Development of a Nitrophobe/Nitrophile Classification for Sand Dunes

COMMISSIONED REPORT

Commissioned Report No. DK1605b Development of a nitrophobe/nitrophile classification for sand dunes.

For further information on this report please contact

Claire Campbell SEPA Corporate Office Strathallan House Castle Business Park STIRLING FK9 4TZ Telephone: 01786 452448 E-mail: [email protected]>

This report should be quoted as:

Jones, L. and Stevens, C. (2017). Development of a nitrophobe/nitrophile classification for sand dunes Scottish Environment Protection Agency Commissioned Report No. DK1605b.

This report, or any part of it, should not be reproduced without the permission of Scottish Environment Protection Agency. This permission will not be withheld unreasonably. The views expressed by the author(s) of this report should not be taken as the views and policies of Scottish Environment Protection Agency.

COMMISSIONED REPORT Summary

Development of a nitrophobe/nitrophile classification for sand dunes

Commissioned Report No. DK1605b Contractor: Jones, L. and Stevens, C. Year of publication: [2017]

Executive Summary Both dry dune habitats and wet dune slack communities of sand dunes are known to be sensitive to N deposition. However, despite their biological importance, they are not currently covered within the nitrophobe/nitrophile assessment methodology. This study aimed to develop indicators for sand dune communities.

We developed lists of nitrophile indicator species, focusing on those above a threshold Ellenberg N value, defined in relation to the 80th percentile Ellenberg N of component NVC communities. Species lists were extracted from the NVC manual for dune habitats.

The method differs slightly from that in Pitcairn et al. (2006) who developed both nitrophile and nitrophobe indicators. In the approach taken here it was decided not to include nitrophobe species, but to focus on nitrophiles expected to increase with eutrophication. The index for any quadrat of plant data is calculated as the sum of percentage cover of listed nitrophile indicator species for the relevant dune habitat type.

UK-wide data were used to develop and test the indicators. This approach is generalizable to any situation in which the dune species have Ellenberg indicator values defined (i.e. primarily temperate Europe). Local benchmarking of indicator values against N deposition would be required to develop absolute values for the metrics in a new situation.

Testing of the indicators against survey data showed a high degree of scatter in individual quadrats, but site means showed broadly positive relationships with N deposition. Testing of the dune slack indicator against an N-gradient in groundwater at a single site also showed some promise. However, this initial testing is based on limited datasets.

We recommend more extensive testing of the indicators against wider datasets, and to consider including terricolous lichens as a component of the indicator in the acidic dune grasslands habitat.

Contents 1 Introduction ...... 5 2 Methods ...... 5 2.1 Categorising dune communities ...... 5 2.2 Selecting indicator species ...... 8 3 Results – selected nitrophile indicator species ...... 9 3.1 Semi-fixed and blow-out communities ...... 9 3.2 Calcareous fixed dune grassland ...... 10 3.3 Acidic dune grassland & dune annuals ...... 11 3.4 Dune slacks...... 12 4 Testing the indicators ...... 13 5 Discussion of indicators and testing ...... 13 6 Conclusions ...... 18 7 References ...... 19

List of Figures Page Figure 1. Frequency distribution of Ellenberg N values, showing mean and 80%ile Ellenberg N for each NVC sand dune community. From Jones et al. 2002. 7 Figure 2. Nitrophile index for quadrats of semi-fixed and blow-out communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site. 14 Figure 3. Nitrophile index for quadrats of calcareous fixed dune grassland communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site. 14 Figure 4. Nitrophile index for quadrats of acidic dune grassland & dune annuals communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site. 15 Figure 5. Nitrophile index for quadrats of dune slacks. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site. 15 Figure 6. Nitrophile index for quadrats of dune slacks, tested against a gradient of nitrate concentration in groundwater at Aberffraw dunes in North Wales. Data from Rhymes et al. (2014; 2015) 16

List of Tables Page Table 1. Groupings of dune NVC communities, 80th percentile Ellenberg N score for each, and threshold Ellenberg N value applied to determine which species were used as indicators for each group. 6 Table 2. Nitrophile indicator species for semi-fixed and blow-out communities (SD7, SD10) 9 Table 3. Nitrophile indicator species for calcareous fixed dune grassland (SD8, SD9) 10 Table 4. Nitrophile indicator species for Acidic dune grassland & dune annuals (SD11, SD12, SD19) 11 Table 5. Nitrophile indicator species for Dune slacks (SD13 – SD17) 12 Table 6. Species positively associated with high N deposition in the Scottish dune repeat survey together with the habitats groups for which they are eutrophic indicators. 16

1 Introduction Both dry dune habitats and wet dune slack communities of sand dunes are known to be sensitive to N deposition (Jones et al. 2004; Jones et al. 2013; Remke et al. 2009a; Rhymes et al. 2014). Demonstrated impacts include shifts in species composition and reductions in plant diversity (Field et al. 2014). Despite their biological importance and high sensitivity to nitrogen deposition, they are not currently covered within the nitrophobe/nitrophile assessment methodology.

Therefore, this project will develop a field-based classification of the response to atmospheric nitrogen of higher plant species and bryophytes typically found in sand dunes in Scotland and the wider UK. It will develop a nitrophobe/nitrophile classification using species with an extreme Ellenberg response, broadly following the methodology used by Pitcairn et al. (2006).

2 Methods 2.1 Categorising dune communities Sand dune communities represent a complex continuum of habitats from dry to wet, high pH to low pH and young to old. This is reflected in the NVC classification which comprises 18 dune communities, including strandline communities (Rodwell, 2000). In order to simplify the development and application of an indicator system, these communities have been grouped according to their likely sensitivity to N. This built on previous work to assess the N sensitivity of dune habitats (Jones et al. 2002), which characterised the distribution profile of Ellenberg N values in each sand dune NVC community, based on their vascular plant composition (Figure 1), calculated for each community based on the species lists provided in Rodwell (2000) and using the Hill-modified Ellenberg N value from PlantAtt (Hill et al. 1999). Using this information as a starting point, communities were grouped according to four main principles:

 similar sensitivity to N as indicated by the Ellenberg N distribution of their component species  similar degree of soil development, mainly separating the younger or more mobile communities from the more stable fixed dune grasslands  soil pH, separating calcareous communities from those which have started to decalcify or those which have acidic sand as parent material, and  any hydrological influence, separating dune slacks from the dry dune communities

When deciding on communities where the group membership was not immediately clear, we looked in more detail at the species composition of the groups, and at the community average Ellenberg R values (as a proxy for soil pH), to make final allocation to groups. The resulting groupings are shown in Table 1.

Strandline communities were excluded as they are exposed to an open N cycle with regular inputs of N from saltspray, seaweed and other marine biological material. The most mobile dune communities were also excluded as they are highly dynamic and the role of N as a factor governing species composition plays a much smaller role than physical disturbance and proximity to the sea. Lastly, infrequent communities such as dune heath and dunes with Juniper were also excluded.

Table 1. Groupings of dune NVC communities, 80th percentile Ellenberg N score for each, and threshold Ellenberg N value applied to determine which species were used as indicators for each group.

Ellenberg NVC N code NVC Community (80th percentile Ellenberg N score) Group name threshold SD2- Honkenya peploides - Cakile maritima strandline community; Strandline SD3 Matricaria maritima - Galium aparine strandline community communities Exclude Elymus farctus ssp. boreali-atlanticus foredune community; Foredune and SD4- Leymus arenarius mobile dune community; mobile dune SD6 Ammophila arenaria mobile dune community communities Exclude

Semi-fixed and blow-out Ammophila arenaria - Festuca rubra semi-fixed dune 6 communities SD7 community (6.0) (SD7, SD10) SD10 Carex arenaria dune community (6.0)

Calcareous SD8 Festuca rubra - Galium verum fixed dune grassland (5.2) fixed dune 6 grassland Ammophila arenaria - Arrhenatherum elatius dune grassland (SD8, SD9) SD9 (6.0) SD11 Carex arenaria - Cornicularia aculeata dune community (3.0) Acidic dune Carex arenaria - Festuca ovina - Agrostis capillaris dune grassland & SD12 grassland (5.0) dune annuals 4 Phleum arenarium - Arenaria serpyllifolia dune annual (SD11, SD12, community Tortulo-Phleetum arenariae (Massart 1908) Br.- SD19) SD19 Bl. & de Leeuw 1936 (4.0)

Sagina nodosa - Bryum pseudotriquetrum dune-slack SD13 community (5.0) Salix repens - Campylium stellatum dune-slack community Dune slacks SD14 (5.0) 6 (SD13-SD17) Salix repens - Calliergon cuspidatum dune-slack community SD15 (5.4) SD16 Salix repens - Holcus lanatus dune-slack community (5.0) SD17 Potentilla anserina - Carex nigra dune-slack community (6.0) Hippophae SD18 Hippophae rhamnoides dune scrub scrub Exclude - Other communities: Dune heath, Dunes with Juniper Other Exclude

Strand line SD 2 Honkenya-Cakile (Southern) SD 3 Matricaria-Galium (Northern) communities mean = 5.94, 80%ile = 7.0 mean = 6.08, 80%ile = 7.0 30 30 25 25 20 20 15 15 10 10 5 5 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

SD 4 Elymus farctus SD 5 Leymus arenarius SD 6 Ammophila arenaria Foredune mean = 5.93, 80%ile = 7.0 mean = 5.44, 80%ile = 7.0 mean = 5.18, 80%ile = 6.4 and mobile 30 30 30 dune 25 25 25 communities 20 20 20 15 15 15 10 10 10 5 5 5 0 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

SD 7 Ammophila-Festuca SD 10 Carex arenaria

mean = 4.14, 80%ile = 6.0 mean = 4.08, 80%ile = 6.0 Semi-fixed 30 30 and blow-out 25 25 communities 20 20 15 15 10 10 5 5 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

SD 8 Festuca-Galium (grazed) SD 9 Ammophila-Arrhenatherum (ungrazed) Calcareous mean = 3.89, 80%ile = 5.0 mean = 4.10, 80%ile = 6.0 fixed dune 30 30 grassland 25 25 20 20 15 15 10 10 5 5 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

SD 11 Carex-Cornicularia SD 12 Carex, Festuca, Agrostis SD 19 Phleum arenaria Acidic dune mean = 2.53, 80%ile = 3.0 mean = 3.37, 80%ile = 5.0 mean = 3.16, 80%ile = 4.0 grassland & 30 30 30 dune annuals 25 25 25 20 20 20 15 15 15 10 10 10 5 5 5 0 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

Dune slacks SD 13 Sagina-Bryum (immature slack)SD 14 Salix-Campylium SD 15 Salix-Calliergon

mean = 3.82, 80%ile = 5.0 mean =3.99, 80%ile = 5.0 mean = 4.31, 80%ile = 5.4 30 30 30 25 25 25 20 20 20 15 15 15 10 10 10 5 5 5 0 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

SD 16 Salix-Holcus (drier, ungrazed) SD 17 Potentilla-Carex nigra (wetter, grazed) SD 18 Hippophae scrub

mean = 3.98, 80%ile = 5.0 mean = 4.34, 80%ile = 6.0 SD 18, mean = 5.00 30 30 Hippophae scrub 30 25 25 25 20 20 20 15 15 15 10 10 10 5 5 5 0 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

Figure 1. Frequency distribution of Ellenberg N values, showing mean and 80%ile Ellenberg N for each NVC sand dune community. From Jones et al. 2002.

The first grouping for which indicators were developed is mobile and semi-fixed dunes. This includes SD7 and SD10, both largely calcareous communities. The second group comprises fixed dune grasslands which are also calcareous SD8 and SD9, the former grazed, the latter ungrazed. The third group comprises the more acidic dune communities SD11 which is strongly acidic and SD12 where surface soils are starting to acidify. It also includes the dune annuals community SD19. A number of combinations were trialled, with preliminary testing against observed data. The testing revealed that although SD12 is often only partly decalcified, the species complement and Ellenberg score profile of the community (Figure 1) was closer to that of the acidic SD11 than the more calcareous SD8 and SD9. Similarly, the species complement of the dune annual community in the NVC has closer affinities to the acidic sand communities than to the calcareous communities. Although the slack communities differed slightly in terms of Ellenberg N profile, they were not considered sufficiently different to further subdivide this group.

2.2 Selecting indicator species The approach of selecting indicator species within each community group differs slightly from earlier work by Pitcairn et al. (2006). It was felt that reliance only on species in the NVC lists which had a constancy value of III or higher (i.e. frequency > 40%) might miss species present at low cover or frequency which may nonetheless become dominant in future due to eutrophication.

In addition, rather than selecting particular species as either nitrophobes or nitrophiles, it was decided to focus solely on the nitrophiles which might expand in cover under more eutrophic conditions. Therefore, the 80th percentile of Ellenberg N value was calculated for each community. This threshold was chosen to include sufficient of the most nitrophilic species from the NVC lists for each community, without creating impractically long lists of species. For consistency, a single representative Ellenberg N value was then chosen across all NVC communities in each group, and all species with an Ellenberg N value equal to or greater than that threshold were selected as nitrophile indicator species for that group. The lists contain both vascular plants and bryophytes, but not lichens. Terricolous lichens were not included as they make up a relatively small proportion of the species complement, with the exception of acidic dune grasslands where they do form an important component of the vegetation. For consistency of approach across communities they were not included in this assessment, but the approach could be modified in future.

3 Results – selected nitrophile indicator species Nitrophile indicator species for each community group are listed below.

3.1 Semi-fixed and blow-out communities

Table 2. Nitrophile indicator species for semi-fixed and blow-out communities (SD7, SD10)

Ellenberg Species N Vascular & other plants Arrhenatherum elatius 7 Cirsium arvense 6 Cirsium vulgare 6 Dactylis glomerata 6 Elytrigia atherica 6 Elytrigia juncea 6 Elytrigia repens 7 Heracleum sphondylium 7 Honckenya peploides 6 Leymus arenarius 6 Potentilla anserina 6 Ranunculus repens 7 Rubus caesius 6 Rubus fruticosus agg. 6 Rumex crispus 6 Senecio vulgaris 7 Sonchus arvensis 6 Sonchus asper 6 Trifolium repens 6 Tripleurospermum maritimum sens.lat. 6 Tussilago farfara 6

Bryophytes Brachythecium rutabulum 8 Bryum argenteum 8 Eurhynchium praelongum 6

3.2 Calcareous fixed dune grassland

Table 3. Nitrophile indicator species for calcareous fixed dune grassland (SD8, SD9)

Ellenberg Species N Vascular & other plants Agrostis stolonifera 6 Arrhenatherum elatius 7 Cirsium arvense 6 Cirsium vulgare 6 Dactylis glomerata 6 Elytrigia atherica 6 Elytrigia juncea 6 Elytrigia repens 7 Equisetum arvense 6 Heracleum sphondylium 7 Leymus arenarius 6 Lolium perenne 6 Myosotis arvensis 6 Plantago major 7 Poa trivialis 6 Potentilla anserina 6 Ranunculus repens 7 Silene latifolia 6 Sonchus oleraceus 7 Torilis japonica 7 Trifolium repens 6

Bryophytes Brachythecium rutabulum 8 Eurhynchium praelongum 6 Plagiomnium rostratum 6 Plagiomnium undulatum 6

3.3 Acidic dune grassland & dune annuals

Table 4. Nitrophile indicator species for Acidic dune grassland & dune annuals (SD11, SD12, SD19)

Ellenberg Species N Vascular & other plants Agrostis capillaris 4 Arenaria serpyllifolia 5 Cirsium arvense 6 Crepis capillaris 4 Erodium cicutarium agg. 4 Euphorbia paralias 5 Festuca rubra agg. 5 Geranium molle 5 Holcus lanatus 5 Lathyrus pratensis 5 Plantago coronopus 4 Plantago lanceolata 4 Poa pratensis sens.lat. 5 Ranunculus repens 7 Senecio jacobaea 4 Trifolium campestre 4 Trifolium dubium 5 Trifolium repens 6 Urtica dioica 8 Viola tricolor 4

Bryophytes Brachythecium albicans 4

3.4 Dune slacks

Table 5. Nitrophile indicator species for Dune slacks (SD13 – SD17)

Ellenberg Species N Vascular & other plants Agrostis stolonifera 6 Alopecurus geniculatus 6 Arrhenatherum elatius 7 Bolboschoenus maritimus 7 Carex hirta 6 Cirsium arvense 6 Elytrigia repens 7 Equisetum arvense 6 Eupatorium cannabinum 7 Glechoma hederacea 7 Iris pseudacorus 6 Juncus gerardii 6 Lolium perenne 6 Lycopus europaeus 6 Persicaria maculosa 7 Phragmites australis 6 Plantago major 7 Poa annua 7 Poa trivialis 6 Potentilla anserina 6 Ranunculus repens 7 Rubus caesius 6 Rumex crispus 6 Salix caprea 7 Solanum dulcamara 7 Sonchus arvensis 6 Trifolium fragiferum 6 Trifolium repens 6

Bryophytes Amblystegium serpens 7 Brachythecium rutabulum 8 Eurhynchium praelongum 6 Plagiomnium rostratum 6

4 Testing the indicators Available data from two targeted gradient surveys and a site-specific study (for dune slacks only) were used to test and refine the indicators. These included a survey in 2006 focusing on dune slack species (Jones 2007) and a survey focusing primarily on decalcified dune grasslands (Field et al. 2014), although incorporating a number of other communities. In these available datasets, there was incomplete representation across all the communities, but they served the purpose of initial testing of the indicators, pending more thorough testing in the companion project RAD031-Botanical Benchmarks (Jones et al. 2017), with more extensive dune datasets. In each figure below, the open circles represent individual quadrats, the black filled circles represent the mean nitrophile score for the site.

Data for the semi-fixed and disturbance communities group was taken from the 2009 survey (Field et al. 2014), only for quadrats recorded as being in the SD7 community. Results from this fairly limited dataset are shown in Figure 2. There is a fairly poor fit.

Data for the calcareous fixed dune grassland group was also taken from the 2009 survey (Field et al. 2014), covering quadrats recorded as matching SD8 and SD9. Figure 3 shows a large scatter among individual quadrats, with the site means showing a trend towards increasing nitrophile index with increasing N deposition. However, the three sites with the largest N deposition show very low index values.

Data for the acidic dune grassland & dune annuals group was also taken from the 2009 survey (Field et al. 2014), covering quadrats recorded as matching SD12. Although individual quadrats show a fair degree of scatter (Figure 4), the site means appear to show a positive relationship with N deposition.

Data for the dune slacks were taken from the 2006 survey (Jones 2007). Although there were fewer sites in this survey, there were far more quadrats taken at each site. Overall, there is a positive trend to the data (Figure 5), with a suggestion that the relationship reaches a plateau above 10 kg N/ha/yr. An additional site-specific dataset from dune slacks at Aberffraw dunes in North Wales was tested, where there is a strong gradient of nitrate concentrations in the groundwater underlying the site, which has been shown to have deleterious effects on the dune slack ecology (Rhymes et al. 2014; 2015). This dataset shows a good relationship with the nitrophile indicator, with a steep increase at low nitrate levels, but reaching a plateau at higher concentrations (Figure 6).

5 Discussion of indicators and testing A number of the eutrophic indicators species identified in this study have previously been identified as responding positively to increased levels of nitrogen, either through experimental N additions or spatial gradients. A repeat survey of 89 coastal dunes in Scotland 34 years after the initial survey revealed a number of species that were positively associated with high N deposition (Pakeman et al. 2016). All of the species most strongly associated with high N deposition in the repeat survey have been identified as eutrophic indicator species in this study. Species correlated with high N deposition in the repeat survey are listed in Table 6.

Figure 2. Nitrophile index for quadrats of semi-fixed and blow-out communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site.

Figure 3. Nitrophile index for quadrats of calcareous fixed dune grassland communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site.

Figure 4. Nitrophile index for quadrats of acidic dune grassland & dune annuals communities. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site.

Figure 5. Nitrophile index for quadrats of dune slacks. Open circles are individual quadrats, filled circles represent the mean of quadrats at each site.

Figure 6. Nitrophile index for quadrats of dune slacks, tested against a gradient of nitrate concentration in groundwater at Aberffraw dunes in North Wales. Data from Rhymes et al. (2014; 2015)

Table 6. Species positively associated with high N deposition in the Scottish dune repeat survey together with the habitats groups for which they are eutrophic indicators.

SD7 + disturbance Acid SD11 + SD12 + Species SD10 SD8 + SD9 Fixed disturbance SD19 Agrostis capillaris 

Agrostis stolonifera 

Arrhenatherum elatius 

Cirsium arvense    Cirsium vulgare  

Dactylis glomerata  

Heracleum sphondylium  

Holcus lanatus 

Potentilla anserina  

Ranunculus repens   

There have been relatively few studies of changes in species composition in response to N deposition in sand dunes, compared with other habitats. In an experimental N addition study on a calcareous fixed dune grassland, Plassmann et al. 2009 found no significant effects of N addition on species composition after two years of N addition at rates of 0, 7.5, 15 kg N ha-1 yr-1, but after eight years of N addition there was an increase in the biomass of bryophytes (Phoenix et al., 2011), although species richness of higher plants was not impacted. Subsequent research suggests that the fixed dune grassland at this site is N and P co-limited (Ford et al. 2016), which may explain the lack of treatment effects. In a modelling study at the same site, Rowe et al. (2011) showed that it might take many decades to achieve species change in response to increased N deposition. In an ammonia exposure experiment conducted using sand dune mesocosms of a calcareous fixed dune grassland vegetation in a gradient away from a poultry unit Dactylis glomerata, Festuca rubra and Plantago lanceolata all responded positively. F. rubra showed significant increases in biomass as did D. glomerata and P. lanceolata which came to dominate mesocosms (Jones et al. 2013). In a different component of the same study, Mohd-Said (1999) also reported increases in these species with N applied at 10 kg N ha-1 yr-1. Other less eutrophic species had lower biomass (Jones et al., 2013).

In dune slacks, there is also some evidence of N effects, although primarily from groundwater nutrients rather than atmospheric N deposition. Although not an atmospheric-focused study, Rhymes et al. (2014) found Trifolium repens was positively associated with elevated N in ground water at Aberffraw dune system in North Wales. Lotus corniculatus, which was not identified as an indicator using this method, was also positively associated with elevated N in ground water suggesting it may also be a eutrophic indicator in dunes. In older studies, Elytrigia juncea has been reported to increase with fertiliser (NPK) addition (Adriani and Terwindt, 1974, Willis, 1965; Pavlik, 1983; Fay and Jeffrey, 1992) together with Sonchus asper, F. rubra, Poa pratensis (Willis, 1965), Crepis capillaris, Geranium molle and Senecio jacobaea (Willis and Yemm, 1961). Changes in the seedbank of dune slacks were observed in a germination study treated with elevated N (Plassmann et al. 2008). Anagallis tenella, Carex viridula, Centaurium littorale, Juncus articulatus, Parnassia palustris and the glaucous sedges (Carex flacca, Carex panacea) all increased their germination success in response to N addition which suggests there could potentially be future changes in their abundance but none of these species have been identified as eutrophic indicators. Plassmann et al. (2008) suggested that germination was enhanced by N additions in these species, but N impacts on subsequent growth of the plants was not studied.

The relative importance of different sources of N to dune slacks has been little studied, but is highly relevant to wetland habitats which could be receiving N from multiple sources in addition to atmospheric N deposition (Rhymes et al. 2015). Further work is required to establish whether the source of N causes differences in ecological impact. A few commissioned studies have started to explore these issues (e.g. Farr & Hall 2014; Farr et al. 2017).

Many of the species identified as eutrophic indicators for sand dune communities in this study have been identified as responding positively to N addition or deposition in other communities. For example, Agrostis capillaris, Festuca rubra and Holcus lanatus were among species identified as increasing with higher N deposition over time in acid grasslands (Dupre et al. 2010). The bryophyte Brachythecium rutabulum has been identified as increasing over time in a period of high N

17 deposition in chalk grassland in The Netherlands (During and Willems, 1986), and was identified as a positive responder to N by Field et al. (2014) in a gradient study on heathlands.

There are also responses that have been observed which are not picked up by these indicators. In a survey of 11 dune systems along a gradient of deposition Jones et al. (2004) found an increase in the height and cover of Ammophila arenaria in mobile and semi-fixed dunes and an increase in cover of Carex arenaria and Hypochaeris radicata. Increases in C. arenaria have also been observed in field and mesocosm studies in dune systems in the Netherlands (e.g. Remke et al. 2009a; Remke et al. 2009b; van den Berg et al., 2005; Adriani and Terwindt, 1974). It is not currently planned to add these species to the indicator lists, but there may be scope to revise them and include other species known to respond to N, similar to the approach followed by Pitcairn et al. (2006).

Testing of the indicators for sand dunes against data from gradient surveys has shown they have some promise in revealing eutrophication signals in the data. However, there is considerable scatter in the relationships and the datasets used to test them so far are quite small when broken down to smaller groups of communities. The factors underlying scatter in the relationships are unclear. Within and across sites, this may include variation in soil pH or organic matter content. It may also include additional drivers of vegetation composition such as climate, hydrology, and site management practices such as grazing intensity. Figure 5 in particular illustrates the high variability in indicator values within a site. This would be expected to be greater for dune slacks than other habitats due to the high variability in hydrological regime, groundwater chemistry, degree of soil development and soil pH in dune slacks of different ages and typologies (Curreli et al. 2013; Davy et al. 2010).

The different habitat types appear to have different sensitivities to nitrogen, reflected in the maximum observed indicator values on the y axis. The relatively low values for the semi-fixed and blow-out communities partly reflect the low total vegetation cover in these habitats where there is still considerable bare-sand present. However, the reason for differences among the other habitats is unclear, and may simply be a result of the relatively small datasets used so far in testing the indicators.

6 Conclusions From the testing so far, the approach seems to work best for acidic dune grasslands, with some indication of a useful relationship at a site level for dune slacks. Until recently, there had been no comprehensive testing of the Pitcairn et al. (2006) indicators approach, apart from the single application to a highly eutrophic gradient away from a poultry unit in their report. This study therefore is one of the first applications against observed N gradients at large scale and in the low- to-middle range of N deposition.

The companion project (RAD031-Botanical Benchmarks) (Jones et al. 2017) also explored benchmarking of a nitrophile index and a suite of other indicators derived from botanical data across a range of habitats, including dune habitats. In the botanical benchmarks study the only dune habitat tested was the acidic dune grassland and annuals, using a wider dataset including sites from the Scottish sand dune survey and CEH data. The results for the nitrophile index showed

18 considerable scatter, with no clear pattern, contrasting to that reported here. The number of acidic dune sites in the UK is rather limited, and full testing of this habitat may need to consider including other sites from the continent to incorporate areas with both lower and higher N deposition. By comparison, the nitrophile-nitrophobe index proved to be a useful metric in other habitats (acid grassland, dry heaths, and wet heaths and bogs).

The reason for this is unclear. It may be that a more nuanced indicator which takes account of known responses to nitrogen, as developed in Pitcairn et al. (2006), would work better than the more objective approach taken here. However, other factors may also play a part including the much greater heterogeneity in dune habitats within and across sites than is typically found in other habitat types. For example, the site-specific data using a nitrogen-gradient within Aberffraw dune slacks in Wales shows that the dune slack metric shows some promise.

Therefore, further testing of the nitrophile indicator for the other dune habitats against larger datasets, and in other situations would help evaluate their usefulness as an indicator of eutrophication. Modification of the acidic dune grasslands species list to include lichens may add additional sensitivity to N responses in this habitat and should be considered.

7 References Adriani, M.J., Terwindt, J.H.J., 1974. Sand stabilisation and dune building, Rijkswaterstaat Communications No. 19, The Hague.

Curreli A., Wallace H., Freeman C., Hollingham M., Stratford C., Johnson H., Jones L. (2013). Eco- hydrological requirements of dune slack vegetation and the implications of climate change. Science of the Total Environment 443, 910-919.

Davy, A.J.; Hiscock, K.M.; Jones, M.L.M.; Low, R.; Robins, N.S.; Stratford, C.. 2010. Protecting the plant communities and rare species of dune wetland systems: ecohydrological guidelines for wet dune habitats. Phase 2. Bristol, UK, Environment Agency, 113pp.

Dupre, C., Stevens, C.J., Ranke, T., Bleeker, A., Peppler-Lisbach, C., Gowing, D.J.G., Dise, N.B., Dorland, E.D.U., Bobbink, R., Diekmann, M., 2010. Changes in species richness and composition in European acidic grasslands over the past 70 years: the contribution of cumulative atmospheric nitrogen deposition. Global Change Biology 16, 344-357.

During, H.J. and Willems, J.H. The impoverishment of the bryophyte and lichen flora of the Dutch chalk grasslands in the thirty years 1953-1983. Biological Conservation 36 (1986) 143-158.

Farr, G & Hall, J. 2014. Atmospheric deposition at groundwater dependent wetlands: implications for effective catchment management and Water Framework Directive groundwater classification in England and Wales. Nottingham, UK, British Geological Survey, 62pp. (OR/14/047)

Farr, G., Hall, J., Jones, L., Whiteman, M., Haslam, A., Philips, N. 2017. Source apportionment of nutrient pressures at groundwater dependent wetlands: Case studies from England and Wales. British Geological Survey, OR/17/021. 84pp.

Fay, P.J., Jeffrey, D.W., 1992. The foreshore as a nitrogen source for marram grass. In: Carter, R.W.G., Curtis, T.G.F., Sheehy-Skeffington, M.J. (Eds.), Coastal dunes. Geomorphology, Ecology and Management for Conservation. Proceedings of the third European dune congress. Galway/Ireland. Balkema, Rotterdam, Galway.

Field, C.D., Dise, N.B., Payne, R.J., Britton, A.J., Emmett, B.A., Helliwell, R.C., Hughes, S., Jones, L., Lees, S., Leake, J.R., Leith, I.D., Phoenix, G.K., Power, S.A., Sheppard, L.J., Southon, G.E., Stevens, C.J., Caporn, S.J.M., 2014. The role of nitrogen deposition in widespread plant community change across semi-natural habitats. Ecosystems 17, 864-877.

Ford, H., Roberts, A., Jones, L. (2016). Phosphorus - nitrogen co-limitation and grazing moderate nitrogen impacts on plant growth and nutrient cycling in sand dune grassland. Science of the Total Environment 542A, 203-209.

Hill, M.O., Mountford, J.O., Roy, D.B., Bunce, R.G.H., 1999. Ellenberg's indicator values for British plants. ECOFACT Volume 2 technical annex. Institute of Terrestrial Ecology, Huntingdon.

Jones, L. (2007). Model based risk assessment of the vulnerability of rare coastal species to N deposition. In: Emmett B.A. (ed) DEFRA Terrestrial Umbrella Final Report 2007. Contract No: EPG 1/3/186.

Jones, L., Nizam, M., Reynolds, B., Bareham, S., Oxley, E., 2013. Upwind impacts of ammonia from an intensive poultry unit. Environmental Pollution 180, 221-228.

Jones, M.L.M., Reynolds, B., Stevens, P.A., Norris, D., Emmett, B.A., 2002. Changing nutrient budgets of sand dunes: Consequences for the nature conservation interest and dune management. 1. A review. Contract Report March 2002. CCW Contract No: FC 73-01-347. CEH Project No: C01919. Centre for Ecology and Hydrology, Bangor, pp. 1-84.

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

Jones, L., Banin, L.F., Bealey, B., Field, C., Caporn, S.J.M., Payne, R., Stevens, C., Rowe, E., Britton, A., Mitchell, R., Pakeman, R., Dise, N., Robinson, E., Tomlinson, S. (2017). Botanical Benchmarks: application of single assessment site-based vegetation survey data in Habitats Regulations Assessment for regulatory decision-making. Scottish Environment Protection Agency Commissioned Report No. DK1605

Mohd-Said, M.N., 1999. Effects of anthropogenic nitrogen inputs on dune grassland. PhD Thesis Thesis, University of Wales, Bangor.

Pakeman, R.J., Alexander, J., Brooker, R., Cummins, R., Fielding, D., Gore, S., Hewison, R., Mitchell, R., Moore, E., Orford, K., Pemberton, C., Trinder, C., Lewis, R., 2016. Long-term impacts of nitrogen deposition on coastal plant communities. Environmental Pollution 212, 337-347.

Pavlik, B.M., 1983. Nutrient and productivity relations of the dune grasses Ammophila-arenaria and Elymus-mollis. 2. Growth and patterns of dry-matter and nitrogen allocation as influenced by nitrogen supply. Oecologia 57 (1-2), 233-238.

Phoenix GK, Emmett BA, Britton AJ, Caporn SJM, Dise NB, Helliwell R, Jones L, Leake JR, Leith ID, Sheppard LJ, Sowerby A, Pilkington MG, Rowe EC, Ashmore MR, Power SA (2012). Impacts of atmospheric nitrogen deposition: responses of multiple plant and soil parameters across contrasting ecosystems in long-term field experiments. Global Change Biology 18(4), 1197-1215.

Pitcairn, C.E.R., Leith, I.D., Sheppard, L. and Sutton, M.A. 2006. Development of a nitrophobe/nitrophile classification for woodlands, grasslands and upland vegetation in Scotland.

Plassmann, K., Brown, N., Jones, M.L.M., & Edwards-Jones, G. (2008). Can atmospheric input of nitrogen affect seed bank dynamics in habitats of conservation interest? The case of dune slacks. Applied Vegetation Science 11(3): 413-420.

Plassmann, K., Edwards-Jones, G., Jones, M.L.M., 2009. The effects of low levels of nitrogen deposition and grazing on dune grassland. Science of the Total Environment 407, 1391-1404.

Remke, E., Brouwer, E., Kooijman, A., Blindow, A., Roelofs, J.G.M., 2009a. Low atmospheric nitrogen loads lead to grass encroachment in coastal dunes, but only on acid soils. Ecosystems 12, 1173-1188.

Remke, E., Brouwer, E., Kooijman, A.M., Blindow, I., Esselink, H., Roelefs, J.G.M., 2009b. Even low to medium nitrogen deposition impacts vegetation of dry, coastal dunes around the Baltic Sea. Environmental Pollution 157, 792-800.

Rhymes, J., Wallace, H., Fenner, N., Jones, L., 2014. Evidence for sensitivity of dune wetlands to groundwater nutrients. Science of the Total Environment 490, 106-113.

Rhymes J., Jones L., Lapworth, D.J., White, D., Fenner, N., McDonald, J.E., Perkins, T.L. (2015). Using chemical, microbial and fluorescence techniques to understand contaminant sources and pathways to wetlands in a conservation site. Science of the Total Environment 511, 703-711.

Rodwell, J.S., 2000. British Plant Communities. Volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge.

Rowe, E.C., Jones M.L.M., Henrys P.A., Smart S.M., Tipping E., Mills R.T.E. &. Evans C. (2011). Predicting effects of N pollutant load on plant species based on a dynamic soil eutrophication indicator. Report to CCW, March 2011.

Siebel, H.N., 1993. Indicatiegetallen van bladen levermossen, IBN-rapport 047, Wageningen. van den Berg, L.J.L., Tomassen, H.B.M., Roelofs, J.G.M., Bobbink, R., 2005. Effects of nitrogen enrichment on coastal dune grassland: a mesocosm study. Environmental Pollution 138, 77-85.

Willis, A.J., 1965. The influence of mineral nutrients on the growth of Ammophila arenaria. Journal of Ecology 53, 735-745.

Willis, A.J., Yemm, E.W., 1961. Braunton Burrows: Mineral nutrient status of the dune soils. Journal of Ecology 49, 377-390.