The species compositions of terrestrial isopod assemblages in 150 public parks in a commuter town in Japan were investigated. Eleven species of terrestrial isopods were present, of which seven were considered native or oriental species, and four were exotic species originally distributed in the Mediterranean and European regions. An exotic species Armadillidium vulgare Latreille, 1804 was found in all parks. Logistic model analysis indicated that the surrounding land use affected the distributions of three native species, Burmoniscus kathmandius (Schmalfuss, 1983), Ligidium koreanum Flasarova, 1972, and Mongoloniscus koreanus Verhoeff, 1930, indicating that landscape properties are important factors that limit the distributions of terrestrial isopods. The present study also showed that the public parks surrounded by forests or semi-natural environments in a commuter town provide habitats for native terrestrial isopods.

GIS, GLM, landscape environments, local environments, logistic distributions

Urbanization has rapidly spread throughout the world and has changed species compositions of regions through decreased diversity of native species (McKinney 2008) and increased numbers of exotics and/or generalist species (Niemelä 1999, Kotze et al. 2011), but the patterns of diversity changes are substantially different depending on biological group and climatic region (Faeth et al. 2011). Thus, understanding and protecting the biodiversity of urban regions is a major ecological concern. Many researchers have contributed to our knowledge of urban biodiversity in recent decades (e.g., McKinney 2008, Jones and Leather 2012, Nielsen et al. 2014), and some suggest that terrestrial isopods are the dominant macro-arthropods in soils of urban regions (Bolger et al. 2000, Smith et al. 2006, Magura et al. 2008a, b, Lee and Kwon 2015). Moreover, it has been observed that urban green spaces in parks provided important habitats that conserve biodiversity in urban environments (Kotze et al. 2011, Barratt et al. 2015). In Japan, many researchers have studied diversity of vegetation, birds, and insects in public parks (e.g., Ishii et al. 1991, Hata et al. 2003, Imai and Nakashizuka 2010, Hattori 2015). However, there has been no comprehensive study of the terrestrial isopods in public parks of Japan.

The Mediterranean region is considered a hotspot of terrestrial isopod diversity (Sfenthourakis and Taiti 2015) and some species that were originally distributed in the Mediterranean and European regions have become dispersed as exotic species throughout the world (e.g., Schmalfuss 2003, Cochard et al. 2010, Hornung et al. 2015). In Japan, seven species that are considered exotic have distribution areas with a potential for expansion (Nunomura 2007, Karasawa and Nakata 2018). These facts suggest that exotic species might already have occupied urban habitats and displaced native species in Japan.

The aims of this study were: 1) to describe the terrestrial isopod fauna in public parks of a commuter town in Japan; 2) to evaluate environments that limit species distributions; and 3) to determine whether public parks in the town provide habitats for native terrestrial isopods.

A total of 17 isopod species was recorded in the city, eleven of which were collected in the public parks. Five species, A. vulgare, Haplophthalmus danicus Budde-Lund, 1880, Porcellio laevis Latreille, 1804, P. scaber Latreille, 1804 and Porcellionides pruinosus (Brandt, 1833), are exotic species in Japan (Karasawa and Nakata 2018). Armadillidium vulgare was found in all parks and has also become the dominant species in the forests. Seven native species were collected in the parks but five of them were found in nine or fewer parks. In contrast, Agnara pannuosa (Nunomura, 1987), Armadillidae sp. 1, Ligidium koreanum Flasarova, 1972, Lucasioides spp. were abundant in the forests. Mongoloniscus koreanus Verhoeff, 1930, M. vannamei (Arcangeli, 1927), P. laevis, and P. scaber were widely distributed in the parks (> 25 parks) but their abundances were low in the forests. Mongoloniscus koreanus was dominant in grassland, as well as A. vulgare (Table 1). Overall, species richness in each park ranged from one to five (Figure 2). At least one exotic species was found in each park, but 80 parks lacked native species. The maximum species richness in a park was four for the native species and three for the exotic species.

Figure 2. 

Frequencies of species richness of native, exotic, and total species in parks

Values of Moran’s I revealed that distribution of P. scaber showed significant spatial autocorrelation (I = 0.027, p < 0.05), while other species and species richness showed no significant spatial autocorrelation. There were differences among species in the spatial scales having the lowest AIC values: A. pannuosa, 25 m; Armadillidae sp. 1, 25 m; B. kathmandius (Schmalfuss, 1983), 50 m; L. koreanum, 500 m; Lucasioides spp., 250 m; M. koreanus, 100 m; M. vannamei, 250 m; P. laevis, 50 m; P. scaber, 500 m; P. pruinosus, 25 m (Supplementary material 1). In addition, the spatial scale of 100 m showed the lowest AIC for the species richness of native and exotic species. Coefficients of variables for the models fitted are described in Table 2. The likelihood ratio test significantly rejected the null model for B. kathmandius, L. koreanum, Lucasioides spp., and M. koreanus, and for the species richness of native species. Conversely, there was no significant difference between the proposal and null models for another species and species richness of exotic species, indicating that their distributions were not explained by the seven environmental variables. Local environment significantly affected the distribution of Lucasioides spp. only: the presence of flower beds may have a negative effect on the presence of Lucasioides spp. The distributions of four native species, A. pannuosa, B. kathmandius, L. koreanum, and M. koreanus were significantly affected by landscape properties, although there was no significant difference between the proposed and null models for A. pannuosa. Agnara pannuosa, B. kathmandius, and L. koreanum were negatively affected by the surrounding residential region, while natural or semi-natural environments (rice paddy, grass and forest) had positive effects on the distribution of M. koreanus. In addition, the model for the species richness of native species showed significant positive values for grass and forest areas, indicating that species richness of native species was high at sites surrounded by natural conditions. However, the model for P. scaber indicated a significant positive value for rice paddy only, although P. scaber showed spatial autocorrelation and the null model was not rejected.

Lee and Kwon (2015) reported that terrestrial isopods comprised approximately 90% of ground arthropods in urban parks of Osaka, a Japanese mega-city, but they did not identify them at species level. The present study found a total of eleven terrestrial isopod species in the public parks of a commuter town, four of which were exotic species. At least one terrestrial isopod species was found in the parks and the maximum species richness was five. These results imply that urban parks in Japan provide important habitats for terrestrial isopods, as observed in another country (Giurginca et al. 2017).

The species compositions in patchily distributed areas such as parks are known to be affected by local environments within the areas and by landscape environments around the parks (Jokimäki 1999, Germaine and Wakeling 2001, Smith et al. 2005). This study indicated that landscape environments play a more important role in determination of species distributions than do local environments; three native species (B. kathmandius, L. koreanum and M. koreanus) preferred parks surrounded by forest or semi-natural environments (rice paddy and grass) to parks located in residential regions. Ligidium koreanum and M. koreanus have become abundant in forests and grassland within the city. In addition, B. kathmandius is widely distributed in East Asia and some Pacific Islands (Schmalfuss1983, Taiti and Ferrara 1991, Karasawa 2016), and is usually found in grasslands of southern regions in Japan (SK, personal observations). Thus, these species are considered grass or forest specialists that may migrate to parks from surrounding semi-natural environments or forests. This finding suggests a hypothesis that the species richness of native species may increase in the parks surrounded by semi-natural and forest environments. The model analysis supported this hypothesis and these observations also indicate that an understanding of the distribution of terrestrial isopods requires evaluation of the surrounding landscape (Dauber et al. 2005).

Four exotic species with original distributions in the Mediterranean and European regions (Schmalfuss 2003, Cochard et al. 2010) were widely distributed in the public parks of the commuter town, but the present study failed to clarify the environmental factors that determined the distribution of exotic species; the null models were not rejected, and P. scaber also tended to be geographically concentrated. Three exotic species, A. vulgare, P. scaber, and P. pruinosus, are considered to be a habitat generalist, an urban specialist and a synanthropic species, respectively (Dangerfield and Telford 1990; Magura et al. 2008b), and they have been commonly found in urban regions (Vilisics et al. 2007, 2012, Vilisics and Hornung 2009, Giurginca et al. 2017). This may explain why human activity plays a more important role than environmental factors for their distributions in the city. Moreover, A. vulgare was distributed in all parks. This species is able to walk 24 m d^-1 and is a potentially able to walk 50 m h^-1 (Furukawa et al. 2017). This high dispersal ability could facilitate the distribution of A. vulgare via roads and residential regions and could be an important factor in the establishment and success of populations over a wide range. The establishment of populations of exotic species in public parks has the potential to be a source of exotic species in the future.

Eleven terrestrial isopod species were found in the urban parks of a commuter town. Native species tended to be distributed in the parks located adjacent to natural environments, while their distributions had little relationship with the local environments within the parks. It is proposed that the location of parks is an important factor to consider in their design to protect the largest number of species of native terrestrial isopods.

We thank members of the Laboratory of Animal Ecology at Fukuoka University of Education for assistance with field surveys, and Dr. Takahito Kuroki for advices on GIS. We also thank the reviewers for their significant comments and efforts towards improving our manuscript. This work was supported by a Grant-in-Aid for Young Scientists (B) Grant Number 26830145. We thank Harry Taylor, PhD, from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.