Effects of the January 2018 storm on blowout development in the Grey Dunes of Texel

(1)

Effects of the January 2018 storm on blowout development in the Grey Dunes of Texel

Revisiting the blowout area of ‘De Nederlanden’ after an intense storm event

BSc-thesis by Max van der Boog Student number: 11002719 First supervisor: Dr. A.M. Kooijman Second supervisor: Dr. ir. J.H. van Boxel

Future Planet Studies, University of Amsterdam Course: BSc Earth Sciences Thesis Project Date: 02-07-2018

Wordcount (excluding references, figure & table captions): 6381 Photo by T.B. Diepenmaat

(2)

Page 2 of 59

Abstract

Grey Dunes in coastal areas of the Netherlands suffer from eutrophication and acidification, caused by deposition of atmospheric nitrogen. Active blowouts in Grey Dunes are known as effective natural measures for enhancing biodiversity by increasing aeolian activity. Active blowouts increase aeolian dynamics in the surrounding areas counteracting soil acidification and eutrophication, in the form of grass-encroachment. Because aeolian activity is driven by wind, this study focusses on the effects of storm events with high wind velocities on blowout areas. The area of ‘De Nederlanden’ is being used as case-study after the large January 2018 storm, which included wind velocities of 31.1 m/s. Three comparisons were made between Active-Stable (2018), Active (2017-2018) and Stable (2017-2018) areas, to see the impact of the storm. The Active-Stable (2018) comparison contained similar results as 2017, with higher grass-encroachment in the stabilized blowout area. Changes between both (2017-2018) comparisons were expected to be equally visible, however these changes did not appear to be present. Non-significant results for important parameters such as pH, bare sand coverage and Ah-layer thickness were found. Together with findings of the historical storm analysis of ‘De Nederlanden’, this study indicates that wind velocities of extreme storm events lead to unfavorable conditions for aeolian dynamics and blowout development in lime-poor Grey Dunes of Texel. However, to extend this proclamation to other areas in the Netherlands, more research on comparable areas needs to be conducted. Nederlandse samenvatting

Grijze duinen zijn van oorsprong gebieden met een hoge soortenrijkdom. Door menselijke activiteiten wordt deze soortenrijkdom bedreigd. Vergrassing door hoge stikstof depositie is een van de belangrijkste oorzaken van deze bedreiging. Een manier om vergrassing tegen te gaan is het activeren van stuifkuilen in de duinen. Stuifkuilen zijn stukken duin met weinig vegetatie, waar wind veel invloed op uitoefent. Deze stuifkuilen verhogen door de toevoer van vers zand de pH van de bodem, wat leidt tot een verhoging in soorten rijkdom. Dit onderzoek is gefocust op het analyseren van de zware storm in januari 2018, om te zien of stormen een positieve invloed hebben op stuifkuilen, en daarbij een verhoging van soortenrijkdom. Het gebied van ‘De Nederlanden’ op Texel wordt gebruikt als onderzoeksgebied. Hierin zullen drie vergelijken worden getrokken, namelijk de stuikuilen in Actief-Stabiel (2018), Actief (2017-2018) en Stabiel (2017-2018). De vergelijking Actief-Stabiel (2018) resulteerde in verwachte uitkomsten, namelijk een hogere mate van vergrassing in het gebied van de stabiele stuifkuil. Duidelijke uitkomsten waren ook verwacht voor de vergelijkingen van Actief (2017-2018) en Stabiel (2017-2018). Echter, niet significante resultaten bij

belangrijke bodemparameters pH, kaal zand oppervlakte en diktes van Ah-lagen zijn tegenstrijdig met vooraf opgestelde hypotheses. Samen met een historische analyse van alle zware stormen in ‘De Nederlanden’ kan worden geconcludeerd dat te hoge windsnelheden, die gepaard gaan met stormen, leiden tot ongunstige condities voor zand transportatie, accumulatie en stuifkuilformatie. Ondanks deze constateringen is vervolgonderzoek in gebieden met vergelijkbare bodem en vegetatie eigenschappen nodig om een oordeel te kunnen geven over andere grijze duingebieden in Nederland.

(3)

Page 3 of 59 Table of Contents Abstract………. 2 Nederlandse samenvatting……….. 2 Abbreviations……… 4 1. Introduction………. 5 2. Study Area………. 6 3. Theoretical Framework………. 6 4. Methodology……… 8

4.1 Experimental set-up in Texel……….. 8

4.2 Laboratory work………. 9

4.3 Statistical and Spatial Analysis……… 10

5. Results……… 10

5.1 Vegetation parameters……… 10

5.2 Soil parameters………. 12

5.3 Aerial photo analysis………. 14

6. Discussion………... 15

6.1 Storm impact on vegetation………. 15

6.2 Storm impact on soil processes……….. 16

6.3 Historical storms and the impact on ‘De Nederlanden’……….. 17

6.4 Implications of study………. 18

7. Conclusions ………... 19

References………... 19

Appendix……… 21

A. Extended methodology – Laboratory calculations ……… 21

B. Statistical analyses tables ……….…. 22

(4)

Page 4 of 59

Abbreviations

BD Bulk Density

C Carbon

CaCO3 Calcium Carbonate

EC Electrical Conductivity

GIS Geographic Information System H2130B Grey Dunes (lime-poor)

km/h Kilometers per hour m/s Meters per second

N Nitrogen NOx Nitrogen oxides NH3 Ammonia OM Organic Matter P Phosphorous pH Scale of acidity rpm Rounds per minute

(5)

Page 5 of 59

1. Introduction

The Netherlands always have, and always will be a nation defending itself against excess water. Twenty-six percent of the total surface of the Netherlands is below sea level, and an extra of twenty-nine percent is vulnerable for flooding by rivers or by sea (IPCC, 2007). Therefore, protection mechanisms in the form of coastal defenses are necessary. An old and profound protection system are the coastal dunes of the Netherlands. The Dutch coastal dunes fulfill several important functions in society, for instance: coastal defense and erosion, nature management, dune stabilization and afforestation, agriculture, tourism, urbanization, recreation and education (Van der Meulen & Salman, 1996). The Wadden Islands in particular, are relying heavily on the functions and services provided by the coastal dunes. Besides the coastal defense function, the nature management and conservation of the ecosystem services of the dunes is essential for the Wadden Islands. As 70 percent of all the national vegetation species and a number of endemic species are found in sand dunes (Veer & Kooijman, 1997). This is also why certain dune areas of the Wadden Islands, among which Grey Dunes are protected by the European Habitat Directive.

Although the Grey Dunes are protected in the Netherlands, they are highly threatened by high levels of atmospheric deposition of Nitrogen (N) and acidifying agents (van der Meulen et al. 1996; Kooijman et al. 2005). This results in soil acidification and eutrophication. Eutrophication of the sand dunes results in encroachment, meaning the dune surface is covered with shrubs, grasses and mosses. This grass-encroachment leads to a dominance of a few tall grass species, which overgrow the smaller pioneer species and results in a limited biodiversity in the Grey Dunes. The overload of N is mainly due to high agricultural activities, which require large amounts of N-rich fertilizers for food production. However, heavy industries and fuel emissions have also contributed to the N overload in the atmosphere (Hicks et al., 2011).

Because of the increasing pressures in the form of soil acidification and eutrophication, the Grey Dunes habitat conditions have changed significantly over the past decades. To restore the Grey Dunes to their natural habitat conditions, several measures were determined (Van der Meulen et al., 1996; Aggenbach et al., 2016). A potential dune restoration measure was the reactivation of dune blowouts. Blowouts are defined as cup or trough shaped erosional depressions in the dune landscape that have a very low vegetation cover (Hesp, 2002). Although they are most common in the foredunes, the blowouts can also occur in stable environments such as the Grey Dunes. The reactivation of these blowouts includes the increase of aeolian activity, which helps to restore the lime buffer of the soil and counteracts soil acidification (van der Meulen et al., 1996). In addition, the increase of aeolian activity would transport newly surfaced and lime-rich sand into the blowout. This fresh sand would serve as a substrate to facilitate pioneer vegetation, which would restore the biodiversity of the coastal dunes.

As aeolian activity plays a major role in this blowout reactivation measure, it is valuable to assess whether strong wind would have positive effects on active, as well as on non-active blowouts. This research focusses on an important blowout area on Texel, in which aeolian activity fluctuates over the years (Aggenbach et al. 2016). The area was studied last year after a few relatively stable years. However, after the storm in January 2018, the situation may be different, and the area of fresh sand much higher than in 2017. Blowouts contribute to improvement of habitat conditions, because fresh sand is deposited in the surrounding areas for at least some time. This means that the soil is rejuvenated, has higher pH, lower amounts of soil organic matter, and suitable habitat conditions for pioneer plant species. The main question this study aims to answer is: What is the impact of the storm in January 2018 on the active and stabilized blowout areas in ‘De Nederlanden’, Texel?

To answer this question, three comparisons are composed between an active blowout area and a stabilized blowout area in 2018, the active blowout area in 2018 and the stabilized blowout area in 2017-2018. After these comparisons the study will be able to distinguish whether the January 2018 storm had more impact in the active blowout area or in the stabilized blowout area. This will help in assessing whether increased windspeeds, in the form of storms, are a positive driver of habitat creation and restoration in the Grey Dunes. The hypothesis of this research is that the increased windspeeds of the January 2018 storm have created higher aeolian activity in the blowout, which is positively effecting habitat conditions of the Grey

(6)

Page 6 of 59

Dunes. This hypothesis will be answered by use of smaller, more easily studied hypotheses, which will be discussed in the results and discussion sections.

2. Study area

The research of this study will take place in the area north of De Koog, which is situated on the Wadden Island Texel (Fig 1). In geological terms, Texel consists predominantly out of Pleistocene clay deposits (Aggenbach et al., 2016). Between the municipalities of Den Hoorn and De Koog, a narrow complex of dune arches was formed. Close to the beaches in the area of De Koog, broad parabolic dunes were created over time. The shape and area around these broad dunes suggests that the aeolian activity in these particular dunes has been high for multiple centuries (Aggenbach et al., 2016). The exact study area in this research is called ‘De Nederlanden’, also visualized in Fig. 1. De Nederlanden is situated on the dune arches with high aeolian activity, north of De Koog. Soil descriptions show this rate of high aeolian activity together with low amounts of calcium carbonates (CaCO3) in the soil profile (Aggenbach et al., 2016). The largest part of De Nederlanden belongs to the habitat

type H2130B Grey Dunes (Van der Meulen, 1996; Aggenbach et al., 2016). Because the Grey Dunes are lime-poor areas, even more aeolian activity is necessary to attract traditional pioneer species. Some of the pioneer species appear in secondary blowouts in which newly surfaced sand increases the base content. These small-scale base rich blowouts facilitate at least 10 extra plant species (Aggenbach, et al., 2016). To see whether habitat conditions improve in active blowout areas, where the aeolian activity is relatively high, this study performs a comparison with an active and a stable blowout area (Fig. 2). In addition, aeolian activity could have increased after the January 2018 storm. To assess the effects of this storm on aeolian activity, and eventually on the improvement of habitat conditions, both areas (Blowout A&B) will be studied again and compared to previous findings on the exact same locations.

3. Theoretical Framework

As mentioned earlier, the high atmospheric nitrogen (N) emissions can negatively influence the functioning of ecosystem services in the coastal dunes. This negative influence is caused by acidification, eutrophication and direct toxicity, which impact dune soil and vegetation (Witz, 2015). Nitrogen occurs mainly in the form of nitrogen oxides (NOx) and ammonia (NH3), originating from industrial activity and agricultural practices

(7)

Page 7 of 59

respectively (Hicks et al., 2011). Although nitrogen deposition levels have decreased, the effects of the nutrient overload are still evident in vulnerable ecosystems (Witz, 2015). The input of nitrogen compounds alters the natural ecological balance of a habitat, resulting in an increase of the species that prefer high rates of nitrogen supply, like grasses, shrubs and mosses. Whereas the more sensitive species, which are often the characteristic pioneer species experience unfavorable effects of the high nitrogen input (Hicks et al., 2011).

The soil in the Grey Dunes of Texel has a low lime content and a low pH in the upper soil layer, which results in an absence of calcium carbonate (CaCO3) buffer capacity of the soil (Kooijman et al., 2005). Without

this buffer capacity the soil acidifies, leading to loss of characteristic plant species. In the reactivation process of blowouts, the underlying soil layer beneath the acidified upper layer, with a higher pH is exposed (ibid.). This newly surfaced sand layer will be transported by wind and deposited on surrounding acidified soils, which partly restores the buffer capacity of this acidified soil by increasing the pH (Kooijman et al., 2012). When the buffer capacity of the soil increases, new pioneer vegetation can be facilitated (van der Meulen et al., 1996). The above described method of increasing the buffer capacity of soils could be difficult for the Grey Dunes in Texel. This is because Grey Dunes are fixed coastal dunes, with low aeolian activity due to stabilization by surrounding vegetation (Witz, 2015; Provoost et al., 2002). The vegetation cover of Grey Dunes usually consists of mosses and lichens, giving Grey Dunes their recognizable grey color (Witz, 2015). Additionally, Grey Dunes are open dune grasslands and form an important habitat in the coastal dune landscape due to their high biodiversity in both plant and animal species (Provoost et al., 2002). To conserve the current biodiversity and to restore some of the lost pioneer species, the Grey Dunes have obtained a priority position in the habitat directive of the European Union (Kooijman et al., 2012; Aggenbach et al., 2016).

One of the measures to restore parts of the habitat conditions is the (re)activation of dune blowouts (Van der Meulen et al., 1996; Aggenbach et al., 2016). By activating blowouts, the aeolian activity is stimulated and the blowout area becomes more dynamic. The aeolian activity is driven by wind, making wind a crucial agent in the reactivation process (Witz, 2015). Blowouts are geomorphological landforms in the dunes, shaped by wind erosion. The reactivation process of a blowout starts with removing the topsoil layer including vegetation, until a depth of 30 to 50 centimeters (Hesp, 2002). The topsoil layer is the less easily eroded layer due to a high soil organic matter content (Kooijman et al., 2005). Besides the artificial method of removing the topsoil layer, the activation process can also occur naturally (Hesp, 2002). This natural occurrence is usually by means of topographic acceleration of the airflow over a dune crest or by intense storm events (Hesp, 2002). The January 2018 storm sets an example of an intense storm on which the effects of transporting dune sand into the blowout will be investigated. For transporting dune sand, a minimum windspeed of 6 m/s is required (van Boxel et al., 1997). In addition, blowout changes show a strong correlation with wind velocities between 6.25 and 12.5 m/s (Jungerius et al., 1991). These wind velocities are critical values for moving sand particles in the range of 0.15 to 0.42 mm. The optimal velocities mostly come along with characteristic southwest winds, which consequently shape the blowouts in similar southwest directions (Fig. 3). Records from the KNMI (2018) show that the storm of January 2018 contained wind velocities up to 112 km/h, equal to 31.1 m/s. These records could in combination with findings of Jungerius et al., (1991) be interpreted as unfavorable wind velocities for blowout formation. This is caused by the complication that small sand particles can hardly attach to extremely high wind velocities, which makes the process of transportation from the erosional walls to the depositional lobe more difficult (Fig. 3). Active blowouts are eroded by wind on the erosional walls and the sand grains are accumulated at the depositional lobe (Witz, 2015; Hesp, 2002). However, a blowout area has a very low pressure which can redirect wind flow in opposite directions, up to 100-degree difference from the original flow path (Witz, 2015). This results in sediment transportation to different parts of the blowout (Witz, 2015; Hesp, 2002). The erosion process causes the blowouts to move further into the landscape in upwind direction (Witz, 2015). Blowouts can reinforce themselves quickly, however the long-term trend shows blowouts will eventually stabilize naturally (Kooijman et al., 2005).

(8)

Page 8 of 59

Fig. 3. Schematic overview of blowouts formation with typical wind flow patterns (Hesp, 2002). 4. Methodology

The methodology section of the research report is divided into multiple divisions, as each division requires certain techniques and methods. This methodology section is largely based on methods described and used by Witz (2015). Because Witz (2015) conducted similar research in another location in the Netherlands, roughly the same methods are also applicable for this research. Along with Witz (2015), other literature studies on coastal dune and blowout areas are intensively used in this research. Results from previous bachelor projects on the same location in Texel also form an informative source, especially for the blowout comparisons after the January 2018 storm.

4.1. Experimental set-up in Texel

During this research an important part consists of fieldwork in the area north of De Koog, Texel. The fieldwork lasts one week, starting from May 4th_{till May 11}th_{. In this week data of the active and the stabilized blowouts}

will be gathered. This data consists out of transects and grid points, and every sample point will be further analyzed. The amount and types of samples for each sample measurement are visualized in table 1. All these sample measurements and points need to be analyzed by similar field forms. Every field form will be digitalized in Microsoft Office Excel after each fieldwork day.

Transects Grid Points

General information - Vegetation coverage (Type & percentage) - Soil profile description

- Two pF-ring soil samples

- Vegetation within 25x25 cm or 50x50 cm plots - GPS location

- Vegetation coverage (Type & percentage)

- Depth and type of topsoil layer - GPS location

Blowout A (Active) 20 samples 150 measurements

Blowout B (Stabilized) 20 samples 150 measurements

Total 40 samples 300 measurements

(9)

Page 9 of 59

In both the active (Blowout A) and the stabilized (Blowout B) areas, two transects are established through the blowout, deposition zones and unaffected dunes further away (Fig. 4). These transects will result in approximately 20 sampling points per blowout, and thus 40 sampling points in total (Table 1). In the sampling points, the soil profile is described in detail and surface characteristics e.g. bare sand cover, moss layer, herb layer and shrub layer are recorded. Moreover, the soil sample for each sampling point is collected with metal pF-rings. The pF-ring samples give more insight in the bulk density, pH, Electrical Conductivity (EC) and Carbon (C) and Nitrogen (N) content of the soil in the transect sampling points. The pF-ring samples will be taken at approximately 0-5 cm below the topsoil layer, with a pF-ring volume of 100 cm3_{per sampling point. The above}

ground vegetation is sampled in plots of 25x25 cm or in 50x50 cm, depending on vegetation coverage. These samples are also considered for the analysis of dry weight samples per m2_{and the C and N content.}

Fig. 4. Transect analysis at active area (Blowout A) Fig. 5. Grid cell analysis at active area (Blowout A)

In addition to the transect sampling points, detailed measurements of surface and soil characteristics are conducted in a grid cell analysis in and around both blowouts (Blowout A & B). Each blowout area will have approximately 150 measurements, resulting in 300 measurements in total (Table 1). The location of each grid point is recorded with a GPS, and similar characteristic surface description as with the transects, together with topsoil soil features are described. The fieldwork data will be digitalized in ArcGIS, to obtain similar patterns like in figures 4 and 5.

4.2. Laboratory work

After the fieldwork in Texel the soil and vegetation samples of the 40 transect points are dried, duplicated and further analyzed in the laboratory. This duplication decreases the probability of errors during the laboratory process and thereby increases the reliability of the lab results. For the soil samples the BD, pH, EC and C and N contents are measured. The vegetation samples will only be measured on dry weight and C and N content. A detailed description of all measurement and lab practices is visualized in table 2 below and in Appendix A.

Types of measurements Outline of laboratory measurements

Bulk Density of soil

20 grams of the original soil sample will be put in the oven for 48 hours at 105 °C. The 20 g subsample is weighted before and after the drying period to calculate the difference in weight between wet and dry sample. This outcome will used to determine the bulk density of the original weight and volume of 100 cm3_{. Calculation of BD is added in Appendix A.}

pH & EC of soil

10 grams of the original soil sample will be mixed with 25 ml of demineralized water, to obtain the optimal 1:2.5 ratio. The resulting substance will be shaken for 2 hours, followed by a resting period overnight and subsequently shaken again 20 minutes on the next day. The values will be measured using pH and EC electrodes.

(10)

Page 10 of 59

C & N content of soil

Two exactly similar (duplicates) subsamples of 20-50 mg are prepared per soil sample and put in the oven for 48 hours at 70 °C. Subsequently, these

subsamples are put in a CHNS analyzer, and both C and N contents will be measured. Calculations of C and N contents of soil are presented in Appendix A. Dry weight of

vegetation

The vegetation samples will be dried, by use of the oven, for 24 hours at 40 °C. The samples are weighted before and after, and the dry material is further ground at a rotational speed 8000 rpm.

C & N content of vegetation

Again, two exactly similar (duplicates) subsamples of 10-15 mg are prepared per vegetation sample and put in the oven for 48 hours at 70 °C. Subsequently, these subsamples are put in a CHNS analyzer, and both C and N contents will be measured. Except moss species and woody material will not be collected in the subsamples, this exclusion might alter the C/N ratio. Calculations of C and N contents of vegetation are visible in Appendix A.

Table 2. Outline of the laboratory measurements including descriptions.

4.3. Statistical and Spatial Analysis

After the fieldwork and laboratory work, all the data will be digitalized in Microsoft Office 2016 Excel. The data will be further analyzed with multiple computational programs like Microsoft Office 2016 Excel, MATLAB and ArcGIS 10.4.1. In Excel the data will be ordered and ‘cleaned’, before running calculations in MATLAB. MATLAB can be used for simple calculations such as means and standard deviations, and for more complex problems like comparisons between two entire datasets. By using MATLAB, a statistical analysis can be made for the comparison of Blowout A & B. To test significant differences between both areas, the independent t-test is used. The independent t-tests are used to make a total comparison between two variables, for instance the two blowout areas. The ANOVA tests are more commonly used for complex comparisons between one response variable and multiple predictor variables, such as different vegetation groups.

To obtain figures and images like in figures 4 and 5, the area also needs to be spatially analyzed. This spatial analysis makes statistical outcomes more explicit and enables to make spatial conclusions of e.g. vegetation cover in specific parts of the blowout areas. ArcGIS 10.4.1 is applied to make a spatial map of the grid points, using the GPS-coordinates which have been noted during the fieldwork. For both blowout areas, a map will be composed to visualize the size of the deflation and deposition zone and depth of accumulated sediment in the deposition zone. This map can be generated by interpolating the grid points to obtain a surface map. By using the ‘Spline’ interpolation function from the ‘Spatial Analysist Tool’ in MATLAB, the surface map can be established.

5. Results

5.1. Vegetation parameters

Differences in the vegetation parameters were significant between the active and stabilized blowout in 2018 (Table 3). Dry weight, C content of vegetation, N content of vegetation, moss and herbs cover all differed significantly between the two zones (Appendix B). None of the vegetation parameters of the active blowout areas in both 2017 and 2018 have proven to be significant, except for shrubs cover (Table 6). For the stabilized blowout area, a significant difference occurred in C/N ratios, moss and shrubs cover, while both C and N content of vegetation, dry weight and herbs cover were not significant (Table 3 & 6).

(11)

Page 11 of 59 Vegetation parameters p-value Active- Stable (2018) Significant Active- Stable (2018) p-value Active (2017-2018) Significant Active (2017-2018) p-value Stable (2017-2018) Significant Stable (2017-2018)

Dry weight 0.000567 Yes 0.6232 No 0.608 No

C content (g/m2) 0.000677 Yes 0.3109 No 0.3321 No

N content (g/m2) 0.0031 Yes 0.845 No 0.9439 No

C/N ratio 0.2123 No 0.4168 No 0.0012 Yes

Table 3. Independent t-tests comparisons for vegetation.

How the vegetation parameters differ significantly is explained with the ANOVA’s in Appendix B and the Tables 4 & 5. To create more reliable results for the C & N variables, the active and stabilized blowout area have been divided into three sub-groups, the unaffected zone, the blowout zone and the accumulation zone. All these zones have been tested with ANOVA’s in Appendix B, showing large differences between the zones. For the Active – Stable comparison, both the C-content and N-content of all zones is significantly higher for the stabilized blowout, even as the dry weight of vegetation. As being mentioned before, the comparison of Active (2017-2018) has not resulted in overall significant differences (Table 3). However, ANOVA results show relatively higher C-contents and N-contents in 2017, except for the blowout zone, in which both contents are higher in 2018. In the stabilized blowout, the C/N ratios were the only significant parameter, showing higher values for all zones in 2017.

Vegetation parameters Mean A 2017 Mean A 2018 Std A 2017 Std A 2018

Dry weight (g) 87.2240 79.6940 56.7096 37.4952 C content (unaffected) (g/m2) 48.7840 38.0598 18.7059 13.9326 C content (blowout) (g/m2) 23.1200 29.4776 19.8802 27.7126 C content (accumulation) (g/m2) 48.7127 37.5298 32.5438 14.1974 N content (unaffected) (g/m2) 1.2216 1.1270 0.3442 0.6375 N content (blowout) (g/m2) 0.9816 1.2754 0.8009 1.2989 N content (accumulation) (g/m2) 1.3630 1.2204 0.8484 0.4020 C/N-ratio (unaffected) 39.8187 7.8574 37.6957 10.6878 C/N-ratio (blowout) 17.2355 11.6079 19.2848 14.0833 C/N-ratio (accumulation) 36.0765 7.5830 30.9015 7.7049

Table 4. Means and standard deviations for the vegetation parameters of the Active blowout zone.

Vegetation parameters Mean S 2017 Mean S 2018 Std S 2017 Std S 2018

Dry weight (g) 151.5440 140.5060 72.7472 61.7841 C content (unaffected) (g/m2) 77.8720 64.1395 32.2714 20.9385 C content (blowout) (g/m2) 51.8400 51.3657 27.0649 20.1575 C content (accumulation) (g/m2) 83.5200 70.8252 39.7506 37.7563 N content (unaffected) (g/m2) 1.8998 2.0382 0.8037 0.4722 N content (blowout) (g/m2) 1.4259 1.6125 0.7753 0.5947 N content (accumulation) (g/m2) 2.0788 1.9823 1.0147 0.9293 C/N-ratio (unaffected) 40.8008 31.6809 5.2324 7.4965 C/N-ratio (blowout) 37.9591 31.9611 7.5413 2.8288 C/N-ratio (accumulation) 40.5505 35.3595 7.0017 3.8095

(12)

Page 12 of 59

The vegetation types also have been recorded and are visualized in Table 6 and in the ANOVA’s and interpolation maps in Appendix B&C. The Active – Stable comparison contains significantly higher values of Moss and Herbs cover for the stabilized blowout (Table 6) (Appendix B). In the Active comparison (2017-2018) Shrubs cover is the only significant vegetation type, showing higher amounts of shrubs in 2018. In addition, Shrubs and Moss cover are both significantly higher in 2018 compared to 2017 in the stabilized blowout (Table 6) (Appendix B).

Vegetation type p-value Active- Stable (2018) Significant Active- Stable (2018) p-value Active (2017-2018) Significant Active (2017-2018) p-value Stable (2017-2018) Significant Stable (2017-2018)

Moss cover (%) 0.0103 Yes 0.3684 No 4.3936e-06 Yes

Herbs cover (%) 0.0349 Yes 1.00 No 0.9665 No

Shrubs cover (%) 0.0664 No 0.0124 Yes 8.0346e-04 Yes

Table 6. Independent t-tests comparisons for vegetation types.

5.2. Soil parameters

Significant differences occurred in soil parameters between the active blowout zone and the stabilized blowout zone in both years (2017 & 2018). The results from the stabilized blowout showed more significant differences compared to the active blowout (Table 7). Furthermore, the difference between both the active and stabilized blowout resulted in many significant differences in the soil parameters.

Soil parameters p-value Active- Stable (2018) Significant Active- Stable (2018) p-value Active (2017-2018) Significant Active (2017-2018) p-value Stable (2017-2018) Significant Stable (2017-2018)

pH 5.0e-06 Yes 0.5197 No 0.0041 Yes

EC (mS/cm) 0.1646 No 0.2823 No 0.0027 Yes

BD (g/cm3) 6.0e-05 Yes 0.0068 Yes 0.0141 Yes

Bare sand coverage (%) 0.0051 Yes 0.8208 No 0.7454 No Ah-layer thickness (cm) 0.1601 No 0.0609 No 0.2575 No

C content (g/m2) 5.0782e-05 Yes 0.7496 No 0.0207 Yes

N content (g/m2) 4.8821e-05 Yes 0.4008 No 0.2792 No

C/N-ratio 0.1408 No 0.2862 No 0.00004 Yes

Table 7. Independent t-tests comparisons for soil.

For the active blowout zone, BD was the only parameter which differed significantly from the data of 2017. Table 8 shows the increase of BD in 2018. In addition, the pH, Bare sand coverage, C/N ratio (unaffected) and C/N ratio (accumulation) have increased in 2018, only these parameters have not been significant (Appendix B). In contrast, the stabilized blowout showed more significant values. pH, BD, C content, C/N ratio (overall), C/N ratio (unaffected), C/N ratio (blowout) and C/N ratio (accumulation) were all significantly higher (Table 6). Whereas, EC was significantly lower in 2018 compared to 2017. In the active blowout both C and N contents have decreased, only these contents have not been significant. Meanwhile both C and N contents have increased in the stabilized blowout, where C content has been tested significant, N content has not. The active blowout appeared to be having increases of C/N ratios in the unaffected and accumulation zone, while the blowout zone decreased sharply compared to 2017. In the stabilized blowout, the C/N ratios of all three zones increased (Table 6 & Appendix B).

(13)

Page 13 of 59

Soil parameters Mean A 17 Mean A 18 Std A 17 Std A 18

pH 6.4945 6.6825 1.0195 0.7968

EC (mS/cm) 76.89 63.715 38.0064 38.3993

BD (g/cm3) 1.1964 1.3610 0.1978 0.1642

Bare sand coverage (%) 23.5 25.8 32.48 31.29

Ah-layer thickness (cm) 4.4626 4.2414 2.976 2.491 C content (g/m2) 808.79 787.67 515.37 558.30 N content (g/m2) 69.20 56.88 47.71 43.84 C/N-ratio (overall) 21.2832 15.8796 21.4914 6.0948 C/N-ratio (unaffected) 10.7954 13.1410 1.8225 0.7137 C/N-ratio (blowout) 63.0864 25.3589 1.7406 7.8859 C/N-ratio (accumulation) 10.8493 13.6774 1.2591 2.5869

Table 8. Means and standard deviations for the soil parameters of the Active blowout zone.

Soil parameters Mean S 17 Mean S 18 Std S 17 Std S 18

pH 5.1330 5.6240 0.5959 0.4015

EC (mS/cm) 78.64 47.875 28.5394 31.9963

BD (g/cm3) 0.9862 1.1325 0.1998 0.1570

Ah-layer thickness (cm) 6.2671 6.3077 3.526 3.317 C content (g/m2) 1400.80 1888.15 586.71 921.63 N content (g/m2) 116.96 135.00 38.74 62.40 C/N-ratio (overall) 11.4816 13.7951 1.9286 1.1283 C/N-ratio (unaffected) 11.1413 13.9385 1.1287 0.6961 C/N-ratio (blowout) 9.8893 12.9048 3.2711 0.6955 C/N-ratio (accumulation) 12.2153 14.0536 1.2591 2.5869

Table 9. Means and standard deviations for the soil parameters of the Stabilized blowout zone.

Besides the statistical analyses on the transect points, spatial analyses have also been conducted with data of the grid points. Ah-layers have been compared between active blowout and stabilized blowout zone (Fig. 7), as well as with spatial maps from 2017 (Fig. 6). Differences between the active and stabilized blowout remained, as the thickness of the Ah-layer seems to be larger in the stabilized blowout (Fig. 7 & 8). However, statistical analysis argue that the differences are not significant. Similar not significant outcomes appear in the Active (2017-2018) and Stable (2017-2018) comparisons (Table 7). This is supported by tables 8 & 9 and the interpolation maps in figures 6 & 7, showing only very small distinctions between the years.

(14)

Page 14 of 59

Fig. 7. Ah-layers for active blowout area (left) and stabilized blowout area (right) in 2018.

5.3. Aerial photo analysis

To gain insight in potential previous storms events in the area of ‘De Nederlanden’, an historical analysis is performed during this study. Aerial photographs of the period 1996-2016 have been studied in ArcGIS and the results are presented in Fig. 7. This figure displays the total bare sand coverage in each year, together with the largest patch of sand, which defines the blowout and close surrounding blowout area. In the period 2006-2009 more aeolian activity appeared in ‘De Nederlanden’, leading to increasing values of bare sand. After a decline in 2010, the highest peak of bare sand coverage was monitored in 2011. Although the bare sand coverage sharply increased in 2011, the surface of the blowout area remained similar to the values of 2010. After 2011 the bare sand coverage fluctuates lightly and reaches an equilibrium around ~40000 m2_{. The blowout area}

reaches high values (~20000 m2_{) in the periods of 2007-2008 and 2010-2011. Since 2011 the blowout surface}

also fluctuates, however the most recent trend from 2014-2016 is negative and shows declining values.

Fig.7. Historical analysis of bare sand coverage and blowout surface (1996-2016) in ‘De Nederlanden’.

To tal s u rfac e o f B are S an d ( m 2) 0 20000 40000 60000 80000 100000 120000 140000 160000 1996 2000 2003 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Historical Analysis of Bare Sand coverage and Blowout

width in (m

2

₎

(15)

Page 15 of 59

In addition to the data from 1996-2016, the data of bare sand coverage 2017 and 2018 is displayed in figures 8 & 9. The active blowout areas still contain high percentages of bare sand near the blowout. Differences are occurring in the accumulation zone, on the right side of the maps; more small patches of 20-30 percent are visible in the 2018 map. The stabilized blowout shows less differences, patches of bare sand are mainly visible in the left corner of the maps. However, despite the fact that the above-mentioned differences are visible in the maps, the statistical values are not significant for bare sand coverage.

Fig. 8. Bare sand coverage for active blowout area (left) and stabilized blowout area (right) in 2017.

Fig. 9. Bare sand coverage for active blowout area (left) and stabilized blowout area (right) in 2018.

6. Discussion

Deposition of atmospheric nitrogen in Grey Dunes causes soil acidification, grass encroachment, and contributes due to enhanced succession to a decline of biodiversity. Because lime-poor Grey Dunes are ecosystems vulnerable to high rates of N deposition, they have been included and protected by the EU Habitat Directive. By providing more knowledge on possible restoration techniques, the Grey Dunes can be better protected in the future. This study confines to the impact of storms on aeolian activity, rejuvenating the soil and counteracting the ongoing succession process caused by atmospheric N deposition. The storm of January 2018 is chosen as case study, in which a natural active blowout zone and a stabilized blowout zone in the lime-poor Grey Dunes of ‘De Nederlanden’, Texel are being reviewed. By comparing this year’s results with last year’s results of the exact same zones, more insights on storm activity on lime-poor dunes are generated and can be used in protecting the vulnerable dunes.

6.1 Storm impact on vegetation

Grass encroachment is one of the negative effects of N deposition (Aggenbach et al., 2016). When an area coops with grass encroachment, this is recognizable by characteristic features of vegetation in N rich soils. To start, this study shows differences in vegetation between the active blowout zone and the stabilized blowout zone in both 2017 and 2018. The results indicate that in the stabilized blowout zone higher grass encroachment was encountered compared with the active blowout zone. Veer & Kooijman (1997) state that

(16)

Page 16 of 59

grass-dominated plots have due competition of light and nutrients a low species diversity for especially mosses and lichens. Veer (1997) adds that tall grass species dominate lichens and low-growing herbs. This is supported by findings of this study in the comparison between active blowout zone and stabilized blowout zone. The interpolation maps of vegetation patterns show that herbs and lichens are more present at the active blowout, and shrubs and grasses are appearing more in the more grass-dominated stabilized zone. The occurrence of more low-growing vegetation in the active blowout is due to higher light and nutrient availability (Veer & Kooijman, 1997). Moreover, in the active area only shrubs cover showed significant differences between both years. The decrease of shrubs in the active area, mainly in the form of dune characteristic Empetrum nigrum and Hippophae rhamnoides can be caused by high aeolian activity (Aggenbach et al., 2016). The stabilized area comparison presented significantly higher values of mosses and shrubs cover. In contrast to the active blowout area, Empetrum nigrum and Hippophae rhamnoides have increased in the stabilized blowout area. The increase in mosses and lichens, and especially lichens-species Cladonia, is somewhat contradicting to the statement of Veer & Kooijman (1997), which argue diversity loss of mosses and lichens in the area with more grass-encroachment. Still, this argument does not point at positive effects of the storm on light availability and nutrient competition, because overall vegetation in the stabilized area has increased and bare sand coverage of the stabilized blowout zone has decreased.

Besides vegetation coverage, biomass is also a proxy to compare vegetation (Veer & Kooijman, 1997). This is supported by Jones et al. (2004), which argue higher atmospheric N inputs, result in higher plant biomass and lower species richness. In this study biomass has been calculated as dry weight of the vegetation. The comparison of active blowout and stabilized blowout backs up the statement of Jones et al. (2004); the dry weight values of the stabilized blowout zone are significantly higher compared to the active zone. However, for both comparisons with 2017 non-significant differences were present in dry weight.

Although biomass differences have not occurred in the Active 2018) and in the Stable (2017-2018) comparisons, N-availability appears to be determined by litter input of the biomass (Kooijman & Besse, 2002). Moreover, N-availability is higher in acidified soils with low rates of decomposition and high soil C/N ratios, increasing the amount of N for the vegetation. Dopheide & Verstraten (1995) add that N-deposition, and thus availability is high in the Wadden area. With these proclamations in mind, the stabilized area should have higher soil C/N ratios and higher N values in the vegetation. In addition, if the storm of January 2018 would have made positive effects on vegetation, the pH should rise and the soil C/N ratios and N values of vegetation decrease. Because soil C/N ratios have been subdivided into three groups, no overall affirmation can be given for the comparison of active blowout zone and stabilized blowout zone. Still,

the unaffected zone as well as the accumulation zone give higher C/N ratios for the stabilized blowout, as expected. Additionally, the N values for the stabilized blowout zone are higher compared to the active blowout zone, showing that more N-preferring vegetation occurs in the stabilized blowout. In the active blowout zone comparison, the pH rose in 2018 compared to 2017, the N values of vegetation did decrease in 2018, except for the accumulation zone and the C/N ratios of soil in which N values of vegetation increased. Unfortunately, all three parameters were not significant, making it impossible to relate the differences between both years to the January 2018 storm. The pH in the stabilized blowout zone did significantly increase, however differences in N values of vegetation for all zones were not significant. Furthermore, the soil C/N ratios were significantly higher for all zones, which indicates the storm seems not to have had the positive effect on vegetation in the stabilized zone.

6.2 Storm impact on soil processes

To continue with the effect of the storm on the soil parameters, the pH has already been discussed briefly in the results section 6.1. pH tends to increase when new fresh bare sand is being deposited (Kooijman et al., 2005). Subsequently, this higher pH counteracts soil acidification, attracts pioneer species and enhances the biodiversity of the area (Kooijman et al., 2005). Positive correlations are found between pH and bare sand cover (Kolb, 2017). Together with the results of this study, it can be concluded that the fresh blown sand of the blowout enhances the pH of the soil in the blowout zone, and thus counteracts soil acidification. The BD

(17)

Page 17 of 59

in this study is significantly higher in the active blowout zone compared to the stabilized blowout zone. This result is in line of expectation, as high BD values indicate coarser structures of the soil (Adam, 1973; Rawls & Walter, 1983; Kolb, 2017). In addition, high BD values illustrate higher sand content. The roots of vegetation or presence of OM instead of sand, loosen the soil and cause lower BD values. Hence, after a storm with high aeolian activity the BD values are expected to be higher in both areas in 2018 compared to 2017. This hypothesis is confirmed with the findings of both comparisons. In both the active blowout as in the stabilized blowout the BD is significantly higher in 2018. When BD increases, the assumption that bare sand cover has increased in 2018 is plausible. However, only for the comparison between the active blowout zone and stabilized blowout zone in 2018 this assumption is confirmed, based on statistical analysis. This difference was already encountered in Kolb (2017) and is caused by the large active blowout. Because bare sand cover has not shown significant differences between 2017 and 2018 in both areas, the Ah-layer thickness of both areas is also expected to be relatively similar. Nonetheless, the Ah layer contains OM, meaning more vegetation, lower pH and lower BD values (Kooijman & Besse, 2002). This means the stabilized blowout zone should contain thicker Ah-layers. Yet, the results in this study show for all three comparisons not significant values, so this hypothesis cannot be confirmed. Even the comparison between active blowout zone and stabilized blowout zone in 2018 gave no significant differences, while the bare sand cover and BD did show negative differences in the stabilized zone. This can potentially be explained by the fact that the transect soil descriptions ranged until approximately 1-meter. In both zones, Ah-layers deeper in the soil (~30 cm) were found. These deeper layers can hardly be affected by one storm, or in the short period of 1 year.

Alongside the older Ah-soils, younger dune soils are characterized by low-growing pioneer species, such as mosses, lichens and small grasses (Grootjans et al., 2013). The blowout zone of the active blowout area shows characteristic signs of younger dune soils, with high pH and high species richness. According to Grootjans et al. (2013), in the older dune soils, further away from the blowout and in the stabilized zone, pH decreases and N-availability of the soil increases. In the comparison of the active blowout zone and stabilized blowout zone a significant higher N-content of the soil is encountered, which confirms the above statement of Grootjans et al. (2013). However, the storm impact on N-content in the soil is minimal, as both comparisons of 2017-2018 are given non-significant values. C-contents of the stabilized zone comparison increased significantly in all three zones. According to Jones et al. (2004), succession is causing more woody species to appear, which makes the litter contain more phenols and organic compounds and decomposition rates slow down. With these signs of low decompositions rates, the N-contents were also expected to increase in 2018. Besides C & N, the EC of a soil provides information about nutrient conditions and acidity of the soil and can be used as parameter for biological activity (Smith & Doran, 1996; Kolb, 2017). Contrastingly than assumed, no significant differences in EC in the active-stable comparison 2018 are present. However, the stabilized blowout comparison (2017-2018) did show significant decreasing values of EC in 2018. Kolb (2017) encountered positive significant correlations between C & N and EC. Even though, the stabilized blowout area has significant decreasing values of EC and increasing values of C & N. According to Kooijman & Besse (2002), this could be due to changes in OM-contents occurring in dune soils.

6.3 Historical storms and the impact on ‘De Nederlanden’

The natural blowout area of ‘De Nederlanden’ is a special phenomenon in the lime-poor and fixed Grey Dunes of Texel. The effects of severe storm events on this active area have also been included in this study. Total bare sand coverage and total blowout surface area have been observed throughout the period of 1996-2016 with an aerial photo analysis. Storms have been recorded for a long time in the Netherlands, the KNMI has recorded heavy storm events (>24.5 m/s) since 1910. By connecting these results with storm data of KNMI (2018), conclusions can be drawn upon the historical effects of heavy storm events on the area of ‘De Nederlanden’. Contrasting from the results in this study, the peaks of bare sand cover, starting in 2006 until 2009 are not caused by a heavy storm, according to data of KNMI (2018). Furthermore, the high peak of bare sand in 2011 is also caused by something different than a heavy storm event. In fact, no heavy storm events occurred in Texel in the period of 2002-2013, meanwhile a lot of fluctuations are visible in the results of the

(18)

Page 18 of 59

aerial photo analysis. In addition, heavy storm events did occur near Texel in 2013, which did not result in more bare sand at the aerial photo of 2014. Other factors, such as grazing have been effective in dry dune grasslands according to Kooijman & de Haan (1995) and could potentially be another causation for the fluctuations of bare sand coverage in ‘De Nederlanden’. However, this is beyond the scope of this paper.

Besides bare sand coverage, the results of this study also show a fluctuating trend of the maximum surface area of bare sand found in each year. In each year (1996-2016), this largest patch of bare sand is presented by the natural active blowout and close surrounding areas (radius of ~25m). The surface of the blowout is changing over time, with the largest surface area in the years 2007-2008 and 2010-2011. These years correspond with the periods of high peak values of bare sand, which means no storm event with windspeeds of 24.5 m/s or higher is causing this increase of the blowout surface area. Explanations are to some extent given by Veer & Kooijman (1997), which discuss the characteristics of Grey Dune soils and vegetation. As mentioned before, natural active blowouts are rare in the lime-poor Wadden area. The absence of calcium-rich particles (CaCO3), and the high N-availability causes soils in the Grey Dunes to stabilize

(Provoost et al., 2004). Veer & Kooijman (1997) add that grass dominated areas have twice the root biomass of open dune grasslands. Because both the active and stabilized blowout zones have high vegetation covers with high grass and scrub encroachment, aeolian activity is limited. Even in the active blowout itself, large roots are visible on the edges of the blowout. These roots also indicate that the blowout is limited to grow in the future, due to the roots holding the sand. The depth of the blowout has not been recorded and is not easily extracted from aerial photographs, however the depth of the active blowout is finite. Due to the already large depth of the blowout, aeolian activity is restrained causing less sand uptake to occur. This process will only increase in the future, when the active blowout is unable to expand in width due to deep rooting and the depth will also reach limits. In addition, the differences of bare sand coverage between 2017-2018 in both blowout areas has been not significant, while a storm of 24.5 m/s or higher did occur in January 2018 (KNMI, 2018). Earlier indicated findings of Jungerius et al. (1991) argue that blowout changes show a strong correlation with wind velocities between 6.25 and 12.5 m/s. This argument implies that extreme high windspeeds (>24.5 m/s) are unfavorable wind velocities for sand particle transportation. Combining arguments of the rooting characteristics and wind velocities, erosional and depositional processes of sand in ‘De Nederlanden’ were limited in 2018.

6.4 Implications of study

This study does contain some implications. To begin, this study is a review of the exact same blowout areas in ‘De Nederlanden’ in 2017. Similar transects and grids are required to perform the comparison, however incomplete forms without GPS-locations have been provided. Resulting in roughly the same transects and grids and not exactly similar sample points. Additionally, to see newly surfaced sand layers in the topsoil, multiple Ah-layers should be noted. Last year’s study did only refer to the total centimeters of Ah-layers in the 20-centimeter topsoil, which made it impossible to compare the different Ah-layers. Thus, to make equal comparisons our Ah-layer results are also total values instead of multiple layers. Another implication is that only 40 soil and vegetation samples were collected from the transects. This relatively low sample size increases the probability of errors, and decreases the statistical power (MacCallum et al., 1996). C-contents, N-contents and C/N ratios are the most influenced parameters by low statistical power, because they have been subdivided into three smaller zones. Along with N, high levels of P play an important function in the process of increasing grass encroachment in the lime-poor Grey Dunes in the Wadden area (Kooijman et al., 2016). This study has not taken into account the effects of the January 2018 storm on P-availability in the soil. However, this is an important indicator for soil characteristics, which should be considered in further research. Further research is also necessary in other Grey Dunes areas along the Dutch coast, because this study only compares two zones in two years. To examine the effects of the January 2018 storm, or other intense storm events, on blowout development in the Netherlands, longer periods and more blowout areas must be compared to come to justified conclusions. In addition, when more studies on intense storm events are conducted, small errors in results due to failing equipment or machinery are more easily prevented. This study

(19)

Page 19 of 59

shows some contrasting results, which could in the case of the C/N analysis be caused by a fault in the CHNS analyzer settings. Uncertainty remains whether the C/N differences between 2017-2018 are caused by the January 2018 storm, or by measurement errors. To conclude, the last implication relies on the subjectivity of ordinations. In both 2017 and 2018 vegetation and bare sand coverage had to be estimated in percentages. Although the estimations were controlled and conducted similarly along the transects, differences in estimations are among the possibilities.

7. Conclusions

Atmospheric N deposition causes acidification and eutrophication of the soil in the lime-poor Grey Dunes in Texel. Differences between active blowout zone and stabilized blowout zone in ‘De Nederlanden’ are clearly visible and correspond with findings of previous studies; higher succession rates lead to more grass encroachment in the stabilized blowout area. The question whether the January 2018 storm enhanced aeolian dynamics, positively effecting both blowout areas, includes some more nuanced conclusions. The active blowout area showed positive effects in the form of more herbs, lichens and mosses and a higher BD, compared to 2017. However, no significant change occurred in bare sand coverage, Ah-layer thickness, C & N variables and pH, indicating few positive effects on soil parameters. On the other hand, the stabilized blowout area displayed more differences, with a significant higher pH, BD, C- content soil and C/N ratio soil. Yet, more N preferring vegetation appeared, in combination with non-significant changing values of bare sand coverage and Ah-layer thickness. In addition, historical analysis on storm events in ‘De Nederlanden’ distinguishes no clear correlation between heavy storm events (>24.5 m/s) and increases of total bare sand coverage and the total blowout zone surface area. Combining these results, the January 2018 storm seems not to have had the expected positive effects on both blowout areas. The aeolian activity might have been increased by the storm, however the non-significant changes of bare sand coverage, combined with the historical findings seem to demonstrate that the extreme January 2018 storm (31.1 m/s) contained unfavorable high wind velocities. Because this is only a study on two blowout areas, more research on comparable lime-poor Grey Dune areas is necessary in order to extend and apply these conclusions to comparable areas in the Netherlands. References

Adams, W. A. (1973). The effect of organic matter on the bulk and true densities of some uncultivated podzolic soils. European Journal of Soil Science, 24(1), 10-17

Aggenbach, C.J.S., Arens, S.M., Kooijman, A.M. & Lammers, W.J. (2016). Beheeradvies activering eolische dynamiek op de Waddeneilanden als PAS-maatregel voor habitattype H2130 Grijze duinen. VBNE,

Vereniging van Bos- en Natuurterreineigenaren Advies OBN-11-DK.

Grootjans, A. P., Dullo, B. S., Kooijman, A. M., Bekker, R. M., & Aggenbach, C. (2013). Restoration of dune vegetation in the Netherlands. In Restoration of coastal dunes (pp. 235-253). Springer Berlin Heidelberg. Haarsma, R., Tijm, S. & van den Brink, H. (2018). Zeer zware storm van 18 januari. KNMI.

Hesp, P. (2002). Foredunes and blowouts: initiation, geomorphology and dynamics. Geomorphology, 48(1-3), 245-268.

Hicks, W. K., Whitfield, C.P., Bealey, W.J., Sutton, M.A. (2011). Nitrogen Deposition and Natura 2000:

Science and Practice in Determining Environmental Impacts. COST729/ Nine/ESF/CCW/JNCC/SEI Workshop Proceedings, published by COST. Available at: http:// cost729.ceh.ac.uk/n2kworkshop

Jones, M. L. M., Wallace, H. L., Norris, D., Brittain, S. A., Haria, S., Jones, R. E. & Emmett, B. A. (2004). Changes in vegetation and soil characteristics in coastal sand dunes along a gradient of atmospheric nitrogen deposition. Plant Biology, 6(05), 598-605.

(20)

Page 20 of 59

Jungerius, P. D., Witter, J. V., & Van Boxel, J. H. (1991). The effects of changing wind regimes on the development of blowouts in the coastal dunes of The Netherlands. Landscape Ecology, 6(1-2), 41-48.

Jungerius, P. D., & Van der Meulen, F. (1989). The development of dune blowouts, as measured with erosion pins and sequential air photos. Catena, 16(4-5), 369-376.

Kolb, E.M. (2017). Active blowouts as management measure for grey dunes on Texel. University of Amsterdam: Faculteit der Natuurwetenschappen, Wiskunde en Informatica. 1-40.

Koninklijk Nederlands Meteorologisch Instituut (KNMI). (2018). Zware stormen in Nederland sinds 1910. Geraadpleegd op 27 juni 2018, van https://www.knmi.nl/nederland-nu/klimatologie/lijsten/zwarestormen. Kooijman, A. M., & De Haan, M. W. A. (1995). Grazing as a measure against grass encroachment in Dutch dry dune grassland: effects on vegetation and soil. Journal of Coastal Conservation, 1(2), 127-134.

Kooijman, A. M., & Besse, M. (2002). The higher availability of N and P in lime‐poor than in lime‐rich coastal dunes in the Netherlands. Journal of Ecology, 90(2), 394-403.

Kooijman, AM., Besse M. & Haak R. (2005). Effectgerichte maatregelen tegen verzuring en eutrofiering in open droge duinen. Eindrapport fase 2. Directie Kennis, Ministerie van Landbouw, Natuur en Voedselkwaliteit, rapport nr. 2005/dk008-O, 158 pp.

Kooijman, A., van der Hagen, H., & Noordijk, E. (2012). Stikstof depositie in de duinen: alles in beeld? Landschap, 29(3), 147-154.

Kooijman, A. M., van Til, M., Noordijk, E., Remke, E., & Kalbitz, K. (2016). Nitrogen deposition and grass encroachment in calcareous and acidic Grey dunes (H2130) in NW-Europe. Biological Conservation. MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological methods, 1(2), 130.

Pluis, J. L. A., & De Winder, B. (1989). Spatial patterns in algae colonization of dune blowouts. Catena, 16(4-5), 499-506.

Provoost, S., Ampe, C., Bonte, D., Cosyns, E., & Hoffmann, M. (2002). Ecology, management and monitoring of dune grasslands in flanders, belgium. Littoral 2002.the Changing Coast, 11-22.

Provoost, S., Ampe, C., Bonte, D., Cosyns, E., & Hoffmann, M. (2004). Ecology, management and monitoring of grey dunes in Flanders. Journal of Coastal Conservation, 10(1), 33-42.

Rawls, W. J. (1983). Estimating soil bulk density from particle size analysis and organic matter content. Soil Science, 135(2), 123-125.

Smith, J. L., & Doran, J. W. (1996). Measurement and use of pH and electrical conductivity for soil quality analysis. Methods for Assessing Soil Quality. Soil Science Society of America (SSSA). Special Publication, (49).

Van Boxel, J. H., Jungerius, P. D., Kieffer, N., & Hampele, N. (1997). Ecological effects of reactivation of artificially stabilized blowouts in coastal dunes. Journal of coastal conservation, 3(1), 57.

Van der Meulen, F., Kooijman, A. M., Veer, M. A. C., & van Boxel, J. H. (1996). Effectgerichte maatregelen tegen verzuring en eutrofiering in open droge duinen. Eindrapportage fase 1.

Van der Meulen, F., & Salman, A. H. P. M. (1996). Management of Mediterranean coastal dunes. Ocean & Coastal Management, 30(2-3), 177-195.

(21)

Page 21 of 59

Veer, M. A. C. (1997). Nitrogen availability in relation to vegetation changes resulting from grass encroachment in Dutch dry dunes. Journal of Coastal Conservation, 3(1), 41-48.

Veer, M. A. C., & Kooijman, A. M. (1997). Effects of grass-encroachment on vegetation and soil in Dutch dry dune grasslands. Plant and soil, 192(1), 119-128.

Witz L., (2015). The potential of small-scale blowout activity for landscape diversity in the Dutch grey dunes. Master Thesis, University of Amsterdam.

Appendix

A. Extended methodology – Laboratory calculations Bulk Density (BD) calculation:

Total wet weight * Dry subsample weight / (Wet subsample weight * Volume) C & N content calculations:

C-content = (BD*Vt) * (%C/100) (in g/m2)

N-content = (BD*Vt) * (%N/100) (in g/m2)

BD = Bulk Density in g/cm3

Vt = soil volume in 10cm topsoil-layer (10*100*100=100000 cm3)

%C = Carbon percentage obtained from CNHS-analyzer %N = Nitrogen percentage obtained from CNHS-analyzer C/N ratios calculations:

(22)

Page 22 of 59

B. Statistical analyses tables Historical analysis tables

Year Bare Sand (m2_{) Mean (m}2₎ _{Std (m}2₎ _{Min (m}2₎ _{Max (m}2₎

1996 44771.113752 243.32127 384.428524 4.35584 3213.120117 2000 57467.977391 181.860688 347.716323 0.054691 2843.459961 2003 42434.693327 70.84256 180.707505 0.140009 2780.659912 2005 22364.483211 96.815945 153.027099 3.150208 1082.349243 2006 65838.343365 116.322161 333.951405 0.239631 4102.106445 2007 81141.329097 117.766806 755.445147 2.220235 18798.013672 2008 83082.187558 129.815918 821.858791 0.054691 19042.058594 2009 82321.547736 136.293953 482.233499 0.070005 6349.115723 2010 57172.354848 114.116477 813.106561 0.746716 17869.359375 2011 150803.071037 218.238887 1454.712414 0.044803 18476.699219 2012 50579.729433 139.338098 403.18335 0.086267 3686.406006 2013 43938.411931 85.483292 344.133746 0.109382 5663.19873 2014 43102.341916 94.939079 721.62957 0.140009 13454.400391 2015 48144.422265 112.224761 495.707791 0.043753 6917.953125 2016 26174.576948 93.480632 312.063539 0.088206 3119.852539 Soil variables

Comparison C/N ratio Active-Stable 2018 Soil

Zones P-value Significant

Unaffected Zone 0.1115 No

Blowout Zone 0.0199 Yes

Accumulation Zone 0.6709 No

Comparison C/N ratio Active 2017- 2018 Soil

Unaffected Zone 0.0279 Yes

Blowout Zone 8.5197e-05 Yes

(23)

Page 23 of 59

Comparison C/N ratio Stable 2017-2018 Soil

Blowout Zone 0.1214 No

Accumulation Zone 0.0034 Yes

Comparison C content Active-Stable 2018 Soil

Comparison N content Active-Stable 2018 Soil

Comparison C content Active 2017-2018 Soil

Comparison N content Active 2017-2018 Soil

Comparison C content Stable 2017-2018 Soil

Comparison N content Stable 2017-2018 Soil

(24)

Page 24 of 59

Means and standard deviations of active blowout for 2017 and 2018. Soil

Soil parameters Mean A 2017 Mean A 2018 Std A 17 Std A 18

pH 6.4945 6.6825 1.0195 0.7968

EC (mS/cm) 76.89 63.715 38.0064 38.3993

BD (g/cm3) 1.1964 1.3610 0.1978 0.1642

C content (g/m2) complete transect 808.79 787.67 515.37 558.30

C content (g/m2) unaffected 937.67 838.16 614.1 628.91

C content (g/m2) blowout 257.16 381.44 89.08 88.25

C content (g/m2) accumulation 950.80 912.45 446.33 589.29

N content (g/m2) complete transect 69.20 56.88 47.71 43.84

N content (g/m2) unaffected 86.68 62.54 51.57 43.42

N content (g/m2) blowout 4.1 16.85 1.51 8.33

N content (g/m2) accumulation 84.92 68.87 33.76 45.41

C/N-ratio complete transect 21.2832 15.8796 21.4914 6.0948

C/N-ratio unaffected 10.7954 13.1410 1.8225 0.7137

C/N-ratio blowout 63.0864 25.3589 1.7406 7.8859

C/N-ratio accumulation 10.8493 13.6774 1.2591 2.5869

Minima and maxima of active blowout for 2017 and 2018. Soil

Soil parameters Min A 17 Min A 18 Max A 17 Max A 18

pH 5.15 5.50 8.70 7.92

EC (mS/cm) 27.60 19.70 160.40 127.40

BD (g/cm3) 0.8098 1.02 1.5745 1.59

Bare sand coverage (%) 0 0 100 100

C content (g/m2) unaffected 288.17 390 1824.18 1923.90

C/N-ratio blowout 61.2903 18.4783 64.7059 32.4561

(25)

Page 25 of 59

Means and standard deviations stable blowout for 2017 and 2018.Soil

Soil parameters Mean S 17 Mean S 18 Std S 17 Std S 18

pH 5.1330 5.6240 0.5959 0.4015

EC (mS/cm) 78.64 47.875 28.5394 31.9963

BD (g/cm3) 0.9862 1.1325 0.1998 0.1570

N content (g/m2) complete transect 116.96 135 38.74 62.40

C/N-ratio blowout 9.8893 12.9048 3.2711 0.6955

C/N-ratio accumulation 12.2153 14.0536 1.2591 2.5869

Minima and maxima of stable blowout for 2017 and 2018. Soil

Soil parameters Min S 17 Min S 18 Max S 17 Max S 18

pH 4.06 4.6 6.13 6.05

EC (mS/cm) 35.8 19.2 139.1 140.9

BD (g/cm3) 0.6009 0.89 1.3378 1.55

Bare sand coverage (%) 0 0 25 50

C/N-ratio blowout 6.7797 12.3668 13.7942 13.9125

(26)

Page 26 of 59

(27)

Page 27 of 59

(28)

Page 28 of 59

(29)

Page 29 of 59

(30)

Page 30 of 59

(31)

Page 31 of 59

(32)

Page 32 of 59

(33)

Page 33 of 59

(34)

Page 34 of 59

(35)

Page 35 of 59

(36)

Page 36 of 59