Introduction

Compared to conventional farming, organic soil management can substantially improve soil structure, especially in terms of organic matter, soil aggregate stability and soil pore characteristics. The beneficial effects of organic farming are said to include better recycling of organic waste, reduced use of synthetic chemicals, reduced incidence of pests and diseases, improvement of plant and animal quality, reduced erosion and run-off, as well as ecological benefits, e.g. greater soil organism diversity (Hole et al. 2005; Papadopoulos et al. 2014). Under intensive conventional farming, soil physical properties can become severely degraded (Conacher and Conacher 1995). The benefits to soil physical properties of adding organic matter and applying good management practices are well documented (Russell and Isbell 1986; Hamblin and Kyneur 1993; Nesbitt and Adl 2014). Soil pH, C:N ratio, particle distribution, bulk density, and soil moisture are often used as indicators for such changes. For example, an enhancement in such soil physical properties was found in an 18-year biodynamic dairy farm in Victoria (Australia) compared to a conventional farm (Lytton-Hitchins et al. 1994). A study in New Zealand by Reganold et al. (1993) supported these findings. However, the improvements are conditional on soil type, climate, regional factors, past land use, types rates of soil amendments and/or practices (Papadopoulos et al. 2014).

In addition to effects on soil physical properties organic management may also support a higher abundance of soil microbes and soil animals. The abundance and diversity of soil animals is often adversely affected by conventional agricultural practices, such as clearing and cultivation (Abbott et al. 1979; Roper and Gupta 1995), and by the use of chemical fertilizers (Marshall 1977) and pesticides (Jackson 1983). Increases in soil biomass and biological abundance, diversity and activity are directly related to increased levels of organic matter and good management practices which, in turn, positively influence soil structure, nutrient cycling and availability, buffering capacity, and pest and disease control in conventional systems (Dick 1992; Pankhurst et al.1997). In semi-arid environments, the role of ants and termites (Lobry de Bruyn and Conacher 1990, 1995) and the importance of litter and organic matter (Spain and Huston 1983) should not be underestimated. Under organic or alternative systems, soil organism abundance and activity have often been found to be superior to those in their conventional counterparts (Fraser et al. 1988; Ryan et al. 1994). However, Daniel et al. (1994) and Penfold (1994) were not able to demonstrate any marked differences between the systems, possibly due to a short period of sampling as well as other factors.

The main goal of the present study was to test the expectation that conversion from conventional to organic farming has a positive effect on the abundance and community structure of soil oribatid mites. We pay particular attention to (1) the responses of the most dominant oribatid species, and (2) the relationships between soil microbial and physicochemical parameters versus oribatid mite abundance that would be valid over both managed cultivation systems.

Materials and methods

Sampling and site description

The experiment ran from October 2012 till March 2013. The experimental station was at Abshway village, Al-Fayoum Governorate, North of Cairo, Egypt. The experimental field (about 2.5 ha) had been cultivated with a variety of crops, including clover, wheat, maize, sunflower and cotton for more than 30 years, in a conventional cropping system. Since 2004, organic farming was established in half of the area (1.25 ha). The application of compost was performed manually every year since 2004 at day 15 before sowing. The added quantity was equivalent to 28.5 m3 per ha. In each field [organic managed system (OMS) and conventional managed system (CMS)], three independent plots (each 40 × 40 m) were laid out with 25 m between them. During sampling, the two fields (OMS and CMS) had uniform vegetation (clover, Trifolium alexandrinum). The experimental organic field was designated as Zafer, code number AO 16 and registered in the Egyptian Center of Organic Agriculture ECOA with international accreditation from EU DAKKAS. Minimum tillage was applied under organic management, however all conventional fields were ploughed. Weed control was performed mechanically in OMS; however, in CMS two herbicides were used: Basagran (active ingredient: bentazon, emulsifiable concentrate 48 %, from BASF), and Fusilade (a.i.: fluazifop-P-butyl, EC 12.5 %, from Syngenta). Herbicide treatments were applied 3× to each plot.

To assess abundance of soil microarthropods, 15 random soil subsamples were collected monthly from each plot (in total 90 samples a month from the two fields).The peripheral edges of the plots were excluded from the sampling scheme to minimize possible edge effects. Sampling was conducted by means of a rectangular metal sampler (10 × 10 cm, driven in the soil to a depth of 12 cm). Soil microarthropod fauna was extracted from the samples using modified Tullgren funnels as described elsewhere (Al-Assiuty et al. 1993). Under the microscope, oribatid mites were separated from other microarthropods, preserved and identified according to Ghilarov and Krivolutsky (1975), Balogh and Balogh (1992) and Subías and Arillo (2001).

Physicochemical and microbial analyses

The analyses of physicochemical and microbial parameters were carried out on ten fresh replicates of subsamples. Before mite extraction from soil samples of each treatment, 10 g fresh soil was taken from each of the field samples during full canopy closure (45 per three plots) then mixed well and we choose ten samples from soil mixture of each treatment to be analyzed.

Organic matter content (OM %) was determined using a muffle furnace at 600 °C for 4 h. The loss in soil sample weight represents the weight of the total organic content which was estimated as a percentage of the dry weight of soil sample. Soil moisture (SM  %) was calculated as water loss/soil dry weight. Carbon (C  %) and nitrogen (N  %) in soil samples were analyzed on a Termo Finnigan Flash EA 1112 Series NC analyzer to determine total C and N. The pH values of the CaCl2-extracts were measured for all soil samples using a Consort p907 pH meter. Pore volume (PV) was calculated using Keen–Raczkowski Box methods used by Sankaram (1966).

The following microbial parameters were measured: microbial biomass (MB, using the fumigation-extraction method according to Joergensen, 1996), fungal density (FD, colony forming units, CFU per ml), relative contribution of fungal species (rc,  % of total number of isolates), fungal species richness (FDI, number of species), number of dominant fungal species (FDS, species representing 5 % or more of the total number of isolates) and fungal species preferred by oribatid mites (PF, identified from other studies, e.g. Afifi, 2010). Fungi were isolated according to Malloch (1997), Warcup (1985), Paul and Clark (1988), and Swanson et al. (1992). One gram from each soil sample was transferred to a sterile test tube containing 10 ml of sterilized distilled water (1:10 dilution), using a flame-sterilized steel spatula. Subsequently, serial solutions were prepared. One ml of each dilution was spread on Petri dishes (9 cm diameter) containing well-dried potato-dextrose agar (PDA). The plates were incubated at 28 ± 2 °C for 3–5 days. Colonies of fungi that developed on the PDA plates were counted as CFU. Discrete fungal colonies were sub-cultured onto fresh medium containing PDA at 50 mg/L (Hansen 1926). Isolated fungi were identified and recorded according to Gilman (1957), Domsch et al. (1980) and Moubasher (1993).

Data analysis

Population densities of oribatid mites were expressed as numbers per m2. Dominance classification was evaluated according to Engelmann (1978). Concentration of dominance was assessed according to Odum (1971) using C′ = ∑(ni/N)2, where ni is the number of individuals per species and N is the total number of all species. Similarities between the treatments were assessed according to Magurran (2004) using Qs = 2 J/a + b as a qualitative measure (where J is the number of shared species in both fields, a is the number of species in field A and b is the number of species in field B) and CN = 2jN/(aN + bN) as a quantitative measure (aN is the number of individuals in field A, bN is the number of individuals in field B and jN is the sum of the lower of the two abundances of shared species in the two fields).

Species diversity values were evaluated using the Shannon–Wiener function (H′) using H = −∑PiInPi and E = H′/InS, where H′ is the diversity index, E is the equitability, Pi is the number of individuals in the i-th species and S is the total number of species. Statistical comparison of Shannon–Wiener function (H′) values in two communities could be made using the t test according to Jayaraman (1999) where t = (H′1–H′2)/S H′1 − H′2, where S H′1 − H′2 is the square root of variance \( \left( {{\text{S}}^{2}_{\text{H}} } \right) \) of the Shannon–Wiener index for each community. To express the linear relationships between OM content and PV, MB and C  %, regression analysis was applied using the linear regression equation Y′ = a + bX where b indicates the average changes in dependent variables for each change of one unit in the independent variables.

A two–way repeated measures ANOVA was applied with management (two levels) as a fixed block factor, time (6 levels), as a repeat-measure fixed factor, and plot as random replicates (IBM SPSS v.19). To fulfill criteria of normal distribution and homogeneity of variances the abundance data were \( \sqrt {\left( {x + 1} \right)} \) transformed. T test was applied to compare the data of physicochemical and microbial analysis after testing the data for normality (Kolmogorov–Smirnov test, software Min tab v.16).

Results

Physicochemical and microbial analysis

Table 1 shows that except for the C/N ratio, all physicochemical and microbial variables differed between the organic management system (OMS) and the conventional management system (CMS). Organic matter content and organic carbon varied considerably between OMS and CMS plots (4.7 and 1.7 % compared to 2.9 and 1.3 %, respectively.(Total nitrogen tended to be greater in OMS than in CMS. Soil pH was slightly alkaline (7.4) in OMS plots and was significantly higher in CMS plots where pH was around 7.7. Soil moisture content was low at the time of sampling, but there was still a statistically significant difference between CMS and OMS plots. Pore volume of the soil was significantly greater in the organic plots than in conventional plots (Fig. 1a), and there was a positive correlation between pore volume and organic matter content in OMS (r = 0.894, t = 5.631, p < 0.05); however, in CMS, this correlation was not significant (r = 0.598, t = 2.109, p > 0.05). Microbial biomass was significantly greater in OMS than in CMS (Table 1). The higher microbial biomass in OMS is consistent with its higher organic carbon content (Fig. 1b); there was a positive correlation between microbial biomass and organic carbon (r = 0.813, t = 3.950, p < 0.05) in OMS, but not in the case of CMS (r = 0.393, t = 1.212, p > 0.05). The ratio of microbial biomass over organic carbon was considerably higher in OMS (6.6) than in CMS (3.9).

Table 1 Overview of mean (±SE, n = 10) physicochemical and microbial data for the two management systems, organic (OMS) and conventional (CMS)
Full size table
Fig. 1
figure 1

Linear regression a between organic matter (%) and pore volume (%) and b between microbial biomass (mg/g) and carbon content (%). OMS organic management systems, CMS conventional management systems

Full size image

Our survey of the fungal communities isolated from soil samples (Table 2) showed that the genus Aspergillus and the species Trichoderma viride dominated in both plots, whereas Rhizoctonia solani, Alterrnaria alternate and Cladosporium cladosporioides were found to be more abundant in OMS. Curvularia lunatus occurred with high frequency in CMS plots. Fungal density (CFU) was significantly higher in OMS than in CMS (Table 1). Likewise, the fungal species diversity of the two community systems was higher in OMS than in CMS (t = 11.119, p < 0.001).

Table 2 Fungal density (colony forming units per ml), relative frequencies (percentage of total number of isolates) and qualitative and quantitative similarities of fungal communities in the organically managed system (OMS) and conventionally managed system (CMS)
Full size table

The qualitative similarity (relative species composition, rsc) of the fungal community between the two managed systems (CMS and OMS) was high (79 %); however, the quantitative similarity between them was relatively low (38.2 %). Accordingly, the difference between the fungal communities resides in the general abundance of fungi rather than in species composition of the community.

Mite abundance and community structure

The total population density of soil oribatids varied widely (3080–7253 individuals per m2) between the two management systems (Fig. 2). A two-way repeated measures ANOVA indicated that land management had a highly significant effect on mite abundance (F1,17 = 58.60, p < 0.001). The analysis also revealed a significant effect of sampling time (F5,13 = 19.72, p < 0.001), and a significant interaction between management and sampling time (F5,13 = 14.23, p < 0.001). These significances are illustrated by Fig. 2, which shows that under both management systems the number of mites reached a peak in December, but the fluctuations in time did not follow the same pattern, explaining the significant interaction between time and management.

Fig. 2
figure 2

Monthly mean (±SE, n = 45) density (no. per m2) for total oribatid mite species (adults) under the two management systems

Full size image

A total of 13 species were counted from the organically managed soils through all sampling dates, whereas 17 species were collected from the conventionally managed system. Scheloribates laevigatus was most frequently found (in 86 and 80 % of the samples) and was most abundant (constituting 30 and 61 % of the total number of individuals) in organically and conventionally managed systems, respectively. With respect to dominance classification (Table 3), the oribatid fauna from organic plots had one species, S. laevigatus, classified as eudominant (30.6 % of the total number of individuals) and three species classified as dominant, Lamellobates h. aegyptica (28.3 %), Aegyptogalumna mastigophora (17.7 %) and Tectocepheus v. velatus (16.5 %). In the conventionally managed system S. laevigatus was also eudominant, with a high contribution of individuals (54 %), whereas only one species, Zygoribatula exarata, was subdominant (9.6 %). Thus, the degree of dominance in CMS was higher than that in OMS (0.32 vs. 0.23). The qualitative similarity of community composition (pooled over all months) between the two managed systems (OMS and CMS) was 80 %. This contrasts with the quantitative resemblance between them being only 41.4 %.

Table 3 Oribatid species and their abundance (no. m−2), relative contribution (rc,  %), species diversity and qualitative and quantitative similarities throughout the period of study in the organically managed system (OMS) and conventionally managed system (CMS)
Full size table

The most commonly encountered oribatid species were Rhysotritia a. ardua, Epilohmannia c. cylindrica, Oppiella nova, T. v. velatus, L. h. aegyptica, S. laevigatus, Scheloribates pallidilus, Z. exarata and A. mastigophora. A substantial amount of variation was observed among the oribatid fauna within the plots, but the most important source of variation was between the two fields, i.e., between the organic and conventional managed systems.Three trends in the abundances of the most common oribatid species associated with management changes could be distinguished: (1) more abundant in CMS than in OMS: E. c. cylindrica, O. nova, Z. exarata and S. pallidilus; (2) more abundant in OMS than in CMS: T. v. velatus, L. h. aegyptica, S. laevigatus and A. mastigophora; and (3) with a time-dependent response, not consistent over the sampling times, overall indifferent to management: R. a. ardua and Xylobates capucinus.

In contrast to the absolute numbers, species diversity of the communities differed between the two management systems (t = 4.179, p < 0.001). Figure 3 shows that Shannon–Wiener diversity (H′) in conventional plots was higher than in organic plots only in the first 2 months, October and November; thereafter the diversities were more or less indistinguishable. The H′ values in CMS fluctuated more strongly over time than H′ in OMS, which indicated a rather steady value over all sampling times.

Fig. 3
figure 3

Monthly changes in species diversity of oribatid mite communities under the two management systems

Full size image

Discussion

Our experimental field design was not optimal because treatments were not replicated over several blocks. On the other hand we repeated the sampling 6 × and added information through analyzing physicochemical and microbial habitat characteristics. In comparing the two fields, it has to be assumed that the differences between them should be attributed to the differences in long-term agricultural practices (Holland et al. 1994). Indeed, large-scale experiments are difficult to replicate. Our study demonstrated that organic farming exerts a significant effect on various physicochemical soil variables, most notably an enhancement of soil organic matter and soil organic carbon content. Both mites and fungi profited from the greater abundance of resources available in organically managed soils. However, this positive effect was not observed for species diversity, or in mites and fungi. The differences between organically management systems and conventional systems were mainly quantitative, whereas the similarity of fungi and oribatid community structure remained large.

Microbial biomass, fungal density, soil moisture and pore volume seem to be the most important factors which promote a higher abundance of mites. Our results are consistent with those of Foissner (1992), Maraun et al. (1998), Coulis et al. (2013) and Al-Assiuty (2014) who found that soil moisture content and microbial biomass were the most important factors affecting the oribatid abundance.

The main difference between the organic and conventional regimes was that under OMS, tillage before planting was minimal and no herbicides were used, whereas CMS fields were ploughed and received two herbicides (Basagran and Fusilade). The lack of herbicide use in OMS led to an increase in weed cover which possibly created a favorable microclimate, e.g. by maintaining a higher humidity in the uppermost layer of soil (Moreby et al. 1994). The abundance and diversity of predatory arthropods within a field has been shown to be closely related to the nature of the surrounding vegetation (Wratten and Thomas 1990; Wratten and Van Emden 1995). Therefore, the frequent application of herbicides in conventional farming systems can reduce plant diversity within the whole agro-ecosystem, and this may reduce populations of beneficial arthropods.

The increase in oribatid mite abundance under organic farming was greatest for those species that already had a high abundance, e.g. T. v. velatus, A. mastigophora, L. h. aegyptica and S. laevigatus. These species are known for their wide ecological amplitude and apparently profit most from increased availability of resources. In contrasts to the main trend, three species of mite showed a lower abundance in the organically farmed field compared to conventional management, viz. E. c. cylindrica, O. nova and Z. exarata. This could be due to competitive pressure and inter-specific differences in sensitivity. Khalil et al. (2009) have indicated that sensitive oribatid species may be replaced by resistant ones without changing the total oribatid mites.

Various studies have reported effects of organic farming on soil-living communities. Several studies demonstrate a higher diversity of certain groups of invertebrates in organic than in conventionally managed fields, e.g. carabid beetles (Kromp 1999; Langmaack et al. 2001), spiders (Feber et al. 1998) and earthworms (Paoletti, 1999b). However, other studies indicate no difference in diversity under organic farming, e.g. Berry et al. (1996) and Blackburn and Arthur (2001) for centipede communities. Physical disturbance of soil, such as tillage, usually has a detrimental effect on the diversity of the soil fauna, especially earthworms (Altieri 1999). However, negative effects of intensive cultivation have also been reported for microarthropods. Alvarez et al. (2001) indicated that the pre-drilling differences in crop management had a significant effect on soil Collembola abundance under an integrated and conventional regime. Soil mite populations were reduced by deep ploughing and the use of heavy machinery (Osler et al. 2008).

Hole et al. (2005) reviewed ten studies that investigated arthropod taxa associated with organic farming, three focusing exclusively on arthropods. Overall the results of all ten studies suggested that organic farming promoted a greater abundance and diversity of arthropods than conventional systems. Minor and Norton (2004) found that compost had a greater effect on mite diversity than chemical fertilizer although both treatments reduced the diversity. Organic farming can lead to higher populations and species diversity of beneficial arthropods (Dritschilo and Wanner 1980; Hokkanen and Holopainen 1986; Kromp 1989; Booij and Noorlander 1992; Moreby et al. 1994).

The generally positive effects of organic farming on soil-living invertebrates are not always easily demonstrable in the studies reported in the literature. This may be due to the fact that population dynamics of soil organisms in agricultural systems depend on a range of factors, such as soil characteristics, climate, type of crop, etc. (Debeljak et al. 2007). Organic farming typically uses long rotations, low tillage, cover crops, and manure or compost amendments, whereas synthetic fertilizers and pesticides are not used. Ecological responses of soil fauna may depend on a combination of several such interacting factors rather than on a single aspect of the management.

A consistently higher abundance of soil animals may also have positive effects on soil functioning. For this reason, soil animals are frequently used as bioindicators for assessing the effect of landscape management and soil quality (Paoletti 1999a). Teuben and Roelofsma (1990) indicated that the functioning of microorganisms was influenced by abiotic factors, substrate quality and biotic components such as soil arthropods. Soil animals change the physical properties of detritus and the chemical quality by gut passage (Gunnarson and Tunlid 1986), selective grazing (Newell 1984) and dispersal of microbial propagules (Visser, 1985). Also, Maraun et al. (1998) demonstrated that oribatid mites accelerate the recovery of the microbial community and are of significant importance for the resilience of the decomposition subsystem. Azarbad et al. (2013) observed a pollution-induced decrease in microbial activity which resulted from both decreased microbial biomass and negative alterations in the metabolism of bacteria and fungi.

In conclusion, the two managed ecological systems showed a clear influence on the population dynamics of oribatid mite species; there was a significantly higher mean number of oribatid mites in the organic system than in the conventional system. This was accompanied by a general increase of soil microbial biomass and fungal abundance. Our study showed that a more sustainable use of agricultural soils may contribute to the protection of a healthy decomposer community, with long-term ecological benefits.