The influence of aeolian activity and the January 18
thstorm on
succession and L. loeselii in dune slacks on the south-western
coast of Texel.
B.Sc. Thesis
By: Stan van Manen
Student number: 10751955
Date: 6-July-2018
First supervisor: Dr. Annemieke Kooijman
Second supervisor: Dr. Gerard Oostermeijer
Abstract
Liparis loeselii is one of the endangered species, growing in dune slack habitats. As it is a pioneer
species, it thrives in early succession stages of dune slacks. When pH drops and the vegetation cover grows during succession, the habitat becomes unsuitable for L. loeselii. Nevertheless, aeolian activity is known to cause rejuvenation in soil and vegetation. Therefore, storm activity may influence the
succession processes in dune slacks. The aim of this research is to analyse the effects of the January 2018 storm on succession and rejuvenation. This will be conducted for two selected dune slacks on the south-western coast of Texel, the Netherlands. Samples will be taken during fieldwork and will be analysed in the laboratory. The results did not prove any significant effect of rejuvenation due to the January 2018 storm. Yet, aeolian activity may influence vegetation and soil succession. First of all, a decrease in pH compared to 2015 was measured for all the transect. Subsequently, bulk density values remained stable for 5 out of 7 transects. Open areas in and around dune slacks may increase the influence of aeolian activity on rejuvenation. Therefore, locations in the research area with lower vegetation covers, nearby footpaths or bare dune slopes are expected to be most affected by incoming fresh sand layers due to aeolian activity. High abundance of L. loeselii is expected to occur at locations with a low vegetation cover and a pH slightly higher than 6.5. This is because the areas with an optimum pH of 6.5 are highly covered with vegetation.
Content
Abstract... 2
1.0 Introduction...3
1.1 Research area...3
1.2 Dune slack succession...5
1.3 Storm discription...6 2.0 Methodology...6 2.1 Fieldwork...7 2.2 Laboratory work...7 2.3 Data analyses...8 3.0 Results...9 3.1 Correlations...9 3.2 T6... 10 3.3 T8... 12 4.0 Discussion...13 5.0 Conclusion...14 References...15
Appendix A: Transect field form...16
Appendix B: Grid point field form...16
Appendix C: Insignificant correlations...17
Appendix D: Maps T6...18
Appendix E: Maps T8...20
Appendix G: MATLAB script...24
1.0
Introduction
The orchid Liparis loeselii (L.) Rich., occurs in various regions throughout North America, central and northern Europe and Russia (Oostermeijer and Hartman, 2014; Kooijman et al., 2016). However, its overall distribution is declining, including in the Dutch regions (Oostermeijer and Hartman, 2014). The species is listed as endangered in the EU-Habitat Directive and has therefore the highest conservation priority for the regions it occurs (European Union, 1992; Oostermeijer and Hartman, 2014). One of the characteristic of the L. loeselii is its occurrence in dune slacks in coastal zones, which is an important habitat and protected by the EU-habitat directive (Oostermeijer and Hartman, 2014; European Union, 1992). L. loeselii is highly dependent on soil properties of these habitats, such as low C content and high pH (Oostermeijer and Hartman, 214). Dune slacks are dynamic habitats and endure succession in both soil and vegetation once they are formed (Kooijman et al., 2016).
There are several studies that conducted research in the abundance of the L. loeselii and its habitat properties. Nevertheless, combined research of dune slack formation, soil succession and the response of characteristic dune slack vegetation is scarce (Kooijman et al., 2016). Yet, this is important, because local and landscape dynamics could cause adverse effects of succession on characteristic species like L. loeselii (Kooijman et al., 2016). One of the landscape dynamics that could possible effect succession and the L. loeselii are storm dynamics. Earlier research on the south-west coast of Texel showed that the L. loeselii values rejuvenated the year after a storm in 2012 (Kooijman et al., 2016). An increasing amount of the inflow of sand due to high aeolian activity could be an important factor in the resetting of succession in dune slacks. The goal of this BSc-project is to analyse the influence of aeolian activity in selected dune slacks on the south-west coast of Texel. Therefore, the main question that this study will answer is: What is the influence of aeolian activity on succession in the dune slacks T6 and T8 in the south-western part of Texel? To answer this main question previous collected data will be compared with new data collected during two weeks of fieldwork.
Further in this thesis a broader description of the concepts will be given. This will be divided into three concepts. Firstly, the research area will be described, then, succession in dune slacks will be given and lastly, a description of the storm of January the 18th will be specified. Subsequently, the methods will
be described. Then, the main results will be given and afterward, the results will be discussed and analysed in the discussion. Finally, a conclusion of this bachelor project will be given.
1.1 Research area
This research will take place on the Hors, which is located at the southern tip of the Wadden island Texel (Fig. 1). The Hors is a shoal which has been attached to the coast in 1749 (Kooijman et al., 2016). The southern coastline of Texel has been growing since the 13th century, due to shoals forming on
seaward side of the ebb-tidal delta, called the Marsdiep (Ballarini et al., 2003). A new shoal is expected to attach to the coast, called the Razende Bol-Noorderhaaks, as it is approaching and has been above the low water line since 1925 (Kooijman et al., 2016). The growing coastline in southwestern direction stimulated the formation of new dune slacks and ridges on the beach plains (Ballarini et al., 2003). When a new shoal connects to the coast, it forms a significant sediment source for dune formation. This sand is captured by vegetation and forms a new dune ridge. This will continue to grow, until a parallel new forming ridge disconnects the older ridge from the sand source (Ballarini et al., 2003). The height of the dune ridges in this area varies from 5 to 15 meters and they remain well preserved due to the growing coastline (Ballarini et al., 2003).
The location of the research area is situated in two dune slacks (Fig. 2). Dune slacks are low-lying areas in coastal dunes with a high water table. During winter and spring the water table is at the soil surface, while in summer it may drop 50-100 centimetres (Grootjans et al., 2002). There is a distinction between primary and secondary dune slacks. Primary dune slacks are characterized by the seclusion of sea water influences due to newly developing dune ridges, while secondary dune slacks are characterized by formation due to sand blow-outs until the water table is reached (Lammerts et al., 2001).
The two selected dune slacks (Fig. 2) for this research are part of an earlier succession series
research conducted by Kooijman et al., 2016; consisting of eleven dune slacks formed between 1973 and 2010. The two dune slacks are formed in 1999 (T6) and 2003 (T8) (Kooijman et al., 2016). They are probably primary dune slacks, as other neighbouring dune slacks form parallel to T6 and T8 (Fig. 2). Dune slack T6, showed peak values of L. loeselii around 2010, then rapidly declined after, however a storm in 2012 triggered rejuvenation. T6 can be separated into three parts:
1) Existing dune slack without rejuvenation and low L. loeselii numbers (T6-T3).
2) Rejuvenated dune slack with high numbers of 6000 L. loeselii around 2015, now rapidly declining (T6-T2).
3) Rejuvenated dune slack with L. loeselii numbers still increasing, possibly due to high aeolian activity (T6-T1).
Dune slack T8 showed peak values of L. loeselii in 2015-2017 and was firstly found in 2010. T8 can be separated into four parts:
1) More or less stable part with approximately 2000 individuals of L. loeselii in 2015-2017 (T8-T1). 2) An area with high numbers of L. loeselii in 2015 but low numbers in 2017.
3) An area with low numbers of L. loeselii in 2015, but high numbers in 2016 and 2017 4) An area with strong increase in numbers of L. loeselii, possibly due to high aeolian activity.
Figure 1, Aerial photograph of 2017 of the southern part of Texel (PDOK, 2018).
1.2 Dune slack succession
Abiotic and biotic factors change over time in a habitat, causing species to disappear and other species to thrive when they are better adapted to the new conditions. This process is called succession. In dune slacks an important cause of succession in the vegetation composition is the groundwater table, depending on the system’s groundwater and precipitation variablity (Grootjans et al., 1991). Furthermore, the accumulation of organic matter in the soil is another cause of succession in dune slacks. In the first year of excistance, vegetation cover is low in dune slacks, resulting in a low production of organic matter (Grootjans et al., 1998). Lastly, pH is an important factor in dune slack succession as well. pH influences nutrient availability via organic matter decomposition. In the dune slacks on the Wadden islands calcium carbonate is the main influencing factor of the soil pH (Grootjans et al., 1995).
The L. loeselii is strongly depending on the stage of the soil succession and the habitat
characteristics. It grows on base-rich or calcareous, nutrient poor coastal dune slacks (Oostermeijer and Hartman, 2014). In addition, it favours habitats with open vegetation, which means in case of the dune slacks, that it prefers early succession stages (Odé and Bolier, 2003). A study by Kooijman et al., 2016 showed that the species peaks in young, lime-poor dune slacks, with a relatively high pH and a low soil organic matter content. Its peak in abundance is approximately after 10 years, at a pH-level of 6.5, and it slowly decreases after that. The life span of the L. loeselii is roughly 34 years, however mowing can increase this. The soil conditions become unsuitable for the plant when pH-H2O is below 5.8, pH-KCI below 5.6 and C-content is higher than 4.3% (Kooijman et al., 2016). A graph made by Oostermeijer and Hartman et al., 2014 confirms that the species thrives with a high pH and a low vegetation cover (Fig. 3). The population abundance decreases again when the pH of the soil gets too low and the vegetation cover too high.
Figure 3, Liparis loeselii population, soil pH and vegetation cover in dune slacks over time (Oostermeijer and Hartman, 2014). 1.3 Storm discription
On Januari the 18th an exceptionaly strong storm disrupted the Netherlands (KNMI, 2018). It is
one of heaviest storms of the past 50 years, with gusts of wind measured up to 143 kilometers per hour. The storm triggered the highest weather alarm possible in the Netherlands, code red. Additionally, the repeating time of a storm like this is estimated on once every eight years (KNMI, 2018). Although the storm was small in size, it caused a lot of damage. Despite the damage, the high windspeeds increaseed the amount of aeolian activity on Texel. This could have increased the amount of sand on the topsoils in the dune slacks, which may rejuvenate the species L. loeselii and renew soil succession (Kooijman et al., 2016).
2.0
Methodology
This research is been conducted in three main stages: a literature study, data gathering and data analysis. The writing process of the final report has been the overlapping part of these stages and has been directed over the whole research period. In table 1 you can find a workflow diagram of the methods stages and the most important activities conducted. Following in this section, a description of the used methods of the data gathering, data analysis and the literature study will be given.
Literature study Data gathering Data analysis
Research Proposal - Introduction - Theoretical Framework Final Thesis - Introduction - Discussion Fieldwork
- Transect point sampling
- Transect point profile description - Grid point profile description
Laboratory work
- Bulk density - pH and EC values - Organic matter content
GIS analysis
- Transect analysis - Grid analysis - Aerial photographs
Statistical analysis
- Linear regression analysis - Boxplot analysis
Discussion
The first part of this research has been conducted as a literature study to get into the main topics. It is important to get an adequate knowledge about the earth scientific and biological processes that formed and are occurring in the given research area. These processes are stated in the introduction of this thesis. Research conducted by earlier studies in this same research area are used for this purpose. However, literature describing dune processes in similar research areas have been used as well.
The second part of this research is the data gathering. This has been divided into two activities; the fieldwork and the laboratory work. Both will be described more detailed in the following part:
2.1 Fieldwork
The fieldwork area contained two selected dune slacks, T6 and T8 (Fig. 2). T6 has been divided into 3 areas (T6-T1, T6-T2, T6-T3), in which each area contained 1 transect with 10 sampling points (Fig. 4a and Appendix D1). T8 has been divided into 2 areas (T8-T1 and T8-T2, T8-T3, T8-T4) in which T8-T1 contained 1 transect and area T8-T2, T8-T3, T8-T4 contained 3 transects, each transect containing 10 sampling points as well (Fig. 4b and Appendix E1).
Figure 4, dune slacks T6 (a) and T8 (b), including the transect points.
The total amount of samples collected during the fieldwork is 70 samples. For each sampling point a field form has been documented (Appendix A). This field form contains information about the GPS location (using a Trimble ArcGIS GPS device), a soil profile description with special attention for thin layers of sand at the top soil, water table height, and a soil surface characteristic description containing
information about coverage of bare sand and coverage of vegetation types.
Additionally, for each transect point two soil samples were taken to measure the bulk density, pH and EC values and the organic matter content in the laboratory. For this purpose, 5 centimetre long metal pF rings, taking 100 centilitres of topsoil, have been used.
Furthermore, for both dune slacks, T6 and T8, grids containing respectively 92 and 180 grid points have been made. A field form (Appendix B), almost similar to the transect field form, was used and contains information about the GPS location, a soil profile description with special attention for thin layers of sand at the top soil, and a soil surface characteristic description containing information about coverage of bare sand and coverage of vegetation types. The location of the transects were based on the location of earlier sample points and expectations of possible rejuvenation locations for the L. loeselii.
2.2 Laboratory work
All the 70 transect samples have been analysed in the laboratory at Science Park Amsterdam, and have been measured on bulk density, Electrical Conductivity (EC), pH and soil organic matter content (SOM). Table 2 gives the methods on how these have been measured. The laboratory and fieldwork results have been digitalized using Microsoft office 365 Excel 2017.
Soil measurement Laboratory method
Bulk Density of soil
The 100 centilitres of the soil sample will be put in the oven for 48 hours at 105°c. After the soil is dried in the oven the sample has been weighted. Knowing the volume and weight of the sample after it has been in the oven, the bulk density has been calculated as:
Dry sample weight Bulk density = ______________________ Volume original sample
pH and EC value of
soil 25 ml of demineralized water has been mixed with 10 grams of the soil sample. Then it has been shaken for 2 hours, followed by a resting period overnight and another 20 minutes in the shaking machine the day after. The values were then measured using a pH and EC electrode.
Soil Organic Matter
content (SOM) The soil samples that remained after the bulk density measurements were used again. The samples were dried again in the 105°c oven for one night. Then 5 grams of soil has been weighted and placed in the oven for 16 hours at 375°c. The soil sample has been weighted again after this process and the SOM has been calculated as:
SOM = Wet sample weight – Dry sample weight
2.3 Data analyses
The data analyses can be divided into the statistical analyses and the GIS analyses. Firstly, by using Excel correlations between variables within the transect and grid points were checked. This has been done by fitting a linear regression model for two selected variables, such as percentage of vegetation cover and the soil organic matter value. If these linear regression models had a R-squared value of ≥0.4, the correlation was considered as valid and calculations for the grid point variable could be made using the given linear regression formula. Correlations with R-squared values of <0.4 were considered to be invalid.
Subsequently, sample values taken in earlier research have been analysed and compared to the 2018 data by using Matlab. Firstly, the data has been checked on outliers, which were taken out of the measurement. Then, boxplots were made to compare our data with earlier research and to analyse differences in variables within the transect dataset. The data that has been made available for dune slack T6, are measurements dating from 2015. Also data from the years 2010 and 2014 have been made available. However, this data is only sampled around T6-T3. For the years 2010, 2014 and 2015 data has been collected in dune slack T8. However, this consists only of 4 data points, two taken around T8-T1 and two taken around T8-T2. The Matlab script can be found in Appendix G.
The spatial GIS analysis has been conducted using ArcGIS 10.5. Aerial photographs of the year 2017 have been retrieved from: Publieke Dienstverlening Op de Kaart (PDOK). The coordinate system that has been used is: Amersfoort RD. The Excel data has been read into the used Geodatabase in ArcMap. Then, polygons where drawn to mark the transect and grid areas. Interpolations of the data were made using the Inverse Distance Weighted (IDW) tool. This tool calculates cell values with a linear weighted interpolation, so the data points that are closest to the cell have a linearly greater influence compared to
points further away. The power parameter used lies between 2 and 3, with 3 as the highest possible value. To fit the interpolations into the fieldwork areas, the tool Extract my Mask was used. Finally, the maps were completed in ArcMap and saved as jpg image.
As the highest and the lowest value of pH and the SOM between the two dune slacks differ a lot, the scales of the maps can be misleading. For T6, the pH scale ranges 5.9 (blue) to 6.9 (red), while for the SOM it ranges from 0.0 (light) to 0.3 (dark). For T8, the scales ranges from 6.3 (blue) to 7.5 (red), while for the SOM from 0.0 (light) to 0.15 (dark). As the colour-schemes in both dune slacks are the same, but the scales differ, a pH or SOM value can represent another colour in both maps. Therefore, to make a comparison between dune slack T6 and T8 a pH map of both dune slack is made with the same pH scale. This was not useful for the SOM, because the scales differed too much to make a useful map. The pH maps with similar scales can be found in Appendix F.
Each map that has been shown in the results can also be seen in the original size, in the appendixes D, E or F.
3.1
Results
3.2 CorrelationsFor the applied linear regression models in the data of dune slack T6, two significant correlations have been found. The first correlation that has been found is related to all the transects in T6 and is the relation between pH and soil organic matter content (Fig. 4a). This shows a decreasing pH for an
increasing SOM content; with a significant R2 value of 0.4345. The second correlation found in the sample date is the correlation between pH and vegetation cover, only for T6-T2 (Fig. 4b). This states a decreasing pH for an increasing vegetation cover; with a significant R2 value of 0.4598. For the linear regression models applied on the T8 data one significant correlation has been found. This correlation is related to the pH and vegetation cover (Fig. 4c). Stating a decreasing pH with an increasing vegetation cover; with a significant R2 value of 0.4967. Other correlations tested by a linear regression model, however insignificant, can be found in Appendix C.
3.3 T6
The following maps establish the spatial distribution of pH (Fig. 5a), vegetation cover (Fig. 5b), SOM content (Fig. 5c) and pH with an optimum for the L. loeselii (Fig. 5d) in dune slack T6. Each map can be found in the original size at Appendix D
Figure 5, Maps of pH (a), Vegetation cover (b), SOM (c) and pH optimum L. loeselii (d) of dune slack T6
The maps show the higher pH-values (≥6.6) for T6-T1 and the western part of T6-T2. Vegetation cover and SOM content are also low in these areas. Lower pH-values (<6.6) with high vegetation cover values and with high SOM contents can be found in T6-T3 and the eastern part of T6-T2. The pH optimum for the L. loeselii lays mainly in the eastern part of T6-T2, however, also in some parts of T6-T3.
In figure 6 the changes in pH and bulk density over the last eight years in T6-T3 can be seen. The changes in pH and bulk density for T6-T1 and T6-T2 can be found in the figure 7 and 8. The pH of the soil in T6-T3 increased between 2014 and 2015. The new data in 2018 states a decrease in median pH of approximately 0.6 since 2015. The bulk density date does not show any big changes for 2018. The pH median in T6-T1 dropped from 7.6 to 6.8 and a slight decrease in bulk density is measured as well (Fig. 7). For T6-T2 a similar drop in pH is measured, although the bulk density stayed equal to 2015 (Fig. 8).
Figure 6, Boxplots of pH (a) and Bulk Density (b) for transect T6-T3
Figure 7, Boxplots of pH (a) and Bulk Density (b) for transect T6-T1
3.4 T8
The following maps establish the spatial distribution of pH (Fig. 9a), vegetation cover (Fig. 9b), SOM content (Fig. 9c) and pH with an optimum for the L. loeselii (Fig. 9d) in dune slack T8. Each map can be found in the original size at Appendix E
Figure 9 , Maps of pH (a), Vegetation cover (b), SOM (c) and pH optimum L. loeselii (d) of dune slack T8
Figure 9a states that the lowest pH-values (<6.6) can be found in T8-T1. This transect completely consists of a vegetation cover of 100 percent and shows the highest SOM values in the dune slack. T8 transects T2-T3-T4 consist of more variation in pH, varying from 6.5 to 7.5. The vegetation cover is less dense throughout these transects too and SOM content is low over the whole area. Higher pH-values and lower vegetation densities occur further in southwestern direction in T8 (Fig. 9a). The optimum pH-values for the L. loeselii were found in T8-T1 and around some spots in T8-T2 and T8-T3 (Fig 9d).
In figure 10 the changes in pH and bulk density can be seen over the last eight years for dune slack T8. The pH-values dropped compared to most transects in T8, after a slight increase from 2014 to 2015. Additionally, the bulk density values only dropped for T8-T1 and stayed almost equal for the other transects.
Figure 10, Boxplots of pH (a) and Bulk Density (b) for dune slack T8 transects T1, T2, T3 and T4.
4.0
Discussion
The results state that pH has decreased since 2015 in both dune slacks, T6 and T8. All the transects showed a lower pH compared to 2015. This indicates that soil succession has probably taken place since then. The significant influence of the January 2018 storm on succession will therefore not be stated. However, the influence of aeolian activity on both dune slacks, is therefore not excluded.
Rejuvenation of L. loeselii due to storm activity occurred in 2012, so influences of wind activity are known (Kooijman et al., 2016). Besides, there are some results in this research that could give an indication of the presence of the influence of aeolian activity on dune slacks. In addition, there are several
complications for this research that made a strong conclusion difficult. In this section, these results and complications will be analysed and discussed.
As stated before, the pH-levels have become lower in all transects, probably because of
succession. It is plausible that an increase in organic matter in the top soil caused this drop. An increase in soil organic matter in the top soil would likely cause a decrease in bulk density, as the ratio of heavier sand particles would have been smaller compared to organic particles (Kooijman et al., 2016). However, the data of 2018 showed a decline in bulk density values for transects T6-T1 and T8-T1, while a stable bulk density value was found for T6-T2, T6-T3, and for T8-T2, T8-T3, T8-T4. Aeolian activity could be a reason for this, as extra sand increases the bulk density. The pH of the soil could be unaffected by this new layer of sand, as it is a fresh new layer and time is needed to influence the pH. This possible increase in pH due to fresh sand was therefore not measurable in the samples yet. In addition, the decrease in bulk density for T8-T1 is explainable as the SOM is much higher compared to the other transects in T8.
The effects of aeolian activity depends on environmental characteristics of the dune slack. Vegetation and bare sand cover of the surroundings could possible influence these effects. After a storm in the winter of 2012, rejuvenation occurred in some parts of dune slack T6 (Kooijman et al., 2016). Rejuvenation of soil succession and L. loeselii was measured in T6-T1 and T6-T2. In T6-T3 no rejuvenation occurred. This is possibly due to a higher vegetation cover around T6-T1 and a lower vegetation cover around T6-T1 and T6-T2 (Fig. 5b). Also, between transects T6-T1 and T6-T2 lays a footpath. This is could possibly increase the amount of fresh sand in these transects (Kooijman et al., 2016). Based on this knowledge, expectations of possibly effected or non-effected locations by aeolian activity can be made for dune slack T8. First of all, T8 is expected to me more influenced by aeolian activity as the vegetation cover is lower in most areas compared to T6, especially around transect T8-T2, T8-T3, T8-T4. However, estimations inside T8 can be made as well. Figure 9b shows that in and around T8-T1 no bare sand is found, as vegetation cover is 100 percent. Aeolian activity is expected to be of less influence on
be effected by aeolian activity. This is not only because the vegetation cover in this area is low, but also because of bare dune slopes directly located at the southern edge of this area (9b).
The age of dune slacks T6 and T8 in 2018 are respectively 19 and 15 (Kooijman et al., 2016). L.
loeselii already passed its peak presence in T6, while T8 is in its last peak presence year (Kooijman et al.,
2016). This, however, is not in line with the data collected and the assumption that peak values of L.
loeselii occur around a pH of 6.5 (Kooijman et al., 2016). Figure 5d shows that the pH-values in most of
dune slack T6 reached this peak value now, or has an even higher pH. The same is occurring in dune slack T8, where most of T8-T1 has a peak pH-value of 6.5, and where transects T8-T2, T8-T3, T8-T4 had a minimum pH-value of 6.5 (Fig. 9d). This could indicate that aeolian activity effected pH in these areas, as lower pH-values where probably expected at this age. Subsequently, in dune slack T6 and T8, where the pH-value is at its optimum for L. loeselii, the vegetation cover is also high. Vegetation cover is almost at every pH-optimum 100 percent. Therefore, high abundance of L. loeselii is not expected in these areas, as it thrives with a lower vegetation cover and lower vegetation heights of other species (Odé and Bolier, 2003);(Oostermeijer and Hartman, 2014). As soils with a pH of 6.5 and a low vegetation cover are scarce the research area, a high abundance of L. loeselii in dune slack T6 and T8 is expected at areas with a higher pH than 6.5 and a less dense vegetation cover.
Currently, the time span of abundance for L. loeselii in dune slacks on Texel is estimated on 34 years (Kooijman et al., 2016). Soil and vegetation succession are important factors in this time span, as pH-values become too low and vegetation cover too high over time (Oostermeijer and Hartman, 2014). However, rejuvenation of soil and vegetation due to aeolian activity can be stimulated. This can be done by making open dune slopes around the dune slacks and lowering vegetation cover and height in the dune slacks. This could increase the amount of fresh sand in dune slacks, distributed by aeolian activity. A sufficient amount of rejuvenation could possible expend the abundance span of L. loeselii in dune slacks.
Measuring the effect of the January 2018 storm was complex. The first and most important reason for this is that sufficient data was missing. The collected data was compared with measurements dating from 2010, 2014 and 2015. Therefore, changes in pH, bulk density or SOM were irrelevant to relate to the January storm. pH could possibly have been increased after the storm, however, data from before the storm was missing in this research. Between 2015 and the January 2018 storm, succession and possibly aeolian activity have taken place. This has probably influenced the soil properties, such as pH and bulk density. The second reason for complexity in this research is caused by the sample points. A lot of sampling points taken in the earlier years where inaccessible during the fieldwork. Therefore, other sampling points are taken, making it less significant to compare new data with the previous years. Additionally, the amount of sample points taken in the earlier research was small compared to 2018. Data collected in dune slack T8 can only be compared with 4 data points, of which two were located in the area of T8-T1 and two were located around T8-T2. Drawing conclusions with such a small sample size is difficult, and so is making assumptions for new transect point located far away from the previous sample points.
5.0
Conclusion
This study did not find any significant effects of the January 2018 storm on rejuvenation or succession in dune slacks on the south-western coast of Texel. It was not possible to draw any hard conclusions on the effects of the January 2018 storm out of the collected data, and the data collected in the years 2010, 2014 and 2015. Although the effects of the storm were not measurable, aeolian activity may effect soil and vegetation succession. For both dune slacks, T6 and T8, a decrease in pH compared to 2015 has been measured. Therefore a decrease in bulk density was also expected. Nevertheless, bulk density remained stable for transects T6-T2, T6-T3 and T8-T2, T8-T3, T8-T4. This could be caused by an increase in fresh deposited sand, possibly due to aeolian activity. Subsequently, some transects in the research area are expected to be more influenced by aeolian activity. Open areas with adequate bare sand coverages, in and around dune slacks, are expected to have an inhibitory effect on succession in combination with aeolian activity. Consequently, locations, in which aeolian activity will have a more significant influence on succession, can be selected in this research area. T8 is expected to be more
influenced by aeolian activity than T6. Especially the south-eastern part of T8, because of a low vegetation cover and the presence of bare dune slopes. In T6 aeolian activity is expected to have more effect around T6-T1 and T6-T2, as they are located next to footpath with bare sand. Furthermore, the highest number of L. loeselii are not expected at a pH around 6.5 in dune slacks T6 and T8, because the vegetation cover is too high for the same areas where a pH of 6.5 is measured. Therefore, higher numbers of L. loeselii are expected at locations with a slightly higher pH, but a lower vegetation cover. Lastly, rejuvenation of soil and vegetation could possibly be stimulated by increasing the amount of open spaces in and around dune slacks.
References
Ballarini, M., Wallinga, J., Murray, A. S., Van Heteren, S., Oost, A. P., Bos, A. J. J., & Van Eijk, C. W. E., 2003. Optical dating of young coastal dunes on a decadal time scale. Quaternary Science Reviews, 22(10-13), 1011-1017.
Grootjans, A.P., Ernst, W.H.O., and Stuyfzand, P.J., 1998, European dune slacks: strong interactions of biology, pedogenesis and hydrology: Trends in Ecology & Evolution, v. 13, p. 96-100.
Grootjans, A.P., Geelen, H.W.T., Jansen, A.J.M., and Lammerts, E.J., 2002, Restoration of coastal dune slacks in the Netherlands: Hydrobiologia, v. 478, p. 181-203.
Grootjans, A.P., Hartog, P.S., Fresco, L.F.M., and Esselink, H., 1991, Succession and Fluctuation in a Wet Dune Slack in Relation to Hydrological Changes: Journal of Vegetation Science, v. 2, p. 545-554.
Grootjans, A.P., Lammerts, E.J., and Van Beusekom, F., 1995, Kalkrijke duinvalleien op de Waddeneilanden: ecologie en regeneratiemogelijkheden: Utrecht, Stichting Uitgeverij Koninklijke Natuurhistorische
Vereniging, 176 p.
KNMI. Haarsma, R., Tijm, S., Van den Brink, H. (2018, 29 January). Zeer zware storm van 18 januari. Retrieved from https://www.knmi.nl/over-het-knmi/nieuws/zeer-zware-storm-van-18-januari.
Kooijman, A. M., Bruin, C. J. W., van de Craats, A., Grootjans, A. P., Oostermeijer, J. G. B., Scholten, R., & Sharudin, R. 2016. Past and future of the EU-habitat directive species Liparis loeselii in relation to landscape and habitat dynamics in SW-Texel, the Netherlands. Science of the Total Environment, 568, 107-117.
Lammerts, E. J., Maas, C., & Grootjans, A. P. 2001, Groundwater variables and vegetation in dune slacks. Ecological Engineering, 17(1), 33-47.
Odé, B., and Bolier, A., 2003, Groenknolorchis op de kaart: Gorteria, v. 29, p. 33-37.
Oostermeijer, J. G. B., & Hartman, Y. 2014. Inferring population and metapopulation dynamics of Liparis loeselii from single-census and inventory data. Acta oecologica, 60, 30-39.
Publieke Dienstverlening Op de Kaart [PDOK]. Luchtfoto’s 2017. Retrieved 26-May-2018 from:
https://www.pdok.nl/nl/actueel/nieuws/artikel/20mrt18-pdok-achtergrondluchtfoto-2017-nu-beschikbaar
Union, E., 1992, Council Directive 92/43/EEC on the Conservation of Natural Habitats and of Wild Fauna and Flora. European Commission, Brussel, Belgium.
Appendix A: Transect field form
TRANSECT FIELDFORM
GPS coordinates:
Dune slack (T6/T8):
Transect (1/2/3/4):
Sampling point:
Date:
Notes:
Samples to take:
Two pF-ring soil samples (100cm
3)
Soil profile description Thickness of top sand layer (mm):
Water Table:
Surface characteristics
Coverage of bare sand (%):
Type(s) of vegetation and % when present
Moss: Herbs: Shrubs:
Vegetation cover (%):
Appendix B: Grid point field form
GRIDPOINT FIELDFORM
GPS coordinates:
Dune slack (T6/T8):
Sampling point:
Date:
Notes:
Topsoil layer
Thickness of top sand layer (mm):
Depth:
Type:
Surface characteristics
Coverage of bare sand (%):
Type(s) of vegetation and % when present
Moss: Herbs: Shrubs:
Appendix C: Insignificant correlations
Appendix D: Maps T6
Appendix D4, pH T6.
Appendix E: Maps T8
Appendix F: pH Maps to compare T6 and T8
Appendix G: MATLAB script
%% Ini T6 % T6 pH 2010 - 2018 pH_2010_T6 = Dataset2015(29:36,14); pH_2010_T6 = cell2mat(pH_2010_T6); pH_2014_T6 = Dataset2015(73:80,14); pH_2014_T6 = cell2mat(pH_2014_T6); pH_2015_T6 = Dataset2015(121:128,14); pH_2015_T6 = cell2mat(pH_2015_T6); % pH T6_T1, T6_T2, T6_T3 pH_2015_T6_T1 = Dataset2015(149:152,14); pH_2015_T6_T1 = cell2mat(pH_2015_T6_T1); pH_2015_T6_T2 = Dataset2015(153:160,14); pH_2015_T6_T2 = cell2mat(pH_2015_T6_T2); pH_2015_T6_T3 = Dataset2015(121:128,14); pH_2015_T6_T3 = cell2mat(pH_2015_T6_T3); T6_T1 = T6(2:11,32); T6_T1 = cell2mat(T6_T1); T6_T2 = T6(12:21,32); T6_T2 = cell2mat(T6_T2); T6_T3 = T6(22:31,32); T6_T3 = cell2mat(T6_T3); % T6 bulk 2010 - 2018 Bulk_2010_T6 = Dataset2015(29:36,19); Bulk_2010_T6 = cell2mat(Bulk_2010_T6); Bulk_2014_T6 = Dataset2015(73:80,19); Bulk_2014_T6 = cell2mat(Bulk_2014_T6); Bulk_2015_T6 = Dataset2015(121:128,19); Bulk_2015_T6 = cell2mat(Bulk_2015_T6); % Bulk T6_T1, T6_T2, T6_T3 Bulk_2015_T6_T1 = Dataset2015(149:152,19); Bulk_2015_T6_T1 = cell2mat(Bulk_2015_T6_T1); Bulk_2015_T6_T2 = Dataset2015(153:160,19); Bulk_2015_T6_T2 = cell2mat(Bulk_2015_T6_T2); Bulk_2015_T6_T3 = Dataset2015(121:128,19); Bulk_2015_T6_T3 = cell2mat(Bulk_2015_T6_T3); BT6_T1 = T6(2:11,35); BT6_T1 = cell2mat(BT6_T1); BT6_T2 = T6(12:21,35); BT6_T2 = cell2mat(BT6_T2); BT6_T3 = T6(22:31,35); BT6_T3 = cell2mat(BT6_T3); % T6 EC 2010 - 2018 EC_2010_T6 = Dataset2015(29:36,13); EC_2010_T6 = cell2mat(EC_2010_T6); EC_2014_T6 = Dataset2015(73:80,13); EC_2014_T6 = cell2mat(EC_2014_T6); EC_2015_T6 = Dataset2015(121:128,13); EC_2015_T6 = cell2mat(EC_2015_T6); % EC T6_T1, T6_T2, T6_T3 EC_2015_T6_T1 = Dataset2015(149:152,13); EC_2015_T6_T1 = cell2mat(EC_2015_T6_T1); EC_2015_T6_T2 = Dataset2015(153:160,13); EC_2015_T6_T2 = cell2mat(EC_2015_T6_T2); EC_2015_T6_T3 = Dataset2015(121:128,13); EC_2015_T6_T3 = cell2mat(EC_2015_T6_T3); ECT6_T1 = T6(2:11,33); ECT6_T1 = cell2mat(ECT6_T1); ECT6_T2 = T6(12:21,33);%% Ini T8 % pH 2010 - 2018 pH_2010_T8 = Dataset2015(41:44,14); pH_2010_T8 = cell2mat(pH_2010_T8); pH_2014_T8 = Dataset2015(85:88,14); pH_2014_T8 = cell2mat(pH_2014_T8); pH_2015_T8 = Dataset2015(133:136,14); pH_2015_T8 = cell2mat(pH_2015_T8); T8_T1 = T8(2:11,32); T8_T1 = cell2mat(T8_T1); T8_T2 = T8(12:21,32); T8_T2 = cell2mat(T8_T2); T8_T3 = T8(22:31,32); T8_T3 = cell2mat(T8_T3); T8_T4 = T8(32:41,32); T8_T4 = cell2mat(T8_T4); % T8 Bulk 2010 - 2018 Bulk_2010_T8 = Dataset2015(41:44,19); Bulk_2010_T8 = cell2mat(Bulk_2010_T8); Bulk_2014_T8 = Dataset2015(85:88,19); Bulk_2014_T8 = cell2mat(Bulk_2014_T8); Bulk_2015_T8 = Dataset2015(133:136,19); Bulk_2015_T8 = cell2mat(Bulk_2015_T8); BT8_T1 = T8(2:11,35); BT8_T1 = cell2mat(BT8_T1); BT8_T2 = T8(12:21,35); BT8_T2 = cell2mat(BT8_T2); BT8_T3 = T8(22:31,35); BT8_T3 = cell2mat(BT8_T3); BT8_T4 = T8(32:41,35); BT8_T4 = cell2mat(BT8_T4); % T8 EC 2010 - 2018 EC_2010_T8 = Dataset2015(41:44,13); EC_2010_T8 = cell2mat(EC_2010_T8); EC_2014_T8 = Dataset2015(85:88,13); EC_2014_T8 = cell2mat(EC_2014_T8); EC_2015_T8 = Dataset2015(133:136,13); EC_2015_T8 = cell2mat(EC_2015_T8); ECT8_T1 = T8(2:11,33); ECT8_T1 = cell2mat(ECT8_T1); ECT8_T2 = T8(12:21,33); ECT8_T2 = cell2mat(ECT8_T2); ECT8_T3 = T8(22:31,33); ECT8_T3 = cell2mat(ECT8_T3); ECT8_T4 = T8(32:41,33); ECT8_T4 = cell2mat(ECT8_T4); %% Bulk T6 % T6-T3 figure(1) subplot(1,2,1);
boxplot([Bulk_2010_T6 Bulk_2014_T6 Bulk_2015_T6], 'Labels', {'2010', '2014', '2015'}) title('Bulk Density T6-T3'); ylabel('g/cm3') ylim([0.2,1.5]); subplot(1,2,2); boxplot([BT6_T3], 'Labels', {'2018'}) title('Bulk Density T6-T3'); ylabel('g/cm3') ylim([0.2,1.5]); figure(2) subplot(1,2,1); boxplot([Bulk_2015_T6_T2], 'Labels', {'2015'}) title('Bulk Density T6-T2'); ylabel('g/cm3') ylim([0.75,1.55]); subplot(1,2,2); boxplot([BT6_T2], 'Labels', {'2018'}) title('Bulk Density T6-T2'); ylabel('g/cm3') ylim([0.75,1.55]); figure(3) subplot(1,2,1); boxplot([Bulk_2015_T6_T1], 'Labels', {'2015'}) title('Bulk Density T6-T1'); ylabel('g/cm3') ylim([1.1,1.55]); subplot(1,2,2); boxplot([BT6_T1], 'Labels', {'2018'}) title('Bulk Density T6-T1');
ylabel('g/cm3') ylim([1.1,1.55]); %% pH T6 % pH T6-T3 figure(2) subplot(1,2,1); boxplot([pH_2010_T6 pH_2014_T6 pH_2015_T6], 'Labels', {'2010', '2014', '2015'}) title('pH T6-T3'); ylabel('pH') ylim([5.8,7.3]); subplot(1,2,2); boxplot([T6_T3], 'Labels', {'2018'}) title('pH T6-T3'); ylabel('pH') ylim([5.8,7.3]); % pH T6-T2 figure(3) subplot(1,2,1); boxplot([pH_2015_T6_T2], 'Labels', {'2015'}) title('pH T6-T2'); ylabel('pH') ylim([6.3,7.9]); subplot(1,2,2); boxplot([T6_T2], 'Labels', {'2018'}) title('pH T6-T2'); ylabel('pH') ylim([6.3,7.9]); % pH T6-T1 figure(4) subplot(1,2,1); boxplot([pH_2015_T6_T1], 'Labels', {'2015'}) title('pH T6-T1'); ylabel('pH') ylim([6.5,7.9]); subplot(1,2,2); boxplot([T6_T1], 'Labels', {'2018'}) title('pH T6-T1'); ylabel('pH') ylim([6.5,7.9]); %% EC T6 figure(4) subplot(1,2,1);
boxplot([EC_2010_T6 EC_2014_T6 EC_2015_T6], 'Labels', {'2010', '2014', '2015'}) title('EC T6 - 10/14/15');
ylabel('EC') ylim([20,600]); subplot(1,2,2);
boxplot([ECT6_T1 ECT6_T2 ECT6_T3], 'Labels', {'T1','T2','T3'}) title('EC T6 - 2018'); ylabel('EC') ylim([20,600]); %% Bulk T8 figure(4) subplot(1,2,1);
boxplot([Bulk_2010_T8 Bulk_2014_T8 Bulk_2015_T8], 'Labels', {'2010', '2014', '2015'}) title('Bulk Density T8 - 10/14/15'); ylabel('g/cm3') ylim([0.9,1.7]); subplot(1,2,2); boxplot([BT8_T1 BT8_T2 BT8_T3 BT8_T4], 'Labels', {'T1','T2','T3','T4'}) title('Bulk Density T8 - 2018'); ylabel('g/cm3') ylim([0.9,1.7]); %% pH T8 figure(5) subplot(1,2,1); boxplot([pH_2010_T8 pH_2014_T8 pH_2015_T8], 'Labels', {'2010', '2014', '2015'}) title('pH T8 - 10/14/15'); ylabel('pH') ylim([6,7.6]); subplot(1,2,2); boxplot([T8_T1 T8_T2 T8_T3 T8_T4], 'Labels', {'T1','T2','T3','T4'}) title('pH T8 - 2018'); ylabel('pH') ylim([6,7.6]); %% EC T8 figure(6) subplot(1,2,1);
boxplot([EC_2010_T8 EC_2014_T8 EC_2015_T8], 'Labels', {'2010', '2014', '2015'}) title('EC T8 - 10/14/15');
ylabel('EC') ylim([20,260]); subplot(1,2,2);
