Interaction studies of vegetation within flow environments are essential for the determination of bank protection, morphological characteristics and ecological conditions for wetlands. This paper uses the MIKE 21 hydrodynamic and salinity model to simulate the hydrodynamic characteristics and salinity transport processes in the Pink Beach wetlands of the Liao River estuary. The effect of wetland plants on tidal flow in wetland areas is represented by a varying Manning coefficient in the bottom friction term. Acquisition of the vegetation distribution is based on Landsat TM satellites by remote sensing techniques. Detailed comparisons between field observation and simulated results of water depth, salinity and tidal currents are presented in the vegetated domain of the Pink Beach wetlands. Satisfactory results were obtained from simulations of both flow characteristics and salinity concentration, with or without vegetation. A numerical experiment was conducted based on variations in vegetation density, and compared with the tidal currents in non-vegetated areas; the computed current speed decreased remarkably with an increase in vegetation density. The impact of vegetation on water depth and salinity was simulated, and the findings revealed that wetland vegetation has an insignificant effect on the water depth and salinity in this wetland domain. Several stations (from upstream to downstream) in the Pink Beach wetlands were selected to estimate the longitudinal variation of salinity under different river runoff conditions; the results showed that salinity concentration decreases with an increase in river runoff. This study can consequently help increase the understanding of favourable salinity conditions for particular vegetation growth in the Pink Beach wetlands of the Liao River estuary. The results also provide crucial guidance for related interaction studies of vegetation, flow and salinity in other wetland systems.
Wetlands are transitional zones between terrestrial ecosystems and aquatic
ecosystems, and have a variety of unique functions, which include the following: providing large amounts of
food, raw materials and water resources for humans, and maintaining the
ecological balance, biodiversity and resources for rare species. The role of
aquatic plants regarding coastal protection from extreme events has consequently become a
recurring question, along with the viability assessment of ecosystem-based
management approaches (Barbier et al., 2008; Temmerman et al., 2013). Coastal
wetlands are mainly distributed in coastal regions in China within eleven specific
areas: Hebei, Liaoning, Shandong, Jiangsu, Zhejiang, Fujian, Guangdong, Hainan,
Taiwan, Tianjin and Guangxi. Coastal wetland ecosystems cover
5.7959
Liao River estuary wetland is located in Panjin City, Liaoning
Province, China, and covers an area of about 451300.5 ha (Zhang et
al., 2009). The main vegetation in the wetland includes
The geographical location (Wang et al., 2017) and satellite image of the Liao River estuary.
Model grids, bathymetry and the validation points.
The validation of tidal level at measured stations.
The validation of current speed (left side) and direction (right side) at measured stations during neap tide.
The validation of current speed (left side) and direction (right side) at measured stations during spring tide. Figures 7 and 8 show the flow field in flood and ebb tide during the neap and spring tide. During the flood tide, as depicted in Figs. 7a and 8a, the general flow of the tidal current well offshore is north-eastward. The main flow flooding into the Liao River divides to go around both sides of Gaizhou Shoal. When the flow reaches the east and north-east of the Gaizhou Shoal, the flow turns to the north-west, forming a mainstream flow from the outside sea into Liao River. West of the Gaizhou Shoal, the water mainly flows to the north and the north-east, and is affected by the delta terrain. During the spring tide period, the Gaizhou Shoal can be swamped by the tidal currents on both sides because of the high tide level. During the ebb tide period, as depicted in Figs. 7b and 8b, due to the development of many shoals, the current at the mouth of the estuary is divided into many branches. West of the Gaizhou Shoal, the current gradually turns from south-westward to southward. East of the Gaizhou Shoal, the tide current flows directly south-east and then turns to the south. Part of the Gaizhou Shoal is high enough to be exposed at the low tide. Overall, the modelled results are acceptable and the model can be used for the following studies. The results of the salinity validation for V1, V2, V3 and V4 between 29 and 30 May 2014 at spring tide are reported in Fig. 9. The model correctly estimates the salinity in the Liao River estuary. The measured salinities were obtained by a Sea-Bird 911 plus conductivity–temperature–depth (CTD) profiler.
Recently, wetland ecosystems have been severely damaged and degenerated through the disproportionate consumption of wetland ecological resources, which in turn has resulted in serious declines in biodiversity and biological resources. A variety of studies on hydrodynamics in estuarine wetlands have been conducted, the most of which have focused on the following aspects: interaction between flow and vegetation, pollutant transport in wetlands and vegetation resistance experiments. Relevant prior research to this particular work mainly revolves around the study of the resistance coefficient of water flow when plants exist; most of this research utilises the Manning coefficient of water resistance (Ree et al., 1958; Chow et al., 1959). In addition, however, some scholars have studied the influence of plants on the structure of water flow, such as changes to flow, turbulence intensity and boundary shear force (Ikeda and Kanazawa, 1996). Considering the height and bending degree of willow species by water during flooding, the vegetation resistance was introduced into the Navier–Stokes equation. Numerical simulations of the three-dimensional flow field of the river and the floodplain wetland were then carried out (Wilson, 2006). Taking the reed community as the subject of their research, Shi et al. (2001) carried out an experiment to investigate the water resistance of non-submerged reeds and the relationship between the density and the resistance of reeds. Based on velocities from laboratory experiments for different water depths, discharges and aquatic vegetation densities, analyses were carried out for the resistance coefficient of vegetation (Li and Zhao, 2004).
The Saint-Venant equation and Nuding model were combined to simulate the
steady-state and unsteady flow in the presence of vegetation in a channel and
to analyse the effects of vegetation cover, beach slope and width on the
cross-section of a river (Helmiö, 2005). A mathematical model that takes advantage of hydraulics
typical of wetlands was used to simulate
a one- and two-dimensional wetland flow in a laboratory experiment and a wetland pond.
(Feng and Molz, 1997). The coupled SWIFT2D surface water and SEAWAT
groundwater migration model was used to simulate the hydrological processes
and salt exchange of the surface water and groundwater in estuaries and
adjacent coastal wetlands (Christian et al., 2000). The two-dimensional
numerical model was used to test the different flow conditions of Zhalong
wetland, and the effect of reed vegetation on the process of storage and
detention in wetlands was comprehensively evaluated (Gu et al., 2006). The
two-dimensional
In general, the majority of studies have focused on the effects of vegetation on fluid movement in flume experiments; few detailed field observations or salinity simulations exist in mudflat–salt marsh ecosystems, especially involving the typical wetland plants of the Liao River estuary. Research on the salinity response of plants to river discharge in wetland waters has not yet been systematically assessed.
In this study, a two-dimensional hydrodynamic and salinity model is used to simulate flow patterns and salinity distribution in the wetland waters of the Liao River estuary. The resistance caused by vegetation is represented by the varying Manning coefficient. This study adopts remote sensing techniques to obtain the spatial distribution of two types of aquatic plants in the Pink Beach wetlands. The numerical model is calibrated and validated against field measurement data; the variation of salinity in the vegetated domain of the Pink Beach wetland is obtained under different runoff conditions.
The MIKE 21 model is one of the most widely used hydrodynamic models (with respects to both domestic and overseas research (Wang et al., 2013)), and was developed by the Danish Hydraulic Institute (DHI). The model is based on the cell-centred finite volume method implemented on an unstructured flexible mesh. It includes hydrodynamic, transport, ecological module/oil spill, particle tracking, mud transport, sand transport and inland flooding modules (Cox, 2003).
The hydrodynamic module is based on the numerical solution of the depth-integrated incompressible flow Reynolds-averaged mass conservation and momentum equations (William, 1980). The governing equations include the following:
The continuity conservation equation:
The coordinates of the monitoring stations.
The fundamental salinity equation is
Flow field during neap tide, interpolated to a regular grid for clarity.
Flow field during spring tide, interpolated to a regular grid for clarity.
The validation of salinity measured at stations during spring tide.
Vegetation classification based on a decision tree.
The Liao River is one of the seven largest rivers in China and is located in
the north of Liaodong Bay. This estuary is a crucial ecological
economic zone and plays an important role in the comprehensive
development and utilisation of marine industry in China. The Liao River
estuary includes the Daliao and Liao rivers (Li et al., 2017). Pink
Beach of the Liao River delta is a marsh wetland covered with
The study domain is located in the north of Liaodong Bay, extending from 40.3032
to 40.7105
To proceed with the numerical simulations, the equations of hydrodynamics require appropriate boundary and initial conditions. A total of three open boundaries and a solid boundary were established. The model was forced at the open boundary, from Huludao to Bayuquan, by a time series of tidal elevations from the TMD (Tide Model Driver) (Padman, 2005). Two flow boundaries in the north of the area are controlled by the discharge. The solid boundary is treated as impermeable with no slip. The salinity data of the open boundary and river discharge are set to 32.8 and 2 psu, respectively, in this model. The initial water level and salinity are 0 m and 32 psu, respectively.
The distribution of the aquatic plants and selected stations at Pink Beach wetlands.
Comparison of the simulated and measured water depths at locations G1 and G2.
Comparison of the simulated and measured salinity concentrations at locations G1 and G2.
Comparisons of the measured and simulated velocities at location G1.
Flow structure of Pink Beach in vegetated
Comparison of water velocity over regions of differing plant density at the Pink Beach wetlands.
Comparison of water depth and salinity in vegetated and non-vegetated areas at location G2.
The simulated salinity concentrations at the five locations displayed in Figure 11.
Simulations were carried out to verify the accuracy of the model. Simulations
with a larger time step caused systematic violations of the Courant number
(i.e. CN > 1), whereas smaller time steps significantly
increased the computational time required. The model is an explicit format with a
maximum time step of half an hour, which is automatically adjusted
according to the CFL (Courant–Friedrichs–Lewy) conditions during the
calculation. The parameter
Simulated salinity under different runoff conditions.
Contour maps of salinity for the period of strongest saltwater intrusion with different rates of runoff.
The water levels and the tidal currents in the study domain were calculated and the results of the numerical simulation were compared with measured data, as shown in Figs. 4, 5 and 6. The model matched the timing of observed tidal water levels at two locations (Fig. 4), with no detectable phase shift in water levels; however the model slightly underestimated the water levels. This may be attributed to the accuracy of the time series of tidal elevations forced by the open boundary and the measured tidal level. The simulated tidal current speeds were approximately consistent with the field data (Figs. 5 and 6). In addition, the simulated direction of tidal currents was also consistent with the measured direction of tidal currents. These satisfactory validation results demonstrate that the proposed model is capable of simulating the flow in the Liao River estuary.
Remote sensing is an effective and powerful way to monitor vegetation status, growth and biophysical parameters (Hunter et al., 2010; DeFries 2008; Ustin and Gamon, 2010). It also allows for frequent acquisitions for multi-temporal studies and the reconstruction of historical time series in a cost-effective way (Coppin and Bauer, 1994; Munyati, 2000). The objective of the present research was, therefore, to adopt remote sensing to obtain information regarding vegetation in the wetlands of the Liao River estuary.
Information on the wetland was acquired on 3 June 2017 from the Landsat8
Operational Land Imager (OLI), provided by the USGS
(
The abovementioned information extraction was carried out as follows:
The effect of vegetation on the hydrodynamics in wetland in estuary areas was
represented by a varying Manning coefficient in the bottom friction term.
The Manning coefficient for vegetation resistance depends on the flow
depths, the number density and the diameter of the vegetation elements (Zhang et
al., 2013). The Manning coefficient
The model simulation was conducted to evaluate estuarine hydrodynamics and
salinity transport in the presence of vegetation at the Pink Beach wetland;
realistic vegetation was incorporated into the model grid. The dominant
vegetation types at the sites (Fig. 11) were
The results regarding flow structure at Pink Beach in the presence and absence of vegetation highlight the relationship between vegetation and currents (Fig. 15). From the numerical experiments, it can be seen that the presence of vegetation increases the resistance of the estuary bed and can effectively reduce the flow velocity. This is because when water flows through the vegetation, momentum and energy are lost, as the drag exerted by vegetation results in decreased flow speed.
Numerical experiments conducted in the wetland were used to investigate the
effect of different plant densities on currents. For the experiments the
It is generally acknowledged that the Liao River estuary is a salt marsh area; therefore the impact of vegetation on water depth and salinity is tested in this paper. Figure 17 presents a time series of water depth and salinity in vegetated and non-vegetated experiments at G2. It can be seen that the effects of vegetation on the water depth and salinity in the wetland domain are not obvious. This is because the tidal wave is a long wave, which has no evident effect on the water depth or the salinity of wetland domains.
Based on measured and simulated salinity data from Liao River estuary
during the spring tide from 26 to 27 July 2017, five stations were selected in the Pink Beach wetland
(from upstream to downstream) as shown in Figure. 11. This was done in order to
analyse the longitudinal distribution of salinity in the tidal cycle under
the same runoff conditions. The simulated salinity data for several stations
along the Liao River from the entrance are given in Figure 18. It can be noted that the salinity
concentration at P1 upstream is far lower than that of P5 downstream. The
salinity concentration in the river can consequently be seen to increase from upstream to
downstream; this is owing to the fact that as dilution by fresh water from upstream increases,
salinity decreases. During the ebb tide, negative tide levels occur with dry domains being noted at
locations G1, G2, P3 and P5. Hence, the salinity concentrations show gaps for
this period. The influence of runoff variation on the salinity distribution
of G1 and G2 in the Pink Beach wetland was analysed under different respective runoff
conditions: 0, 30, 101, 285 and 450 m
In this study, the MIKE 21 model is used to simulate the hydrodynamic
characteristics and salinity transport processes in the Pink Beach wetlands
of the Liao River estuary. The model couples hydrodynamic and salinity modules
with the salt marsh plant effects. The spatial discretization of the primitive
equations is performed using a cell-centred finite volume method in the
horizontal plane with an unstructured grid of triangular elements. Landsat
images are applied to differentiate the wetland vegetation types in the study region.
Based on the obvious spectral distinction of vegetation, a
decision tree containing a number of decision rules is designed to classify
different types of vegetation cover; the Liao River estuary is classified
into water body, shoal, and major wetland vegetation types, (e.g.
The model is tested by simulating the water level, tidal current and salinity concentration in Liao River estuary, and the results are consistent with the measured data. The tidal flats are periodically exposed above the surface of the water in the study area. Numerical predictions indicate that vegetation imposes a significant influence on flow dynamics. The existence of vegetation is associated with lower flow velocities, as vegetation can modify the flow structure and lead to energy dissipation. By analysing the longitudinal variation of salinity in the Pink Beach wetland, we found that salinity gradually increased from upstream to downstream. The effect of runoff on salinity distributions in the wetland is fairly distinct. When the river discharge is low, less freshwater is mixed into the system and salinity is higher. These results are important for understanding wetland dynamics and salinity transport processes. Furthermore, they contribute to an improved comprehension of suitable circumstances for vegetation growth in the Pink Beach wetland. More generally, this study can provide an important scientific basis for wetland conservation and restoration.
The research data are provided in the Supplement.
The supplement related to this article is available online at:
The authors declare that they have no conflict of interest.
This work was supported by the National Natural Science Foundation of China (51779039), the Wetland Degradation and Ecological Restoration Program of Panjin Pink Beach (PHL-XZ-2017013-002), the Fund of Liaoning Marine Fishery Department (201725), the Open Fund of the State Key Laboratory of Hydraulics and Mountain River Engineering (SKHL1517). Edited by: John M. Huthnance Reviewed by: two anonymous referees