Response of earthworm communities to soil disturbance: Fractal dimension of soil and species’ rank-abundance curves

Response of earthworm communities to soil disturbance: Fractal dimension of soil and species’ rank-abundance curves

Applied Soil Ecology 43 (2009) 83–88 Contents lists available at ScienceDirect Applied Soil Ecology journal homepage:

231KB Sizes 5 Downloads 56 Views

Applied Soil Ecology 43 (2009) 83–88

Contents lists available at ScienceDirect

Applied Soil Ecology journal homepage:

Response of earthworm communities to soil disturbance: Fractal dimension of soil and species’ rank-abundance curves Andre´s Duhour a,b,c,*, Cristina Costa d, Fernando Momo a,c, Liliana Falco c, Leonardo Malacalza c a

Instituto de Ciencias, Universidad Nacional de General Sarmiento, J.M. Gutie´rrez 1150, C.P. 1613, Los Polvorines, Buenos Aires, Argentina CONICET, Avenida Rivadavia 1917 – CP C1033AAJ – Ciudad Auto´noma de Buenos Aires, Buenos Aires, Argentina c INEDES, Universidad Nacional de Luja´n, Ruta 5 y Avenida Constitucio´n, C.P. 6700, Luja´n, Buenos Aires, Argentina d Edafologı´a, Departamento de Tecnologı´a, Universidad Nacional de Luja´n, Ruta 5 y Avenida Constitucio´n, C.P. 6700, Luja´n, Buenos Aires, Argentina b



Article history: Received 4 November 2008 Received in revised form 5 June 2009 Accepted 12 June 2009

Soil structure degradation and its relationship with soil fauna communities is a crucial issue in soil management. The aim of this work is to analyze soil fractal dimensions and the earthworm community structure along a perturbation gradient in a Typic Argiudoll soil with different combinations of annual crops and pastures. Samples were taken in four sites: a natural grassland (NAT) without cultivation for 30 years, and three pastures of 1–3 years (P1, P2, P3). Five undisturbed soil blocks were taken per site in order to evaluate soil structural complexity. Two fractal dimensions were measured on images of polished sections of the blocks: the mass fractal dimension of pores (Dmp), and the mass fractal dimension of soil solid complement (Dms). Biomass, number and species composition of earthworm communities were assessed by hand-sorting 15 samples of 25 cm  25 cm  20 cm on each site. Species rank vs. abundance distribution were fitted to geometric series and broken stick models. Shannon index and expected species number by rarefaction (E(S)) were calculated. Both soil solid and pore space evaluated may be considered as fractals in all sites analyzed. We found that the fractal dimension of pores was significantly (p < 0.03) higher for NAT and P3 (1.63  0.03 and 1.62  0.02, respectively) than for P1 (1.43  0.05). Species richness was P1: 3, P2: 7, P3: 7, NAT: 9. Species abundance vs. rank curves shows better fit to the geometric series model in all sites. We found the highest values of Dmp in sites with the higher diversity and species richness of earthworms, suggesting a trend between the observed biological activity and the fractal dimension of the habitat. ß 2009 Elsevier B.V. All rights reserved.

Keywords: Perturbation Soil structure Fractal dimension Earthworm community structure Crop rotations Abundance-rank curves

1. Introduction The increase of disturbances due to growing agricultural practices causes strong modifications in soil ecosystems. Over the past 25 years, the Argentine Pampas have suffered an agriculturization process, characterized by the advance of annual crops over different environments in competition with previous and traditional land uses such as the rotation of agriculture with grazing (Teubal, 2003; Manuel-Navarrete et al., 2009). The observed changes have modified soil structure, water balance, plant biomass production and, consequently, may have affected soil communities such as earthworm assemblages (Decae¨ns and Jime´nez, 2002; Casas, 2005).

* Corresponding author at: Instituto de Ciencias, Universidad Nacional de General Sarmiento, J.M. Gutie´rrez 1150, C.P. 1613, Los Polvorines, Buenos Aires, Argentina. Tel.: +54 11 44697543; fax: +54 11 44697501. E-mail address: [email protected] (A. Duhour). 0929-1393/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.apsoil.2009.06.004

Agroecosystems are complex systems where tools derived from fractal geometry can be used to integrate soil processes by allowing quantification of soil structure (Anderson et al., 1998; Gime´nez et al., 1998; Pachepsky et al., 1999; Filgueira et al., 1999). Fractal dimensions have been used for habitat characterization and for the analysis of modifications produced in or by different communities of soil organisms. Kampichler (1999) outlines the potential of fractal concepts for different studies of soil fauna, reviewing a fractal approach to describe nematode movement patterns and analyze the impact of habitat complexity on abundance: body size distribution of soil microarthropods. Finlay and Fenchel (2001) suggest that observed species richness and abundance of soil protozoa are explained by the soil having a ˜ ach and Le Comber (2004) analyze the fractal structure. Roman correlation of fractal dimensions with other geometric measures of Thomomys bottae burrows. Earthworms are one of the most important components of the soil fauna. These organisms are considered ecosystem engineers because of their action on the soil habitat and the modulation of resources for other organisms (Lavelle, 2002). There are numerous


A. Duhour et al. / Applied Soil Ecology 43 (2009) 83–88

studies of the relationship between earthworms and soil structure which apply a variety of methods of analysis such as the characterization of burrows by means of CAT scan technology (Joschko et al., 1994; Langmaak et al., 1999; Je´gou et al., 2002), measurement of bulk density, aggregate stability or mean weight diameter of aggregates (Degens, 1997; Decae¨ns et al., 2002), as well as quantification of changes in pore shape or distribution (Binet and Curmi, 1992; VandenBygaart et al., 2000; Lamande´ et al., 2003). In Argentina, earthworm studies have focused on the relationships of different soil chemical parameters and earthworm abundance (Momo et al., 1993) or in the relationships of soil use and management with the earthworm community structure (Falco et al., 1995; Clemente et al., 2003) rather than on the observation of effects on soil structure properties. Studies of the pore system complexity in soils with different earthworm communities are still lacking. The quantification of the action of earthworms in a complex soil environment using fractals could allow a better understanding of the reciprocal effects between earthworm community and soil structure. In general, sites with natural or artificial pastures present a relatively lower level of soil perturbation than sites exclusively used for annual crops. This situation allows a reconstitution of community biomass and species richness of earthworms (Decae¨ns and Jime´nez, 2002) and therefore may show differences in soil structure (Joschko et al., 1994). The aim of this work was to analyze and relate soil fractal dimensions and earthworm community structure along a perturbation gradient encompassing different rotations of annual crops and pastures on a Typic Argiudoll. 2. Methods 2.1. Study site The present work was carried out in Luja´n, Buenos Aires province, Argentina. The climate of the region is temperate humid with a mean annual precipitation of 1000 mm and a mean annual temperature of 17 8C. Soils are Typic Argiudolls, with silt loam texture in the surface horizon (Table 1). Four sampling sites were selected differing in their land use histories along a perturbation gradient. This gradient was established considering the time (from the sampling moment to the past) with permanence of pasture and the number of annual crops in the use history – a site with longer pasture use and fewer annual crops was considered less perturbed. Given this definition, the perturbation ranking is P1 > P2 > P3. A fourth site (NAT) was sampled and considered the least disturbed, i.e. reference site (see Table 2). The three pasture sites are located on the Experimental dairy farm of the National University of Luja´n (34850 S and 598340 W); the sites have areas ranging from 8 to 12 ha. Soil management was made by conventional tillage with plow and disk harrow. The pastures composition was the following – P1: Trifolium pratense, Trifolium repens, Lotus corniculatus, Dactylis glomerata, Bromus catharticus and Lolium perenne. P2: Medicago sativa, T. pratense, T. repens, D. glomerata, B. catharticus, and L. perenne. P3: M. sativa, T. pratense, D. glomerata, and B. catharticus. All pastures were grazed by dairy cows with the exception of P1 that was mechanically harvested. The fourth site (NAT) is located about 10 km W in Cortines (398550 2500 S – 608470 1500 W); some of the present species were: B. catharticus, Festuca arundinacea, L. perenne, Lolium multiflorum, T. repens, T. pratense, Taraxacum officinale and Rumex crispus. This site was selected as a reference because it has been a natural grassland for the last 30 years. At each site, one hectare was sampled using a regular grid with a separation interval of 25 m. The grid was placed 6 m from the

boundary of the site after randomly selecting an initial corner. In each site earthworm community characteristics and fractal parameters of soil structure were assessed in the Autumn of 2000. 2.2. Fractal analysis of soil structure After discarding the upper 5 cm of the soil surface to avoid roots, five undisturbed blocks were sampled randomly at the grid points by cutting 8-cm-side portions with the aid of a shovel and a knife. Each block was impregnated under vacuum with polyester resin following Fitzpatrick (1984). The resin mixture was composed by 53.4% of a polyester resin, 45% of a diluent (styrene monomer), 0.5% of a catalyst (methyl ethyl ketone), 0.1% of a fluorescent dye (Amarillo Oracet 8GF – Ciba Geigy1) and 1% of white pigmented paste. Blocks were placed in a vacuum desiccator, covered with the resin mixture and maintained under 76 mmHg for 1 h. Curing started 48 h after impregnation. A vertical face of each block was cut with a diamond saw, polished with sandpaper of decreasing grit size, and then photographed with a digital camera under ultraviolet light, obtaining a pixel resolution of 70 mm. Image manipulation was made with the ImageJß program (Rasband, 1997–2005). Images were taken in color and converted to gray scale using the 8-bit function. Then, binary images were obtained using the threshold function of the program and comparing visually the original gray level image with the binary obtained; we made in some cases minor adjustments to the threshold automatically selected to improve the segmentation of structures of interest (Gime´nez et al., 1997a) (Fig. 1). Fractal dimensions were calculated in ImageJ with the FracLacß plug-in by the box-counting method (Karperien, 2005). This method estimates the fractal dimension by placing over the image of interest grids of different box sizes (e) and counting the boxes that contain foreground pixels for each grid (N(e)). The fractal dimension was estimated as the slope of the linear regression line between log(e) and log(N(e)). Selected grid sizes ranged from 2 pixels to 45% of the size of the image. The grid position was changed randomly four times and images were characterized with the mean of four measurements of fractal dimensions. Two fractal dimensions were estimated: the mass fractal dimension of the solid component (Dms) and the mass fractal dimension of the pore component (Dmp). 2.3. Earthworm community structure analysis For earthworm community characterization, samples were taken at fifteen randomly selected points of the grid, not necessarily at the same points of the soil blocks previously taken. At each point, soil portions of 25 cm  25 cm  20 cm were handsorted (Zicsi, 1958). Collected earthworms were weighted, counted and preserved. They were identified taxonomically according to Righi (1979). The rarefied expected number of species (E(S)) and Shannon (H0 ) index were estimated as measures of change in species composition across sites. E(S) is a method to obtain the expected number of species if all samples were of the same size. We used the lowest number of individuals recorded as the sample size in the calculation. 39 8 2 N  Ni > > > < X> 6 7= n 7 ¼ Expected number of species;   EðSÞ ¼ 16 4 5> > N > > ; : n where E(S) is the expected number of species, n the sample size selected, N total number of individuals collected, and Ni number of individuals of species i (Hurlbert, 1971).

A. Duhour et al. / Applied Soil Ecology 43 (2009) 83–88


Table 1 Morphological, chemical and physical characteristics of the surface horizon of the studied sites. Site

Thickness (cm)


Organic carbon (%)

Total nitrogen (%)

P1, P2, P3 NAT

32 24

Fine subangular blocky Granular

1.65 1.9

0.170 0.198

Texture Clay (%)

Silt (%)

Sand (%)

26.6 27.4

59.2 60.0

14.2 12.6

Table 2 Land use history of the tree pastures selected at the National University of Luja´n site. Abbreviations and scientific names: PP: permanent pastures, Rg: ryegrass (Lolium sp.), Av: Avena sativa, Mill: Panicum sp., Mz: Zea mays, Sorg: Sorghum bicolor, Moha: Setaria italica. Site


P3 P2 P1













Av/Mill Rg Av/Mz

PP Mz Mz

PP Mz Mz

PP PP Sorg

PP PP Moha



Av/Mz Mz PP

PP Rg Mz

PP PP Rg/Sorg



Ninety five percent confidence intervals were estimated for H0 by the jackknife method, and values with confidence intervals that do not overlap were considered statistically different (Southwood and Henderson, 2000). H0 ¼ 


pi log2 ð pi Þ Shannon index


where pi the number of individuals of species i over the total of individuals of the sample, S the species richness. Taking into account that the Shannon index estimation may vary with sample size, evenness (H0 /Hmax) and a redundancy measure were added for better support to the conclusions (Hurlbert, 1971) R ¼ 1  V; V ¼

where i is the species rank, Ni the number of individuals of rank i, NT the total number of individuals, S is the observed species richness, and k the preemption coefficient, that ranges between 0 and 1, and represents the proportion of resources taken by each species. The relative goodness of fit was evaluated by means of the Akaike Information Criterion (AIC). Model fitting and goodness of fit estimations were made with the ‘vegan’ R-package (Oksanen et al., 2008).

H0  Hmin ¼ Redundancy; Hmax  Hmin

where H0 is the observed value, Hmax was calculated as log2(S), Hmin is the value of H0 if one species was represented by N  (S  1) individuals and the other species by one individual each (Hurlbert, 1971). The community structure may be analyzed using rankabundance curves (Southwood and Henderson, 2000). Models of abundance vs. rank can provide information about the resource partition between species. Several theoretical models have been proposed for rank-abundance curves, especially for communities with few species taxonomically related (guilds), as is the case of earthworm assemblages (Tokeshi, 1990; Fattorini, 2005). Among them, Motomura’s geometric series (GS) and MacArthur’s broken stick (BS) models represent extreme conditions in which a single resource can be distributed (apportioned) among competing species. The relative fitting of the observed distributions to these models can be used to distinguish communities along a perturbation gradient (Fattorini, 2005). The GS model is typical of high dominance communities where resources are hierarchically apportioned whereas the BS model fits better the situation of equitable communities, whose resources are randomly assigned (May, 1975; Tokeshi, 1990; Fattorini, 2005). For this analysis, the logarithm of abundance of each species was plotted against the species rank and the plots were fitted to the abundances predicted by the geometric series and broken stick models: i1

N i ¼ NT kð1  kÞ Ni ¼


S X i¼1

1 i

Geometric series

Broken stick

Fig. 1. One of the images used for fractal dimension calculations from site P2. This image shows pores in black and was used to obtain Dmp, then inverted to calculate Dms.

A. Duhour et al. / Applied Soil Ecology 43 (2009) 83–88


Table 4 Species found at each studied site.

Differences in fractal dimensions, biomass and abundance between sites were determined by the Kruskal–Wallis test (p < 0.05). The Nemenyi non-parametric test of multiple comparisons was performed (p < 0.05) (Zar, 1996). The R program was used for all the analyses (R Development Core Team, 2008). 3. Results 3.1. Fractal analysis The soil solid and pore spaces may be considered as fractals in all sites analyzed (Table 3). Kruskal–Wallis test results are significant among the sites analyzed in both fractal dimensions of solid and pore spaces (Dms: p < 0.03; Dmp: p < 0.03). The mass fractal dimensions of the solid space (Dms) is greater than 1.8 in all sites (range: 1.82–1.86) showing the highest value at the NAT site. We find statistical differences only between the NAT and P2 sites (Table 3). The mass fractal dimensions of the pore system (Dmp) ranged between 1.32 and 1.70 (Table 3). The highest values of Dmp are found at the sites NAT and P3, following a similar trend with the selected perturbation gradient. Sites NAT/P3 and P1 show significant differences in Dmp values.






Aporrectodea caliginosa Aporrectodea rosea Aporrectodea trapezoides Octolasion cyaneum Octolasion tyrtaeum Bimastos parvus Dendrodrilus rubidus Microscolex dubius Microscolex phosphoreus Dichogaster sp. Eukerria saltensis






  * *




Dominant species.

Table 5 AIC values for the goodness of fit of the GS and BS models to the abundance-rank curves in all sites analyzed. For the GS model, values of the coefficient k are presented. Lower values of the AIC (*) indicate better fit to the model. Site

P1 P2 P3 NAT

3.2. Earthworm community structure Earthworms’ biomass vary between 17.8 and 72.1 g/m2, the highest corresponding to the less disturbed pasture site (P3), whose values are significantly different from those of the P2 and P1 sites, whereas biomass from the NAT site is statistically different from the values of the P1 site only. The number of earthworms per square meter varied between 302.9 and 702.9, the highest being the P3 site, with values significantly different from the values of site P2 (Table 3). Eleven species were found in all the study (Table 4). With respect to the origin and ecological category of the species found, those belonging to the Microscolex genus are native and epigeic– endogeic. The rest of the species found are European and North-American introduced by humans in the region and have endogeic habits. The native species Microscolex dubius is dominant in P1 and P3 and codominant in P2 with M. phosphoreus, a native species also. The exotic Aporrectodea caliginosa is dominant in NAT. The most disturbed site (P1) showed an earthworm community with three species and the lowest diversity indexes. The highest richness – nine species – is observed in the less disturbed NAT site (Tables 3 and 4). The E(S) for each site is estimated using the number of individuals recorded in the P1 site and has a similar trend to S and the Shannon index; these measures are lowest in the P1 site.

Geometric series

Broken stick




0.849 0.570 0.650 0.355

13.75* 38.90* 29.79* 55.22*

29.83 58.04 107.28 56.29

Intermediate values are found at P2 and P3 and the highest in NAT (Table 3). The analysis of species abundance vs. rank models shows a better fit to the geometric series model in all sites indicated by the lower AIC obtained (Table 5). However, differences may be observed in the k coefficient – indicator of dominance – of the geometric series model that represents the proportion of resources apportioned by each species. Fig. 2 shows the observed abundance and the GS model fitted to data. 4. Discussion 4.1. Fractal dimensions and soils A fractal dimension is a measure of the degree of occupation that a geometric object makes of the space (Frontier, 1987). For an object embedded in a plane, like photographs of soil sections, the fractal dimension ranges between 1 and 2, the topological dimensions of line and plane, respectively. The solid and pore spaces of the Typic Argiudoll soils analyzed in this work across a disturbance gradient always show a fractal structure. Mass fractal dimensions of both solid and pore spaces are in the range

Table 3 Dms, Dmp, biomass, abundance, observed species richness (S) in samples, estimated species richness (E(S)) and Shannon index (H0 ). Different letters indicate statistically significant differences (p < 0.05). For each measure mean values and standard errors are presented. For H0 jackknife estimations and confidence intervals are presented. P1 Fractal dimensions Dms Dmp Biomass and density Biomass (g/m2) Individuals per m2 Diversity and richness Observed richness (S) E(S) (rarefaction) H0 (jackknife estimate) Evenness Redundancy (R)


1.85  0.01ab 1.43  0.05b 17.8  5.3a 426.7  67.3ab 3 3 0.69  0.25a 0.43 0.78


1.83  0.01b 1.54  0.01ab 27.9  6.1ab 302.9  49.2a 7 6.80  0.43 1.70  0.34b 0.55 0.74


1.84  0.01ab 1.62  0.02a

1.86  0.01a 1.63  0.03a

72.1  1.6c 702.9  110.3b

51.1  7.4bc 536.5  76.1ab

7 5.89  0.75 1.70  0.23b 0.54 0.67

9 8.98  0.12 2.60  0.37c 0.71 0.60

A. Duhour et al. / Applied Soil Ecology 43 (2009) 83–88


Fig. 2. Relative abundance of species (Log) vs. rank. Dots: observed values; line: GS model.

referenced by other authors (Gime´nez et al., 1997b; Anderson et al., 1998). Fractal measures of pore space (Dmp) show a wider range than solid space measures (Dms), this fact coincides with observations reported by Gime´nez et al. (1997b). The pore space (Dmp) follows the same tendency of the pre-selected levels of disturbance. The two calculated fractal measures describe complementary geometrical properties of soil having different responses to perturbation gradient. Moreover, the fractal dimension of soil pores is a measure of the heterogeneity or complexity of the arrangement of pore space, which is the habitat of earthworms (Frontier, 1987). Agricultural practices disrupt macropores and produce compaction that increases contact among aggregates and therefore reduces the proportion of air-filled pores in soils. The higher homogeneity of pore sizes may be related to disruption produced by plowing, and by the reduction of biological activity. Alternatively, the higher Dmp may be related to the addition of biologically generated structures along several years. In consequence, in soils with pastures, structure is recovered through the action of plant roots and macrofauna, which by producing interconnected galleries and channels increase the proportion of macropores (Decae¨ns and Jime´nez, 2002). It is expected that soils with structure formed by large and continuous pores would have greater Dmp values than soils with small and isolated pores (Anderson et al., 1998; Gime´nez et al., 1997a). 4.2. Earthworms, porosity and diversity Among the different soil organisms, earthworms are recognized as one of the main factors in producing macroporosity, and are therefore considered ecosystem engineers (Lavelle, 2002). In consequence, earthworms may be one of the main factors for improving soil porosity in the analyzed sites and could be responsible for the high values in Dmp in the less disturbed soils. Earthworm biomass and abundance are negatively affected by increasing agricultural pressure (Edwards and Bohlen, 1996; Paoletti, 1999). In sites with fertilized pastures, the higher food quality and availability could lead to high earthworm biomass and abundance (Clemente et al., 2003). In this work all sites have high values of biomass and abundance (>17 g/m2 and 300 ind/m2).

Despite the high abundance of earthworms in all sites, the recently established pasture P1 has low species richness with significant dominance of one species. The species richness values of P2 and P3 could be related to a tendency to increase their diversity, approaching communities without cropping activities, such as NAT (Table 2). Under the conditions of this study, in 2–3 years after pasture establishment there is a recovery of a significant number of species. The same trend is observed in the Shannon index that exhibits its maximum value at the NAT site. We find an increase both in the species richness and the community evenness (Table 3). The presence of fractal structure in soil pore and solid system may correlate with the increase of the species richness of earthworms as observed in this work, but it is not clear whether it is because earthworms can ‘see’ more habitat space or because this space is more complex. A more straightforward relationship was previously observed between the fractal nature of soil habitat and the number of individuals of different size (Kampichler, 1999; Finlay and Fenchel, 2001). More investigation is needed for elucidate this topic, exploring the relationship between body size and number of individuals of different earthworm species. The GS model provides a better fit in all situations analyzed, suggesting that resources are hierarchically distributed within the community. The value of the k coefficient in the GS model is a measure of dominance and behaves as complementary of evenness. It shows a trend to increase in the same way that the perturbation level. These results could indicate that one or a few dominant species of earthworms may be responsible of the observed fractal pattern. We might propose that, in general, there is an interaction between soil macrofauna and soil structure in which both were mutually transformed. Once a perturbation ceases, both earthworm community and soil structure increase their complexity. The data suggest a link between Dmp as measure of the pore system complexity and diversity and species richness of earthworms, meaning that a more complex soil environment could favor a more diverse earthworm community. The presence of larger pores and a more continuous pore system may offer niches for more species of earthworms. In effect, we have found more equitable communities and a change in the pattern of dominance in the GS model across the perturbation gradient that could be related to the fractal pattern of soil pore space.


A. Duhour et al. / Applied Soil Ecology 43 (2009) 83–88

For ecosystem engineers such as earthworms, communities with higher species richness may exert a variety of effects in soil pore system complexity that configures the conditions to maintain higher values of the respective fractal dimensions. The disruption of soil structure produced by agricultural practices may affect soil functions, conditioning the development of inhabiting soil fauna. 5. Conclusions In this work we have studied the relationship between earthworm community and fractal dimension of solid and pore components of soil. We have found the higher values of Dmp in sites with the higher diversity and species richness of earthworms, suggesting a trend between the observed biological activity and fractal measures of the habitat. There could be an association between the degree of perturbation of the soil and the activity and complexity of earthworm guild. There is also mutual influence between complexity of the pore system (that offers habitats of different structures) and earthworm activity that modifies soil pore systems changing in turn conditions for other organisms. Changes in these variables have been observed both as variations in the soil fractal structure and of composition, abundance and distribution functions of earthworm communities. The relative weight of biological activity in soil structure, in relation to the presence and abundance of earthworms must be the subject of future experimental studies. Fractal dimensions of soils can be used as good indicators of the perturbation degree because they represent the result of both mechanical disruption effects and biological modification of the pore system. Acknowledgements We want to thank to Dr. Daniel Gime´nez from Rutgers University for his critical reading of the manuscript and his helpful comments. This work was supported by a fellowship from the National Council of Scientific Research (CONICET, Argentina). References Anderson, A.N., Mc Bratney, A.B., Crawford, J.W., 1998. Application of fractals to soil studies. Advances in Agronomy, vol. 63. Academic Press, New York, 1–76. Binet, F., Curmi, P., 1992. Structural effects of Lumbricus terrestris (Oligochaeta: Lumbricidae) on the soil-organic matter system: micromorphological observations and autoradiographs. Soil Biol. Biochem. 24 (12), 1519–1523. Casas, R.R., 2005. Efectos de la intensificacio´n agrı´cola sobre los suelos. CienciaHoy 15 (87), 44–45. Clemente, N.L., Lo´pez, A.N., Vincini, A.M., Castillo, H.A., Carmona, D.M., Manetti, P.L., San Martino, S., 2003. Abundancia de megadrilos (Annelida: Oligochaeta) en diferentes sistemas de produccio´n. Ciencia del Suelo. 21 (2), 35–43. Decae¨ns, T., Jime´nez, J.J., 2002. Earthworm communities under an agricultural intensification gradient in Colombia. Plant Soil 240, 133–143. Decae¨ns, T., Asakawa, N., Galvis, J.H., Thomas, R.J., Ame´zquita, E., 2002. Surface activity or soil ecosystem engineers and soil structure in contrasted land use systems of Colombia. Eur J. Soil Biol. 38, 267–271. Degens, B.P., 1997. The contribution of carbohydrate C and earthworm activity to the water-stable aggregation of a sandy soil. Aust. J. Soil Res. 35, 61–71. Edwards, C.A., Bohlen, P.J., 1996. Biology and Ecology of Earthworms, third ed. Chapman and Hall, London, 426 pp. Falco, L., Momo, F., Craig, E., 1995. Asociaciones de lombrices de tierra y su relacio´n con la cobertura vegetal en suelos forestados de Argentina. Rev. Chil. Hist. Nat. 68, 523–528. Fattorini, S., 2005. A simple method to fit geometric series and broken stick models in community ecology and island biogeography. Acta Oecol. 28, 199–205.

Filgueira, R.R., Fournier, L.L., Sarli, G.O., Arago´n, A., Rawls, W.J., 1999. Sensitivity of fractal parameters of soil aggregates to different management practices in a Phaeozem in central Argentina. Soil Till. Res. 52, 217–222. Finlay, B.J., Fenchel, T., 2001. Protozoan community structure in a fractal soil environment. Protist 152, 203–218. Fitzpatrick, E.A., 1984. Micromorphology of soils. Cambridge University Press, 433 pp. Frontier, S., 1987. Applications of fractal theory to ecology. In: Legendre, P., Legendre, L. (Eds.), Developments in Numerical Ecology, vol. G14. NATO ASI Series, 585 pp. Gime´nez, D., Allmaras, R.R., Nater, E.A., Huggins, D.R., 1997a. Fractal dimensions for volume and surface of interaggregate pores – scale effects. Geoderma 77, 19–38. Gime´nez, D., Perfect, E., Rawls, W.J., Pachepsky, Y., 1997b. Fractal models for predicting soil hydraulic properties: a review. Eng. Geol. 48 (3), 161–183. Gime´nez, D., Allmaras, R.R., Huggins, D.R., Nater, E.A., 1998. Mass, surface, and fragmentation fractal dimensions of soil fragments produced by tillage. Geoderma 86, 261–278. Hurlbert, S.H., 1971. The nonconcept of species diversity: a critique and alternative parameters. Ecology 52, 577–586. Je´gou, D., Brunotte, J., Rogasik, H., Capowiez, Y., Diestel, H., Schrader, S., Cluzeau, D., 2002. Impact of soil compaction on earthworm burrow systems using X-ray computed tomography: preliminary study. Eur. J. Soil Biol. 38 (3–4), 329–336. Joschko, M., Wendroth, O., Rogasik, H., Kotzke, K., 1994. Earthworm activity and functional and morphological characteristics of soil structure. In: Transactions of the 15th World Congress in Soil Science, Acapulco, 4a, pp. 144–163. Kampichler, C., 1999. Fractal concepts in studies of soil fauna. Geoderma. 88, 283–300. Karperien, A., 2005. FracLac Advanced User’s Manual. Charles Sturt University, Australia. Lamande´, M., Hallarie, V., Curmi, P., Pe´re`s, G., Cluzeau, D., 2003. Changes of pore morphology, infiltration and earthworm community in a loamy soil under different agricultural managements. Catena 54 (3), 637–649. Langmaak, M., Scharader, S., Rapp-Bernhardt, U., Kotzke, K., 1999. Quantitative analysis of earthworm burrow systems with respect to biological soil-structure regeneration after soil compaction. Biol. Fert. Soils 28, 219–229. Lavelle, P., 2002. Functional domains in soils. Ecol. Res. 17, 441–450. Manuel-Navarrete, D., Gallopı´n, G.C., Blanco, M., Dı´az-Zorita, M., Ferraro, D., Herzer, H., Laterra, P., Murmis, M., Podesta´, G.P., Rabinovich, J., Satorre, E., Torres, F., Viglizzo, E.F., 2009. Multi-causal and integrated assessment of sustainability: the case of agriculturization in the Argentine Pampas. Environ. Dev. Sustainability 11, 621–638. May, R.M., 1975. Patterns of species abundance and diversity. In: Diamond, M.L., Cody, J.M. (Eds.), Ecology and Evolution of Communities. Harvard University Press, Cambridge, MA, pp. 81–120. Momo, F., Giovanetti, C.M., Malacalza, L., 1993. Relacio´n entre la abundancia de distintas – especies de lombrices de tierra (Annelida, Oligochaeta) y algunos para´metros fı´sico-quı´micos en un suelo tı´pico de la estepa pampeana. Ecologı´a Aust. 3, 7–14. Oksanen J., Kindt R., Legendre P., O’Hara B., Simpson G.L., Stevens M.H.H., 2008. vegan: Community Ecology Package. R package version 1.11-4,, Pachepsky, Y., Crawford, J.W., Rawls, W.J., 1999. Fractals in soil science. Geoderma 88 (special issue). Paoletti, M.G., 1999. The role of earthworms for assessment of sustainability and as bioindicators. Agric. Ecosyst. Environ. 74, 137–155. R Development Core Team, 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3900051-07-0, URL Rasband, W.S., 1997–2005. ImageJ, U.S. National Institutes of Health, Bethesda, Maryland, USA, Righi, G., 1979. Introduccio´n al estudio de las lombrices del suelo (Oligoquetos Megadrilos) de la Provincia de Santa Fe (Argentina). Coleccio´n Climax, Revista de la Asociacio´n de Ciencias Naturales del Litoral. 10, 89–155. ˜ ach, S.S., Le Comber, S.C., 2004. Measure of pocket gopher (Thomomys bottae) Roman burrow geometry: correlates of fractal dimension. J. Zool. Soc. Lond. 262, 399– 403. Southwood, T.R.E., Henderson, P.A., 2000. Ecological methods, 3rd ed. Blackwell Science, Oxford, 592 pp. Teubal, M., 2003. Soja transge´nica y modelo agroalimentario argentino. Realidad Econ. 196, 52–74. Tokeshi, M., 1990. Niche apportionment or random assortment: species abundance patterns revisited. J. Anim. Ecol. 59, 1129–1146. VandenBygaart, A.J., Fox, C.A., Fallow, D.J., Protz, R., 2000. Estimating earthworminfluenced soil structure by morphometric image analysis. Soil Sci. Soc. Am. J. 64, 982–988. Zar, J., 1996. Biostatistical analysis. Prentice-Hall, New Jersey, EEUU, 662 pp. Zicsi, A., 1958. Determination of number and size of sampling unit for estimating lumbricid populations of arable soils. In: Murphy, P.W. (Ed.), Progress in Soil Zoology. Butterworths, London, pp. 68–71.