1 Introduction

North Sea estuaries represent highly dynamic environments and host specialised flora and fauna which have to cope with fluctuating conditions in relatively short time scales such as large salinity and temperature gradients, turbidity, and tidal currents. Within such a dynamic environment, various physiotopes and habitats exist in close neighbourhood, building a valuable and sensitive ecosystem of high connectivity (Elliott et al. 2019). Besides the term physiotope, the concepts habitat, ecotope, and biotope are often used to identify biological units of an ecosystem. In this study, the term physiotope solely defined by abiotic parameters (Farina 1998) was applied to describe such units. The reason for this is that the terms biotope, habitat (Natuhara 2006), and ecotope (Forman 1995) require information about biological factors which were not available for the time of the reference scenario or at the middle of the twentieth century.

Additional to their ecological value, estuaries are of high economic relevance with important harbour cities and touristic infrastructure (e.g. Meire et al. 2005; Van der Belt and Costanza 2011; Kennish 2019). This exerts pressure on such ecosystems, and against this background, Kennish (2019) grouped anthropogenic activities on estuarine environments into stressors that (1) change water quality, that (2) driven by physical factors which can have an impact on physiotopes, and (3) both at the same time. Such physical engineering activities include the construction of embankments, dykes and flood barriers, as well as the modification of estuaries by dredging fairway channels to provide safe navigation but often requiring long-term maintenance dredging. These interventions alter the hydrology and can influence ecosystems in an estuary (e.g. Meire et al. 2005; Dugan et al. 2011; Kennish 2019; Vauclin et al. 2020).

For example, the channelization of streams frequently applied in combination with the construction of groynes and dams for impoundments often results in an increase of the flow velocity (Kennish 2019). Such higher currents in a channelized estuary can change the patterns of erosion and sedimentation. From a civil engineer’s perspective this was often a desired effect (e.g. Franzius 1888) but inevitably had led to changes in the grainsize distribution of sediments (Franzius 1888; Hjulström 1935; Vauclin et al. 2020). In their study, Vauclin et al. (2020) investigated the influence of infrastructures such as dams, dikes, and groynes on sedimentation patterns. Interpreting grain size data from cores, the authors observed that after the implementation of such infrastructure homogenous sediments had replaced more diverse substrates. This is of relevance because the sediments provide the substrate for physiotopes and habitats (Kusnierz and Sivers 2018) of an estuary. Another effect of channelizing estuaries and higher currents is often an increased concentration of suspended matter like in the Humber (Morris and Mitchell 2013). Consequently, such a combination of higher currents and changing sediment grain size effects the occurrence of physiotopes (Kennish 2019). In addition, the diking of supratidal areas inevitably reduced the areas of many physiotopes directly (Kennish 2019). However, directly does not mean at once. Townend (2004) worked on changes in estuaries and figured out that the extent of physiotopes can react on hydraulic engineering activities on different timescales.

Considering this context, it was the aim of this study to quantify temporal changes in the extent of shallow water physiotopes in a series of North Sea estuaries, including the Elbe, Humber, Scheldt, and Weser. There are many studies which focus on wetland loss due to anthropogenic impacts such as Decleer et al. (2016) or Davidson (2014). Studies partly integrating intertidal areas can be found as well (e.g. Mander et al. 2007) but rare are studies integrating shallow water areas (e.g. Meire et al. 2005). However, our study integrates supratidal, intertidal, and subtidal physiotopes and, additionally, considers their position in the estuary with respect to salinity zones ranging from freshwater to polyhaline. We consider this to be important because these areas distinguished according to water level and salinity provide living space to different biological communities (e.g. Kennish 2019). The authors think that these physiotopes should be distinguished to better describe and assess changes on estuarine ecosystems.

Within this study, physiotope areas were quantified for different times and compared to a reference scenario in the nineteenth century prior to extended engineering activities in the estuaries. Through a comprehensive analysis of maps from different times compared with the time of implementation of engineering activities, the influence of such activities against the reference scenario was quantified. Additionally, the calculated recent extent of physiotopes serves as a tool to assess the impact of future changes.

2 Methods

Engineering activities with potential influence on hydraulic conditions in the estuaries like diking, land reclamation, channelizing anabranches, deepening fairways, cutting of tributaries, and constructing weirs were listed chronologically because changes on the hydrology can influence the sedimentation (Hjulström 1935).

Another focus was to map and measure the size of living space for organisms in the estuaries. Such living space is distinguished into units such as biotopes, habitats, or physiotopes. While the terms biotope, habitat (Natuhara 2006), and ecotope (Forman 1995) require a range of descriptive parameters including information about biological factors, the term physiotope is defined exclusively by abiotic criteria (Farina 1998). Since for the time of our reference scenario at the end of the nineteenth century and for the time at the mid of the twentieth century only abiotic parameters were available, the term physiotope was applied in this study to describe ecological units.

2.1 Time steps

We compared the extent of physiotopes within the four estuaries at three times: (1) at the end of the nineteenth century/early twentieth century, (2) at the mid of the twentieth century, and (3) in recent times at the end of the twentieth century. The situation at the end of the nineteenth century/the early twentieth century was used as reference status prior to most changes due to hydraulic engineering construction. For the reference scenario information on water depth in the respective maps were needed. A time step after 1950 in the middle of the twentieth century was the next choice because after the 2nd World War bigger ships were designed, requiring greater water depth (Table 1).

Table 1 Development of container vessels acc. to Tran and Haasis (2015)
Full size table

The age of the used nautical charts differed in accordance with the times of mapping the respective estuaries and their fairways (Table 2). The maps, nautical charts and aerial photographs (see detailed reference list in the annex) were digitised and geo-referenced using ArcGIS version 10.0. The extent of physiotopes within different salinity zones along the rivers was calculated by creating polygons. The systematic mistake for the calculated areas was below 0.15%, calculated as deviation of the sum of all areas in a salinity zone divided by the area of the salinity zone before separation into respective areas.

Table 2 Age of nautical charts considered for calculating physiotope areas
Full size table

2.2 Salinity zonation according to the Venice system (Caspers 1959)

Each investigated estuary was subdivided into salinity zones according to the Venice system (Caspers 1959). The basis to this was respective median values of chlorinity over 6 years (2004–2009) calculated from measurements within routine monitoring programmes along the rivers (Geerts et al. 2012). So freshwater, oligo-, meso-, and polyhaline zones were distinguished (Table 3). This zonation was also applied retroactively to the previous time steps.

Table 3 Lateral extension of salinity zones of the estuaries of Elbe, Humber, Scheldt, and Weser based upon the Venice system (Geerts et al. 2012)
Full size table

2.3 Physiotope areas

Within each salinity zone, physiotope areas were distinguished at different depth intervals, including below mean low water level (MLWL), mean high water level (MHW), and above MHW, e.g. sub-, inter-, and supratidal areas; synonymously, intertidal areas are often called tidal flats (Table 4). Within the supratidal areas marshes, summer polder, and stagnant water were distinguished. ‘Marsh’ was characterised as a supratidal area without a summer dike which is a dike lower than storm surge water levels, and ‘summer polder’ represents areas protected by a summer dike and ranged towards the main dike (Schuchardt et al. 2007). The ‘stagnant water’ classification is related to physiotopes located in dike forelands. Unfortunately, the information needed to determine ‘stagnant water’ and the ‘summer polder’ physiotopes were only available for the Sea Scheldt and the Weser estuaries.

Table 4 Classification of physiotope types
Full size table

3 Engineering activities with influence on the hydrology of the estuaries of Elbe, Humber, Schelde, and Weser

In geological time scales, North Sea estuaries have changed as the result of climate change after the Weichselian ice age (appr. 115,000 until 11,700 years before present (Cohen et al. 2013)). The melting of glaciers and inland ice shields flooded the North Sea basin due to a sea level rise of approximately 120 m. Since the commencement of sea-level monitoring in the nineteenth century, the measured relative sea level rise has comprised climatic and isostatic movements in the North Sea (e.g. Kiden et al. 2002), but with effects found to differ regionally. For example, in the German Bight, the mean high-water level rose approximately 25 cm while the mean low-water level increased by 10 cm/100 years since 1890 (Fickert and Strotmann 2007). For the Humber area, a minor increase in the rate of relative sea-level rise has been assessed at 4 mm yr−1 up to 12 mm yr−1 by end of the twentieth century (Environmental Agency 2011). Additionally, the patterns of precipitation can change, with such changes influencing the hydraulic situation in estuaries (Atkins 2002; Schuchardt and Schirmer 2005; Meire et al. 2005).

In addition to such naturally occurring processes, the four estuaries have been historically modified by hydraulic engineering activities, mainly land reclamation, to protect the landscape and human settlements from flooding and further measures to allow and maintain navigation. An important recent driver for change has been the development of increased vessel sizes which require maximum draughts of up to 16 m in the main fairways (Table 1), necessitating engineering work to ensure sufficient fairway depth. In addition to deepening of waterways, many tributaries were cut off from the tidal regime by the installation of weirs (Table 5). This combination of interventions has led to changes in hydrology resulting in increasing tidal ranges (Fig. 1) along the estuarine system (e.g. Franzius 1888; Meire et al. 2005; Schuchardt et al. 2007). The effects on the hydrology of estuaries from engineering measures to protect the land from flooding and to improve navigation are described in numerous publications (e.g. Franzius 1888; Meire et al. 2005; Elsebach et al. 2007; Fickert and Strothmann 2007; Schuchardt et al. 2007; Elliott et al. 2019; Vauclin et al. 2020).

Table 5 Hydraulic engineering activities in chronological order with potential influence on the hydrology of the estuaries
Full size table
Fig. 1
figure 1

Tidal range (A, main figure) from the nineteenth century to recent times at monitoring gauges in estuaries (data from Schuchardt et al. (2007) for Elbe and Weser and Meire et al. (2005) for the Sea Scheldt and for the Humber from ABPmer (2004)). The location of all estuaries can be found in the map (B)

Full size image

In Table 5 engineering activities which could have had an impact on the hydrology of the investigated estuaries are listed chronologically. Additionally, the point in time of the maps that were used to calculate the extent of physiotopes is inserted as well to see which maps can present the effects of implemented hydraulic engineering activities on the physiotopes.

4 Spatial change of littoral physiotopes

Over time the overall size of the shallow water physiotopes in the estuaries of the Elbe, Sea Scheldt, and Weser showed decreases in a range between 4 and 36%, whereas the respective areas of the Humber estuary remained quite stable since the nineteenth-century baseline (Table 6). Since the physiotope types are distinguished with a vertical component with respect to the water depth and a horizontal component with respect to an increasing salinity towards the mouth of the estuary, first, both aspects are presented separately.

Table 6 Extent of physiotopes (ha) from nineteenth century until present times–summing up the areas of sub-, inter-, and supratidal
Full size table

4.1 Extent of physiotopes distinguished with respect to water depth from the nineteenth century towards the end of the twentieth century

Without consideration of salinity along the contemporary estuaries more than half of the shallow water, physiotope areas are covered by subtidal physiotopes. The values range from 69% in the Sea Scheldt and 50% in the Elbe while the extent of intertidal areas varies between 17% in the Sea Scheldt and 40% in the Weser. So, the supratidal areas cover less than 12%. Comparing this with the values in the nineteenth century, it becomes obvious, that except of the Humber the areas of the littoral physiotopes decreased; however, mainly at the expense of supratidal areas which shrinked by more than half of the areas; especially in the Sea Scheldt the loss of the supratidal areas reached a value of 79% (Table 6).

The losses in the intertidal were below 11% and the changes in the subtidal varied in wide ranges from close to zero in the Sea Scheldt up to 7% in the Elbe. But by distinguishing the subtidal physiotopes into the parts deep water, slope, and shallow water it becomes obvious that the overall category ‘subtidal’ comprises opposing trends. For example, in the Sea Scheldt, the development in the subtidal was mainly related to the increase of deep water whereas slope and shallow water physiotopes as part of subtidal areas decreased. Altogether, a loss of intertidal areas was calculated like in the Weser and the Elbe.

4.2 Change of physiotopes distinguished with respect to salinity zones from the nineteenth century towards the end of the twentieth century

Considering the salinity reveals that the greatest loss of physiotopes took place along the freshwater and oligohaline zones. Since the nineteenth century in the Elbe, the Sea Scheldt, and the Weser between 33 and 56% of the freshwater areas were lost. The oligohaline zones showed a spatial reduction in a comparable range of 38 to 22%. The size of the mesohaline zone decreased to a lesser extent between below 18% whereas areas of polyhaline physiotopes remained nearly unchanged (Table 7).

Table 7 Extent of physiotopes (ha) in the estuaries subdivided in salinity zones according to Caspers (1959)
Full size table

4.3 Change of physiotope areas in freshwater and oligohaline zone

The loss of freshwater areas in the Elbe, the Sea Scheldt, and the Weser estuary summed up to many thousands of hectares (Fig. 2); however, allocated differently with respect to water depth. While the Weser showed a loss distributed equally to all physiotope types, the freshwater zone of the Sea Scheldt was characterised by an enormous loss of supratidal areas, and in the Elbe, a gain of intertidal areas was compensated by loss in supratidal and intertidal.

Fig. 2
figure 2

Changes of freshwater physiotopes areas (%) from the nineteenth century compared to recent times (the numbers integrated in the graph are changes expressed in hectare)

Full size image

The oligohaline zone of the Elbe estuary was dominated by a decrease in about 70% of the supratidal area accompanied by an increase of intertidal (Fig. 3). Subtidal areas decreased as well. The oligohaline zone of the Weser showed that the subtidal areas decreased by 50%, with supratidal also decreasing while the intertidal physiotope types showed a slight increase in extent. Again, the Sea Scheldt was dominated by a significant loss of supratidal areas.

Fig. 3
figure 3

Change of physiotope areas (%) in the oligohaline zones from the nineteenth century compared to recent times (the numbers integrated in the graph are changes expressed in hectare)

Full size image

5 Discussion

This study quantified the change of littoral physiotopes in estuaries opposed to engineering activities with potential impact on the hydrology of estuaries. In a similar approach based on more than 180 reports Davidson (2014) showed that more than 50% of wetlands were lost worldwide since the eighteenth century. On a more detailed scale, Decleer et al. (2016) compared the spatial extent of habitats in Flanders at different times from 1950 towards 2005. Both, Davidson (2014) and Decleer et al. (2016) remarked that estimations on area loss, subdivided into different habitat types are very rare in literature. However, an evidence-based management of an estuary needs exact data, e.g. on the extent of its physiotopes. Accordingly, our project provides data on a detailed scale for four estuaries in the North Sea region and their development since the end of the nineteenth century. The digitization of sea charts of the estuaries but at different times has enabled us to quantify the development of areas subdivided into supratidal, intertidal and subtidal. Considering the salinity combined with the water depth the study has facilitated a comparison of physiotope patterns over different periods and for different salinity zones along the estuaries.

The consideration of the salinity zone is an important information because the loss of areas was most distinct in freshwater and oligohaline zones. Our calculated loss might be a minimum loss because over time, straightening and deepening of fairways could have led to a movement of the saline waters upstream (e.g. Schuchardt et al. 2007). Accordingly, it can be assumed that upstream freshwater or oligohaline zones had a wider extent in times prior to engineering interventions altering the hydraulic conditions and our calculated losses tend to underestimate the changes.

Another focus of this study was to see whether hydraulic engineering activities influenced the distribution and extent of physiotopes. It is known that, e.g. the straightening of branches in an estuary to a main fairway and deepening of such fairways to ease the navigation of deep vessels has caused higher tidal propagation and increased tidal range. This had been shown for the Elbe (Fickert and Strotmann 2007; Kerner 2007), the Weser (Schuchardt et al. 2007) and the Sea Scheldt (Meire et al. 2005). In the studied estuaries most of those engineering activities started in the second half of the nineteenth century except of the Humber (Table 5).

In the Humber, as summarised by Morris and Mitchell (2013), most changes on the estuary’s shape had been taken place between Roman times and the beginning of the eighteenth century. As a consequence, the area of tidal Humber has been reduced from over 90,000 ha to an area of approximately 30,000 ha since the early 1700s (Boyes and Elliott 2006). Hence, the maps of the physiotopes along the Humber evaluated in this study considered a time span after most engineering activities with an impact on the extent of littoral physiotopes. Nevertheless, the data represent the actual distribution of physiotopes and they can be applied as a reference to assess spatial changes of physiotopes areas of the Humber in the future.

5.1 Overall trends considering physiotopes

The data show a loss of shallow water areas within most of the investigated estuaries except the Humber. In the other estuaries investigated in this study the biggest relative loss was observed from the freshwater and oligohaline and to a lesser extend in the mesohaline zone (Fig. 4). In contrast, the polyhaline areas remained relatively stable in extent over time.

Fig. 4
figure 4

Changes (%) of littoral physiotope areas in salinity zones compared to their extent end of the nineteenth century. The Elbe, Sea Scheldt and Weser show the highest losses of areas in the inner zones (freshwater and oligohaline) of the estuaries, and closer to the mouth (mesohaline) less areas diminished

Full size image

Historically most harbours were developed further upstream in the estuaries where better shelter against storms was naturally provided. This advantage of upstream harbours became less important in the twentieth century when the vessels became larger, and ultimately the upstream position of these ports often became a problem or constraint, as larger vessels less affected by storms also required greater water depths and widths. To guarantee the access of vessels to important upstream harbour infrastructure in the estuaries, capital and maintenance dredging practices have been adopted. In the course of capital dredging, the channels in the estuaries were straightened and deepened. Both factors brought about hydrological changes such as higher currents and increasing tides upstream and due to the position of the important harbours in the freshwater or oligohaline zones, the upstream zones were more effected than downstream parts of the estuary.

A comparison of the different littoral areas showed that the supratidal physiotope areas decreased the most (Fig. 5). Such a loss of supratidal areas could be attributed to the construction of dikes. Reasons for diking were land reclamation and later, the protection of arable land and infrastructure from flooding. Due to data availability being only for the Sea Scheldt, the supratidal areas could be distinguished in sub-categories (Fig. 6) and one can see that stagnant water and summer polder areas were lost completely with time because the main dike was positioned as close as possible towards the river.

Fig. 5
figure 5

Development of littoral physiotopes from the end of the nineteenth century to recent times. In all estuaries, the supratidal shows the highest diminishing

Full size image
Fig. 6
figure 6

Loss of supratidal (graph A) and subtidal (graph B) physiotopes along the Sea Scheldt (%), calculated as difference of areas in nineteenth century. Minus area mid of twentieth century and minus recent areas

Full size image

The extent of physiotopes in the oligohaline zones showed the same trends (Fig. 4). The supratidal areas were affected by different hydraulic engineering such as straightening the fairways, protecting the foreland of dikes with summer dikes, land reclamation by new dikes, and higher tidal ranges upstream. Focusing on the supratidal, one can see that this occurred in the Elbe, the Weser, and the Sea Scheldt (Table 6), but mostly along the Sea Scheldt. Here, the supratidal areas were reduced by 78% from the nineteenth century to recent times. All supratidal physiotopes were affected and summer polder and stagnant water areas were lost completely (Fig. 6).

5.2 Processes stimulating the observed spatial change of physiotopes in an estuary

The geological substrate underlying an estuary can be seen as a starting point to form physiotopes (Kusnierz and Sivers (2018). Based on that, the hydrology can also influence the physiotopes (Simon et al. 2017), because the velocity of currents determines which grain size of sediments is deposited or eroded (Hjulström 1935). Hence, engineering intervention altering the hydrology of estuarine systems will lead to changes in sedimentation and on the distribution of physiotopes as well. Often higher currents were a favoured effect from a hydraulic engineer’s perspective because the currents reduce sedimentation in the fairways (e.g. Franzius 1888). However, the elevated current velocity and tidal range increased the risk of flooding in upstream parts of the estuaries (e.g. Schuchardt et al. 2007). Consequently, upstream weirs were constructed and tributaries were cut off. The erosive power of the currents was controlled by groynes, training walls and shoreline structures. Again, these measures caused higher current velocities (Elsebach et al. 2007) and further penetration of the tides into the tidal river catchment (Fig. 1) (e.g. Meire et al. 2005; Fickert and Strotmann 2007; Schuchardt et al. 2007).

Of course, the question rises if this study can prove that a hydraulic engineering measure had influenced the physiotope patterns in an estuary. Solely based on our data, this is possible as far as the diking of supratidal areas is concerned. Interventions like straightening and deepening fairways, cutting of tributaries and installing weirs or groynes can change the hydrology in estuaries and hence the extent of physiotopes on different time scales. However, Townend (2004) worked on such changes in estuaries and figured out that a response on engineering activities such as channelizing can be expected in a range of years up to decades. Consequently, it needs more than evaluating maps of two to three time points to prove and figure out the concrete effects of a single engineering measure on the distribution of physiotopes.

Hence, our study could not identify the response of single hydraulic activity. Nevertheless, the study shows that changes in the distribution of physiotopes can be registered after important engineering measures in the estuaries. Remarkably, we observed nearly no spatial change of physiotopes in the Humber. Here, the most impacting engineering activities had been implemented prior to the studied maps. In the other estuaries of Elbe, Sea Scheldt, and Weser the reference scenario was a baseline prior to the most important engineering measures with an impact on the hydrology. This difference between maps of the Humber and the other estuaries enhances the plausibility that the mentioned engineering activities are the reason for the observed changes in the distribution of the physiotopes within the estuaries of the Elbe, the Sea Scheldt and the Weser.

To see the effects of a single engineering activity needs a dedicated monitoring programme and a more detailed look in time and space on a project scale. Examples of an appropriate scale are the investigation at Harriersand in the Weser estuary (Fig. 7) and the development of the Durme (Fig. 8), a tributary of the Sea Scheldt. Both are impressive cases showing the consequences of diking, straightening and deepening the main fairways on the extent of physiotopes. At Harriersand the originally braided fairways of the Weser were transformed into a single and deeper fairway by straightening and diking. Additionally, a former anabranch was enclosed by dikes and an island was created and protected with dikes (Fig. 7).

Fig. 7
figure 7

The Weser at Harrier Sand in the nineteenth century and today (2008). In former days the river meandered in a wider river bed while now the water passes in a straightened deep canal. Broad parts of the former river bed are sheltered with dikes or transformed into polders by summer dikes

Full size image
Fig. 8
figure 8

The change of physiotopes along the Durme from 1930 to 2001 shows an impressive loss of areas, especially the summer polders and stagnant waters are affected

Full size image

Looking at the Durme (Fig. 8), the effects of straightening and diking led to a significant loss of supratidal summer polder and stagnant waters until recent times. In both examples, the loss of shallow water physiotopes was inevitably linked to hydraulic engineering projects.

6 Conclusions and outlook

This study exhibits the temporal changes of littoral physiotopes in estuaries at the North Sea. Our approach to map such physiotopes at different time steps since the end of the nineteenth century allowed the quantification of their spatial alteration. As one important result, the collected data, maps and created shape files represent a comprehensive reference scenario for assessing the spatial effects of future management options or interventions in the Elbe, Humber, Sea Scheldt, and Weser estuaries.

Additionally, the consideration of salinity zones made obvious that most changes took place in the upstream parts of the estuaries in freshwater and oligohaline zones. Here the highest spatial loss along the estuaries was determined. This is a relevant aspect of this study because areas in each salinity zone host different biological communities. At the first glance, it may be logical to focus environmental management of estuaries on areas where the greatest spatial losses were calculated. However, such a spatial focus should consider the connectivity of habitats and potentially the continuum of ecosystems and their services along an estuarine system in particular.

For two reasons we consider this an important aspect, first, because the next generation of container vessels is pending, with a likely further increase in size. This raises the question if a further deepening of fairways still remains an appropriate management action given the loss of physiotope areas that might occur, and second, such losses might be compensated spatially, e.g. by creating new habitats in other places (e.g. Mander et al. 2007). However, this is a complicated task if the habitat connectivity, particularly between different salinity zones is to be maintained. Since a connectivity among different salinity zones is vital for a sustainable and resilient estuary (Elliott et al. 2019), it is recommended that the maintenance or enhancement of such connectivity is a priority for future management and compensatory interventions.