q., Chen, Z.Y., Gan, X.J., Wu, J.H., Zhao, B., Ma, Z.J., Cheng, X.L., Jiang, L.F., Chen, J.K., 2009. Spartina alterniflora invasions in the Yangtze river estuary, China: an overview of current status and ecosystem effects. Ecol. Eng., 35(4), 511–520 Marani, M., Da Lio, C., D'Alpaos, A., 2013. Vegetation engineers marsh morphology through multiple competing stable states. Proc. Natl. Acad. Sci. 110 (9), 3259–3263. Meire, D.W., Kondziolka, J.M., Nepf, H.M., 2014. Interaction between neighboring vegetation patches: impact on flow and deposition. Water Resour. Res. 50 (5), 3809–3825. Mitchener, H., Torfs, H., 1996. Erosion of mud/sand mixtures. Coast. Eng. 29 (1–2), 1–25. Mullan Crain, C., Bertness, M.D., 2006. Ecosystem engineering across environmental gradients: implications for conservation and management. Bioscience 56 (3), 211–218. Murray, A.B., Paola, C., 2003. Modelling the effect of vegetation on channel pattern in bedload rivers. Earth Surf. Process. Landf. 28 (2), 131–143. Murray, A., Knaapen, M., Tal, M., Kirwan, M., 2008. Biomorphodynamics: physical–biological feedbacks that shape landscapes. Water Resour. Res. 44 (11), W11301. Orton, G., Reading, H., 1993. Variability of deltaic processes in terms of sediment supply, with particular emphasis on grain size. Sedimentology 40 (3), 475–512. Partheniades, E., 1965. Erosion and deposition of cohesive soils. J. Hydraul. Div. 91 (1), 105–139. Patterson, S., McKee, K.L., Mendelssohn, I.A., 1997. Effects of tidal inundation and predation on Avicennia germinans seedling establishment and survival in a sub-tropical mangal/salt marsh community. Mangrove Salt Marshes 1 (2), 103–111. Peralta, G., van Duren, L.A., Morris, E.P., Bouma, T.J., 2008. Consequences of shoot density and stiffness for ecosystem engineering by benthic macrophytes in flow dominated areas: a hydrodynamic flume study. Mar. Ecol. Prog. Ser. 368, 103–115. Pestrong, R., 1965. The development of drainage patterns on tidal marshes. Stanford University publications 10(2). Phillips, J.D., 1995. Biogeomorphology and landscape evolution: the problem of scale. Geomorphology 13 (1), 337–347. Schoelynck, J., De Groote, T., Bal, K., Vandenbruwaene, W., Meire, P., Temmerman, S., 2012. Self-organised patchiness and scale-dependent bio-geomorphic feedbacks in aquatic river vegetation. Ecography 35 (8), 760–768. Schwarz, C., Ysebaert, T., Zhu, Z., Zhang, L., Bouma, T.J., Herman, P.M.J., 2011. Abiotic factors governing the establishment and expansion of two salt marsh plants in the Yangtze estuary, China. Wetlands 31 (6), 1011–1021. Schwarz, C., Ye, Q., Wal, D., Zhang, L., Bouma, T., Ysebaert, T., Herman, P., 2014. Impacts of salt marsh plants on tidal channel initiation and inheritance. J. Geophys. Res. Earth Surf. 119 (2), 385–400. Scott, M.L., Friedman, J.M., Auble, G.T., 1996. Fluvial process and the establishment of bottomland trees. Geomorphology 14 (4), 327–339. Stallins, J.A., 2006. Geomorphology and ecology: unifying themes for complex systems in biogeomorphology. Geomorphology 77 (3), 207–216. Sun, S.C., Cai, Y.L., Liu, H., 2001. Biomass allocation of Scirpus mariqueter along an elevational gradient in a salt marsh of the Yangtse river estuary. Acta Bot. Sin. 43 (2), 178–185. Sun, S.C., Cai, Y.L., An, S.Q., 2002. Differences in morphology and biomass allocation of Scirpus mariqueter between creekside and inland communities in the Changjiang estuary, China. Wetlands 22 (4), 786–793. Tal, M., Paola, C., 2007. Dynamic single-thread channels maintained by the interaction of flow and vegetation. Geology 35 (4), 347. Temmerman, S., Bouma, T.J., van de Koppel, J., van der Wal, D.D., De Vries, M.B., Herman, P.M.J., 2007. Vegetation causes channel erosion in a tidal landscape. Geology 35 (7), 631–634. van de Koppel, J., Bouma, T.J., Herman, P.M.J., 2012. The influence of local-and landscapescale processes on spatial self-organization in estuarine ecosystems. J. Exp. Biol. 215 (6), 962–967. van Hulzen, J., Van Soelen, J., Bouma, T., 2007. Morphological variation and habitat modification are strongly correlated for the autogenic ecosystem engineer Spartina anglica (common cordgrass). Estuar. Coasts 30 (1), 3–11. van Ledden, M., van Kesteren, W.G.M., Winterwerp, J.C., 2004. A conceptual framework for the erosion behaviour of sand-mud mixtures. Cont. Shelf Res. 24 (1), 1–11. van Wesenbeeck, B.K., Crain, C.M., Altieri, A.H., Bertness, M.D., 2007. Distinct habitat types arise along a continuous hydrodynamic stress gradient due to interplay of competition and facilitation. Mar. Ecol. Prog. Ser. 349, 63–71. van Wesenbeeck, B.K., van de Koppel, J., Herman, P.M.J., Bertness, M.D., van der Wal, D., Bakker, J.P., Bouma, T.J., 2008a. Potential for sudden shifts in transient systems: distinguishing between local and landscape-scale processes. Ecosystems 11 (7), 1133–1141. van Wesenbeeck, B.K., van de Koppel, J., Herman, P.M.J., Bouma, T.J., 2008b. Does scale dependent feedback explain spatial complexity in salt marsh ecosystems? Oikos 117 (1), 152–159. Vandenbruwaene, W., Temmerman, S., Bouma, T.J., Klaassen, P.C., De Vries, M.B., Callaghan, D.P., Van Steeg, P., Dekker, F., Van Duren, L.A., Martini, E., 2011. Flow interaction with dynamic vegetation patches: implications for biogeomorphic evolution of a tidal landscape. Journal of Geophysical Research: Earth Surface 116 (F1) (2003–2012). Vandenbruwaene, W., Bouma, T.J., Meire, P., Temmerman, S., 2012. Bio-geomorphic effects on tidal channel evolution: impact of vegetation establishment and tidal prism change. Earth Surf. Process. Landf. 38 (2), 122–132.

C. Schwarz et al. / Geomorphology 250 (2015) 298–307 Wang, C., Temmerman, S., 2013. Does biogeomorphic feedback lead to abrupt shifts between alternative landscape states?: an empirical study on intertidal flats and marshes. J. Geophys. Res. 118 (1), 229–240. Wang, Q., An, S.Q., Ma, Z.J., Zhao, B., Chen, J.K., Li, B., 2006. Invasive Spartina alterniflora: biology, ecology and management. Acta Phytotaxonomica Sin. 44 (5), 559–588. Wang, R.Z., Yuan, L., Zhang, L.Q., 2010. Impacts of Spartina alterniflora invasion on the benthic communities of salt marshes in the Yangtze estuary, China. Ecol. Eng. 36 (6), 799–806. Winterwerp, J.C., Van Kesteren, W.G.M., 2004. Introduction to the Physics of Cohesive Sediment Dynamics in the Marine Environment. Elsevier Science.

307

Wolanski, E., Boorman, L., Chícharo, L., Langlois-Saliou, E., Lara, R., Plater, A., Uncles, R., Zalewski, M., 2004. Ecohydrology as a new tool for sustainable management of estuaries and coastal waters. Wetl. Ecol. Manag. 12 (4), 235–276. Zong, L., Nepf, H., 2010. Flow and deposition in and around a finite patch of vegetation. Geomorphology 116 (3), 363–372. Zong, L., Nepf, H., 2012. Vortex development behind a finite porous obstruction in a channel. J. Fluid Mech. 691, 368–391.