Communities of soil macrofauna in green spaces of an urbanizing city at east China

We evaluated the diversity of soil macrofauna communities inhabiting urban green spaces in Yancheng City, an urbanizing city located east of China. In the end of April 2011, the taxonomic richness, abundance and composition of soil macrofauna communities were assessed and compared among ﬁ ve types of green space (poplar forest, rapeseed farm, grassland in park, lawn and nursery garden) and three depth layers in the soil, with taxonomic resolution attained at the order level. Taxonomic richness (orders) and abundance were signiﬁ cantly different among green spaces. Diversity indices (Margalef’s taxonomic richness R and Shannon-Weaver diversity index H’) were higher in poplar forest, grassland in park and nursery garden than in rapeseed farm and lawn. Taxonomic richness (Chao 2) showed a similar trend. There were signiﬁ cant effects of green space type and soil layer, which showed a signiﬁ cant interaction affecting macrofauna composition. We recommend that urban green spaces can be used for maintaining biodiversity, not only for landscape purposes.


INTRODUCTION
Green spaces are par ticularly impor tant in urban environments where the soil biota may enhance the environmental quality of the city through degradation of pollutants and reduction of surface water run-off due to the development and preservation of soil structure (Smith et al. 2006).However, urban biodiversity is subject to a number of natural processes and anthropogenic activities, making it dif ficult to identify a general conser vation strategy (Scalenghe & Marsan 2009).Studies on aboveground invertebrates of habitat fragments in urban areas have identified several factors that influence species diversity, including: fragment size, degree of isolation, fragment age, edge effects, disturbance levels, habitat characteristics, and invasion by exotic species (Bolger et al. 2000, Gibb & Hochuli 2002).However, data on below-ground taxa are scarce (Rossi et al. 2006).In many cities, the majority of urban green spaces are domestic with human made environment patches.These unique environments, although highly modifi ed and disturbed, have recently been identifi ed as an important source of native biodiversity (Gaston et al. 2004).Land use has a strong infl uence on the overall abundance, diversity and community composition of soil macrofauna (Barros et al. 2002, Bar rios et al. 2005).Significant dif ferences have been detected among the urban green spaces on the vegetation cover (Ge et al. 2005), and the soil macrofauna in the green spaces can be used to evaluate the health of the urban ecosystem (McIntyre 2001).
The world is undergoing the largest wave of urban growth in histor y.In 2008, for the first time in histor y, more than half of the world's population was living in towns and cities (UNFPA 2010).Urbanization in China is taking place at an unprecedented pace and will continue over the next decades.The level of urbanization in China has risen from 18 % in 1978 to 30 % in 1995 and to 39 % in 2002 (Song & Ding 2007), and then 47 % in 2010(UNDESA 2010).It is expected that China will quadruple its total GDP and reach 55 % of urbanization by 2020 (Song & Ding 2007).In the urbanization, the land use types in the cities would be changed eventually, and then the green spaces is a ver y important kind of land use for the health of the cities (McIntyre 2001), and therefore the biodiversity in the green spaces of the cities should be evaluated with the urban sprawl.
In this study, we used different diversity i n d i c e s a n d s t r u c t u r e c o m p o s i t i o n t o compare the macrofauna communities in fi ve green spaces of Yancheng City, Jiangsu Province, China, aiming to apply the results to the planning of urban green spaces and the conservation of biodiversity in cites.

Study area and green spaces
Yancheng City is located on the west-Pacifi c coast at the transition between subtropical and temperate zones in Jiangsu Province, eastern China.Annual rainfall, primarily between June and August, averages 900 to 1100 mm.Average daily temperatures were 14.6 to 15.9 ºC (Committee of Jiangsu Provincial Atlas Compilation 2004).The urban area of Yancheng City has risen from 70.6 km 2 in 2006 to 118 km 2 in 2011, while some suburban areas have turned to be urban in recent years.
Along Xindu Road, five types of green space (see Table 1) were selected at the end of April 2011 in Yancheng City, and Xindu Road was a main traffi c road passing through the old area and new area.The habitats were selected on the basis of the function of green spaces in the city, such as typical green spaces (the river green belt, park and lawn for road greenization) and patches embedded in the city following with the urbanization (nursery garden and farm).
The maximum distance among different green spaces was about 6 km.Yancheng is a city diked from the sea, and their soils corresponded to Fluvisols (FAO/ UNESCO Taxonomy) or Inceptisols (Soil Taxonomy).

Sampling and identifi cation
A sample plot was settled at each type of green space, where fi ve soil blocks of 20 cm × 20 cm to 15 cm depth were collected and sorted.Sampling units were located 5 m apart and distributed randomly in the plot.The soil block was removed from the ground, divided into three layers of 5 cm depth (I: 0-4.9 cm, II: 5-9.9 cm and III: 10-15 cm), and hand-sorted for soil macrofauna.Each soil block was hand-sorted for about 60 min and any macrofauna encountered was counted for assessing abundance, preserved in 70 % ethanol, and identifi ed to order level (Pauli et al. 2011).

Data analysis
The taxonomic richness was used in the analysis of soil macrofauna communities with the macrofauna identified at the order/class level (Pauli et al. 2011).We also used some general diversity indices such as Shannon-Weaver diversity index H' (Shannon & Weaver 1949), Margalef's richness index R (Margalef 1957) and Pielou's evenness index J (Pielou 1966) based on the communities' composition data sets for each soil block (n = 5 for each type of green space).In addition, Chao 2 richness index was employed to estimate the taxonomic richness with a 95 % confi dence interval for each green space using bootstrap replicates method (Hammer et al. 2001).Based TABLE 2 Number of soil macrofauna individuals (identifi ed to order level) collected in the studied green spaces.

Order
Poplar Based on the taxonomic richness and abundance data sets for each block with diffrent layers, two-way ANOVAs of taxonomic richness and abundance were used to determine the effects of green spaces types and of soil layers.
Based on an Euclidean distance matrix (75 × 18) created from the communities' composition data, ANOSIM (with 9999 permutations) was used for testing the statistical signifi cance of the factors green space type (n = 5) and soil layer (n = 3).

RESULTS
A total of 18 soil macrofauna taxa were identifi ed in the investigation: arthropods (14 orders), molluscs (two orders) and annelids (two orders).Hymenoptera (all of them ants) and Isopoda were the dominant taxa (Table 2).
The soil macrofauna communities found in nursery garden, grassland of park and poplar forest had the higher taxonomic richness, abundance, R and H' values.Lowest values occurred in lawn and rapeseed farm (Fig. 1).
Chao 2 richness index show that the highest value also occur red in nurser y garden, grassland in park and poplar forest, the lowest value occurred in lawn and rapeseed farm (Fig. 2).One-way ANOVA results for the variation in abundance, taxonomic richness, diversity, richness and evenness of soil macrofauna among fi ve different types of green space.

DISCUSSION
The green spaces sampled in this study had a high taxonomic richness and a wide abundance range of soil macrofauna (Table 1).Thus, our results supported the idea that the type of green space has a significant effect on the composition of soil macrofauna communities, which was similar to other studies in urban green spaces (Ge et al. 2005, Smith et al. 2006).The results indicated that the higher diversity would occur in the complexity habitat, which could be determined by the soil quality and vegetation enviromental characters (Mathieu et al. 2005).In the post researches, it has been reported that the soil quality and soil macromorphology could be the factors determining the biodiversity which were affected by urbanization (Manta et al. 2002, Acosta et al. 2011).
showed the similar trend, and then the highest value occured at the grassland in park.Our results indicated that in the urbanizing city, complex habitats can offer a higher diversity of soil macrofauna, and that the effect of trees on biodiversity should be especially considered in the planning of green spaces (Binkley & Giardina 1998, Barbier et al. 2008).
The soil macrofauna distribution (including taxonomic richness and abundance) was af fected significantly by the type of green space, soil layer and the interaction.The results showed that not only green space but also the soil depth was strongly related to patterns of soil macrofauna distribution and community composition (Gibb & Hochuli 2002, Pauli et al. 2011).In addition to individual factors, the interaction between type of green space and soil layer was an important factor which could determine the ver tical distribution patterns.The signifi cant interaction indicated that soil macrofauna ver tical distribution was also affected by the change of soil layers which collaborated with types of green spaces.
It has been reported that soil macrofauna in urban green spaces is dif ferent to the macrofauna in the forest or in the countryside (Smith et al. 2006, Scalenghe & Marsan 2009).Never theless, the green space should be used for maintaining the biodiversity in urban ecosystem, not only for designing them as landscape.Our results indicated that the types of green spaces should be the main driving force for the biodiversity variation in the study.

Fig 2 :
Fig 2: Taxonomic richness of different green spaces as estimated by Chao 2 with a 95 % confi dence interval.Means with different letters are signifi cantly different (if the confi dence intervals overlap, they are not significantly different; α = 0.05).

TABLE 1
Characteristics of the green spaces selected in the study.
the same data sets, a one-way ANOVA was used to compare taxonomic richness and abundance among green spaces, and a Tukey test was used if signifi cant differences occurred. on

TABLE 4
ANOSIM results for the composition of macrofauna communities in different types of green spaces and soil layers.Resultados de ANOSIM para la composición de las comunidades de la macrofauna en diferentes tipos de espacios verdes y capas del suelo.