- Email: [email protected]
The effects of anthropogenic disturbances and hydrological activity of a river on soil Collembola communities in an urbanized zone
European Journal of Soil Biology 82 (2017) 116e120
Contents lists available at ScienceDirect
European Journal of Soil Biology journal homepage: http://www.elsevier.com/locate/ejsobi
The effects of anthropogenic disturbances and hydrological activity of a river on soil Collembola communities in an urbanized zone Kamil Rzeszowski Museum and Institute of Zoology PAS, Wilcza St. 64, 00-679 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 February 2017 Received in revised form 11 September 2017 Accepted 12 September 2017
Soil Collembola communities inhabiting urban zones are highly modified by anthropogenic disturbances. This effect has not yet been studied along an urban river, where hydrological processes play an important role. The main objective of the present study was to test the hypothesis that the impact of anthropogenic disturbances on the variation in Collembola communities is distinctly different in the riparian area of a river and in nearby area not influenced by processes associated with river activity, i.e. exposed only to urbanization. The study was performed in Warsaw and surroundings (Poland) along the Vistula River. Sampling was conducted in three locations, varying with respect to the level of anthropogenic disturbances e in the city center (high level of anthropogenic disturbances), at the edge of city (moderate level of anthropogenic disturbances) and outside of the city (low level of anthropogenic disturbances), each in the area located in front of the levee (prevalent influence of hydrological processes) and behind the levee (lack of river influence), in spring and autumn. The level of anthropogenic disturbances was associated with a significant variation in the Collembola communities only in the area located behind the levee. Season was associated with significant community variation only in the area located in front of the levee, which can be induced by variable hydrological processes. It can be concluded that anthropogenic disturbances do not affect Collembola communities in the riparian area of an urban river, which reflects the high resistance of communities associated with inundated soils to disturbances in a city. © 2017 Elsevier Masson SAS. All rights reserved.
Handling editor: Stefan Schrader Keywords: Urbanization Disturbances River Riparian area Collembola
The impact of disturbances on biodiversity is one of the main topics of ecological research (review in Ref. [33]). In urbanized zones, such anthropogenic disturbances as habitat loss, habitat fragmentation, changes in climate and modifications of biogeochemical cycles are prevalent [13]. They significantly affect soil biotopes [18]; therefore, the concept of urban soils emphasizes the role of intensive, non-agrarian human activities in their formation and their unique properties in comparison to nonurban ones [5]. In this way, specific human-associated environmental conditions prevalent in cities distinctly shape the communities of soil fauna, including Collembola [24]. Until now, modifications of Collembola communities under different intensities of anthropogenic disturbances in highly urbanized zones have never been studied in habitats associated with large rivers. Rivers situated in cities are highly polluted and modified with respect to e.g. water level [2]. Their riparian areas are often considerably disturbed as a consequence of conversion to other types of land use, increased erosion
E-mail address: [email protected]. https://doi.org/10.1016/j.ejsobi.2017.09.004 1164-5563/© 2017 Elsevier Masson SAS. All rights reserved.
of the river channel, increased rate of sediment deposition and changes in hydrology [17]. It is well documented that communities of soil Collembola inhabiting riparian areas are shaped to a large extent by hydrological processes associated with the river [22,28], such as seasonal floods (‘flood pulse concept’, [12]) and the movement of the ground-water level (‘flow pulse concept’, [31]). Therefore, the main objective of the present study was to test the hypothesis that the impact of anthropogenic disturbances present in a city on the variation in soil Collembola communities is distinctly different in the riparian area of a river where hydrological processes are prominent and in nearby area not influenced by such processes (i.e. exposed mainly to urbanization). The study was performed in central Poland, along the Vistula River located in Warsaw and its surroundings. Warsaw is a typical central European city with an area of about 512 km2 and a population of 2.5 million inhabitants. The Vistula River is the longest river draining to the Baltic Sea. The annual rainfall in the river basin is about 600 mm per year, with a summer maximum. Floods in the Vistula are generated by intense rains, snowmelt and ice-related phenomena. They are seasonally variable. During the growing
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
season, they often occur from late June to the end of July due to intense rainfall, but they are very rare in the second half of April and in all of May [4]. The Vistula crosses the middle of the city of Warsaw, from south to north, along a length of approximately 25 km. In the urbanized zone, the river and its valley are highly degraded, especially at the city center. The river channel inside the city is substantially narrowed by protective flood dikes (the so-called ‘Warsaw corset’). It was also considerably modified by the extensive dredging of sand and gravel for the post-World War II reconstruction of the city [16]. Outside the city, flood protection is less extensive, so the river and its riparian area are notably less degraded. A large floodplain has not developed, however, at the north of the city, floods can occasionally inundate the entire flood terrace [9]. Collembola sampling was conducted at three locations e in the city center (centered at 52140 22.6800 N, 2120 23.0800 E), where the level of anthropogenic disturbances was high; at the edge of city, approximately 10 km south from the city center (centered at 52100 22.9400 N, 21 70 39.5100 E), where the anthropogenic disturbance level was moderate, and outside the city, in an agricultural zone, approximately 20 km north of the city center, (centered at 52 220 45.7800 N, 20 480 56.8000 E) where the anthropogenic disturbance level was low. At each research location, 6 research sites were sampled. Three of them were established in an area situated in front of the levee, directly near the river, in the riparian area. The maximal distance from the river was approximately 200 m, but typically it was much smaller. In this area, Collembola communities were impacted both by anthropogenic disturbances, such as the degradation of riparian habitat in the city center and by hydrological processes frequently associated with the river activity, such as floods [12,] and changes in soil humidity and aeration induced by movements of the groundwater [31]. The other sites were established in an area situated behind the levee, at a typical distance of approximately 500 m from the river. In this area, the above-mentioned river-associated processes were absent, so Collembola communities were shaped mainly by urbanization-related disturbances such as extensive soil degradation by diverse anthropogenic activities (soil alkalization, contamination and enrichment with concrete debris, etc.) that led to formation of urban technosols [14]. The distance between sites along the river was approximately 900 m at each location. Overall 18 sites were sampled. In both areas, research sites were replicated three times at each level of the anthropogenic disturbances. This sampling design allowed representative results for whole of studied region to be obtained. Sampling was performed twice e in late spring (May) and early autumn (September) of 2016, in order to isolate the seasonal variability of Collembola communities associated with e.g. variable hydrological processes. A full list of the research sites with geographical coordinates and sampling dates is provided in Table A1 (Appendix A, on-line supplementary materials). In the sites situated in the area located in front of the levee, the vegetation has the character of a riparian forest belonging to the Salicion albae association. It was highly degraded in the city center, where only a narrow tree belt along the river was preserved. The riparian area itself was mainly devoid of trees and was highly degraded by construction work, especially near bridges. Bare patches of soil and concrete debris were frequently observed. At the edge of the city, riparian forest was better preserved and relatively broad, but it was fragmented into large patches. The riparian area was frequently (but not completely) devoid of trees and was slightly degraded by recreational activities near the hiking trail. It was slightly littered, the vegetation was degraded and soil was compressed locally. Outside the city, the riparian forest was locally narrow, but generally it was in natural condition, with very dense
117
shrubs and a rich herbaceous layer, also in riparian area. The effects of anthropogenic activities were not obvious. In the area located behind the levee the vegetation resembles meadows. In the city center, it was characterized by isolated street lawns similar to the Arrhenatherion alliance, which belong to the Arrhenatheretalia order. The vegetation in these lawns was generally low and sparse. It was regularly mowed during the maintenance of urban greenery. At the edge of the city, such lawns were generally similar, but they were much larger and less isolated. They were not subject to mowing. Vegetation was high and dense. Outside the city, vegetation has the character of abandoned pastures established after the cessation of agricultural activities approximately 20 years ago and belong to the Cynosurion alliance from the Arrhenatheretalia order. Human activities were not obvious, therefore, the vegetation was mainly in a natural state. The general appearance of each type of research site is shown in Figure A1 (Appendix A, on-line supplementary materials). Soils at each location were developed on sandy, Holocene alluvial deposits in both areas. The area located in front of the levee soil belonged to the fluvisols, according to the FAO World Reference Base for Soil Resources [11]. It was similar at each location. In the area located behind the levee, the soil was degraded in the city center and at the edge of the city. In both of these locations, it can be classified as technosols. It was affected by e.g. a high accumulation of various technogenic deposits, such as concrete debris and limestone rubble in the profile. Soil degradation was notably higher in the city center, where the presence of rubble was very obvious. Outside of the city, the soil was very similar to that found in the riparian area. At each research site, three soil cores with an area of 5 cm in diameter and 10 cm in depth were taken at random during each sampling occasion. Collembola were extracted from soil cores using a modified MacFadyen high gradient extractor for a maximum of 10 days. The obtained specimens were identified to species level and deposited at the Museum and Institute of Zoology, PAS. Collembola communities were characterized by basic ecological indices e total abundance (A), species richness (S), Shannon diversity index (H0 ) and Pielou evenness index (J0 ). To check whether basic community indices could be ordinate to the studied factors, a canonical correspondence analysis (CCA) was applied [30]. The calculated model was constrained by ‘season’ (spring and autumn), ‘river activity’ (areas situated in front of and behind the levee) and ‘disturbances’ (high, moderate and low level of anthropogenic disturbances). To identify the relationships between environmental factors and variation in the Collembola communities at the level of species composition, a variation decomposition within series of CCA and partial CCA (pCCA) analyses were applied. First, an ordination model constrained by ‘season’, ‘river activity’ and ‘disturbances’ was calculated for the whole dataset. It was intended to isolate the influence of river-associated processes to variation in Collembola communities in the whole studied region. Second, two separate ordination models were calculated, individually for areas situated in front of and behind the levee, constrained by ‘disturbances’ and ‘season’. They allow to examine the impact of the studied factors in the presence and absence of hydrological processes associated with river activity. In each ordination model, the pure effect of a particular factor was assessed when the remaining factors were removed and used as covariates. Treatment of a particular factor alternatively in the analysis as an explanatory variable or covariate allowed the interpretation of the effect of a given factor independently of interactions with other factors [1]. All statistical analyses were performed with a data matrix in which the values of Collembola species abundances in each cell were calculated from the sum of individuals counted in three
118
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
samples obtained at each research site, in order to avoid pseudoreplications. Different sampling occasions (spring and autumn) at the sites were treated as replicates. Before the calculation of the ordination models, Collembola species data were extrapolated to be expressed as the number of individuals per one square meter and log (x10 þ 1) transformed. Community indices was also log (x10 þ 1) transformed. Analyses were performed with interspecies correlation scaling, species scores divided by SD and centering by species, not standardized by sample. In all analyses, environmental factors were coded as a series of dummy variables. The significance of the models was estimated by the Monte Carlo permutation test with 499 randomizations. The significance level was p < 0.05. All calculations were performed using Canoco 5. A total of 67 Collembola species were identified. 21 of them were found only in the area situated in front of the levee. The 14 other species were found only behind the levee. A full list of species in each studied areas and their mean abundances are provided in Table A2 (Appendix A, on-line supplementary materials). Ecological indices of sampled communities (Table A3, Appendix A, on-line supplementary materials) were not ordinated according to the studied environmental factors (value of the all canonical CCA axes, ltrace ¼ 0.002; F-ratio statistics F ¼ 0.866; p ¼ 0.530; for the CCA correlation biplot, see Figure A2, Appendix A, on-line supplementary materials). Despite this, according to the ordination model calculated for the whole examined region, anthropogenic disturbances, river activity and season significantly affected Collembola communities at the level of species composition. All canonical axes explained more that 14.5% of community variation. The pure effect of river activity was significant and explained more that 4.2% of community variation. The pure effect of season was also significant and explained more that 4.6% of variation, while the pure effect of anthropogenic disturbances was insignificant (Table 1). In the area situated in front of the levee, the influence of anthropogenic disturbances and season on Collembola communities was insignificant. The pure effect of season has, however, significance and explained more than 8.7% of community variation (Table 1). Further pCCA results for this area showed that occurrence €rner, 1901 and Folsoof such species as Pseudachorutes parvulus Bo mia manolachei Bagnall, 1939 was assonated with spring. At the same time occurrence of Lepidocyrtus lignorum (Fabricius, 1793) was associated with autumn (Figure A3, Appendix A, on-line supplementary materials). Simultaneously, the pure effect of anthropogenic disturbances was insignificant (Table 1). In the area located behind the levee, the influence of anthropogenic disturbances and season was significant and all canonical axes explained more than 21.6% of Collembola community variation. The pure effect of anthropogenic disturbances was highly significant and explained more that 15.5% of variation, while the pure effect of season was
insignificant (Table 1). Further pCCA results showed that many of the Collembola species were highly associated with particular levels of anthropogenic disturbances. For example, the occurrence of Isotoma anglicana Lubbock, 1862 was highly associated with a high level of anthropogenic disturbances, while the occurrence of Mesaphorura macrochaeta Rusek, 1976 was linked to a low level of disturbances (Figure A4, Appendix A, on-line supplementary materials). The results of CCA analysis obtained for the whole examined region showed that river-associated hydrological processes notably affected Collembola communities at the level of species composition. Similar results were obtained in the more natural environment, on the extensive river floodplain [29]. Therefore, the natural activity of the river seems to have a universal impact on soil Collembola communities under different environmental conditions and led to the formation of its very specific species composition. Within the group of species found only in the riparian area, there were many species characteristic for inundated soil habitats, for example Anurida ellipsoides Stach, 1949 e frequently associated with streams and swamps, Anurida tullbergi Schott, 1891 e associated with littoral areas, Pratanurida cassagnaui Rusek, 1973 e a species associated with beach meadows [7], Oligaphorura schoetti (Lie-Pettersen, 1896) e mainly mountainous species, in lowlands it is frequently found in the riparian areas of large rivers [28], Ballistura schoetti (Dalla Torre, 1895) e a species mainly associated with seashores and also occurring near large inland rivers [19], Isotoma riparia (Nicolet, 1842) e an obvious riparian species, and Sminthurides aquaticus (Bourlet, 1843) e a species frequently found near the shores of waterbodies [8]. Simultaneously, the ordination models calculated separately for each studied areas showed that the response of Collembola communities to the anthropogenic disturbances highly differs in the riparian area of the river and in the nearby area isolated from its influence. The significant response of Collembola communities to anthropogenic disturbances in the city and its surroundings that was clearly observed behind the levee is in accordance with previous studies that showed large differences in community composition between urban, suburban and adjacent areas [10,25,27]. Such species associated with different levels of anthropogenic disturbances, like I. anglicana and M. macrochaeta, are clearly sensitive to urbanization-associated disturbances in soil physicochemical properties, e.g. the content of soil nutrients and pH, as was extensively demonstrated in the previous study performed in Warsaw [23]. The lack of a community response to anthropogenic disturbances in the riparian area of the river showed, however, that in this area, Collembola communities are not sensitive to this, despite the high degradation of the riparian forest in the city center. As mentioned above, these communities have a very specific species
Table 1 Impact of river activity, anthropogenic disturbances and season on variation of Collembola communities indicated by variation decomposition within series of CCA a pCCA analyses in the entire examined region of Warsaw and surroundings and separately in areas situated in front of the levee and behind the levee. ltrace e value of the all canonical CCA axes; F e value of the F-ratio statistic; p-value e significance level of the effect tested by Monte Carlo permutation test. Examined area
Explanatory variables
Covariates
In front of the levee and behind the levee (entire region) Disturbances, River activity, Season No Season Disturbances, River activity Disturbances, River activity, Season River activity Disturbances, Season In front of the levee Disturbances, Season No Season Disturbances Disturbances Season Behind the levee Disturbances, Season No Season Disturbances Disturbances Season Significant p-values are in bold (p < 0.05).
ltrace Explained variance (%) F
p-value
1.336 0.381 0.606 0.346 1.324 0.534 0.780 1.299 0.413 0.865
0.002 0.008 0.060 0.022 0.122 0.048 0.562 0.008 0.140 0.024
14.53 4.62 7.17 4.22 19.18 8.73 12.26 21.68 8.10 15.55
1.318 1.503 1.197 1.364 1.107 1.338 0.978 1.292 1.234 1.289
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
composition. The results obtained showed, therefore, that such communities associated with inundated soils are highly resistant to anthropogenic disturbances in city. The lack of a community response to the anthropogenic disturbances in the riparian area can be possibly explained by the effective dispersal of Collembola along the river. Collembola have a high ability to passively disperse by water [3], therefore they can be transported during flooding events. In this way, it is possible that Collembola communities inhabiting riparian area were structured by metacommunity-type dynamics along the river [15], therefore many well-dispersed species can be found both in natural and degraded locations. Local communities may not be, therefore, differentiated by habitat degradation associated with different levels of anthropogenic disturbances. The high spatial dynamics of Collembola communities inhabiting heterogeneous riparian areas was shown at a small scale by Russell et al. [21] and by Russell and Griegel [20], therefore, the lack of response to anthropogenic disturbances indicated in the present study indirectly suggests that such mechanisms may also operate at the regional scale of a city and its surroundings. Rivers can therefore serve as corridors for Collembola, reducing the negative effects of habitat degradation (e.g. fragmentation of the riparian forest) in the human-dominated landscape [32]. Such a role of the river in an urbanized zone was also proposed by Zapparoli [34] for all of the insect fauna. Unfortunately, the dynamics of Collembola communities along rivers has never been studied at a large spatial scale. The present results showed that this issue obviously needs further research. The response of Collembola communities to season also differed in each studied areas. The lack of seasonal variability in communities inhabiting the area situated behind the levee indicates that Collembola communities not influenced by river-related hydrological processes are stable in an annual time-frame. The seasonal dynamics of Collembola communities in urbanized zones is generally poorly studied. Some seasonal fluctuations were reported for community structure [6] and for total abundance of Collembola [26] for sites situated exclusively within cities. The results of the present study showed, however, that such seasonal variations are too small to be detectable at the regional scale of the city and its surroundings. The significant response to season observed in the Collembola communities inhabiting the area situated in front of the levee can be induced by the distinct impact of such hydrological processes as seasonal floods. Small-scale flooding events occurred during the present study, since some sites examined as dry in spring were partially inundated in autumn. Such an effect of season was also ska et al. documented on the extensive river floodplain by Sterzyn [29], where it was highly significant. The low significance of this factor in the present study indicate, however, that seasonal hydrological processes have a less pronounced effect on soil Collembola communities in the riparian area of a river with a complex, flood protection infrastructure. Such processes, ever with low significance can however reduce impact of anthropogenic disturbances to an undetectable level. It is also important to note that all of studied environmental factors do not affected basic ecological indices of Collembola communities. The observed effects were therefore associated only with the impact of hydrological processes on the characteristic communities inhabiting the riparian area in front of the levee and with changes in the occurrence pattern of individual species induced by anthropogenic disturbances behind the levee. 1. Conclusions The results of the present study showed that the response of Collembola communities to anthropogenic disturbances in an
119
urbanized zone and its surroundings is strongly differentiated between the riparian area of the river and a nearby area not impacted by river-associated hydrological processes. Collembola communities were not disturbed by habitat degradation in the riparian area of the river, which reflects the high resistance of Collembola communities associated with inundated soils to anthropogenic disturbances within city. Simultaneously, the seasonal variability of communities seems to be driven mainly by river-associated hydrological processes and is not detectable in sites isolated from them. Conflict of interest The author declare that have no conflict of interest. Acknowledgements I thank to Wanda Weiner for taxonomic verification of material and to Barbara Przybylska for language corrections. Moreover, I thank two anonymous Reviewers for their very useful comments. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.ejsobi.2017.09.004. References [1] D. Borcard, P. Legendre, P. Drapeau, Partialling out the spatial component of ecological variation, Ecology 73 (1992) 1045e1055. [2] C.A. Collier, M.S.D. Neto, G.M.A. Aretakis, R.E. Santos, T.H. de Oliveira, J.S. Mourao, W. Severi, A.C.A. El-Deir, Integrated approach to the understanding of the degradation of an urban river: local perceptions, environmental parameters and geoprocessing, J. Ethnobiol. Ethnomed 11 (2015) article 69. [3] S.J. Coulson, I.D. Hodkinson, N.R. Webb, J.A. Harrison, Survival of terrestrial soil-dwelling arthropods on and in seawater: implications for trans-oceanic dispersal, Funct. Ecol. 16 (2002) 353e356. [4] J. Cyberski, M. Grzes, M. Gutry-Korycka, E. Nachlik, Z.W. Kundzewicz, History of floods on the river Vistula, Hydrol. Sci. J. 51 (2006) 799e817. [5] W.R. Effland, R.V. Pouyat, The genesis, classification, and mapping of soils in urban areas, Urban Ecosyst. 1 (1997) 217e228. [6] I. Eitminaviciute, Microarthropod communities in anthropogenic urban soils. 2. Seasonal dynamics of microarthropod abundance in soils at roundabout junctions, Entomol. Rev. 86 (2006) 136e146. [7] A. Fjellberg, The Collembola of Fennoscandia and Denmark. Part I: Poduromorpha. Fauna Entomologica Scandinavica, Volume 35, Brill, Leiden, Boston, €ln, 1998. Ko [8] A. Fjellberg, The Collembola of Fennoscandia and Denmark. Part II: Entomobryomorpha and Symphypleona. Fauna Entomologica Scandinavica, Volume 42, Brill, Leiden, Boston, 2007. _ [9] M. Gutry-Korycka, A. Magnuszewski, J. Suchozebrski, W. Jaworski, M. Marcinkowski, M. Szydłowski, Numerical estimation of flood zones in the Vistula River valley, Warsaw, Poland, in: S. Demuth, A. Gustard, E. Planos, F. Scatena, E. Servat (Eds.), Climate Variability and Change - Hydrological Impacts (Proceedings of the Fifth FRIEND World Conference Held at Havana, Cuba, November 2006) vol. 308, IAHS Publication, 2006, pp. 191e195. IAHS. [10] C. Huang, T. Wang, Y. Luo, S. Chen, K.C. Kim, Assessing Collembola biodiversity under human influence at Three Gorges Area, China, Environ. Entomol 42 (2013) 214e222. [11] IUSS Working Group WRB, World Reference Base for Soil Resources 2014. International Soil Classification System for Naming Soils and Creating Legends for Soil Maps, World Soil Resources Reports No. 106, Food and Agriculture Organization of the United Nations, Rome, 2014. [12] W.J. Junk, P.B. Bayley, R.E. Sparks, The flood-pulse concept in river floodplain systems, in: D.P. Dodge (Ed.), Proceedings of the International Large River Symposium, Canadian Special Publication of Fisheries and Aquatic Sciences 106, Department of Fisheries and Oceans, Ottawa, 1989, pp. 110e127. [13] I. Kowarik, Novel urban ecosystems, biodiversity, and conservation, Environ. Pollut. 159 (2011) 1974e1983. [14] W. Kwasowski, Soils of traffic areas in Warsaw, in: P. Charzynski, P. Hulisz, R. Bednarek (Eds.), Technogenic Soils of Poland. Polish Society of Soil Science, . 2013, pp. 207e229. Torun [15] M.A. Leibold, M. Holyoak, N. Mouquet, P. Amarasekare, J.M. Chase, M.F. Hoopes, R.D. Holt, J.B. Shurin, R. Law, D. Tilman, M. Loreau, A. Gonzalez, The metacommunity concept: a framework for multi-scale community
120
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
ecology, Ecol. Lett. 7 (2004) 601e613. [16] A. Magnuszewski, S. Moran, Vistula River bed erosion processes and their influence on Warsaw's flood safety, in: Y.J. Xu (Ed.), Sediment Dynamics from the Summit to the Sea (Proceedings of a Symposium Held in New Orleans, Louisiana, USA, 11e14 December 2014) vol. 367, IAHS Publication, 2014, pp. 147e154. IAHS. [17] National Research Council, Riparian Areas. Functions and Strategies for Management, National Academy Press, Washington DC, 2002. [18] S.T.A. Pickett, M.L. Cadenasso, Altered resources, disturbance, and heterogeneity: a framework for comparing urban and non-urban soils, Urban Ecosyst. 12 (2009) 23e44. [19] M. Potapov, Isotomidae. Synopses on Palaearctic Collembola, vol. 3, Abhand€ rlitz 73, 2001, pp. 1e603. lungen und Berichte des Naturkundemuseums Go [20] D.J. Russell, A. Griegel, Influence of variable inundation regimes on soil Collembola, Pedobiologia 50 (2006) 165e175. [21] D.J. Russell, A. Hauth, O. Fox, Community dynamics of soil Collembola in floodplains of the upper rhine valley, Pedobiologia 48 (2004) 527e536. [22] D.J. Russell, H. Schick, D. N€ ahrig, Reactions of soil Collembolan communities to inundation in floodplain ecosystems of the Upper Rhine Valley, in: G. Broll, W. Merbach, E.-M. Pfeiffer (Eds.), Wetlands in Central Europe, Soil Organisms, Soil Ecological Processes and Trace Gas Emissions, Springer-Verlag, Berlin, Heidelberg, 2002, pp. 35e70. _ [23] K. Rzeszowski, P. Zadrozny, P. Nicia, The effect of soil nutrient gradients on Collembola communities inhabiting typical urban green spaces, Pedobiologia (2017), http://dx.doi.org/10.1016/j.pedobi.2017.06.003. In press. [24] L. Santorufo, J. Cortet, C. Arena, R. Goudon, A. Rakoto, J.-L. Morel, G. Maisto, An assessment of the influence of the urban environment on collembolan communities in soils using taxonomy- and trait-based approaches, Appl. Soil Ecol.
78 (2014) 48e56. [25] L. Santorufo, J. Cortet, J. Nahmani, C. Pernin, S. Salmon, A. Pernot, J.L. Morel, G. Maisto, Responses of functional and taxonomic collembolan community structure to site management in Mediterranean urban and surrounding areas, Eur. J. Soil. Biol. 70 (2015) 46e57. [26] L. Santorufo, C.A.M. van Gestel, A. Maisto, Sampling season affects conclusions on soil arthropod community structure responses to metal pollution in Mediterranean urban soils, Geoderma 226e227 (2014) 47e53. ska, Communities of Collembola in natural and transformed soils of [27] M. Sterzyn the linden-oak-hornbeam sites of the Mazovian Lowland, Fragm. Faun. 34 (1990) 165e262. ska, Assemblages of Soil Collembola in Wetlands in the Floodplains [28] M. Sterzyn of Some Polish Rivers, Museum and Institute of Zoology PAS, Warszawa, 2009. ska, J. Shrubovych, I. Kaprus, Effect of hydrologic regime and forest [29] M. Sterzyn age on Collembola in riparian forests, Appl. Soil Ecol. 75 (2014) 199e209. [30] P. Smilauer, J. Leps, Multivariate Analysis of Ecological Data Using CANOCO 5, second ed., Cambridge University Press, Cambridge, 2003. [31] K. Tockner, F. Malard, J.V. Ward, An extension of the flood pulse concept, Hydrol. Process. 14 (2000) 2861e2883. [32] C.C. Vos, H. Baveco, C.J. Grashof-Bokdam, Corridors and species dispersal, in: K.J. Gutzwiller (Ed.), Applying Landscape Ecology in Biological Conservation, Springer-Verlag, New York, 2002, pp. 84e104. [33] P.S. White, A. Jentsch, The search for generality in studies of disturbance and ecosystem dynamic, in: K.J. Esser, W. Kadereit, W. Beyschlag (Eds.), Progress in Botany, Genetics, Physiology, Systematics, Ecology, vol. 62, Springer-Verlag, Berlin, Heidelberg, 2001, pp. 399e450. [34] M. Zapparoli, Urban development and insect biodiversity of the Rome area, Italy, Landsc. Urban. Plan. 38 (1997) 77e86.
Contents lists available at ScienceDirect
European Journal of Soil Biology journal homepage: http://www.elsevier.com/locate/ejsobi
The effects of anthropogenic disturbances and hydrological activity of a river on soil Collembola communities in an urbanized zone Kamil Rzeszowski Museum and Institute of Zoology PAS, Wilcza St. 64, 00-679 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 February 2017 Received in revised form 11 September 2017 Accepted 12 September 2017
Soil Collembola communities inhabiting urban zones are highly modified by anthropogenic disturbances. This effect has not yet been studied along an urban river, where hydrological processes play an important role. The main objective of the present study was to test the hypothesis that the impact of anthropogenic disturbances on the variation in Collembola communities is distinctly different in the riparian area of a river and in nearby area not influenced by processes associated with river activity, i.e. exposed only to urbanization. The study was performed in Warsaw and surroundings (Poland) along the Vistula River. Sampling was conducted in three locations, varying with respect to the level of anthropogenic disturbances e in the city center (high level of anthropogenic disturbances), at the edge of city (moderate level of anthropogenic disturbances) and outside of the city (low level of anthropogenic disturbances), each in the area located in front of the levee (prevalent influence of hydrological processes) and behind the levee (lack of river influence), in spring and autumn. The level of anthropogenic disturbances was associated with a significant variation in the Collembola communities only in the area located behind the levee. Season was associated with significant community variation only in the area located in front of the levee, which can be induced by variable hydrological processes. It can be concluded that anthropogenic disturbances do not affect Collembola communities in the riparian area of an urban river, which reflects the high resistance of communities associated with inundated soils to disturbances in a city. © 2017 Elsevier Masson SAS. All rights reserved.
Handling editor: Stefan Schrader Keywords: Urbanization Disturbances River Riparian area Collembola
The impact of disturbances on biodiversity is one of the main topics of ecological research (review in Ref. [33]). In urbanized zones, such anthropogenic disturbances as habitat loss, habitat fragmentation, changes in climate and modifications of biogeochemical cycles are prevalent [13]. They significantly affect soil biotopes [18]; therefore, the concept of urban soils emphasizes the role of intensive, non-agrarian human activities in their formation and their unique properties in comparison to nonurban ones [5]. In this way, specific human-associated environmental conditions prevalent in cities distinctly shape the communities of soil fauna, including Collembola [24]. Until now, modifications of Collembola communities under different intensities of anthropogenic disturbances in highly urbanized zones have never been studied in habitats associated with large rivers. Rivers situated in cities are highly polluted and modified with respect to e.g. water level [2]. Their riparian areas are often considerably disturbed as a consequence of conversion to other types of land use, increased erosion
E-mail address: [email protected]. https://doi.org/10.1016/j.ejsobi.2017.09.004 1164-5563/© 2017 Elsevier Masson SAS. All rights reserved.
of the river channel, increased rate of sediment deposition and changes in hydrology [17]. It is well documented that communities of soil Collembola inhabiting riparian areas are shaped to a large extent by hydrological processes associated with the river [22,28], such as seasonal floods (‘flood pulse concept’, [12]) and the movement of the ground-water level (‘flow pulse concept’, [31]). Therefore, the main objective of the present study was to test the hypothesis that the impact of anthropogenic disturbances present in a city on the variation in soil Collembola communities is distinctly different in the riparian area of a river where hydrological processes are prominent and in nearby area not influenced by such processes (i.e. exposed mainly to urbanization). The study was performed in central Poland, along the Vistula River located in Warsaw and its surroundings. Warsaw is a typical central European city with an area of about 512 km2 and a population of 2.5 million inhabitants. The Vistula River is the longest river draining to the Baltic Sea. The annual rainfall in the river basin is about 600 mm per year, with a summer maximum. Floods in the Vistula are generated by intense rains, snowmelt and ice-related phenomena. They are seasonally variable. During the growing
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
season, they often occur from late June to the end of July due to intense rainfall, but they are very rare in the second half of April and in all of May [4]. The Vistula crosses the middle of the city of Warsaw, from south to north, along a length of approximately 25 km. In the urbanized zone, the river and its valley are highly degraded, especially at the city center. The river channel inside the city is substantially narrowed by protective flood dikes (the so-called ‘Warsaw corset’). It was also considerably modified by the extensive dredging of sand and gravel for the post-World War II reconstruction of the city [16]. Outside the city, flood protection is less extensive, so the river and its riparian area are notably less degraded. A large floodplain has not developed, however, at the north of the city, floods can occasionally inundate the entire flood terrace [9]. Collembola sampling was conducted at three locations e in the city center (centered at 52140 22.6800 N, 2120 23.0800 E), where the level of anthropogenic disturbances was high; at the edge of city, approximately 10 km south from the city center (centered at 52100 22.9400 N, 21 70 39.5100 E), where the anthropogenic disturbance level was moderate, and outside the city, in an agricultural zone, approximately 20 km north of the city center, (centered at 52 220 45.7800 N, 20 480 56.8000 E) where the anthropogenic disturbance level was low. At each research location, 6 research sites were sampled. Three of them were established in an area situated in front of the levee, directly near the river, in the riparian area. The maximal distance from the river was approximately 200 m, but typically it was much smaller. In this area, Collembola communities were impacted both by anthropogenic disturbances, such as the degradation of riparian habitat in the city center and by hydrological processes frequently associated with the river activity, such as floods [12,] and changes in soil humidity and aeration induced by movements of the groundwater [31]. The other sites were established in an area situated behind the levee, at a typical distance of approximately 500 m from the river. In this area, the above-mentioned river-associated processes were absent, so Collembola communities were shaped mainly by urbanization-related disturbances such as extensive soil degradation by diverse anthropogenic activities (soil alkalization, contamination and enrichment with concrete debris, etc.) that led to formation of urban technosols [14]. The distance between sites along the river was approximately 900 m at each location. Overall 18 sites were sampled. In both areas, research sites were replicated three times at each level of the anthropogenic disturbances. This sampling design allowed representative results for whole of studied region to be obtained. Sampling was performed twice e in late spring (May) and early autumn (September) of 2016, in order to isolate the seasonal variability of Collembola communities associated with e.g. variable hydrological processes. A full list of the research sites with geographical coordinates and sampling dates is provided in Table A1 (Appendix A, on-line supplementary materials). In the sites situated in the area located in front of the levee, the vegetation has the character of a riparian forest belonging to the Salicion albae association. It was highly degraded in the city center, where only a narrow tree belt along the river was preserved. The riparian area itself was mainly devoid of trees and was highly degraded by construction work, especially near bridges. Bare patches of soil and concrete debris were frequently observed. At the edge of the city, riparian forest was better preserved and relatively broad, but it was fragmented into large patches. The riparian area was frequently (but not completely) devoid of trees and was slightly degraded by recreational activities near the hiking trail. It was slightly littered, the vegetation was degraded and soil was compressed locally. Outside the city, the riparian forest was locally narrow, but generally it was in natural condition, with very dense
117
shrubs and a rich herbaceous layer, also in riparian area. The effects of anthropogenic activities were not obvious. In the area located behind the levee the vegetation resembles meadows. In the city center, it was characterized by isolated street lawns similar to the Arrhenatherion alliance, which belong to the Arrhenatheretalia order. The vegetation in these lawns was generally low and sparse. It was regularly mowed during the maintenance of urban greenery. At the edge of the city, such lawns were generally similar, but they were much larger and less isolated. They were not subject to mowing. Vegetation was high and dense. Outside the city, vegetation has the character of abandoned pastures established after the cessation of agricultural activities approximately 20 years ago and belong to the Cynosurion alliance from the Arrhenatheretalia order. Human activities were not obvious, therefore, the vegetation was mainly in a natural state. The general appearance of each type of research site is shown in Figure A1 (Appendix A, on-line supplementary materials). Soils at each location were developed on sandy, Holocene alluvial deposits in both areas. The area located in front of the levee soil belonged to the fluvisols, according to the FAO World Reference Base for Soil Resources [11]. It was similar at each location. In the area located behind the levee, the soil was degraded in the city center and at the edge of the city. In both of these locations, it can be classified as technosols. It was affected by e.g. a high accumulation of various technogenic deposits, such as concrete debris and limestone rubble in the profile. Soil degradation was notably higher in the city center, where the presence of rubble was very obvious. Outside of the city, the soil was very similar to that found in the riparian area. At each research site, three soil cores with an area of 5 cm in diameter and 10 cm in depth were taken at random during each sampling occasion. Collembola were extracted from soil cores using a modified MacFadyen high gradient extractor for a maximum of 10 days. The obtained specimens were identified to species level and deposited at the Museum and Institute of Zoology, PAS. Collembola communities were characterized by basic ecological indices e total abundance (A), species richness (S), Shannon diversity index (H0 ) and Pielou evenness index (J0 ). To check whether basic community indices could be ordinate to the studied factors, a canonical correspondence analysis (CCA) was applied [30]. The calculated model was constrained by ‘season’ (spring and autumn), ‘river activity’ (areas situated in front of and behind the levee) and ‘disturbances’ (high, moderate and low level of anthropogenic disturbances). To identify the relationships between environmental factors and variation in the Collembola communities at the level of species composition, a variation decomposition within series of CCA and partial CCA (pCCA) analyses were applied. First, an ordination model constrained by ‘season’, ‘river activity’ and ‘disturbances’ was calculated for the whole dataset. It was intended to isolate the influence of river-associated processes to variation in Collembola communities in the whole studied region. Second, two separate ordination models were calculated, individually for areas situated in front of and behind the levee, constrained by ‘disturbances’ and ‘season’. They allow to examine the impact of the studied factors in the presence and absence of hydrological processes associated with river activity. In each ordination model, the pure effect of a particular factor was assessed when the remaining factors were removed and used as covariates. Treatment of a particular factor alternatively in the analysis as an explanatory variable or covariate allowed the interpretation of the effect of a given factor independently of interactions with other factors [1]. All statistical analyses were performed with a data matrix in which the values of Collembola species abundances in each cell were calculated from the sum of individuals counted in three
118
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
samples obtained at each research site, in order to avoid pseudoreplications. Different sampling occasions (spring and autumn) at the sites were treated as replicates. Before the calculation of the ordination models, Collembola species data were extrapolated to be expressed as the number of individuals per one square meter and log (x10 þ 1) transformed. Community indices was also log (x10 þ 1) transformed. Analyses were performed with interspecies correlation scaling, species scores divided by SD and centering by species, not standardized by sample. In all analyses, environmental factors were coded as a series of dummy variables. The significance of the models was estimated by the Monte Carlo permutation test with 499 randomizations. The significance level was p < 0.05. All calculations were performed using Canoco 5. A total of 67 Collembola species were identified. 21 of them were found only in the area situated in front of the levee. The 14 other species were found only behind the levee. A full list of species in each studied areas and their mean abundances are provided in Table A2 (Appendix A, on-line supplementary materials). Ecological indices of sampled communities (Table A3, Appendix A, on-line supplementary materials) were not ordinated according to the studied environmental factors (value of the all canonical CCA axes, ltrace ¼ 0.002; F-ratio statistics F ¼ 0.866; p ¼ 0.530; for the CCA correlation biplot, see Figure A2, Appendix A, on-line supplementary materials). Despite this, according to the ordination model calculated for the whole examined region, anthropogenic disturbances, river activity and season significantly affected Collembola communities at the level of species composition. All canonical axes explained more that 14.5% of community variation. The pure effect of river activity was significant and explained more that 4.2% of community variation. The pure effect of season was also significant and explained more that 4.6% of variation, while the pure effect of anthropogenic disturbances was insignificant (Table 1). In the area situated in front of the levee, the influence of anthropogenic disturbances and season on Collembola communities was insignificant. The pure effect of season has, however, significance and explained more than 8.7% of community variation (Table 1). Further pCCA results for this area showed that occurrence €rner, 1901 and Folsoof such species as Pseudachorutes parvulus Bo mia manolachei Bagnall, 1939 was assonated with spring. At the same time occurrence of Lepidocyrtus lignorum (Fabricius, 1793) was associated with autumn (Figure A3, Appendix A, on-line supplementary materials). Simultaneously, the pure effect of anthropogenic disturbances was insignificant (Table 1). In the area located behind the levee, the influence of anthropogenic disturbances and season was significant and all canonical axes explained more than 21.6% of Collembola community variation. The pure effect of anthropogenic disturbances was highly significant and explained more that 15.5% of variation, while the pure effect of season was
insignificant (Table 1). Further pCCA results showed that many of the Collembola species were highly associated with particular levels of anthropogenic disturbances. For example, the occurrence of Isotoma anglicana Lubbock, 1862 was highly associated with a high level of anthropogenic disturbances, while the occurrence of Mesaphorura macrochaeta Rusek, 1976 was linked to a low level of disturbances (Figure A4, Appendix A, on-line supplementary materials). The results of CCA analysis obtained for the whole examined region showed that river-associated hydrological processes notably affected Collembola communities at the level of species composition. Similar results were obtained in the more natural environment, on the extensive river floodplain [29]. Therefore, the natural activity of the river seems to have a universal impact on soil Collembola communities under different environmental conditions and led to the formation of its very specific species composition. Within the group of species found only in the riparian area, there were many species characteristic for inundated soil habitats, for example Anurida ellipsoides Stach, 1949 e frequently associated with streams and swamps, Anurida tullbergi Schott, 1891 e associated with littoral areas, Pratanurida cassagnaui Rusek, 1973 e a species associated with beach meadows [7], Oligaphorura schoetti (Lie-Pettersen, 1896) e mainly mountainous species, in lowlands it is frequently found in the riparian areas of large rivers [28], Ballistura schoetti (Dalla Torre, 1895) e a species mainly associated with seashores and also occurring near large inland rivers [19], Isotoma riparia (Nicolet, 1842) e an obvious riparian species, and Sminthurides aquaticus (Bourlet, 1843) e a species frequently found near the shores of waterbodies [8]. Simultaneously, the ordination models calculated separately for each studied areas showed that the response of Collembola communities to the anthropogenic disturbances highly differs in the riparian area of the river and in the nearby area isolated from its influence. The significant response of Collembola communities to anthropogenic disturbances in the city and its surroundings that was clearly observed behind the levee is in accordance with previous studies that showed large differences in community composition between urban, suburban and adjacent areas [10,25,27]. Such species associated with different levels of anthropogenic disturbances, like I. anglicana and M. macrochaeta, are clearly sensitive to urbanization-associated disturbances in soil physicochemical properties, e.g. the content of soil nutrients and pH, as was extensively demonstrated in the previous study performed in Warsaw [23]. The lack of a community response to anthropogenic disturbances in the riparian area of the river showed, however, that in this area, Collembola communities are not sensitive to this, despite the high degradation of the riparian forest in the city center. As mentioned above, these communities have a very specific species
Table 1 Impact of river activity, anthropogenic disturbances and season on variation of Collembola communities indicated by variation decomposition within series of CCA a pCCA analyses in the entire examined region of Warsaw and surroundings and separately in areas situated in front of the levee and behind the levee. ltrace e value of the all canonical CCA axes; F e value of the F-ratio statistic; p-value e significance level of the effect tested by Monte Carlo permutation test. Examined area
Explanatory variables
Covariates
In front of the levee and behind the levee (entire region) Disturbances, River activity, Season No Season Disturbances, River activity Disturbances, River activity, Season River activity Disturbances, Season In front of the levee Disturbances, Season No Season Disturbances Disturbances Season Behind the levee Disturbances, Season No Season Disturbances Disturbances Season Significant p-values are in bold (p < 0.05).
ltrace Explained variance (%) F
p-value
1.336 0.381 0.606 0.346 1.324 0.534 0.780 1.299 0.413 0.865
0.002 0.008 0.060 0.022 0.122 0.048 0.562 0.008 0.140 0.024
14.53 4.62 7.17 4.22 19.18 8.73 12.26 21.68 8.10 15.55
1.318 1.503 1.197 1.364 1.107 1.338 0.978 1.292 1.234 1.289
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
composition. The results obtained showed, therefore, that such communities associated with inundated soils are highly resistant to anthropogenic disturbances in city. The lack of a community response to the anthropogenic disturbances in the riparian area can be possibly explained by the effective dispersal of Collembola along the river. Collembola have a high ability to passively disperse by water [3], therefore they can be transported during flooding events. In this way, it is possible that Collembola communities inhabiting riparian area were structured by metacommunity-type dynamics along the river [15], therefore many well-dispersed species can be found both in natural and degraded locations. Local communities may not be, therefore, differentiated by habitat degradation associated with different levels of anthropogenic disturbances. The high spatial dynamics of Collembola communities inhabiting heterogeneous riparian areas was shown at a small scale by Russell et al. [21] and by Russell and Griegel [20], therefore, the lack of response to anthropogenic disturbances indicated in the present study indirectly suggests that such mechanisms may also operate at the regional scale of a city and its surroundings. Rivers can therefore serve as corridors for Collembola, reducing the negative effects of habitat degradation (e.g. fragmentation of the riparian forest) in the human-dominated landscape [32]. Such a role of the river in an urbanized zone was also proposed by Zapparoli [34] for all of the insect fauna. Unfortunately, the dynamics of Collembola communities along rivers has never been studied at a large spatial scale. The present results showed that this issue obviously needs further research. The response of Collembola communities to season also differed in each studied areas. The lack of seasonal variability in communities inhabiting the area situated behind the levee indicates that Collembola communities not influenced by river-related hydrological processes are stable in an annual time-frame. The seasonal dynamics of Collembola communities in urbanized zones is generally poorly studied. Some seasonal fluctuations were reported for community structure [6] and for total abundance of Collembola [26] for sites situated exclusively within cities. The results of the present study showed, however, that such seasonal variations are too small to be detectable at the regional scale of the city and its surroundings. The significant response to season observed in the Collembola communities inhabiting the area situated in front of the levee can be induced by the distinct impact of such hydrological processes as seasonal floods. Small-scale flooding events occurred during the present study, since some sites examined as dry in spring were partially inundated in autumn. Such an effect of season was also ska et al. documented on the extensive river floodplain by Sterzyn [29], where it was highly significant. The low significance of this factor in the present study indicate, however, that seasonal hydrological processes have a less pronounced effect on soil Collembola communities in the riparian area of a river with a complex, flood protection infrastructure. Such processes, ever with low significance can however reduce impact of anthropogenic disturbances to an undetectable level. It is also important to note that all of studied environmental factors do not affected basic ecological indices of Collembola communities. The observed effects were therefore associated only with the impact of hydrological processes on the characteristic communities inhabiting the riparian area in front of the levee and with changes in the occurrence pattern of individual species induced by anthropogenic disturbances behind the levee. 1. Conclusions The results of the present study showed that the response of Collembola communities to anthropogenic disturbances in an
119
urbanized zone and its surroundings is strongly differentiated between the riparian area of the river and a nearby area not impacted by river-associated hydrological processes. Collembola communities were not disturbed by habitat degradation in the riparian area of the river, which reflects the high resistance of Collembola communities associated with inundated soils to anthropogenic disturbances within city. Simultaneously, the seasonal variability of communities seems to be driven mainly by river-associated hydrological processes and is not detectable in sites isolated from them. Conflict of interest The author declare that have no conflict of interest. Acknowledgements I thank to Wanda Weiner for taxonomic verification of material and to Barbara Przybylska for language corrections. Moreover, I thank two anonymous Reviewers for their very useful comments. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.ejsobi.2017.09.004. References [1] D. Borcard, P. Legendre, P. Drapeau, Partialling out the spatial component of ecological variation, Ecology 73 (1992) 1045e1055. [2] C.A. Collier, M.S.D. Neto, G.M.A. Aretakis, R.E. Santos, T.H. de Oliveira, J.S. Mourao, W. Severi, A.C.A. El-Deir, Integrated approach to the understanding of the degradation of an urban river: local perceptions, environmental parameters and geoprocessing, J. Ethnobiol. Ethnomed 11 (2015) article 69. [3] S.J. Coulson, I.D. Hodkinson, N.R. Webb, J.A. Harrison, Survival of terrestrial soil-dwelling arthropods on and in seawater: implications for trans-oceanic dispersal, Funct. Ecol. 16 (2002) 353e356. [4] J. Cyberski, M. Grzes, M. Gutry-Korycka, E. Nachlik, Z.W. Kundzewicz, History of floods on the river Vistula, Hydrol. Sci. J. 51 (2006) 799e817. [5] W.R. Effland, R.V. Pouyat, The genesis, classification, and mapping of soils in urban areas, Urban Ecosyst. 1 (1997) 217e228. [6] I. Eitminaviciute, Microarthropod communities in anthropogenic urban soils. 2. Seasonal dynamics of microarthropod abundance in soils at roundabout junctions, Entomol. Rev. 86 (2006) 136e146. [7] A. Fjellberg, The Collembola of Fennoscandia and Denmark. Part I: Poduromorpha. Fauna Entomologica Scandinavica, Volume 35, Brill, Leiden, Boston, €ln, 1998. Ko [8] A. Fjellberg, The Collembola of Fennoscandia and Denmark. Part II: Entomobryomorpha and Symphypleona. Fauna Entomologica Scandinavica, Volume 42, Brill, Leiden, Boston, 2007. _ [9] M. Gutry-Korycka, A. Magnuszewski, J. Suchozebrski, W. Jaworski, M. Marcinkowski, M. Szydłowski, Numerical estimation of flood zones in the Vistula River valley, Warsaw, Poland, in: S. Demuth, A. Gustard, E. Planos, F. Scatena, E. Servat (Eds.), Climate Variability and Change - Hydrological Impacts (Proceedings of the Fifth FRIEND World Conference Held at Havana, Cuba, November 2006) vol. 308, IAHS Publication, 2006, pp. 191e195. IAHS. [10] C. Huang, T. Wang, Y. Luo, S. Chen, K.C. Kim, Assessing Collembola biodiversity under human influence at Three Gorges Area, China, Environ. Entomol 42 (2013) 214e222. [11] IUSS Working Group WRB, World Reference Base for Soil Resources 2014. International Soil Classification System for Naming Soils and Creating Legends for Soil Maps, World Soil Resources Reports No. 106, Food and Agriculture Organization of the United Nations, Rome, 2014. [12] W.J. Junk, P.B. Bayley, R.E. Sparks, The flood-pulse concept in river floodplain systems, in: D.P. Dodge (Ed.), Proceedings of the International Large River Symposium, Canadian Special Publication of Fisheries and Aquatic Sciences 106, Department of Fisheries and Oceans, Ottawa, 1989, pp. 110e127. [13] I. Kowarik, Novel urban ecosystems, biodiversity, and conservation, Environ. Pollut. 159 (2011) 1974e1983. [14] W. Kwasowski, Soils of traffic areas in Warsaw, in: P. Charzynski, P. Hulisz, R. Bednarek (Eds.), Technogenic Soils of Poland. Polish Society of Soil Science, . 2013, pp. 207e229. Torun [15] M.A. Leibold, M. Holyoak, N. Mouquet, P. Amarasekare, J.M. Chase, M.F. Hoopes, R.D. Holt, J.B. Shurin, R. Law, D. Tilman, M. Loreau, A. Gonzalez, The metacommunity concept: a framework for multi-scale community
120
K. Rzeszowski / European Journal of Soil Biology 82 (2017) 116e120
ecology, Ecol. Lett. 7 (2004) 601e613. [16] A. Magnuszewski, S. Moran, Vistula River bed erosion processes and their influence on Warsaw's flood safety, in: Y.J. Xu (Ed.), Sediment Dynamics from the Summit to the Sea (Proceedings of a Symposium Held in New Orleans, Louisiana, USA, 11e14 December 2014) vol. 367, IAHS Publication, 2014, pp. 147e154. IAHS. [17] National Research Council, Riparian Areas. Functions and Strategies for Management, National Academy Press, Washington DC, 2002. [18] S.T.A. Pickett, M.L. Cadenasso, Altered resources, disturbance, and heterogeneity: a framework for comparing urban and non-urban soils, Urban Ecosyst. 12 (2009) 23e44. [19] M. Potapov, Isotomidae. Synopses on Palaearctic Collembola, vol. 3, Abhand€ rlitz 73, 2001, pp. 1e603. lungen und Berichte des Naturkundemuseums Go [20] D.J. Russell, A. Griegel, Influence of variable inundation regimes on soil Collembola, Pedobiologia 50 (2006) 165e175. [21] D.J. Russell, A. Hauth, O. Fox, Community dynamics of soil Collembola in floodplains of the upper rhine valley, Pedobiologia 48 (2004) 527e536. [22] D.J. Russell, H. Schick, D. N€ ahrig, Reactions of soil Collembolan communities to inundation in floodplain ecosystems of the Upper Rhine Valley, in: G. Broll, W. Merbach, E.-M. Pfeiffer (Eds.), Wetlands in Central Europe, Soil Organisms, Soil Ecological Processes and Trace Gas Emissions, Springer-Verlag, Berlin, Heidelberg, 2002, pp. 35e70. _ [23] K. Rzeszowski, P. Zadrozny, P. Nicia, The effect of soil nutrient gradients on Collembola communities inhabiting typical urban green spaces, Pedobiologia (2017), http://dx.doi.org/10.1016/j.pedobi.2017.06.003. In press. [24] L. Santorufo, J. Cortet, C. Arena, R. Goudon, A. Rakoto, J.-L. Morel, G. Maisto, An assessment of the influence of the urban environment on collembolan communities in soils using taxonomy- and trait-based approaches, Appl. Soil Ecol.
78 (2014) 48e56. [25] L. Santorufo, J. Cortet, J. Nahmani, C. Pernin, S. Salmon, A. Pernot, J.L. Morel, G. Maisto, Responses of functional and taxonomic collembolan community structure to site management in Mediterranean urban and surrounding areas, Eur. J. Soil. Biol. 70 (2015) 46e57. [26] L. Santorufo, C.A.M. van Gestel, A. Maisto, Sampling season affects conclusions on soil arthropod community structure responses to metal pollution in Mediterranean urban soils, Geoderma 226e227 (2014) 47e53. ska, Communities of Collembola in natural and transformed soils of [27] M. Sterzyn the linden-oak-hornbeam sites of the Mazovian Lowland, Fragm. Faun. 34 (1990) 165e262. ska, Assemblages of Soil Collembola in Wetlands in the Floodplains [28] M. Sterzyn of Some Polish Rivers, Museum and Institute of Zoology PAS, Warszawa, 2009. ska, J. Shrubovych, I. Kaprus, Effect of hydrologic regime and forest [29] M. Sterzyn age on Collembola in riparian forests, Appl. Soil Ecol. 75 (2014) 199e209. [30] P. Smilauer, J. Leps, Multivariate Analysis of Ecological Data Using CANOCO 5, second ed., Cambridge University Press, Cambridge, 2003. [31] K. Tockner, F. Malard, J.V. Ward, An extension of the flood pulse concept, Hydrol. Process. 14 (2000) 2861e2883. [32] C.C. Vos, H. Baveco, C.J. Grashof-Bokdam, Corridors and species dispersal, in: K.J. Gutzwiller (Ed.), Applying Landscape Ecology in Biological Conservation, Springer-Verlag, New York, 2002, pp. 84e104. [33] P.S. White, A. Jentsch, The search for generality in studies of disturbance and ecosystem dynamic, in: K.J. Esser, W. Kadereit, W. Beyschlag (Eds.), Progress in Botany, Genetics, Physiology, Systematics, Ecology, vol. 62, Springer-Verlag, Berlin, Heidelberg, 2001, pp. 399e450. [34] M. Zapparoli, Urban development and insect biodiversity of the Rome area, Italy, Landsc. Urban. Plan. 38 (1997) 77e86.