Salt marsh ecosystems are known for their unstable conditions, characterized by large fluctuations and gradients in various environmental parameters, which could give rise to different stressors (e.g., Fagherazzi et al., 2012). Important stressors encompass shifts in temperature and salinity, as well as specific hydrodynamic conditions including storm-tide frequency, exposure to wave energy, and deposition of sediment event layers. Although salt marsh systems are generally well adapted to such an environment, extreme variations in abiotic factors may constrain the optimal or tolerance ranges of the system (Davy, 2002; Davy et al., 2011). The natural feedbacks between plant growth, sedimentation and morphodynamics, which allows a salt marsh to adapt to sea-level rise, have been altered at many coasts by human activities, such as diking, drainage, and transition to pasture land (Kirwan and Megonigal, 2013). While natural stressors already affect the salt marsh ecosystem, anthropogenic impacts further contribute to these pressures. However, our understanding of how these human-altered salt marshes influence various ecosystem components and salt marsh organisms remains incomplete.

Most of today’s mainland salt marshes in the German Wadden Sea have been affected through land reclamation measures (Meier, 2004). Of these, the most important measures include the construction of groin fields made of brushwood fences in the foreshore of the dikes, which ensures a hydrodynamically calmed settlement space in which fine sediment can be deposited. This strategy results in accelerated accretion of the terrain, and thereby promoting the colonization of the tidal flats by halophytic pioneer plants (Hofstede and Schirmacher, 1996; Gettner, 2003; Hofstede, 2003). Given the economic viability as agricultural land and for resource extraction, the reclaimed land was then traditionally used for livestock grazing as well as haymaking and harvesting marsh sods (Meier, 2004; Esselink et al., 2009; Esselink et al., 2017). However, some measures changed the morphology of the salt marshes (such as diking and drainage), others influenced the local erosion and sedimentation processes (such as groins or dredging channels for navigation purposes) or even changed the ecological and biological conditions of the marshes (such as disrupting natural tidal patterns, water movement, and nutrient cycling; Suchrow et al., 2012; Kirwan et al., 2016a; Siemes et al., 2020; Gläsner, 2023). Although intensive foreshore management measures in Schleswig-Holstein (Germany) have been reduced since the establishment of the Wadden Sea National Park in 1985 CE and are nowadays mainly restricted to the areas near the dike base, the formerly cultivated salt marshes still show a rather poor conservation status according to the EU Habitats Directive (LLUR, State Office for Agriculture, Environment and Rural Areas, 2020). Yet the regularly flooded salt marshes provide various ecosystem services and therefore urgently need to be protected, conserved, and ecologically restored (Bromberg Gedan et al., 2009; Barbier et al., 2011). In principle, massive dikes that border the shoreline are the most common coastal protection structures in northern Germany (Kelletat, 1992). On the seaward side of the dikes, however, salt marshes provide regulating services by serving as a natural buffer zone against heavy swells and storm surges from the North Sea and by attenuating incoming wave energy and current-induced shear stress (Feagin et al., 2009; Kirwan and Megonigal, 2013; Möller et al., 2014). Thus, salt marshes likewise contribute to coastal protection and play an essential role for future-proof and sustainable coastal protection concepts as so-called ‘Nature-based Solution’ (NbS) approach (Narayan et al., 2016; Schoonees et al., 2019). This principle involves using green and soft infrastructure, such as salt marshes, to supplement or replace conventional coastal protection structures, thereby simultaneously enhancing the ecological value of the coast while preserving its protective function (Pontee et al., 2016). Another important regulating ecosystem service is the incorporation of organic carbon-rich suspended matter during flooding of the marshes at high and storm tides, making salt marshes one of the most important ‘Blue Carbon’ players (Kathilankal et al., 2008; Mcleod et al., 2011; Kirwan and Mudd, 2012; Müller et al., 2018).

The resilience of salt marshes depends largely on their ability to grow vertically by incorporating suspended sediments introduced by recurrent flooding around the mean high water level (MHW; Van Proosdij et al., 2006), ensuring sediment accretion rates higher than sea-level rise (SLR), and on their potential to migrate laterally further inland as sea level rises (Allen, 1990; Andersen et al., 2011; Kirwan and Megonigal, 2013; Kirwan et al., 2016b). In fact, an important response, and thus the underlying natural long-term stability of salt marshes, through dynamic adaptation to changing hydrodynamic conditions, is their migration to adjacent uplands (salt marsh transgression; Miller et al., 2021; Walters et al., 2021). A slowed SLR combined with sufficient sediment supply is the key driver of salt marsh formation, both by creating accommodation and by filling the newly formed space through sediment accretion (Watson et al., 2014; Fagherazzi et al., 2020). Nevertheless, coastal salt marshes are in urgent need of conservation and protection (Esselink et al., 2009), as ecosystems are expected to be more frequently and intensively exposed to the effects of extreme high sea levels (ESLs) under the influence of global warming by the year 2100 CE (Arns et al., 2015; Arns et al., 2017; IPCC, 2019; Lang and Mikolajewicz, 2020; IPCC, 2022). Although salt marshes are, in principle and to some degree, capable of creating new equilibria between their vertical growth and water level fluctuations through self-reinforcing interactions between plant communities and morphodynamics (Fagherazzi et al., 2020; Van de Ven et al., 2023), there is an upper limit to optimal SLR rates for salt marsh growth (Morris et al., 2002). If SLR rates are too high, the vegetation will no longer be able to achieve sufficient elevation gain to be still in equilibrium with sea level; a limitation that is further exacerbated when regional sediment availability is low (Morris et al., 2002; Bartholdy et al., 2010; Fagherazzi et al., 2013; Fagherazzi et al., 2020). Furthermore, the rate of transgression of salt marshes also depends on local morphology, aboveground vegetation, and engineered barriers in the hinterland (Kirwan et al., 2016a; Fagherazzi et al., 2020).

In this context, the question remains as to how and to what extent anthropogenically modified coasts and their cultivated ecosystems can withstand more frequent and more intense occurring ESLs and the associated increased run-up heights of storm surges. This question arises given that in the human-altered salt marshes on the Wadden Sea coast, natural features such as a varying morphology and diverse plant communities are no longer present, but instead the salt marshes are characterized by flattened and drained areas with altered soil chemical properties and mature successional stages of mono-dominant plant species (e.g., Van Wijnen et al., 1997; Morris et al., 2002; Bromberg Gedan et al., 2009; Davy et al., 2009). According to the Flora-Fauna-Habitat Directive (or Habitats Directive), such cultivated salt marshes show considerable deficits with regard to their conservation status, and it is assumed that this in turn may degrade the ecosystem services of the salt marshes in the long term (Wessels, 2021). Furthermore, the artificially straightened marsh areas that lie well above the MHW level hardly experience any tide-induced sedimentation and are thus almost detached from the marine environment – a circumstance whose long-term effects are not yet fully understood (Broome et al., 1988; Doody, 2008; Bromberg Gedan et al., 2009; Bunzel et al., 2021). Ultimately, the massive dikes prevent the spatio-temporal dynamics of the salt marshes accompanying SLRs (Kirwan et al., 2016a). Accordingly, today’s cultivated salt marshes along the German Wadden Sea coast are suspected of being particularly vulnerable under rising sea levels, as coastal narrowing and habitat loss can occur if the salt marshes are prevented from migrating upland by anthropogenic structures (Van der Wal and Pye, 2004; Pontee, 2011), and at the same time experience hardly any marine sediment input (Bunzel et al., 2021). Human interventions have thus significantly affected the natural physico-chemical gradients as well as the plant communities of today’s Wadden Sea salt marshes, and the cultivation of salt marshes will also have a considerable impact on their future development and survival.

To better assess the evolution of salt marshes under climatic and anthropogenic influences, it is crucial to reconstruct salt marsh evolution over the past 20^th century, when climatic and non-climatic factors increasingly influence coastal ecosystems (He and Silliman, 2019). Tidal flats and salt marshes are colonized by a variety of benthic foraminiferal species. Benthic foraminifera are unicellular organisms that can account for up to 50% of eukaryotic biomass in coastal areas (Moodley et al., 2000). Along the tidal frame, foraminiferal taxa exhibit distinct vertical zonation, and similar to salt marsh plants, they closely follow physico-chemical gradients along salt marsh successional stages (such as salinity, substrate, and pH that are linked to inundation and elevation of the terrain; Scott and Medioli, 1986; Scott and Leckie, 1990; Müller-Navarra et al., 2016). In the differentiation and taxonomic identification of the foraminiferal species, the test material and wall structure are of particular relevance. The test materials can be i) non-mineralized and organic, ii) agglutinated (i.e., they collect sediment particles from the surrounding habitat), or iii) consisting of calcium carbonate (porcellaneous or hyaline; Marszalek et al., 1969; Goldstein, 1999). These tests have a high potential to be preserved in the sediment for a long time, making foraminifera good bioindicators (e.g., Marszalek et al., 1969; Francescangeli et al., 2016). In the intertidal zone, the calcareous foraminifera occur preferentially in the tidal flats, which are very frequently to permanently flooded, while in the adjacent vegetated salt marshes, which are rarely to not flooded at all, primarily agglutinated foraminifera live (Horton et al., 1999a). In this regard, dead or fossilized tests of benthic foraminifera deposited at the coastal environments and integrated into the sedimentary archives provide information that can be used to reconstruct past ecosystems (Alve et al., 2009).

Although the use of foraminifera to reconstruct environmental conditions in different salt marsh areas has proven successful, comparatively little information is available from the Wadden Sea on the distribution of recent and fossil salt marsh foraminifera. This may be due to the fact that the cultivated salt marshes along the German Wadden Sea coast lack a pronounced landward rise in surface elevation, resulting not only in the absence of distinct vertical salt marsh zonation but also in the absence of foraminiferal zonation with respect to the tidal frame (Müller-Navarra et al., 2016), complicating the use of foraminifera as a proxy for reconstructing past salt marsh development and environmental change. Therefore, the question arises how past environmental changes and the associated responses of a coastal ecosystem can be reconstructed and assessed in this region using intertidal benthic foraminifera. Commonly, a general differentiation between tidal influences and marsh development can be reconstructed based on the relative abundance of calcareous tidal flat and agglutinated salt marsh foraminifera. However, as foraminifera are considered to be one of the microorganism groups most sensitive to coastal environmental degradation (Capotondi et al., 2015), intertidal foraminifera may also be suitable for assessing environmental and anthropogenic stressors on (cultivated) Wadden Sea salt marshes during the 20^th century. Hence, further information on changes in the ecosystem state can be derived by abnormalities in the test morphology of benthic foraminifera, as foraminifera can encounter unfavorable environmental conditions by modifying their test shapes (e.g., Geslin et al., 2000; Geslin et al., 2002; Polovodova and Schönfeld, 2008; Brunner et al., 2013; Martínez et al., 2023).

In this study, the historical benthic foraminiferal records of two sedimentary salt marsh archives from Schleswig-Holstein, central Wadden Sea, have been evaluated. The selected study sites have been heavily modified over the last century by human activities such as foreshore management measures and coastal protection structures. Today, the two sites differ such that since the establishment of the Wadden Sea National Park, many management activities have largely ceased at one site, while the other site continues to be managed. Since both sites have undergone different histories of intensive land use and degrees of modification (e.g., drainage frequency, grazing intensity, times of dike construction), whereas the influence of long-term climate change and associated storm surge dynamics along the central Wadden Sea coast should have exerted similar environmental stress at both sites, this study is twofold: i) to test whether intertidal benthic foraminifera (calcareous and agglutinated indicator species) can be used to reconstruct and to assess the evolution of local coastal ecosystems in the Wadden Sea area over the past century, and ii) to investigate how cultivated salt marshes in the Wadden Sea have responded to different non-climatic (foreshore management and coastal protection measures) but similar natural external influences (such as hydrodynamics), potentially indicated by test modifications of representative foraminifera in response to stress.

In Schleswig-Holstein, Germany, most of the mainland salt marshes have been anthropogenically modified to some degree (Figure 1A). This includes the typical use of groins made out of brushwood fences to promote foreshore development. The fences are installed on the tidal flats in front of the salt marshes, where they form a hydrodynamically calmed settlement space in which fine sediments can be deposited (Hofstede, 2003; Doody, 2008). At the same time, it is a common practice to construct drainage ditches on salt marshes, which improves soil aeration and reduces soil waterlogging, thus favoring the occurrence of species that are well adapted to dry to moderately moist and well-drained soils, such as couch grass (Elymus spp.; Bockelmann and Neuhaus, 2002). In this way, a mature vegetation stage of the salt marsh succession is achieved at an accelerated rate (Schröder et al., 2002; Rupprecht et al., 2015). The drainage ditches also reduce tidal currents so that fine sediments are deposited within the artificial channels. For maintenance, it is necessary to dredge the material accumulated in the ditches, subsequently placing the material on the surrounding salt marsh areas (Hofstede, 2003; Stock and Maier, 2016). In this way, drained groin fields in the foreshore enable a spatial expansion of the land–sea transition, and equally lead to improved accretion of the terrain and acceleration of salt marsh maturing (Erchinger, 1971; Von Lieberman et al., 1998). The areas are usually grazed, with mainly sheep being used for this purpose (pers. comm. LKN.SH, 2022). Since 1985 CE, however, German intertidal systems are under national nature protection and belong to the Wadden Sea National Park (Stock, 2003; Esselink et al., 2009). With the aim of ecologically restoring the salt marsh areas, the maintenance of artificial drainage ditches has been abandoned in many areas and grazing reduced or stopped. Since the original goal of land reclamation has been given up, no new groins are being built on the Schleswig-Holstein mainland coast today and the maintenance of existing groins has been gradually reduced since the 1980s (Esselink et al., 2009).

However, the investigated coastal salt marsh north of the embanked Friedrichskoog polder (Figure 1C), located in Dithmarschen of the federal state Schleswig-Holstein, is still actively managed, leaving the area streaked by drainage ditches and intensively grazed by sheep (Bunzel et al., 2020). Thus, the ~6 km² territory in the northern part of the polder is characterized by a short-grazed, monotonous vegetation consisting of a few grass species (i.e., Festuca rubra and Puccinellia maritima) and a flat but elevated surface morphology lacking the typical and naturally pronounced landward surface increase (Stock et al., 2005; Bunzel et al., 2020). As the salt marsh is bordered by a dike in the south and due to its unsheltered exposition to wave action, the seaward edge of the marsh is further characterized by a prominent erosional cliff (Figure 2, bottom). Due to the proximity to the Elbe estuary, the coastal waters near Friedrichskoog also contain high amounts of suspended sediment, causing a rapid infilling of the local artificial drainage ditches in the salt marshes during high tides. Consequently, the network of parallel ditches is dredged regularly every two to three years at Friedrichskoog (pers. comm. LKN.SH, 2020). The average tide levels recorded at the nearby tide gauge (Friedrichskoog Hafen, no. 110021) for the observation period 1986–2018 CE, provided by the Federal Maritime and Hydrographic Agency of Germany (BSH), are as follows: highest astronomical tide (HAT): 2.1 m above NHN (‘Normalhöhennull’ datum level); mean high water spring (MHWS): 1.7 m NHN; mean high water (MHW): 1.6 m NHN; mean low water (MLW): −0.4 m NHN; and mean low water spring (MLWS): −0.3 m NHN. Since the surface of the salt marsh at Friedrichskoog is nowadays 2.8 m NHN and thus well above the level of MHWS (Figure 2, bottom), the marsh is only flooded during storm tides when the water level is exceeding MHW by 1.5 m and severe storms push the water ashore (Gerber et al., 2016).

To assess the extent to which foraminiferal data reflect local environmental factors and site-specific conditions caused by different land use histories, or are more likely to reflect super-regional factors such as storm climate variability, a second salt marsh site was considered for comparison (Bay of Tümlau), with the data collected there being based on the studies by Müller-Navarra et al. (2019) and made available by the authors for this work. The Bay of Tümlau is located on the west coast of the Eiderstedt peninsula in North Frisia, Schleswig-Holstein (Figure 1B). The salt marshes in the Bay of Tümlau were also intensively managed in the last century, and the drainage ditches were maintained (dredged) every three to seven years (pers. comm. LKN.SH, 2017). Since the founding of the National Park, however, these foreshore management measures have been largely limited to the area near the dike, leaving the salt marshes largely untouched (Hofstede and Schirmacher, 1996). As a result, the artificial drainage ditches have filled themselves in over time or have been actively filled in and are now only relics, as are the brushwood fences on the tidal flats in front of the salt marshes (Esselink et al., 2017). Compared to the salt marsh site at Friedrichskoog, there is also a return of more diverse plant communities (i.e., the Atriplex portulacoides/Artemisia-type, Atriplex portulacoides/Puccinella-type and Festuca rubra) as well as a few naturally meandering tidal creeks in the Bay of Tümlau (Stock et al., 2005). Erosion also takes place in the Bay of Tümlau on the seaward side of the salt marsh, as can be seen from an erosional cliff, which is, however, far less pronounced and steep compared to Friedrichskoog (Figure 2, top). The average tide levels recorded at the nearby tide gauge (Tümlauer Hafen, no. 110016) for the observation period 2001–2013 CE, provided by the BSH, are as follows: HAT: 2.0 m NHN; MHWS: 1.6 m NHN; MHW: 1. 4 m NHN; MLW: −1.7 m NHN; MLWS: no data available. The surface of the salt marsh in the Bay of Tümlau is located 2.1 m NHN, which is also well above the MHWS level, meaning that flooding of the salt marsh can only occur during storm tides (Figure 2, top). For more details of the regional setting, see Müller-Navarra et al. (2016, 2019).

This study builds on the investigation of the 1.16 m long sediment-filled and U-shaped plastic profile (‘U-channel’) named GeoHH-FK that was retrieved from the erosional cliff face at the seaward side of the salt marsh at Friedrichskoog in November 2016 CE (54°2’35.02’’N, 8°52’20.41’’E; Figure 1C). For further details see Bunzel et al. (2020). The 1.06 m long sediment-filled U-channel from the Bay of Tümlau (TB13-1) was retrieved from the local erosion cliff by Müller-Navarra et al. (2019) in August 2013 CE (54°21’51.07’’N, 8°40’35.09’’E; Figure 1B). The terrain heights of the salt marshes and position of the sampled erosional cliffs were surveyed with the Leica Geosystems GNSS field controller (Viva Uno CS10) equipment. The surveyed data were referred to the nearby triangulation stations (no. 191908800, Friedrichskoog; no. 161803110, Bay of Tümlau), and post-processing was realized with the Leica Geo Office 8.3 software package. The resulting vertical uncertainty is less than 1 cm. Details for the triangulation station were provided by the State Office for Surveying and Geoinformation Schleswig-Holstein (LVermGeo SH).

The obtained sediment sequence GeoHH-FK was sampled at 0.5 cm spacing, in which every second sample was considered for foraminiferal analyses, giving 116 samples in total with a sediment volume of ~1.75 cm³ each. Sediment sequence TB13-1 was sampled in 1 cm steps and performed by Müller-Navarra et al. (2019), resulting in a sediment volume of ~2.56 cm³ per sample. After sampling took place, all samples were preserved in 96% ethanol to avoid degradation of the organic components and buffered with a carbonate solution to prevent dissolution of the calcareous components. Counting of foraminiferal individuals was carried out on subsamples (splits) from the wet-sieved 63–500 µm size fraction. The splits were obtained by means of a wet-splitter (Scott and Hermelin, 1993), into which each sample was placed for at least one hour to allow the sediment to fully and evenly settle down, producing eight equal splits per sample. The number of splits that need to be counted for their foraminiferal content depends on the minimum number of total individuals (Schönfeld et al., 2012). As foraminiferal species richness is rather low in salt marshes (Murray, 2006), foraminiferal analyses within this study are based on a target value of at least 100 individuals per split, since a number of 100 counted foraminiferal individuals is sufficient to provide statistical reliability, which is at least valid for those species making up >5% of the total fauna (Fatela and Taborda, 2002; Kemp et al., 2020). Taxonomical identification of the foraminifera from the GeoHH-FK samples was mainly based on Feyling-Hanssen (1972); Wright et al. (2011); Milker et al. (2015); Müller-Navarra et al. (2016), and Schönfeld (2018). As the analyzed sediments are rich in different genera whose individuals were often poorly preserved and seemed to be reworked, identification up to species level has often been impossible. Hence, corresponding individuals were grouped by genus, from which most genera making up less than 1% of the total fauna, while the total sum of this group accounted for 3% on average. This concerns, among others, genera such as Bolivinita, Bulimina, Cibicides, and Cibicidoides.

Based on the observations of Müller-Navarra et al. (2017), who identified abnormally grown tests of Entzia macrescens in active salt marshes on the island of Sylt, regular and irregular tests of E. macrescens were also distinguished from each other in this study and counted separately. The irregularly grown individuals were referred to as E. macrescens_irregular. To avoid the quantitative integration of potential intra-specific morphological variations, following Alve (1991), only severe test irregularities were counted for this study (Figures 3A–C). High-resolution images of E. macrescens_irregular originating from Friedrichskoog were generated by using a Keyence VHX-6000 microscope (Figures 3A–C). The preparation of the sediment samples from the Bay of Tümlau and the identification of the local foraminifera species were carried out by Müller-Navarra et al. (2019), who then also provided the foraminiferal counting data for this study and for the analyses that follow herein. Due to the different average quantities of individuals within a group, i.e., E. macrescens_irregular versus E. macrescens, as well as their trends over the entire period under consideration, the ratio between abnormal and regular E. macrescens tests was also calculated. Since the distribution of foraminifera is closely dependent on various abiotic factors (such as substrate, organic matter, salinity, oxygen or pH), which in turn are significantly controlled and constrained by tides and currents in the intertidal region (Camacho et al., 2015; Francescangeli et al., 2017), the relationship between the biota (foraminifera) and abiotic factors (coastal flooding) was statistically evaluated. In more detail, cross-correlations between absolute abundances of E. macrescens_irregular and storm tide frequency in the German Bight were calculated for both sites (Friedrichskoog and Bay of Tümlau) to compare the temporal variability and potential lags and leads between the occurrence of abundant abnormal tests and years with more frequent storm tide events. Cross-correlation analyses (with t-tests) were performed by using the PAleontological STatistics (PAST) software version 4.02 (Hammer et al., 2001) and on the previously resampled records (Δt = 1 yr) by applying a simple linear interpolation processed with the AnalySeries software 2.0, version 05/2005 (Paillard et al., 1996). Required age model compilations for the sediment successions (GeoHH-FK and TB13-1) build on the establishment of an integrated stratigraphic framework for active foreshore salt marshes, in which age markers were mainly obtained by radionuclides (¹³⁷Cs) and the human-induced mercury (Hg) pollution history (for further details see Bunzel et al., 2020).

Since the North Sea coast is basically exposed to several storm surges per year (Gerber et al., 2016), but the flood risk itself additionally depends on the prevailing wind direction and strength during a storm surge, both parameters were considered in this study. In addition, it is pointed out in literature that when assessing the impact of storm surges, strong wind conditions should last for at least six hours to account for both the different timing of MHW along the North Sea coast (Gerber et al., 2016) and the tidal phase, as there is also a positive relationship between rising tides and storm surges (Horsburgh and Wilson, 2007). In order to define meteorologically forced changes in sea level (storm surge events) that should have flooded the coasts with high probability, the German Meteorological Service (DWD) supplied historical wind data (for the stations at Husum, Büsum, and Cuxhaven) and the BSH provided seawater level observations (for the tide gauge at Cuxhaven; Figure 1A). All datasets were filtered to only consider severe wind conditions (hereafter referred to as storminess) and high seawater levels (hereafter referred to as storm tides) in this study. These include the daily mean values of the westerly winds (≥180° to 360° azimuth), which have wind speeds of ≥7 Beaufort (Bft), and the annual number of extreme sea level highs where the observed seawater level (daily peak values) exceeds the MHW by ≥1.5 m (Gerber et al., 2016). A correction with respect to tidal predictions, i.e., the subtraction of the full tidal cycles from the observed seawater level (Pugh, 1987), was not carried out for the Cuxhaven tide gauge record, because the tide surge interaction is rather complex and the generation of surges is associated with rising tides (Horsburgh and Wilson, 2007). Accordingly, the high seawater levels shown and discussed in here are the result of both storm surges and tides, i.e., storm tides.

Within the sediment sequence GeoHH-FK from Friedrichskoog, a total of 48 foraminiferal taxa were identified (27 taxa at genus level and another 21 taxa at species level). Of these, the majority belong to the recent calcareous benthic tidal flat foraminifera (on average ~71%), mainly consisting of Ammonia spp., Cribroelphidium selseyense, and Haynesina germanica. It was observed that the highest number of calcareous foraminifera occurred in the lower part of the sedimentary sequence (below 86 cm depth and chronologically before 1940 CE), with absolute abundance here reaching up to 3,255 specimens cm⁻³ (Figure 4A). The total amount of agglutinated foraminifera averaged 20% of the total fauna, with a maximum density of 795 specimens cm⁻³. Among the group of agglutinated foraminifera, E. macrescens is the most dominant species (on average ~13%), reaching a maximum density within the sediment of about 615 specimens cm⁻³ (Figure 4B). In contrast to the group of calcareous foraminifera, the agglutinated foraminifera represented by E. macrescens are somewhat more evenly distributed across the sediment sequence GeoHH-FK in terms of their absolute abundance, but slightly more E. macrescens individuals occurred in the central part of the sequence and after 1930 CE (Figure 4B). The separately counted foraminifera, i.e., the abnormally grown tests of E. macrescens, reached values of up to 250 specimens cm⁻³ at Friedrichskoog, which accounts for >30% of the total fauna in the respective sediment samples. The density of abnormally grown tests successively increased from about 78 cm depth in the sediment sequence (which chronologically corresponds to about the 1950s) until maximum values are reached at 43 cm depth (about the mid-1980s). Subsequently, the values drop back to the previous low level (Figure 5B).

For the sediment sequence TB13-1 from the Bay of Tümlau, Müller-Navarra et al. (2019) identified 24 foraminiferal taxa in total (12 taxa at genus level and another 12 taxa at species level). The proportion of calcareous (on average ~49%) and agglutinated (on average ~48%) foraminifera of the total fauna is almost equal here. According to the authors, the most common calcareous species are Ammonia batava, Cribroelphidium selseyense, Cribroelphidium williamsoni, and H. germanica, while the most abundant agglutinated species are B. pseudomacrescens, E. macrescens, and Trochammina inflata. Similar to the sequence of GeoHH-FK, it was observed for TB13-1 that the highest number of calcareous foraminifera occurred in the lower part of the sedimentary sequence, below about 90 cm depth (also corresponding approximately to the time before 1940 CE). The calcareous foraminiferal density in TB13-1 is with maximum 20,000 specimens cm⁻³ considerably higher than in GeoHH-FK (Figure 4A). With about 700 specimens cm⁻³, average abundances of agglutinated foraminifera within sediment sequence TB13-1 are similar to those of GeoHH-FK. The same applies to E. macrescens that reached up to 685 specimens cm⁻³ (on average ~37%). However, the absolute abundance trends of E. macrescens over time are not consistent with GeoHH-FK. For example, TB13-1 shows slightly higher densities from about 67 cm depth bottom-up (starting from the second half of the last century; Figure 4B). Further details on the foraminifera faunal composition found in the Bay of Tümlau can be obtained from Müller-Navarra et al. (2019). Among the abnormally grown E. macrescens, up to 30 specimens cm⁻³ could be identified at the Bay of Tümlau, which accounts for >9% of the total fauna in the respective sediment samples. Thus, the relative abundance of E. macrescens_irregular in sequence TB13-1 is much lower than in GeoHH-FK. Nevertheless, the trend over time is very similar to that of GeoHH-FK, so that the density of individuals increases gradually, i.e., from the mid-1960s to the 1990s (Figure 5B), corresponding to a depth range of 67 to 26 cm, reaching a maximum in the late 1980s and followed by relatively low values until today. However, E. macrescens_irregular appeared somewhat later in the Bay of Tümlau (in the mid-1960s), whereas at Friedrichskoog an increase of E. macrescens_irregular was already recorded from the first half of the 20^th century.

The results of the statistical cross-correlation analyses between E. macrescens_irregular (specimens density at both study sites) and tide gauge records of storm tides (number of events per year) showed that the best correlations (Bay of Tümlau: p-value = 0.005; Friedrichskoog: p-value = 0.001) were obtained when the highest numbers of abnormally grown foraminifera occurred three to five years after a period of frequent storm tides (Bay of Tümlau: lag = −3; Friedrichskoog: lag = −5).

The evaluation of benthic foraminiferal records from two different Wadden Sea salt marshes in Schleswig-Holstein (Germany) contributes to a better understanding of the linkages and responses between foreshore cultivation and changes in storm climate. The main conclusions are as follows:

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: PANGAEA Open Access library: https://doi.pangaea.de/10.1594/PANGAEA.927336.

The manuscript presents research involving animals (single-celled protists) that does not require ethical approval for its study.

Field work, sample preparation and foraminifera analyses were carried out by DB, who also designed the study, wrote the main body of the article and designed the figures. FF contributed to the discussion of the foraminifera data. YM and GS contributed to field work and were involved in all data discussions, providing critical comments and major revisions during the preparation phase of the article. All authors contributed to the article and approved the submitted version.