Biological Invasions

, Volume 14, Issue 3, pp 567–577

Humans introduce viable seeds to the Arctic on footwear

Authors

  • Chris Ware
    • The University Centre in Svalbard (UNIS)
  • Dana M. Bergstrom
    • Australian Antarctic Division, Department of the Environment, Water, Heritage and the Arts
  • Eike Müller
    • The University Centre in Svalbard (UNIS)
    • The University Centre in Svalbard (UNIS)
    • Tromsø University Museum
Original Paper

DOI: 10.1007/s10530-011-0098-4

Cite this article as:
Ware, C., Bergstrom, D.M., Müller, E. et al. Biol Invasions (2012) 14: 567. doi:10.1007/s10530-011-0098-4

Abstract

Expanding visitation to Polar regions combined with climate warming increases the potential for alien species introduction and establishment. We quantified vascular plant propagule pressure associated with different groups of travelers to the high-Arctic archipelago of Svalbard, and evaluated the potential of introduced seeds to germinate under the most favorable average Svalbard soil temperature (10°C). We sampled the footwear of 259 travelers arriving by air to Svalbard during the summer of 2008, recording 1,019 seeds: a mean of 3.9 (±0.8) seeds per traveler. Assuming the seed influx is representative for the whole year, we estimate a yearly seed load of around 270,000 by this vector alone. Seeds of 53 species were identified from 17 families, with Poaceae having both highest diversity and number of seeds. Eight of the families identified are among those most invasive worldwide, while the majority of the species identified were non-native to Svalbard. The number of seeds was highest on footwear that had been used in forested and alpine areas in the 3 months prior to traveling to Svalbard, and increased with the amount of soil affixed to footwear. In total, 26% of the collected seeds germinated under simulated Svalbard conditions. Our results demonstrate high propagule transport through aviation to highly visited cold-climate regions and isolated islands is occurring. Alien species establishment is expected to increase with climate change, particularly in high latitude regions, making the need for regional management considerations a priority.

Keywords

AlienNon-indigenousDispersalGerminationHuman-mediated dispersalPropagule pressure

Introduction

Until recently in the high-Arctic and Antarctic, two processes have maintained ecological integrity: low frequency of human-mediated dispersal, and the prevailing climate, both of which are rapidly changing (Convey et al. 2006; Elven et al. 2011). Seed dispersal by humans and cargo is to some degree documented for the Antarctic (Whinam et al. 2005; Frenot et al. 2005; Lee and Chown 2009a, b; SCAR 2010), and has been the subject of management development by the Antarctic Treaty Parties (Australia and SCAR 2007). In contrast, no such quantification exists for the Arctic, where few biosecurity measures are currently employed.

The total number of archaeophytic, persistent, and transient alien plants in the Arctic is low, constituting a very low to zero proportion of the regional Arctic floras (Elven et al. 2011). Exceptions exist in some of the millennium-old Viking settlements and more recent Russian settlements where non-indigenous plants are more prevalent. In some Arctic settlements, sometimes very far to the north, casual introductions are quite frequent (Liška and Soldán 2004; Elven et al. 2011). For other taxa, records of established alien species exist, although these are also few. A vole is known to be established on Svalbard (Fredga et al. 1990) while the first records of a non-native crustacean and species of kelp have been made at lower Arctic latitudes (Ashton et al. 2008).

Elsewhere in the world, humans and their associated activities have been demonstrated to be effective vectors of unintentional species transfer, providing carriage for plants (Clifford 1956; Powell 1968; Falinski 1972; Higashino et al. 1983; Whinam et al. 2005; Lee and Chown 2009a; Wichmann et al. 2009), arthropods (McCullough et al. 2006; Hughes et al. 2010), bacteria (Curry et al. 2002; Drake et al. 2007; Hughes et al. 2010; McNeill et al. 2011), terrestrial vertebrates (Gillespie 1985), and marine organisms (Carlton 1985; Gollasch 2002). Generally, the little amount of research concerning invasion processes in the Arctic is biased towards post-invasion—a trend identified globally (Puth and Post 2005). To date, the only attempt to quantify the significance of a pathway of species introduction to the Arctic focused on ship-mediated introductions to Alaska (Hines and Ruiz 2000).

Human activity in the Arctic has rapidly increased over the past 40 years (Kaltenborn 2000; Forbes et al. 2004). Between 1995 and 2004 there was a 255% increase in the number of tourists visiting Svalbard (Governor of Svalbard 2006), while Greenland recorded a 500% increase over the same time period (Statistics Greenland 2009). While the tourism sector is increasing rapidly, so too are other travel sectors such as that associated with science. Polar scientists often visit and work in several alpine or high latitude environments, and may move frequently between them (e.g. Whinam et al. 2005), increasing the chances of introducing organisms pre-adapted to Arctic environmental conditions.

While dispersal is a critical step in species invasions, the Arctic climate also presents a significant barrier to species colonization (Alsos et al. 2007). The effect of climate on new species colonization is complex, and varies at different stages of colonization (Shevtsova et al. 2009). For plants, the initial bottleneck of colonization may be germination. Low temperatures have been shown to limit germination in Arctic plant species (Sørensen 1941, Müller et al. 2011) suggesting also that germination of introduced alien plant species would be similarly impaired. Despite this, many temperate grassland, shrub and herbaceous species have been shown experimentally to be capable of germination at surface temperatures commonly recorded in the Arctic today (Baskin and Baskin 1998; Trudgill et al. 2000). Indeed, Arctic summer surface temperatures can be several degrees warmer than those reported from meteorological stations recorded at two meters above the surface (Scherrer and Körner 2010). Furthermore, it is possible that seeds introduced today, capable of lying dormant in soil for many years (Thompson et al. 1997), may be capable of germination under future climates.

Alien plants are widely documented at higher latitudes (Alaska, sub-Antarctic), and are relatively easy to monitor and identify (compared to e.g. bacteria, arthropods, fungi). For these reasons, and due to the incomplete knowledge of other taxonomic groups in the Arctic (Coulson and Refseth 2004; Elvebakk and Prestrud 1996; Alsos et al. 2009), plant seeds make appropriate exemplars to investigate the extent to which new species could be transported to, and survive in, Arctic regions. Humans can carry a high plant propagule load on footwear (Clifford 1956; Powell 1968; Falinski 1972; Higashino 1983; Whinam et al. 2005; Lee and Chown 2009a; Wichmann et al. 2009; McNeill et al. 2011), from which seeds can disperse (Lee and Chown 2009a; Wichmann et al. 2009; Pickering and Mount 2010; Ware and Bergstrom, unpublished data), and as such, humans are likely to introduce alien seeds while traveling. Here, by using footwear as a pathway of introduction, we investigated the threat of species transfer to Svalbard. We asked the following questions: (1) What is the size and composition of the seed load being carried to Svalbard on travelers’ footwear? (2) What factors explain the number of seeds on footwear? (3) Could seeds transported to Svalbard on footwear germinate under current Svalbard conditions? Based on the results of these investigations, implications for management are discussed.

Methods

Location

Around 60% of the Svalbard archipelago (74°–81°N and 10°–30°E) is covered in ice, leaving coastal pockets suitable for vegetation (Jónsdóttir 2005). Three of the five Arctic bioclimatic subzones identified by Walker et al. (2005) are present in Svalbard: polar desert (subzone A), northern Arctic tundra (subzone B) and middle Arctic tundra (subzone C). Mean air temperatures of the warmest month in each zone respectively are between 1–2.5°C, 2.5–4°C, and 4–6°C (Elvebakk 2005). There are 165 native plant species (Alsos et al. 2011). Around 60 non-indigenous plant species have been recorded around the main settlements (Liška and Soldán 2004), of which around 28–37 are established or are frequently re-introduced (Elven and Elvebakk 1996; Elven et al. 2011).

Most travelers to Svalbard travel by plane, arriving at the major airport located in Longyearbyen, on the island of Spitsbergen. During 2008, 68,901 travelers flew into Longyearbyen (Governor of Svalbard, personal communication) with more than 90 % of these typically arriving over the tourist season (March–September) (Governor of Svalbard 2006). Many travelers join expedition ships at the local port, exploring the archipelago by ship.

Footwear sampling

We sampled the footwear of 259 travelers arriving at Svalbard Airport between 20 June and 28 September 2008. Around 75% of travelers arriving in Svalbard were wearing footwear with soles capable of carrying substantive quantities of soil, such as those typical of hiking/running shoes (Ware, unpublished data); only these travelers were asked to participate in the survey. We scraped off any soil attached to participants’ footwear using a stiff-bristled brush and forceps, scrubbed the shoe sole, and inspected the shoe lacing and tongue for biological material. Footwear was cleaned until all visible material was removed, and material was collected in plastic bags. A sampling unit was considered as a pair of shoes and in instances where travelers arrived with two pairs of shoes (i.e. one pair in their luggage) the second pair was considered a separate sample. In between samples, sampling equipment was cleaned thoroughly and visually inspected for dirt and propagules so as to avoid sample contamination. Samples were tagged with unique identifiers. We sorted samples into the following categories with the aid of a dissecting microscope (3×): seeds; plant and invertebrate fragments (bryophyte fragments, leaves, macroscopic invertebrate parts); soil (organic material); and non-organic material (highly variable, but commonly including metal and plastic fragments, chewing gum, and feathers). Total seed and bryophyte fragment numbers were tallied. Seeds were identified to the lowest taxonomic group possible using an identification guide (Cappers et al. 2006) and online resources (Kirkbride et al. 2006). Families were crosschecked against the most invasive families listed within the Global Invasive Species Database (ISSG 2010). While bryophyte fragments collected may have been capable of vegetative growth, we excluded these from substantive analysis due to the difficulties associated with their identification (e.g. Rowntree et al. 2010). We considered a focus on vascular plants to be a priority, owing to their significance in the global invasive flora (e.g. ISSG 2010). Soil samples were stored at −20° prior to sample sorting, and weighed following sorting using Metter Toledo scales.

Germination

We placed collected seeds on filter paper (grade 1, Whatman, Maidstone, UK) moistened with distilled water via a wick attached to a reservoir. Seeds were then kept in a phytotron chamber at 10°C, under 24-h light (approximately 40 μmol m−2 s−1 at seed surface, 35 W fluorescent tube, 840 HE (Osram, Munich, Germany)) to simulate ambient average summer soil conditions in Svalbard. A temperature of 10°C was selected as this best reflects average soil surface records from a number of favorable Svalbard sites; the temperature was fixed as this reflects the relatively low standard temperature deviation recorded from these sites (Müller, unpublished data). Seeds were monitored for germination (protrusion of a radical) for 48 days.

Traveler statistics

Participants also completed a questionnaire (linked to their footwear sample using a unique identifier), categorizing themselves as a tourist, scientist, businessperson, resident, or student. Furthermore, they indicated whether, and when, they had last cleaned their footwear; whether they had used their footwear in the 3 months prior to traveling; and in what type of habitat they had used their footwear (forested, alpine, rural, or urban areas) over the three previous months.

Statistical analysis

To determine the relationship between the weight of soil collected from footwear and the number of seeds found, we fitted a generalized linear model (GLM) with a quasipoisson error distribution and logarithmic link function, and an overdispersion characteristic. From this, we also calculated seeds per 1 g of soil. To test for correlation between seed load and the two explanatory variables, traveler categories and previous footwear use, we fitted GLMs separately. Again, we fitted the GLMs using a quasipoisson error distribution with a logarithmic link function and an overdispersion characteristic. We began by fitting maximum models containing all predictor variables and interaction terms. Model simplification was then achieved by removing variables and interaction terms stepwise. Model fit was assessed using analysis of variance (ANOVA) tests to determine whether simplified models significantly increased deviance; the significance of difference was assessed using F tests. Where deviance was not increased (i.e. P < 0.05), variables or interaction terms were omitted from further modeling. To determine the effectiveness of traveler footwear cleaning prior to travel, we used the Wilcoxon rank sum test with continuity correction. For all mean values calculated in our analyses, standard errors were also calculated (±SE).We used the statistical and programming package R (version 2.10.0, R Development Core Team 2008) to carry out all analyses.

Results

Footwear samples

Overall, 40% of the 259 footwear samples examined contained seeds. A total of 1,019 seeds were collected representing a mean of 3.9 (±0.8) seeds per traveler, or 9.9 (±1.1) seeds per traveler that had seeds attached to their footwear. The maximum number of seeds found in a single sample was 117 (Table 1), and 26 samples (10%) contained 10 or more seeds. In addition, we also found 465 bryophyte fragments in the samples representing a mean of 1.8 (±0.6) fragments per traveler. A mean of 0.27 g (±0.06) of soil was found on a pair of footwear, with a range: 0–9.9 g (35% of footwear did not contain any soil). The amount of soil present on footwear was significantly correlated with the number of seeds present (F = 165, P < 0.05, 166 df). Where soil was present in a sample, there was an average of 2.9 ± 1.2 seeds per gram of soil. We did not find any live invertebrates, eggs, or larvae in the samples.
Table 1

Summary of footwear samples, and survey information collected from travelers arriving to Svalbard

Traveler category

n

% Contaminated

% With soil

Total seeds

Mean seeds

Bootstrapped 95% CI

Max. seeds per sample

% Cleaned

Tourist

170

41

69

631

3.71

2.3, 6.6

117

21

Scientist

37

57

65

212

5.73

2.8, 12.0

62

19

Student

28

36

46

98

3.50

1.1, 9.1

39

21

Business

19

47

63

59

3.11

1.0, 7.7

25

10

Resident

5

20

40

19

3.80

0.0, 7.6

19

0

Total

259

48

65

1,019

3.93

2.8, 6.1

20

CI confidence interval

The majority of the identified seeds collected were grasses (60%), with 17 Poaceae species identified (Table 2). Other seeds present were tree, herb and sedge seeds, with a proportion (13%) unable to be identified (Fig. 1). Four of the herb species and three grass species identified from our samples have already established as alien species in Svalbard. Only two possible native species were found (Table 2).
Table 2

Number of seeds of native or alien species found on traveler’s footwear arriving in Svalbard, and seed germination percentages

Group and family

Taxa

Native

Alien

Unidentified species

Life form

% Germination

Gymnosperm

      

Cupressaceae

Thuja plicata

 

1

 

P

 

Angiosperms—dicotyledons

      

Apiaceae

Torilis japonica

 

1

 

A

 

Asteraceaea

Unidentified

  

1

  

Taraxacum sp.b

 

9

 

P

100

Betulaceae

Betula pubescens

 

143

 

P

1

Brassicaceaea

Unidentified

  

1

  

Erucrastrum sp.

 

1

   

Isatis sp.

 

7

   

Nasturtium microphyllum

 

23

 

P

9

Caryophyllaceae

Unidentified

     

Cerastium brachypetalum

 

8

 

A

100

Cerastium glomeratum

 

12

 

A

100

Ericaceae

Vaccinium sp.

  

5

  

Fabaceaea

Astragalus glycyphyllos

 

1

 

P

 

Medicago falcata

 

1

 

A/P

 

Papavaceaea

Unidentified

  

1

  

Papaver sp.

  

3

  

Plataginaceae

Plantago major

 

27

 

P

7

Polygonaceae

Unidentified

  

4

 

100

Polygonum aviculare

 

3

 

A/P

 

Rumex sp.

  

1

  

Rumex crispus

 

1

 

P

 

Ranunculaceaea

Ranunculus sp.

  

1

  

Ranunculus acris

 

5

 

P

 

Rosaceaea

Geum macrophyllum

 

1

 

P

 

Geum rivale

 

1

 

P

 

Angiosperms—monocotyledons

      

Cyperaceae

Carex sp.

  

14

  

Carexacutiformis

 

1

 

P

 

Juncaceaea

Unidentified

  

3

  

Juncus sp.

  

6

  

Juncus effuses

 

10

 

P

60

Juncus pygmaeus

 

22

 

A

 

Juncaginaceae

Triglochin maritima

 

20

 

P

 

Poaceaea

Unidentified

  

73

 

4

Agrostis sp.

  

1

 

100

Agrostis stolonifera

 

24

 

P

38

Alopecurus pratensis

 

6

 

P

50

Ammophila arenaria

 

2

 

P

50

Bromopsis sp.

 

1

   

Bromus sp.

 

2

   

Bromus hordeaceus

 

4

 

A

75

Calamagrostis pseudophragmites

 

2

 

P

 

Dactylis glomerata

 

2

 

P

100

Deschampsia sp.

  

1

  

Deschampisa caespitosa

 

3

 

P

 

Avenella flexuosa

 

89

 

P

45

Festuca sp.

  

21

 

5

Festuca lemanii

 

1

 

P

100

Festuca rubra

1

  

P

 

Holcus lanatus

 

1

 

P

100

Hordeum sp.

  

2

 

50

Lolium perenne

 

2

 

A/P

50

Phleum pratense ssp. pratense

 

2

 

P

 

Phleum pratense ssp. serotinum

 

1

 

P

 

Poa sp.

  

41

 

39

Poa annua

 

36

 

A

58

Poa trivialis

 

180

 

P

57

Poa pratensis

62

  

P

10

Trisetum flavescens

 

3

 

P

 

Unidentified seeds

   

118

 

7

Totals

 

63

663

293

 

26

Numbers in bold indicate those alien species that have already established on Svalbard. Life form: perennial (P), annual (A) or both (A/P) according to www.plants.usda.gov

aFamilies that are those identified as the most invasive worldwide (as per Pyšek 1998) or families with a high number of invasive species (ISSG 2010)

bTaraxacum sp. are not currently listed as a Svalbard alien, but are present in the Russian settlement of Pyramiden (Ware, pers. obs.) and Barentsburg (Alsos, pers. obs.)

https://static-content.springer.com/image/art%3A10.1007%2Fs10530-011-0098-4/MediaObjects/10530_2011_98_Fig1_HTML.gif
Fig. 1

Percentage of total propagules by growth form collected from the footwear of travelers to Svalbard

Germination

Of the total 1,019 seeds tested for germination, 266 (26%) germinated under the test conditions. Taraxacum sp. (n = 9), Cerastium brachypetalum (n = 8), C. glomeratum (n = 12), and Dactylis glomerate (n = 2) all recorded 100% germination, while Deschampsia flexuosa, Poa annua and P. trivialis all recorded germination >40% (Table 2).

Traveler statistics

Category of traveler arriving to Svalbard had no significant effect on the number of seeds imported on footwear (Table 1). Twenty percent of participants reported cleaning their footwear prior to travel: of this percentage, 49% contained seeds. There was no evidence that footwear cleaning by participants lowered the number of seeds transported on shoes (W = 5,615.5, P > 0.05).

Fifty seven per cent of the participants had used their footwear in forests, while 30% had used their footwear in alpine regions (Fig. 2). A GLM that included use of footwear in both forested and alpine areas provided the best fit to the data (Table 3), demonstrating that footwear previously used in these two habitats contained a significantly higher seed load (F = 11.06, P = 0.001, 257 df).
https://static-content.springer.com/image/art%3A10.1007%2Fs10530-011-0098-4/MediaObjects/10530_2011_98_Fig2_HTML.gif
Fig. 2

Percentage of total seeds by previous use collected from the footwear of travelers to Svalbard

Table 3

The results of ANOVA tests comparing different generalized linear models investigating the effects of where footwear had been previously used (forest, alpine, rural or urban habitats) on the number of seeds affixed to footwear

Model variables

Deviance residuals (max–min)

Deviance on df

∆ Deviance

P

Seeds = forest × alpine × rural × other

(−5.0398 to 2.8748)

2,932.4 on 244

Seeds = forest + alpine + rural + other

(−4.249 to 2.876)

3,294.4 on 254

−362.00

0.1325

Seeds = forest + alpine + rural

(−4.076 to 2.892)

3,314.6 on 255

−20.2

0.4263

Seeds = forest + alpine

(−3.8360 to 3.1327)

3,331.0 on 256

−16.4

0.4785

Seeds = forest

(−3.423 to 3.423)

3,370.6 on 257

−39.6

0.272

Discussion

This study demonstrates that people arriving in Svalbard pose an identifiable hazard to the local environment through the introduction of alien plant seeds that are capable of germination even under current climatic conditions. Travelers are providing the means to increase the plant species pool capable of reaching the Arctic, a trend identified already in the sub-Arctic (Carlson and Shephard 2007). The seed load per person transferred to the Arctic is similar to that being introduced by expeditioners to the Antarctic, and the same types of species are being transported (Lee and Chown 2009a). Our findings support those of others demonstrating that humans are capable of translocating many of the world’s widespread alien plant species (Pickering and Mount 2010).

Our analysis demonstrated that footwear previously used in forested or alpine areas carried significantly higher numbers of seed than that used in rural or urban areas. Few studies have attempted to investigate previous use as a factor predisposing an item to contamination. McNeill et al. (2011) found golfing footwear to be the most highly contaminated item in a study of the footwear of arriving airplane passengers in New Zealand, while Whinam et al. (2005) found many Antarctic expeditioners had recently used their clothing in natural environments. The positive relationship between outdoor use and clothing and equipment contamination is logical, and our study reaffirms the notion that these items provide the greatest biosecurity hazard.

The strong association between the presence of soil and incidence of seeds and bryophyte fragments is consistent with other studies (Hughes et al. 2010; McNeill et al. 2011), and highlights the potential for any clothing and equipment capable of carrying soil to mediate alien organism introduction. The mean number of seeds found per gram of soil reported here (2.9 ± 1.2) is comparable to that found in soil attached to the footwear of arriving aircraft passengers to New Zealand (2.5 ± 0.37 per 1 g soil—McNeill et al. 2011). As with McNeill et al. (2011), our seed counts may be underestimated owing to imperfect visual searches in our samples, while counts would be slightly inflated by the few occasions where seeds were found in the absence of soil (e.g. on footwear lacing or tongue).

The sampled seed load contained a number of cosmopolitan species, and eight of the 17 families identified belonged to those families ranked as most invasive at a global scale (Pyšek 1998). Considering the dominance of Poaceae seeds found in connection with other human-mediated seed dispersal studies (e.g. Schmidt 1989; Hodkinson and Thompson 1997; Lee and Chown 2009a), and the wide geographic range of establishment that some Poaceae species have achieved (i.e. Poa annua and Poa trivialis have both established in the Antarctic—Frenot et al. 2005; Hughes et al. 2010), the finding that Poaceae seeds dominated our samples was not unexpected.

Our relatively high germination rates indicate that germination may not be a barrier to establishment in Svalbard for many non-indigenous species. Germination occurred rapidly under the test conditions, with 87% of those that germinated doing so within 14 days, and the remainder within 48 days—well within the growing season. These germination results are based on present climatic means; however, temperature increases of 0.61°C per decade are expected for the period 1961–2050 (Hanssen-Bauer 2002) which would likely favour the germination of more northerly plants if introduced to Svalbard (e.g. Trudgill et al. 2000; Milbau et al. 2009).

Further improving the chances of successful establishment are the few samples that contained many propagules. One sample contained over 100 seeds, and 26 samples had 10 or more seeds (10%). If these were dispersed into suitable habitat, the effects of propagule pressure would increase the likelihood of their successful establishment (i.e. Williamson 1996; Lockwood et al. 2005, 2007; Colautti et al. 2006). Similarly, where seeds are pre-adapted to the Svalbard climate, the potential for establishment is greater. Many participants had used their footwear in either northern boreal forested regions or alpine regions, and many of the species identified from shoe samples are found in these habitats. While we would expect these species to be better adapted to the challenges of establishing in Svalbard, establishment of other more generalist species cannot be precluded. Indeed, Barbarea vulgaris ssp. arcuata and species of the Tarxacumruderalia aggregate—both generalist European natives—have established on Svalbard and are spreading locally (Alsos, pers. obs.).

As our data were collected over a summer period, they were not suited to testing the effects of seasonality on seed load (no data were collected during winter or spring). There was however, a large variation in seed loading (see CIs—Table 1), and our models failed to account for parts of this variation (Table 2). Factors such as footwear use at times of seed production and dispersal may then be important considerations in more precisely modeling human-mediated seed influx to Svalbard.

While recognizing the above caveat, it is possible to project an estimate of a yearly seed load introduced on footwear based on visitor numbers alone. For the year of the study (2008) 68,901 people arrived at the Svalbard airport, equating to an estimated yearly seed load of 270,000 from travelers’ footwear alone. In addition, approximately 30,000 cruise ship passengers land on Svalbard. In a separate preliminary study we sampled the footwear of three hundred cruise ship passengers landing on Svalbard and found just 21 seeds and bryophyte fragments (Ware, unpublished data), suggesting that cruise ship passengers may contribute a smaller propagule load to the region. Many other pathways of species introduction to Svalbard exist, including cargo, planes, scientific equipment and the clothing and personal equipment of travelers (e.g. Whinam et al. 2005; Barnes et al. 2006; Lee and Chown 2009a, b). From these, organisms of other taxa may be transferred. Thus, the total propagule load being introduced to Svalbard would be considerably higher than estimated here.

Management implications

Our study suggests that modern aviation, as the means by which tourism has achieved its rapid increase, has the potential to increase the pressure of plant species introduction to highly visited cold-climate regions and isolated islands globally. Studies elsewhere demonstrate that footwear and the soil attached to it are furthermore capable of carrying a variety of other taxa (e.g. McNeill et al. 2011). While we found no evidence that footwear could transport live invertebrates, eggs, or larva, we did not analyze collected samples for the presence of bacteria or fungi which may have been present. The question of whether alien plants can establish on Svalbard requires further investigation (being dependent on a variety of factors including soil moisture, aspect, and season); however, our germination results, and the artificial ranges achieved by other introduced plants at high latitudes, suggest that a more conservative approach to regional biosecurity need be considered if the ecological and genetic integrity of the local flora is to be maintained. As many other organisms can be transported in association with soil, measures taken to reduce the seed load imported to Svalbard will also reduce the hazard of other organisms being introduced.

Measures to address the introduction of seeds via footwear exist. The ineffectiveness of footwear cleaning by travelers participating in this study suggests that educating travelers of the need to clean footwear prior to arrival may not be effective alone. While the effectiveness of educating Antarctic expeditioners to clean footwear and personal equipment of seeds and contaminants was apparent (Bergstrom, pers. obs.), this may have been due to the combination of follow-up inspections. Beyond education, more stringent management measures could include the adoption of a biosecurity policy at entry points to Svalbard, such as those in place in New Zealand (Biosecurity New Zealand 2010), and for Antarctic tour operators (IAATO 2010). Any management interventions in Svalbard would ideally be positioned within a more comprehensive framework, incorporating all pathways and vectors of introduction, and all organisms, especially pathogens.

Our study makes a case for a more precautionary approach to the management of alien species in Svalbard. The pathway analysis described here suggests that more non-indigenous plant species can be expected in Svalbard if measures to prevent their introduction are not considered. Moreover, our study implies that isolated islands or regions worldwide, which are experiencing similar increases in human traffic as is occurring in Svalbard, may be exposed to similar hazards or even greater hazards if climate is less of a limiting factor.

Acknowledgments

We thank the many people who assisted in footwear sampling at the Svalbard airport, and the 259 participants who kindly gave up their shoes for cleaning after arrival. We would also like to thank the staff at the Svalbard Airport for allowing us to conduct the study on airport premises. Financial assistance for this project was received from the Svalbard Environmental Protection Fund. This project is closely aligned with the IPY Aliens in Antarctica Project. Finally, we wish to extend our thanks to three anonymous reviewers for their comments on the manuscript.

Copyright information

© Springer Science+Business Media B.V. 2011