Tidal flats are valuable habitats and are important for coastal protection. The total area of tidal flats is decreasing worldwide due to various problems like sea level rise, coastal squeeze, subsidence by gas extraction and erosion initiated by man made constructions. The construction of a storm surge barrier and compartmentalization dams in the Eastern Scheldt in the 1980s is one example of a man made structure that resulted in a change in hydrodynamic conditions of the Eastern Scheldt estuary and hence the sediment equilibrium. As a result, channels are filling in and tidal flats inside the estuary are eroding. Nourishing tidal flats with sediment might be a promising solution to mitigate these effects.
To test this approach, a small area of the Galgeplaat, a tidal flat in the Eastern Scheldt, was nourished in 2008 with 130.000 m3 sand dredged from adjacent channels over a total area of 150.000 m2. The processes of sediment distribution on the flats and benthic recolonization are coupled and interact with each other. Therefore the design challenge is to find an optimum to reduce the initial impact of the nourishment on the benthic fauna, while optimizing the distribution of the sand over the tidal flat by wind and waves and the subsequent recovery of benthic life.
The saltmarshes and intertidal flats in the Eastern Scheldt are important and unique resting and feeding areas for seals and wading birds such as oystercatchers and curlews. These shallow areas also provide important additional protection against coastal flooding because they attenuate wind waves before they reach the dikes.
The construction of the storm surge barrier in the Eastern Scheldt has changed the hydrodynamics of the system. Tidal flow velocities in the channels have decreased by some 30% and the sediment transport capacity, important for accretion of the intertidal flats, has shrunk by as much as 75%. Moreover, the storm surge barrier itself acts as an almost perfect block to sediment exchange between the outer delta and the basin. As a consequence, the estuary lacks the sediment transport capacity as well as the sediment availability to maintain the original balance between erosion and sedimentation. While erosion by waves generated inside the estuary continues, the mechanisms building up the tidal flats have dramatically weakened. As a result of this imbalance, saltmarshes and tidal flats in the estuary are eroding. Between 1986 and 2010 the intertidal area has decreased by 1380 ha and the flats are reducing by approximately 1 cm in height per year (De Ronde, et al., 2011). Moreover, wave action tends to erode the higher parts of the flats hence flattening the relief, which further reduces the foraging time for birds.
Without human intervention, the intertidal area is expected to decrease by about 50 ha to 100 ha per year in the coming years. From the 11,000 ha of intertidal area in 1986 only 75% (circa 9,000 ha) will remain by 2060 (De Ronde, et al., 2012).
The reduction of the intertidal area in the Eastern Scheldt is undesirable for both nature and safety reasons. Rijkswaterstaat (Dutch National Water Board) defined several types of mitigation measures (Van Zanten and Adriaanse, 2008):
Nourishing tidal flats is one of the solutions to limit the loss of intertidal areas. Nourishment of beaches is increasingly used to combat shoreline erosion, but in intertidal habitats so far nourishments have been hardly applied (Ysebaert et al., 2009). Little is known about the effectiveness, feasibility and costs of these measures. The Directorate-General Rijkswaterstaat of the Ministry of Infrastructure and the Environment, the authority responsible, decided to carry out a pilot nourishment on the Galgeplaat.
The Galgeplaat nourishment
The Galgeplaat is one of the eroding intertidal flats in the Eastern Scheldt. Between 1985 and 2001 the area of the Galgenplaat shrunk from 1000 to 964 ha and its mean height was lowered by 0.33 m (approx. 1 cm/year) (Van Zanten and Adriaanse, 2008).
In 2008 a trial nourishment was carried out on the Galgeplaat by Rijkswaterstaat, with the aim to reduce the pace of erosion. The Galgeplaat was chosen because it is monitored and investigated over the years, so there is a significant amount of background knowledge. In addition, planned maintenance dredging of the adjacent channels for shipping meant enough sediment was readily and easily available for this trial nourishment.
The objective of Rijkswaterstaat is to answer the following questions with this experimental nourishment:
In addition, Building with Nature took advantage of the opportunity to use this innovative nourishment to gain new knowledge on the driving processes of a tidal flat and the integration of morphological and biological processes. This is done by assessing and modelling the ecological and morphological impact and effectiveness of the nourishment in preserving valuable intertidal areas. The main objectives were:
Building with Nature seized this opportunity to use this innovative nourishment to gain new knowledge on the driving processes of tidal flat morphology and ecology and the mutual interaction of morphological and biological processes. This is done by monitoring, modelling and analyzing the ecological and morphological impact of the nourishment and assessing its effectiveness in preserving this and other valuable intertidal areas. The main objectives were:
The activities in this project contribute to the BwN-programme objectives by generating ecosystem knowledge, design rules and practical expertise.
The data used in this Building with Nature case are the monitoring data from Rijkswaterstaat and additional benthic samples collected and analysed by Building with Nature. In addition, the data produced by the Argus database are essentially geo-referenced JPEG-pictures (or movies), saved in the Argus-database together with information on time and camera settings.
Consequences for fauna
The intertidal areas in the Eastern Scheldt are rich in benthic flora and fauna. They form a food source for higher trophic levels like fish and birds. One of the important intertidal benthic species is the cockle Cerastoderma edule. Since 1992 the policy is to reserve the cockles primarily as food for waders, except when high the stocks are high. Other characteristic benthic macrofauna species include worms like the lugworm Arenicola marina, bivalves like Macoma balthica, and in the shallow pools crustaceans like Crangon crangon.In recent decades the Pacific Oyster Crassostrea gigas has spread in the Eastern Scheldt, and now covers about 10% of the intertidal area.
Most benthic fauna have specific preferences regarding soil composition, hydrodynamics and emergence time. The latter is changing when the tidal flats are lowered and flattened due to erosion. Between 2001 and 2045 the emergence time is expected to reduce from almost 9 hours to less than 5 hours per day (Van Zanten and Adriaanse, 2008). The loss of intertidal area will cause a substantial change in species composition in this area. Macrofauna living in the lower intertidal areas (e.g. the Pacific Oyster) will take advantage of this temporary increase in living area, but species preferring higher areas will lose their habitat. This will influence their condition and the size of the population will decline (Van Zanten and Adriaanse, 2008).
The erosion of tidal flats also has an effect on birds foraging on them (mainly waders). As the area and the relief of the flats decrease, birds have less time to forage. The number of waders is predicted to decrease drastically when the intertidal area disappears (Van Zanten and Adriaanse, 2008; De Ronde et al., 2012).
Although the nourishment will reduce the erosion rate of the intertidal flats and increase the emergence time in that area, it will bury and kill all benthic fauna within its footprint. After some time the benthic fauna is expected to recolonise the area. Both benthic recolonisation and changes in physical processes have been monitored and analysed, before and after the Galgeplaat nourishment.
Results of benthic sampling
Sampling just before the start of the nourishment (June 2008) at locations within the planned nourishment area and in the surrounding area revealed similar chlorophyll-a concentrations, total biomass, species richness (i.e. total number of species present) and total density of benthic macrofauna. The chlorophyll-a concentration, a measure for the presence of microphytobenthos, dropped drastically after the nourishment and recovery to pre-nourishment values was still ongoing after 2 years. The nourishment killed almost all benthic macrofauna within its footprint. Shortly after the nourishment, in September and October 2008, only few organisms were observed in the samples, mainly being adult mud snails (Hydrobia ulvae), migrating from the surrounding undisturbed area onto the nourishment. This species was very common in the summer of 2008 and is capable of dispersing over long distances by e.g. floating. In 2009 we observed a further recolonisation of the nourishment by benthic macrofauna. The density matched the density of the reference area, but in both areas it was not as high as prior to the nourishment. The low density observed in the reference stations in 2009 and 2010 was due to the almost complete absence of Hydrobia ulvae, by far the most dominant species in the summer of 2008. Hydrobia ulvae is known to show great year-to-year variation in the whole Eastern Scheldt, and therefore these fluctuations are not abnormal (Troost and Ysebaert, 2011). Both biomass and species richness was in 2009 still lower on the nourishment. One year later, in July and October 2010, both biomass and species richness were similar to the reference area, but species composition and dominance still differed. Data from 2011 will be presented by Mesel et al. (in prep) as soon as available. The recolonisation of the benthic fauna on the nourishment showed a very patchy pattern, with some sites having a relatively rich fauna, whereas others showed hardly any macrofauna. The latter were mainly situated on the higher northern part of the nourishment, where the sediment dries out faster during low tide.
Based on images from the Argus-Bio station we were able to link the occurrence of wet and dry areas to the position of the sampling locations for the benthic macrofauna. The wet and dry areas refer to the state of the sediment during low tide. Some sites drained quickly after the water had retreated, while at others remained wet for a longer period of time. The recovery of the benthic macrofauna was different in the wet and dry areas. More particularly, the number of species, biomass and total density of the macrofauna recolonizing the nourishment was higher in the wet areas.
Further information on the relationship between sand nourishments and ecology can be found in Knowledge - Biogeomorphological effects of nourishment, NL.
Consequences for morphology
Wave and current action were expected to spread the nourished sediment and heighten the surrounding area. On a time scale of a couple of years the nourishment would compensate the erosion of a large part of the tidal flat and keep the flat above low water.
Results field measurements
Initially, the bed level has been raised by the nourishment from - 0.5 m NAP to + 0.5 NAP on average. After two years, only minor morphological changes of the nourishment were observed. Some redistribution of sediment occurred, but the overall change in sediment volume of the nourishment was only some 2 % (Holzhauer, et al., 2009 and 2010, Van der Werf, et al., 2011).
Field measurements showed that the maximum height of the nourishment slowly decreased in time. Erosion occurred mostly on the higher part in the north, whereas the southeastern part experienced some sedimentation.
Most of the eroded sediment is being transported to the north east and deposited along the edge of the nourishment.
The silt content of the sediment dropped directly after the nourishment, but returned to pre-nourishment values within a year (Holzhauer et al., 2010).
Results morphological modelling
The observed changes in hydraulic, morphological and relevant abiotic parameters provide information on the effects of the nourishment. An objective of the Building with Nature programme was to develop a modelling tool to support the planning and design of future nourishment, based on the observations from this one. Using the model bed roughness, developed under this project, and others such as the ANT-project (Cronin, 2011; Das, 2010), the effects of location, dimensions and sediment type of a nourishment were described including the influence of wave and current action. Simulations show that locally generated waves in combination with tidal currents play an important role in the transport of sediment around the flat, whereas tidal currents alone have hardly any effect on erosion. Qualitatively, the model is reproducing the general erosion and sedimentation patterns on the nourishment, as the sediment spreads in a northerly direction onto the intertidal flat, in line with the cumulative sedimentation/erosion plot from field measurements.The model was also used to estimate the effect of a hypothetical nourishment elsewhere on the Galgeplaatmodel. In terms of volume and height changes around the nourishment, placing it at the northern part of the Galgeplaat had a much larger effect than at the current location. This shows that the bed level and the location of a nourishment on the intertidal flat can play an important role in its morphological effects. The impact of the nourishment on existing morphodynamics was also investigated with the model. Although in reality the nourishment is undergoing less morphological change than expected, the model is qualitatively reproducing the sand redistribution pattern and the trend is also correctly simulated at several monitoring locations on the nourishment. Despite the overestimation of the rate of change, the model is a useful tool to examine the processes determining the redistribution of nourished sand and to test nourishment scenarios. Suspended sediment transport patterns correlate with the significant wave height around the nourishment, which indicates that the wave-induced near-bed orbital velocities influences the transport more than the current velocities alone.
The inclusion of biological features, such as oyster reefs, also had an impact on the simulation of current magnitudes over the intertidal flat and hence the patterns of sedimentation/erosion. As these features were included by modifying the bed roughness, only one type of effect is taken into account, the effect on the roughness height of the bed. Other effects are, for instance, sediment trapping, vertical mixing of sediments and modification of sediment properties.
Consequences for safety
The loss of intertidal area also involves safety issues. When dikes border on wide or deep waters like de Eastern Scheldt, waves can run up high during storms. When the flats erode, become lower and finally disappear, those dikes will be exposed to higher waves and so may need upgrading in the coming decades (Van Zanten and Adriaanse, 2008).
Primary purpose of the Galgeplaat pilot was to investigate whether nourishment can compensate the erosion of a tidal flat. Furthermore, the pilot offers the opportunity to gain a better insight into the biological and morphological development of an intertidal flat and the relationship between biotic and abiotic parameters. This will help improving the design of new measures with respect to the mitigation of the ongoing erosion of tidal flats in estuaries like the Eastern Scheldt in the future.
At the planning and design phase, the challenge was to meet a number of conditions:
One important argument in the design of the nourishment was the availability of sand from dredging activities in nearby navigation channels. Apart from this, the location for the nourishment on the Galgeplaat was chosen on the basis of the following criteria. It was:
From August to September 2008 130,000 m³ of sand was deposited on the Galgeplaat in a low circular mound, creating a kind of sand reservoir (figure 9).
First, a protective ring of sand was built, approximately 1 m high with a diameter of 450 m. The ring was made by bulldozers with sediment taken from the flat itself. At the south east side an opening was created for discharging excess water during the nourishment operation. At the inner side of the opening a settling-basin was excavated, in order to reduce the content of suspended matter in the effluent. The opening was oriented towards the nearby channel (Brabantsche Vaarwater).
This ring was filled with sand during the flood phase of the tidal cycle and spread by bulldozers during the ebb phase. The time phasing for the nourishment activities was based on the water level reaching the top of the ring wall. This allowed for a construction process with a controlled dredging-induced turbidity in the surroundings.
During and directly after the nourishment the concentration of suspended matter in the water around the tidal flat was monitored continuously at five locations, enabling an intervention if needed. The maximum sediment concentration in the water column near the commercial mussel beds was set at 50 NTU over and above the background sediment concentration in the water column. This maximum was exceeded twice during the nourishment activities. Both times the nourishment activities were allowed to proceed after an analysis of the situation by Rijkswaterstaat.
The sand for the nourishment, dredged from two adjacent channels, was slightly coarser than that on the tidal flat. This resulted in a small increase of the median grain size at the nourished location, but not in a change of the relative grain size distribution (Holzhauer and van der Werf et al., 2010).
The surface of the nourishment was not horizontal, but mildly sloping, with a maximum at NAP + 0.4 m and descending to NAP - 0.6 m in the south (figure 10). This was an unintended consequence of a lack of sediment at the end of the construction phase.
The development of the nourishment and its impact on the morphological and biological processes has been and will be further monitored for several years, primarily by Rijkswaterstaat. Additionally Building with Nature has taken extra samples of the benthic macrofauna and has installed an Remote monitoring of bio -and morphological developments - ArgusBio for morphological imaging and close-up images of birds, and other ecological features like the presence of algal mats and microphytobenthos.
A detailed monitoring program was set up to follow the morphological and ecological development of the nourishment on the Galgeplaat in space and time by Ramaekers, et al., (2007).
Morphological developments were monitored every month in the first year and later every third month, through visual inspections at the edge of the nourishment, sedimentation-erosion plots at 14 locations along three transects and elevation measurements with RTK-DGPS with a spatial resolution of 25 m. Wave and current velocities were measured with an ADCP during a month just after placement of the nourishment. Additionally a Waverider was installed 200 m southwest of the Galgeplaat in order to measure the wave climate continuously. During construction suspended matter in the water column was measured in the channels around the tidal flat. Permanent hydrodynamic and meteorological measurement stations located in and around the Eastern Scheldt were used for data on water level, wind speed and wind direction respectively (Holzhauer and Van der Werf, 2009; Holzhauer et al., 2010, Van der Werf et al., 2011).
Ecological measurements included regular sampling of benthic macrofauna, sediment characteristics (grain size) and chlorophyll-a to track the benthic recolonisation in time. Sampling sites on the nourishment (n = 10) are compared with reference stations (n=6) in nearby undisturbed sediments. Macrofauna samples consisted of three cores (3 × 0.005 m2) pushed 30 cm into the sediment within a 1-m radius of the sample site, located with a GPS. The macrofauna samples were sieved over a 1-mm mesh, fixed with 4% buffered formalin and stained with Rose Bengal, after which specimens were determined to the species level. Sediment samples for grain size and chlorophyll-a concentrations were taken with a 1-cm diameter tube pushed 3 cm and 1 cm into the sediment respectively. Samples were taken in June 2008 (before the nourishment), and shortly after the completion of the nourishment in September and October 2008. In 2009 and 2010 sampling was done in April, July and October. In July and October 2009-2012 additional samples (n=25 in total) were taken on the nourishment to get a better picture of the spatial patterns of recolonisation on the nourishment (De Mesel and Ysebaert, 2011, De Mesel et al., in prep).
Morphological Modelling of the Galgeplaat
The processes occurring on the intertidal flats and their effects on accretion and erosion are still not well understood, due to their highly variable and dynamic nature. Therefore in order to gain more understanding on these processes and how they affect the morphological development of the estuary, a depth-averaged, two-dimensional horizontal (2DH) Delft3D-FLOW hydrodynamic model was created for the Galgeplaat within the context of the ANT project (figure 12). Delft3D-WAVE (SWAN) was used to simulate waves on the same computational grid which was nested in a larger wave domain. In their turn, these coupled hydrodynamic and wave models were nested in a larger domain, the KustZuid (South Coast) model (figure 13). This larger model simulates the hydrodynamics (including waves) of the southern part of the North Sea, Western Scheldt and Eastern Scheldt.
The sediment transport and morphology module can be run online with Delft3D-FLOW and supports both bed-load and suspended load transport of non-cohesive sediments and suspended load of cohesive sediments. Here, uniform sand fraction with a grainsize of 200µm was applied.The nested models were set up for a spring and winter spring/neap cycle in April and November 2009, respectively, and for a longer period from September to December 2009 to investigate tje wind and wave forcing responsible for sedimentation and erosion. Once these forcing factors had been established, different simulations were set-up including the nourishment in the model, in order to examine the morphological effect at its current location and elsewhere on the Galgeplaat. Ecological effects also play a strong role on the morphological developments of intertidal flats. Oyster reefs and mussel beds modify current patterns and influence sediment movement. Algal and diatom mats play a role in stabilising the bed, especially during summer. Further model simulations were performed including these ecological effects, particularly investigating the roughness effects of biological features on current magnitudes. At locations where large patches of oysters exist, the simulation of current magnitudes in the model was improved with respect to measurements. This in turn reduced tje erosion on the intertidal flats. More details can be found in Cronin, 2012.
Several lessons-learned have been obtained on the field of morphology, ecology dynamics and (predictive) modelling
The nourishment has a relatively long-lasting effect on the height and emergence time at the nourished area itself. On the other hand, the nourishment only has a limited effect on the surrounding area on the flat. Several factors can explain this limited redistribution:
The nourishment buried and killed all benthic macrofauna. The recolonisation of the nourished area started directly after the nourishment was finished. After 2-3 years, the recolonisation appears to be most successful on the lower parts of the nourishment. The recolonisation seems to be driven by:
Although the emergence time was negatively correlated with benthic recolonisation on the Galgeplaat nourishment, this factor seems less relevant and correlated with the water content (dry vs wet areas) of the sediment. Shaping the nourishment with bulldozers is not recommendable, as this might additionally compact the sediment.
EcoShape pilot project