Effects of fungicides and biofungicides on population density and community structure of soil oribatid mites

Science of the Total Environment 466–467 (2014) 412–420

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Effects of fungicides and biofungicides on population density and community structure of soil oribatid mites Abdel-Naieem I.M. Al-Assiuty a, Mohamed A. Khalil a, Abdel-Wahab A. Ismail b, Nico M. van Straalen c,⁎, Mohamed F. Ageba a a b c

Department of Zoology, Faculty of Science, Tanta University, Tanta, Egypt Plant Pathology Research Institute, Agriculture Research Center, Giza, Egypt Department of Ecological Science, VU University Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands

H I G H L I G H T S • A field study showed that two biofungicides have little side-effects on soil invertebrates. • Chemical fungicides affected density and egg number in a species-specific manner. • Oribatid mites represent a sensitive indicator for effects of pesticides on the soil ecosystem.

a r t i c l e

i n f o

Article history: Received 2 June 2013 Received in revised form 17 July 2013 Accepted 17 July 2013 Available online 10 August 2013 Editor: Damia Barcelo Keywords: Community Density dependence Egg number Fungicides Oribatid mites Soil

a b s t r a c t To compare the side-effects of chemical versus biofungicides on non-target organisms in agricultural soil, a study of population structure, spatial distribution and fecundity of oribatid mites, a diverse and species-rich group of microarthropods indicative of decomposer activity in soil was done. Plots laid out in agricultural fields of a research station in Egypt, were cultivated with cucumber and treated with two chemical fungicides: Ridomil Plus 50% wp (active ingredients = metalaxyl and copper oxychloride) and Dithane M-45 (active ingredient = mancozeb), and two biofungicides: Plant Guard (containing the antagonistic fungus Trichoderma harzianum) and Polyversum (containing the fungi-parasitic oomycete Pythium oligandrum). All treatments were done using both low-volume and high-volume spraying techniques to check whether any effects were dependent on the method of application. Oribatid mite communities were assessed from soil core samples collected during the growing season. Total abundance of oribatids was not different across the plots, but some species decreased in number, while one species increased. Species diversity and community equitability decreased with the application of chemical and biofungicides especially when using high-volume spraying. In control plots most oribatid species showed a significant degree of aggregation, which tended to decrease under fungicide treatment. Ridomil Plus, Plant Guard and Polyversum had a negative effect on the gravid/ungravid ratio of some species. Egg number averaged over the whole adult population was not directly related to the application of chemical and biofungicides but it showed a species-specific relationship with population density. In general biofungicides had a smaller effect on population size and community structure of oribatid mite species than chemical fungicides. The results indicate that biofungicides may be the preferred option when aiming to prevent side-effects on sensitive groups among the species-rich soil detritivore community. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Fungicides, herbicides and insecticides are used in agriculture to maximize crop production; they are credited for increasing food production and helping to protect man and animals against crop loss and diseases. When pesticides are sprayed, depending on the crop's canopy, a significant amount of the dose may reach the soil. Under normal

⁎ Corresponding author. Tel.: +31 20 5987070; fax: +31 20 5987123. E-mail address: [email protected] (N.M. van Straalen). 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.07.063

agricultural practice, pesticide residues will be mixed into the top 15 cm layer of soil, the region of greatest activity of microorganisms and soil fauna (Blasco and Picó, 2009; Van Straalen and Van Rijn, 1998). Most studies of pesticide effects on soil communities have focused on earthworms and Collembola, and few on soil-living mites. This contrasts with their great diversity, abundance and functional significance (Al-Assiuty et al., 1993; Seastedt, 1984). Oribatida (also known as Cryptostigmata) are one of the most numerically dominant arthropod groups in the organic horizons of most soils, where their densities can reach more than a hundred thousand individuals per square meter (Norton, 1990). Oribatida have a great variety of feeding niches

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

413

Fig. 1. Showing the layout of plots used in this study to assess effects of fungicide applications. Please note that length and width are drawn to different scales. Treatments are indicated by capital letters: D = Dithane M-45, R = Ridomil Plus, T = Plant Guard, P = Polyversum, C = control (water only); methods of application are indicated by subscripts: Hv = high volume, applied with a knapsack sprayer and Lv = low volume, applied with a motorized knapsack.

and are actively involved in decomposition of organic matter and in nutrient cycling (Behan-Pelletier, 1999). Most oribatid mites can feed on a variety of fungal species (Maraun et al., 1998; Schneider et al., 2004; Schneider and Maraun, 2005). The spectra of fungi preferred by different species partly overlap and it is still unknown to what extent interspecific differences contribute to the coexistence of many seemingly similar species in a community (Maraun et al., 1998; Schneider and Maraun, 2005). Giving their dependence on fungal and other organic resources in the soil, oribatids will be affected by pesticides not only directly, but also indirectly, through shifts in the below-ground microbial community. Several species have been found to be insensitive to insecticides, e.g., to chlorpyrifos (Stark, 1992). Other mites are positively affected by pesticides and still others are negatively affected, e.g., in response to the natural insecticide azadirachtin (Stark, 1992), copper and p-nitrophenol (Parmelee et al., 1993), and lindane (Scholz-Starke et al., 2013). Yet there are indications that oribatid mites as a group are more sensitive to toxic substances than other soil invertebrates (Streit, 1984; Lebrun and Van Straalen, 1995; Denneman and Van Straalen, 1991; Van Straalen et al., 1989). The goal of the present study was to test whether biofungicides would have less severe side-effects on sensitive soil microarthropods than conventional chemical fungicides. In addition we tested two spraying devices differing in the amount of fungicide product applied per hectare, to investigate whether in addition to type of fungicide, possible side-effects could be minimized by a reduction of spray volume. In this way we aimed to develop the most environment-friendly method for plant protection in Egyptian cucumber culture. We assessed

population structure, spatial distribution and reproductive status of oribatid mites as indicators of side-effects on the soil ecosystem. 2. Materials and methods 2.1. Study site The experiment was carried out on a 24 × 32 m field at the experimental farm of Gemmeiza Agriculture Research Station, 15 km to the north-east of the city of Tanta, Governorate of Gharbia, Egypt (30° 49′ 14.62″ N, 31° 7′ 11.06″ E) in 2005. The soil was a clayey loam, pH 7.4–7.6; average soil temperature varied between 21 and 23 °C; daily average air temperature between 25 and 29 °C in the period of sampling. The field was divided into four equal blocks (I, II, III and IV, each 6 × 32 m), and each block further subdivided into 2 × 8 m plots (Fig. 1). Cucumbers were planted across the entire block with 50 cm distance in between them; a total of 64 plants were growing in each plot. Plant foliage initially covered about 10% of the surface area, increasing to 90% after one and a half month. 2.2. Application of fungicides Two chemical fungicides were used in this study, Ridomil Plus 50% wp (active ingredients = metalaxyl-M and copper oxychloride) and Dithane M-45 80% wp (active ingredient = mancozeb, with manganese and zinc). The two biofungicides were Plant Guard (containing the antagonistic fungus Trichoderma harzianum), and Polyversum (containing the fungi-parasitic oomycete Pythium oligandrum). Both the conventional

Table 1 Active ingredients and application rates of the four fungicides used in this study. Fungicide treatments were repeated six times during the growing season of cucumber (3 months). Fungicide product

Active ingredient

Concentration of a.i. in product

Tank concentration of product

Ridomil Plus wp

Metalaxyl Copper oxychloride Mancozeb, ethylene-bisdithiocarbamate Manganese Zinc Trichoderma harzianum Pythium oligandrum

15% 35% 62% 16% 2% 30 × 106 cells/mL No less than 105 cfu/g

1.5 g/L

Dithane M-45 wp

Plant Guard Polyversum

wp, wettable powder; a.i., active ingredient; cfu, colony-forming units; Lv, low volume; Hv, high volume.

3 g/L

3 mL/L 1.5 g/L

Field application rate of a.i. Lv (475 L/ha)

Hv (950 L/ha)

0.11 kg/ha 0.25 kg/ha 0.88 kg/ha 0.23 kg/ha 0.03 kg/ha 43 × 109 cells/ha 71 × 106 cells/ha

0.21 kg/ha 0.50 kg/ha 1.77 kg/ha 0.46 kg/ha 0.06 kg/ha 86 × 109 cells/ha 143 × 106 cells/ha

414

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

Table 2 List of oribatid mite species and their dominance in plots sprayed with two chemical fungicides, Ridomil Plus and Dithane M-45, throughout the sampling period. Species

Scheloribates laevigatus Scheloribates pallidulus Zygoribatula dermatoglypha Rhysotritia ardua ardua Xylobates capucinus Tectocepheus sarekensis Lamellobates hauseri aegyptica Niloppia sticta Striatoppia papillata Anchipteria aegyptica Total number per m2 (mean ± standard error)

RLv

Control RC No.

%

907.0 32.6 A 381.9 13.7 B 318.3 11.4 B 302.4 10.9 B 779.7 28.0 B – – 31.8 1.1 D – – 31.8 1.1 D 31.8 1.1 D 2785 ± 1301

No.

RHv r%

238.7 10.0 B 413.7 17.7 B 95.5 4.1 D 477.4 20.4 B 1082 46.3 A – – 31.8 1.4 D – – – – – – 2339 ± 1510

No.

Control DC r%

111.4 3.4 D 556.9 17.1 B 63.7 2.0 D 429.6 13. B 2100 64.4 A – – – – – – – – – – 3262 ± 1743

No.

DLv %

1416.2 33.7 A 509.20 12.1 B 302.3 7.2 C 572.6 13.6 B 1178 28.0 B 31.8 0.8 E 31.8 0.7 E 63.7 1.5 D 95.5 2.3 D 4201 – 4201 ± 1711

No.

DHv %

270.0 7.6 C 556.9 15.7 B 270.0 7.61 C 556.9 15.7 B 1814 51.1 A 16.0 0.45 E 31.8 0.9 E 31.8 0.9 E – – 3547 – 3547 ± 1630

No.

%

206.9 3.9 D 668.3 12.6 B 238.60 4.49 D 700.2 13.1 B 3437 64.7 A – – 31.8 0.6 E 31.8 0.6 E – – 5315 – 5315 ± 2106

Plot designations: R sprayed with Ridomil Plus and D: sprayed with Dithane M-45; subscripts indicate type of spraying, Lv: low volume and Hv: high volume. Two sets of control plots (RC and DC) were sprayed with water only. No.: number per meter squared, %: relative contribution in community; dominance class is indicated by capital letters, A eudominant: over 30% of individuals, B dominant: 30–10% of individuals, C sub-dominant: 10–5% of individuals, D minor: 5–1% of individuals, and E rare: less than 1% of individuals.

fungicides and the biofungicides are normally directed against downy mildew and other fungal diseases in cucumber, at rates recommended by the Gemmeiza Agriculture Research Station in Egypt, shown in Table 1. Two spraying devices were used namely, (a) a knapsack sprayer, providing a high spray volume (Hv) with an easy-action pump handle and a metal hand lance tube fitted with a hollow-cone nozzle, which produced relatively large spray droplets (98 droplets/cm2 of 198 μm diameter), delivering 950 L/ha (1.5 L/plot), and (b) a motorized knapsack sprayer (Lv), which used a 2 hp, 2-stroke, 35 cm3 engine, operating at 6000 rpm and providing an air speed at outlet of about 75 m/s, producing small spray droplets (163 droplets/cm2 of 145 μm in diameter), delivering pesticide at low volume, 475 L/ha (0.75 L/plot). The low volume application allowed a saving of about 50% in terms of fungicide product compared to the Hv sprayer. Still the fungicidal action is assumed to be comparable, because low volume spray concentrates the fungicide on plant foliage and decreases the amount lost to the soil. Consequently, low volume spray reduces environmental pollution while also decreasing costs (Moustafa, 2003). Each of the four fungicides was applied using both spraying devices, allowing a complete comparison over all combinations. In accordance with agricultural practice, fungicide sprays were applied twice a month during the growing season (3 months). 2.3. Sampling and microarthropod enumeration Each of the 48 plots was sampled twice a month throughout the growing season from March to June, till the crop was full-grown (a total of six times). Eight random soil cores were taken per plot 7 days after spraying with fungicide, using a cylindrical metal corer

(10 cm diameter, 78.5 cm2 surface area, 8 cm depth). The total area of soil sampled per plot per sampling occasion was 0.0268 m2; this figure was used to convert counts of mites to densities per m2. Peripheral parts of the plots were excluded from sampling to minimize possible edge effects. Oribatid mites were extracted from these soil cores using Tullgren-type extraction funnels. Preparation and identification of species were undertaken as explained in detail elsewhere (Al-Assiuty et al., 1993). For the most common oribatid species reproductive potential was assessed by noting the number of eggs in the body. Eggs were easily visible since all mites were cleared and mounted on microscopic slides to allow reliable species identification. 2.4. Data analysis Temporal trends in oribatid densities were insignificant and ignored in this study; to minimize the occurrence of zeroes for the rare species, and to avoid pseudoreplication by repeated sampling from the same plot, the data for the eight cores from each plot were pooled over the six sampling occasions and considered a single observation per plot; this leaves four true replicates (plots) for each of the 12 treatments (cf. Fig. 1). Differences in the number of individuals per plot were tested using a Kruskal–Wallis test. A χ2 test was used to compare the fraction of “gravid” individuals (containing one or more eggs) between treatments (Sokal and Rohlf, 1995). Similarity between mite communities exposed to different treatments was assessed using the Cλ index of Morishita (1959). Community diversity was measured using Simpson's diversity index (Di), equitability (E, an index sensitive to changes in dominance structure), the Shannon-Wiener index (H), and Shannon equitability (J), an index sensitive to changes in rare species (Pielou,

Table 3 List of oribatid mite species and their dominance in plots sprayed with two biofungicides, Plant Guard and Polyversum, throughout the sampling period. Species

Scheloribates laevigatus Scheloribates pallidulus Zygoribatula dermatoglypha Rhysotritia ardua ardua Xylobates capucinus Tectocepheus sarekensis Lamellobates hauseri aegyptica Niloppia sticta Striatoppia papillata Total number per m2 (mean ± standard error)

TLv

Control TC No.

%

652.4 16.6 B 461.5 11.7 B 270.5 6.9 C 747.9 19.0 B 1591 40.5 A 31.8 0.8 E 31.8 0.8 E 31.8 0.8 E 111.4 2.8 D 3930 ± 1939

No.

THv %

477.4 13.6 B 397.8 11.3 B 143.2 4.1 D 541.0 15.4 B 1798 51.1 A 31.8 0.9 E – – 31.8 0.9 E 95.5 2.7 D 3517 ± 1681

No.

Control PC %

525.1 12.4 B 318.3 7.5 C 111.4 2.6 D 461.5 10.9 C 2705 63.9 A – – – – 31.8 0.8 E 79.6 1.9 D 4233 ± 1906

No.

PLv %

827.5 18.6 B 572.9 12.9 B 191.0 4.3 D 716.1 16.1 B 1957 44.1 A 31.8 0.7 E 31.8 0.7 E 31.8 0.7 E 79.6 1.8 D 4440 ± 2271

No.

PHv %

700.2 16.9 B 397.8 9.6 C 127.3 3.1 D 636.5 15.4 B 2067 50.0 A 31.8 0.8 E 31.8 0.8 E 31.8 0.8 E 111.4 2.7 D 4137 ± 2102

No.

%

827.5 11.7 B 350.1 4.9 D 111.4 1.6 D 668.3 9.4 C 4662 65.8 A 15.9 0.2 E 63.7 0.9 E 143.2 2.0 D 238.7 3.4 D 7081 ± 3617

Plot designations: T sprayed with Plant Guard and P sprayed with Polyversum; subscripts indicate type of spraying, Lv: low volume and Hv: high volume. Two sets of control plots (TC and PC) were sprayed with water only. No.: number per meter squared, %: relative contribution in community; dominance class is indicated by capital letters, A eudominant: over 30% of individuals, B dominant: 30–10% of individuals, C sub-dominant: 10–5% of individuals, D minor: 5–1% of individuals, and E rare: less than 1% of individuals.

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

415

Fig. 2 (continued).

1974). Spatial clustering of species was assessed using the index λ, defined as λ = s/√x, where s is the standard deviation and x is the arithmetic mean density. Significance of λ was tested using χ2 with n − 1 degrees of freedom where χ2 = (n − 1)λ2 (Debauche, 1962). The relationship between egg number and population density was studied by plotting the data in a correlation diagram and testing the significance of Pearson's correlation coefficient (Sokal and Rohlf, 1995). The relative dominance of each species was classified according to the dominance classes of Engelmann (1978): (A) eudominant: species comprising over 30% of the total number of individuals, (B) dominant: 30–10% of individuals, (C) sub-dominant: 10–5% of individuals, (D) minor: 5–1% of individuals and (E) rare: less than 1% of the total number. 3. Results 3.1. Density and species composition

Fig. 2. Average population densities of the five most common oribatid species: (a) Scheloribates laevigatus, (b) Scheloribates pallidulus, (c) Zygoribatula dermatoglypha, (d) Rhysotritia ardua ardua, (e) Xylobates capucinus, and (f) the total number of oribatids, as a function of fungicide treatments using low-volume and high-volume applications. Error bars indicate standard errors of the mean.

Ten species of oribatid mites belonging to nine genera were extracted from the investigated plots (Tables 2, 3). Mean density ( X  SE ) of total oribatid mites in the control plots varied from 2785 ± 1301 to 4440 ± 2271 individuals/m2. Scheloribates laevigatus, Scheloribates pallidulus, Rhysotritia ardua ardua and Xylobates capucinus were dominant species in the control plots, contributing more than 10% to the total number of individuals, cf. Table 2. Zygoribatula dermatoglypha was subdominant in the control plots, while Anchipteria aegyptica, Tectocepheus sarekensis, Lamellobates hauseri aegyptica and Niloppia sticta were recorded as minor or rare species in all control and treated plots. Mean densities of the total oribatids in plots treated with chemical fungicides (Ridomil Plus and Dithane M-45) were generally lower than the control when the low-volume sprayer device was used but higher than the control when using the high volume spray (Tables 2, 3). The

416

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

same trend was valid for the biofungicide-treated plots (Plant Guard and Polyversum). However, these differences were not statistically significant when applying a Kruskal–Wallis test to the total number of individuals (Table 4). Focusing on the densities of separate species, significant effects of chemical fungicides were shown in various cases (Fig. 2). The application of the fungicide Ridomil Plus, using both low-volume (Lv) and high-volume (Hv) devices, had a significant negative effect (p b 0.05) on the densities of S. laevigatus and Z. dermatoglypha. Application of Ridomil using a Hv sprayer, caused a significant increase (p b 0.05) in the density of X. capucinus which became eudominant (64% of total abundance) in comparison with 28% and 46% in the control and the low-volume treatment, respectively (Table 2, Fig. 2). In plots treated with Dithane M-45 at low volume, S. laevigatus decreased and became subdominant (8% compared with 34% in the control plot) (Table 2) Moreover, it became a minor species in the case of Hv application. The species S. pallidulus, Z. dermatoglypha and R. a. ardua showed a high degree of constancy maintaining the same degree of dominance as in the control. X. capucinus was the only oribatid species that showed an increase in its abundance; it represented about 51% of the total oribatid individuals in the Lv treatment. The oribatid mite densities showed no significant response to Plant Guard, using both the high volume and the low volume sprayers. (Table 3, Fig. 2). In plots treated with Polyversum at low volume, changes were likewise small. X. capucinus maintained the same position, and its relative contribution increased to 50%, while S. laevigatus, and R. a. ardua were still recorded as dominant species (16.9% and 14.5%, respectively). S. pallidulus became subdominant (9.6%). The densities of these three species showed a decreasing trend, but this was not statistically significant, compared to the control. Under the Hv treatment, oribatids showed variable responses to Polyversum. X. capucinus increased in density and represented the only eudominant species in these plots with a relative contribution of 65% (Table 3). S. laevigatus, Z. dermatoglypha, T. sarekensis and L. hauseri aegyptica had the same degree of dominance (0.7%) as in the control plots. S. pallidulus and R. a. ardua became minor and subdominant species (4.9% and 9.4, respectively). Table 4 shows a summary of the effects, using the Kruskal–Wallis test applied to the five most dominant species. Obviously the chemical fungicides have species-specific effects, where decreases of Z. dermatoglypha and S. laevigatus densities are compensated by an increase in X. capucinus. In addition, no negative effects are seen for the biofungicides (Fig. 2).

(1.637, 1.634, 1.577 and 1.546 in the four controls). Species diversity reached its lowest value in the RHv plot (Ridomil, high volume).

3.2. Effects on species diversity

4. Discussion

Two diversity indices (Simpson's D and Shannon's H′) were used to detect and evaluate the effect of fungicide application on the community level. Fig. 3 shows that species diversity and equitability in all cases tended to decrease in comparison to the controls, with an intermediate effect in the low-volume fungicide treatments, and the strongest effect in the high-volume treatments. Shannon's diversity index ranged between 1.044 and 1.637 decits and its evenness ranged between 0.552 and 0.787. The highest diversity (H′) was recorded in the control plots

Our results show that the effect of fungicides on the abundance and species diversity of oribatid mites varies greatly depending on taxon and fungicide as well as on the mode of application. Most likely, these effects are not due to direct toxicity of fungicides to the mites, but they are mediated by changes in the fungal community of the soil (as an essential food item for soil microarthropods). Some authors have attempted to assess the community structure of fungi along with microarthropod densities in pesticide impact studies (Scholz-Starke

3.3. Effects on community similarity The similarity between the communities across treatments was higher in the biofungicide-treated plots than in the chemical fungicide treated plots (Table 5). The similarity between PC and PLv plots was the highest of all (Cλ = 0.992) while the lowest degree of similarity was recorded between RC and RHv (Cλ = 0.670). This indicates that biofungicides had a smaller effect on community structure than chemical fungicides and that the strongest effect was due to Ridomil Plus at high volume. 3.4. Effects on spatial distribution The aggregation patterns of the oribatids responded in different ways to chemical and biofungicide application. All species showed a significant aggregation in the control plots which tended to decrease in the plots treated with fungicides towards a random distribution in some but not all species (data not shown). X. capucinus responded in the reverse direction: it tended to become more aggregated in the treatments. This was also the only species that increased in numbers due to the fungicides (Fig. 2). Over all species, a decrease of aggregation was generally correlated with a decrease of density. 3.5. Effects on egg number Fig. 4 shows the correlation between egg number per individual and oribatid mite abundance. When viewed over all 12 treatments, a positive correlation (r = 0.791, p b 0.05) was observed in the case of S. laevigatus. However, the other species showed no clear correlation between abundance and egg number (p N 0.05). Fig. 5 shows the rate of egg-containing individuals expressed as gravid to ungravid ratio for five species in each treatment. Chi-square tests for the ratio between mites with eggs and without eggs showed that Ridomil Plus, Plant Guard and Polyversum had a negative effect on the proportion of eggcontaining individuals in S. laevigatus, while Dithane M-45 had no significant effect. The data show that S. laevigatus and Z. dermatoglypha were the species that responded most strongly to the application of fungicides, while S. pallidulus, R. a. ardua and X. capucinus seemed to be less affected.

Table 4 Kruskal–Wallis tests for differences in the number of mites per sample between treated plots and control plots for five selected oribatid mite species found in all four blocks. Plots sprayed with Ridomil Plus

Plots sprayed with Dithane M-45

Plots sprayed with Plant Guard

Plots sprayed with Polyversum

Species

H

P

H

P

H

P

H

P

Scheloribates laevigatus Scheloribates pallidulus Zygoribatula dermatoglypha Rhysotritia ardua ardua Xylobates capucinus Total

11.792 0.651 7.100 1.244 5.108 1.504

0.003* 0.772 n.s. 0.029* 0.357 n.s. 0.078 n.s. 0.471 n.s.

8.517 0.097 0.137 0.792 6.065 1.241

0.014* 0.952 n.s. 0.934 n.s. 0.673 n.s. 0.048* 0.538 n.s.

0.158 0.220 1.655 0.776 1.633 0.561

0.924 n.s. 0.896 n.s. 0.437 n.s. 0.678 n.s. 0.422 n.s. 0.755 n.s.

0.650 0.063 0.172 0.017 6.046 3.818

0.723 n.s. 0.969 n.s. 0.917 n.s. 0.992 n.s. 0.040* 0.148 n.s.

The data are based on two spraying methods and a control. H, Kruskal–Wallis test statistic (2 degrees of freedom), P, level of significance.

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

417

Fig. 3. Simpson's D (■) and Shannon's H (▲) diversity indices, and their associated evenness (△) and equitability indices (□), calculated for oribatid mite communities treated with (A) Ridomil Plus, (B) Dithane 45-M, (C) Plant Guard, and (D) Polyversum.

et al., 2013), but causal relationships between the occurrence of fungal species and specific microarthropods are difficult to establish. In addition to microbial communities, knowledge of feeding habits of the microarthropods is also crucial to understand the species-specific responses (Al-Assiuty et al., 1993; Al-Assiuty and Khalil, 1995), as is knowledge on the direct toxicity of the active ingredients. Metalaxyl, copper and mancozeb are known to be toxic to earthworms (De Silva et al., 2010; Xu et al., 2011), but their toxicity to oribatid mites has, as far as we know, never been tested. To conclude, our data are the outcome of complex interactions between microbial change, feeding preference and direct toxicity. Generally, the two chemical synthetic fungicides, Ridomil Plus and Dithane M-45 had a significant influence on the community structure of oribatid mites in comparison with the biofungicides Plant Guard and Polyversum. However, this influence varied from negative effects on species such as S. laevigatus and Z. dermatoglypha, to positive effects on X. capucinus. In addition, S. pallidulus and R. a. ardua were not affected at all by the application of any of these fungicides. Consequently, there were no effects on the total density of oribatids. Similar species-

specific responses were seen in insecticide and herbicide studies. For example, Al-Assiuty and Khalil (1995) found that multiple application of insecticides (profenofos, chlorofluazuron, fenvalerate) had a significant negative effect on the abundance of S. laevigatus and Zygoribatula exarata, while R. a. ardua appeared to be tolerant species to the treatments and was not influenced either by the kind of insecticide nor by its dose. In a study on the herbicide atrazin a similar conclusion was drawn (Sabatini et al., 1979). The fact that the same oribatid genera appear to be responsive (Scheloribates, Zygoribatula) or non-responsive (Rhysotritia) to both fungicides and insecticides seems to indicate a general syndrome of species-specific sensitivity or robustness towards habitat disturbance. The increased total abundance of mites in plots treated with highvolume spray was due to the high abundance of X. capucinus, an insensitive species which reached about 66% of the total abundance of the oribatid mites. A tendency for high dominance and low local diversity is also seen in other microarthropod communities. For example the species Carabodes willmanni, an oribatid mite living inside the thallus of Cladonia growing on healthy soil in Brittany, North-

Table 5 Similarity indices (Cλ) of the oribatid mite communities between differently treated plots. Comparison

Cλ

Comparison

Cλ

Comparison

Cλ

Comparison

Cλ

RLv vs RC RHv vs RC RLv vs RHv

0.812 0.670 0.945

DLv vs DC DHv vs DC DLv vs DHv

0.777 0.674 0.972

TLv vs TC THv vs TC TLv vs THv

0.975 0.903 0.972

PLv vs PC PHv vs PC PLv vs PHv

0.992 0.913 0.956

Plot designations: R sprayed with Ridomil Plus, D: sprayed with Dithane M-45, T sprayed with Plant Guard, and P sprayed with Polyversum; subscripts indicate type of spraying, Lv: low volume and Hv: high volume. Four sets of control plots (RC, DC, TC and PC) were sprayed with water only.

418

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

Fig. 4. Relationship between population density and egg number for five oribatid mite species sampled from cucumber fields: (A) Scheloribates laevigatus, (B) Scheloribates pallidulus, (C) Zygoribatula dermatoglypha, (D) Rhysotritia ardua ardua and (E) Xylobates capucinus.

West France (Bellido, 1990) comprised more than 95% of the whole microarthropod community in this habitat. A low local diversity of oribatid communities was also reported for Pinus silvestris soils which were dominated by a single species, Tectocepheus velatus (Schenker, 1984). Beta diversity is an important aspect of oribatid mite communities, implying that only large-scale surveys can reveal the complete species richness of a particular region (Zaitsev et al., 2013). Our study showed that species diversity and equitability indices were higher in control plots than in treated plots. In other words, the community structure of oribatid mites became more biased towards specific species under fungicide application. Studies documenting the effects of pesticides or soil pollution on oribatid mite communities have shown that the changes in abundance and dominance underlying total density and species richness may be quite complicated (Khalil et al., 2003, 2009; Weigmann, 1984; Weigmann and Kratz, 1987). There seems to be a large variety in responses to soil factors among oribatids, and the overall effects can only be understood in terms of shifts

among individual species (Weigmann and Kratz, 1987; Wallwork, 1983; Hågvar, 1984). Many soil invertebrates tend to show a heterogenous spatial distribution and mites are no exception. Both exogenous and endogenous factors may cause clumped distributions. Our results in the fungicidefree plots showed that the five common oribatid species have indeed a tendency to aggregate. Earlier research has shown a similar pattern of aggregation in soil-living mites (Al-Assiuty and Khalil, 1995). In our study, most of the species had an aggregated distribution in the controls and tended to move towards a more random distribution in the treated plots, however, one species responded in the reverse direction: it tended to become more aggregated in the treatments. This was also the only species that increased in numbers due to the fungicides. This supports the common view that aggregation in soil invertebrates is tightly linked to population dynamics (Ettema and Wardle, 2002). A remarkable result from our study was that Ridomil Plus, Plant Guard and Polyversum, but not Dithane M-45, had a negative effect on the rate of gravid individuals of S. laevigatus. S. pallidulus,

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

419

Fig. 5. Gravid to ungravid ratios of the five oribatid mite species sampled from cucumber fields: (A) Scheloribates laevigatus, (B) Scheloribates pallidulus, (C) Zygoribatula dermatoglypha, (D) Rhysotritia ardua ardua and (E) Xylobates capucinus in relation to chemical and biofungicide application.

Z. dermatoglypha and X. capucinus did not show such a response, while R. a. ardua showed a negative response only towards the application of the biofungicide Polyversum. These species-specific responses may relate to the feeding habits of the mites. Gut content analysis has shown that S. laevigatus is a euryphagic species (Khalil et al., 2011). It feeds on a large number of resources such as fungal hyphae, multicellular fungal spores, and unicellular fungal spores. In this way, it can easily profit from the quality of the habitat if this contains a large array of food items and so there is a positive correlation between abundance and egg number. Consequently, when its habitat is deteriorated by fungicides, the egg number decreases. Other oribatids, such as Xylobates lophotricus have a more narrow diet, causing fierce intraspecific competition. At low density, these animals may be relieved from competition resulting in an increase of fertility; in such species the correlation between egg number and density was shown to be negative rather than

positive (Al-Assiuty et al., 1993; Khalil et al., 2011). This pattern was also expected for S. pallidulus, another species with a narrow diet, but in the present study it showed a non-significant correlation. In general, increased reproductive output with decreasing density may act as a compensation mechanism for toxicity and is often observed in population studies with toxicants, e.g. by Noël et al. (2006). Our data suggest that such effects will be strongest in species with narrow diets that are limited by interspecific competition at high densities. It must be emphasized that the number of eggs, averaged over the whole adult population, does not directly relate to fertility. The same number of eggs may be produced and laid in a short time, or they may be carried for a long period and laid when conditions are favorable (Cortet et al., 2002; Stamou and Sgardelis, 1989). In addition, the average egg number is influenced by the sex ratio (Luxton, 1981). Interspecific differences in egg number may also be due to temporal patterns

420

A.I.M. Al-Assiuty et al. / Science of the Total Environment 466–467 (2014) 412–420

in reproductive activity differing between species, as well as other aspects of reproduction biology. In summary, our study showed that biofungicides had a smaller effect on community structure of oribatid mite species than chemical fungicides. Especially Plant Guard did not have any significant density effects at all, while Polyversum caused an increase in one of the species. There were no differences between the spraying methods for these biofungicides. Therefore, we argue that biofungicides may be an environment-friendly alternative to chemical fungicides, when side-effects on soil-living communities are to be avoided. Conflict of interest The authors declare that there is no conflict of interest with respect to this paper. Acknowledgments The authors are grateful to three anonymous reviewers that commented upon an earlier version of the manuscript. References Al-Assiuty AIM, Khalil MA. The influence of insecticide-pheromone substitution on the abundance and distributional pattern of soil oribatid mites. Exp Appl Acarol 1995;19: 399–410. Al-Assiuty AIM, Bayoumi BM, Khalil MA, Van Straalen NM. Egg number and abundance of ten Egyptian oribatid mite species (Acari: Cryptostigmata) in relation to habitat quality. Eur J Soil Biol 1993;29:59–65. Behan-Pelletier VM. Oribatid mite biodiversity in agroecosystems: role for bioindication. Agric Ecosyst Environ 1999;74:411–23. Bellido A. Caractéristiques biodémographiques d'un acarine oribate (Carabodes willmanni) des pelouses xérophilles. Can J Zool 1990;68:2221–9. Blasco C, Picó Y. Prospects for combining chemical and biological methods for integrated environmental assessment. Trends Anal Chem 2009;28:745–57. Cortet J, Ronce D, Poinsot-Balanguer N, Beaufreton C, Chabert A, Viaux P, Cancela de Fonseca JP. Impacts of different agricultural practices on the biodiversity of microarthropod communities in arable crop systems. Eur J Soil Biol 2002;38:239–44. De Silva PMCS, Pathiratne A, Van Gestel CAM. Toxicity of chlorpyrifos, carbofuran, mancozeb and their formulations to the tropical earthworm Perionyx excavatus. Appl Soil Ecol 2010;44:56–60. Debauche HR. The structural analysis of animal communities of the soil. In: Murphy P, editor. Progress in soil zoology. London: Butterworths; 1962. p. 10–25. Denneman CAJ, Van Straalen NM. The toxicity of lead and copper in reproduction tests using the oribatid mite Platynothrus peltifer. Pedobiologia 1991;35:305–11. Engelmann HD. Zur Dominanzklassifizierung von Bodenarthropoden. Pedobiologia 1978;18:378–80. Ettema C, Wardle DA. Spatial soil ecology. TREE 2002;17:177–83. Hågvar S. Six common mite species (Acari) in Norwegian coniferous forest soils. Relations to vegetation types and soil characteristics. Pedobiologia 1984;27:355–64. Khalil MA, Al-Assiuty AIM, Abdel-Lateif HM, Abd-Allah SM. The effect of motor exhausts on the diversity and abundance of soil oribatid mite communities in roadside verge soil. Egypt J Appl Sci 2003;18(12B):776–92. Khalil MA, Janssens TKS, Berg MP, Van Straalen NM. Identification of metal-responsive oribatid mites in a comparative survey of polluted soils. Pedobiologia 2009;52:207–21.

Khalil MA, Al-Assiuty AIM, Van Straalen NM. Egg number varies with population density; a study of three oribatid mite species in orchard habitats in Egypt. Acarologia 2011;51: 251–8. Lebrun P, Van Straalen NM. Oribatid mites: prospects for their use in ecotoxicology. Exp Appl Acarol 1995;19:361–79. Luxton M. Studies on the oribatid mites of a Danish beech wood soil IV. Developmental biology. Pedobiologia 1981;21:312–40. Maraun M, Migge S, Schaefer M, Scheu S. Selection of microfungal food by six oribatid mite species (Oribatida, Acari) from two different beech forests. Pedobiologia 1998;42:232–40. Morishita M. Measuring of interspecific association and similarity between communities. Mem Fac Sci Kyushu Univ Ser E (Biol) 1959;3:65–80. Moustafa MSH. Effect of applicator-types on fungicide efficiency and environmental pollution. Proc. 11th Annual Conference of the Misr Society of Agricultural Engineering; 2003. p. 735–46. Noël HL, Hopkin SP, Hutchinson TH, Williams TD, Sibly RM. Towards a population ecology of stressed environments: the effects of zinc on the springtail Folsomia candida. J Appl Ecol 2006;43:325–32. Norton RA. Acarina: Oribatida. In: Dindal DL, editor. Soil biology guide. New York: Wiley; 1990. p. 779–803. Parmelee RW, Wentsel RS, Phillips CT, Simini M, Checkai RT. Soil microcosm for testing the effects of chemical pollutants on soil fauna communities and trophic structure. Environ Toxicol Chem 1993;12:1477–86. Pielou EC. Population and community ecology. Principles and methods. New York: Gordon & Breach; 1974. Sabatini MA, Pederzoli A, Fratello B, Bertolani R. Microarthropod communities in soil treated with atrazine. Boll Zool 1979;46:333–41. Schenker R. Spatial and seasonal distribution patterns of oribatid mites (Acari: Oribatida) in a forest soil ecosystem. Pedobiologia 1984;27:133–49. Schneider K, Maraun M. Feeding preferences among dark pigmented fungal taxa (Dematiacea) indicate limited trophic niche differentiation of oribatid mites (Oribatida, Acari). Pedobiologia 2005;49:61–7. Schneider K, Migge S, Norton RA, Scheu S, Langel R, Reineking A, Maraun M. Trophic niche differentiation in soil microarthropods (Oribatida, Acari): evidence from stable isotope ratios (15N/14N). Soil Biol Biochem 2004;36:1769–74. Scholz-Starke B, Beylich A, Moser T, Nikolakis A, Rumpler N, Schäffer A, Theißen B, Toschki A, Roß-Nickoll M. The response of soil organism communities to the application of the insecticide lindane in terrestrial model ecosystems. Ecotoxicology 2013;22:339–62. Seastedt TR. The role of microarthropods in decomposition and mineralization processes. Ann Rev Entomol 1984;29:25–46. Sokal RR, Rohlf FJ. Biometry. The principles and practices of statistics in biological research 3rd ed. New York: WH Freeman and Company; 1995. Stamou GP, Sgardelis SP. Seasonal distribution patterns of oribatid mites (Acari: Cryptostigmata) in forest ecosystem. J Anim Ecol 1989;58:893–904. Stark JD. Comparison of the impact of a neem seed-kernel extract formulation ‘Margosan-O’ and chlorpyrifos on non-target invertebrates inhabiting turf grass. Pestic Sci 1992;36:293–9. Streit B. Effect of high copper concentrations on soil invertebrates (earthworms and oribatid mites): experimental results and a model. Oecologia (Berlin) 1984;64: 381–8. Van Straalen NM, Van Rijn JP. Ecotoxicological risk assessment of soil fauna recovery from pesticide application. Rev Environ Contam Toxicol 1998;154:83–141. Van Straalen NM, Schobben JHM, De Goede RGM. Population consequences of cadmium toxicity in soil microarthropods. Ecotoxicol Environ Saf 1989;17:190–204. Wallwork JA. Oribatids in forest ecosystems. Ann Rev Entomol 1983;28:109–30. Weigmann G. Structure of oribatid mite communities in the soil of urban areas. Acarology 1984;6:917–23. Weigmann G, Kratz W. Oribatid mites in urban zones of West Berlin. Biol Fertil Soils 1987;3:81–4. Xu P, Diao J, Liu D, Zhou. Enantioselective bioaccumulation and toxic effects of metalaxyl in earthworm Eisenia foetida. Chemosphere 2011;83:1074–9. Zaitsev AS, Van Straalen NM, Berg MP. Landscape geological age explains large scale spatial trends in oribatid mite diversity. Landsc Ecol 2013;28:285–96.