当前位置:首页 >> 农学 >>

6 Collembola and mites in plots fertilised with different types of green manure


Pedo biologia, 44, 556–566 (2000) ? Urban & Fischer Verlag http://www.urbanfischer.de/journals/pedo

Collembola and mites in plots fertilised with different types of green manure
J?rgen Aagaard Axelsen1 and Kristian Thorup Kristensen2
National Environmental Research Institute, Dpt. of Terrestrial Ecology, Vejlsoevej 25, DK 8600 Silkeborg, Denmark 2 Danish Institute of Agricultural Sciences, Dept. Fruit, Vegetable and Food Science, P.O. Box 102, DK 5792 Aarslev, Denmark
1

Accepted: 22. Februray 2000

Summary
The spring and summer abundance of soil living mites and Collembola was investigated in organically grown field plots which had been covered with either the nitrogen catch crops winter rye, hairy vetch, fodder radish and a control (stubble). The catch crops were incorporated in the soil shortly before sowing of spring barley with undersown clover grass. Microarthropods were extracted from 10 cm deep soil samples and were taken in May, June and August. The densities of both microarthropods were extremely high with up to 120,000 Collembola m-2 and 90,000 mites m-2. The highest densities of Collembola were found in the plots with fodder radish as a catch crop, and the most abundant species were Tullbergia sp., Isotoma notabilis and Folsomia fimetaria. The mite fauna consisted mainly of Mesostigmatic and Prostigmatic mites and was more abundant in the catch crop plots than in the control plots in early June. The input of organic matter from the catch crops is supposed to be part of the reason for the high microarthropod densities, but the barley with undersown clover grass may also play a role. Key words: Collembola, mites, green manure, catch crop, soil fauna, organic farming

Corresponding author: J?rgen Aagaard Axelsen
0031–4056/00/44/05–556 $ 12.00/0

Green manuring and Collembola and mites

557

Introduction
There has recently been a strongly increased consumer interest for organically grown farm products in Denmark, and many consumers are willing to pay more for such products. The reason for this is twofold: firstly, the consumers want to get food without pesticide residues, and secondly, to support a more “natural” agriculture. The latter means that the consumers expect there to be more natural life, including soil fauna, in organically grown fields than in conventionally grown fields. It is difficult to make good comparisons between the two farming types, because the two types often differ in their choice of crop rotation. However, a few strong comparisons have been made revealing an increase in the number and biomass of earthworms in organic and biodynamic fields compared to conventionally grown fields. The comparison was made in fields with the same crop placed in identical crop rotations, thus, the differences were only due to the type of farming (Pfiffner & Mader 1997). Several authors have found higher microbial biomass in organic soils than in conventional soils, but the differences were attributed to management strategies and crop rotations used in organic farming (Bolton et al. 1985; Fraser et al. 1988; Heinonen-Tanski 1990). Similar effects can be expected for Collembola and some mites, because they generally feed on microorganisms. However, so far nobody has found proof for higher densities of Collembola and mites in soils from organically driven farms, but the higher incidence of clover-grass, which has a high density of Collembola and mites (Krogh 1994), will most likely increase the average density of these species in organic farms. Presently, most Danish organic farms are dairy farms where clovergrass leys, farmyard manure and slurry add substantial amounts of organic matter to the soil, and thereby substrate to the soil microflora and microfauna. With a further development of organic farming, more farms with a low stocking-rate or even without any animal production at all will be grown organically. On such farms there are little or no animal manure, and clovergrass leys do no fit naturally into the crop rotations. Green manures and nitrogen catch crops will often be grown on such farms to secure the N supply and to keep N losses low. This will constitute a main difference between organic and conventional farms with few or no animals. Many of the green manures and catch crops will be grown only in the autumn after the harvest of a main crop, i.e. during a short season with relatively unfavourable growing conditions. Still, autumn grown green manures and catch crops have been found to reduce nitrogen losses significantly, which secures a higher N supply for succeeding crops (Thorup-Kristensen 1994; Thorup-Kristensen & Bertelsen 1996), and they can be assumed to have many other effects on the soil. The application of organic matter to agricultural fields increases the densities of almost all members of the detritus food web, including Collembola and mites (Primentel & Warneke 1989). Further, plant cover during autumn and winter gives a protective soil cover which can also be expected to improve the living conditions for the soil microfauna (Wardle 1995). Consequently, it can be expected that growing and incorporating nitrogen catch crops or green manures will increase the densities of these groups in the soil. The aim of this study was to investigate whether: 1 the density of Collembola and mites are affected by growing green manure and catch crops. 2 there are significant differences in the effect of different species of catch crop in their impact on these microarthropod groups.

558

J?rgen Aagaard Axelsen and Kristian Thorup Kristensen

Materials and Methods
Field experiment. The experimental area was located at the Research Centre Aarslev (10°27’E, 55°18’N), on a Typic Agrudalf soil containing 1.7 % C, 0.17 % N, 15 % clay, 27 % silt and 55 % sand in the plough layer. The soil pH was 7.0. The experimental design was a randomised complete block, with three replicates. The catch crop plots were 5 by 10 metres. The catch crop species were winter rye (Secale cereale), fodder radish (Raphanus sativus) and hairy vetch (Vicia villosa). Apart from these species, a control treatment without catch crops was included. The experiment was established within an organic crop rotation on 31 July 1996. The catch crops were grown after the harvest of green peas. The pea residues were rotovated into the soil, and the soil was ploughed before establishment of the catch crops. The fallow plot was kept free of vegetation by aid of a few superficial harrowings during the autumn. The catch crops were rotovated into the soil (down to 15 cm depth) in late March the subsequent spring, and spring barley with undersown clover grass were sown around 20 April. Samples of the catch crops were taken in mid November and late March, right before catch crop incorporation. Plant samples were taken by cutting the catch crops just below ground. The plant material was washed to remove the attached soil. At each sampling plant material from one m2 was sampled. At the first sampling date in November, root material was sampled by excavating a soil block of 0.12 by 0.60 m, and 0.2 m deep. Roots were washed from this soil block. The root and aboveground plant materials were air dried at 80 °C for 20 hours.

Microarthropod sampling
Soil samples were taken with a metal core sampler to a depth of 10 cm. The soil samples were transferred directly into two plastic tubes with a height of 5 cm and a diameter of 6 cm. Thus, it was possible to keep the two depths 0–5 cm and 5–10 cm separate. The tubes were covered with plastic lids during transportation. The microarthropods were extracted from the soil cores over a week in a high gradient extractor. The extraction started at a temperature of 30°C and the temperature was increased stepwise to 60°C. The animals were stored in glycerin until identification. Collembola were identified to species level according to Fjellberg (1980), and mites to groups. Samples were taken three times during the growth season in 1997, viz. 16 May, 4 July and 12 August. Three samples (from both horizons) were taken from each of the three replicates of a treatment, i.e. 9 samples per treatment from each horizon on every sampling day.

Statistical analysis
All statistical analyses for differences between treatments were performed by analysis of variance (one way) when the data could not be rejected as fitting to a normal distribution. Tests for normality were both Shapiro-Wilks test, standardised skewness test, standardised kurtosis test and a goodness-of-fit test. If normality of the data were rejected by one of these tests, the nonparametric Kruskal-Wallis test was used instead. When the Kruskal-Wallis test showed a significant treatment effect a box and whisker plot was used to identify which treatments differed from each other. In case of ANOVA the Multiple range test was used for this purpose

Results
Fodder radish produced much more organic matter during the autumn, and took up more total N than the two other crops (Table 1), though rye had the highest root production followed by hairy vetch. The N concentration was highest in hairy vetch, fol-

Green manuring and Collembola and mites

559

Table 1. The shoot dry matter in November and March and the root dry matter in March of the three green manure species Shoot DM (Mg ha-1) N kg ha-1 DM (Mg ha-1) % in DM November 1996 1.68 4.09 1.91 5.06 4.75 3.04 0.55*** 0.49*** 1.03 0.72 0.41 0.77* 2.51 3.23 2.16 0.27*** N % in kg ha-1 DM March 1997 3.34 89 3.74 87 2.66 48 0.57** 28***

Rye Hairy vetch Fodder radish LSD Root Rye Hairy vetch Fodder radish LSD

69 97 145 29** 26 23 9 16*

2.66 2.39 1.61 0.63*

lowed by rye and the lowest concentration was found in fodder radish. Fodder radish did not survive the winter, and at the time of incorporation in the spring, it had lost much of its biomass and N content. The two other crops survived the winter, and at incorporation time rye and hairy vetch showed the highest biomass and N content. The total density (0–10 cm) of both Collembola and mites was extraordinarily high in all plots but especially in the green manure plots where densities up to 118,000 Collembola and 89,000 mites per m-2 where found. The density of Collembola was significantly dependent on the treatment (P < 0.001) with the fodder radish plots having higher densities on the 4 June than both the two other green manure types and the

Fig. 1. The mean densities of Collembola on the three sampling dates in the four treatments. The vertical bars show the standard error on the mean. Results that are significantly different, are marked with different letters

560

J?rgen Aagaard Axelsen and Kristian Thorup Kristensen

Fig. 2. The mean densities of mites on the three sampling dates in the four treatments. The vertical bars show the standard error on the mean. Results that are significantly different, are marked with different letters

fallow plots. On 12 August the Collembola density in the fodder radish plots only was significantly higher than the fallow plot (P < 0.005). The densities in the other green manure plots were higher than in the fallow plots but the differences were not significant (Fig.1). In addition, the density of mites was significantly affected by the treatment by being lower in the fallow plot than in the green manured plots. On the first sampling date the mite density in the fodder radish plots were significantly higher than in both the hairy vetch and fallow plots (P < 0.001). On the 4 June there were no significant differences between the green manured plots, but the rye and hairy vetch plots had higher densities than the fallow plots (P < 0.05), while there were no significant differences on the 12 August (Fig. 2). It is remarkable that in the green manured plots the strongest increase in density of both Collembola and mites occur between 16 May and 4 July, while the strong increase in the fallow plots are delayed and occur between 4 June and 12 August. The same trend is seen for the mites, where a steep increase is seen between 16 May and 4 June and a decrease in density is observed between 4 June and 12 August in the green manured plots. In the fallow plot there is a continuous increase throughout the growth season, and all treatments reached the same level on 12 August. The Collembola were distributed on 26 species and the mites on four groups (Table 2). Five species of Collembola (Tullbergia spp., Isotoma notabilis, Isotoma anglicana (0–5 cm), Willemia sp. (5–10 cm) and Folsomia fimetaria) were dominating with the two smaller ones Tullbergia spp. and Isotoma notabilis as the most dominating ones (Fig. 3). On the first sampling date in the middle of May the only treatment difference on the Collembola species level was a significantly lower density of I. notabilis in the fallow plots in both soil layers 0–5 cm (P < 0.05) and 5–10 cm (P < 0.05) (Fig. 3a + b). This difference was still found on 4 June (P < 0.001 in both layers), but the most remarkable result from the 4 June sampling was the significantly higher density of F. fimetaria in fodder radish (P < 0.001 in upper layer and P < 0.01 in lower

Green manuring and Collembola and mites

561

Table 2. The total numbers of the Collembola species and mite groups Date 16 May Depth 0–5 cm 5–10 Mites: Prostigmatic 291 217 Mesostigmatic 302 136 Astigmatic 131 80 Cryptostigmatic 45 37 Total Mites 769 470 Collembola: Tullbergia spp. 265 211 Isotoma notabilis 127 72 Folsomia fimetaria 77 224 Isotoma anglicana 27 1 Willemia sp. 35 45 Sminthuridae 26 10 Pseudosinella alba 0 0 Entomobrya sp. 0 0 Onychiurus sp. 11 13 Smithurinus elegans 1 2 Neelus minutus 0 1 Lepidocyrthus lanuginosus 0 0 Lepidocyrtus cyaneus 0 0 Smithurinus viridis 2 0 Pseudosinella decipiens 0 0 Isotomurus palustris 0 0 Hypogastrura sp. 0 0 Folsomia quadriculata 0 0 Tullbergia quadrispina 0 0 Total Collembola 571 579 4 June 0–5 cm 5–10 4128 343 61 10 4542 542 585 388 10 82 147 21 4 9 5 9 1 1 4 0 1 0 3 0 1812 1593 339 57 32 2021 1206 689 1387 1 280 26 15 2 35 0 14 0 2 0 3 0 0 0 0 3660 12 August 0–5 cm 5–10 2576 255 4 2 2837 1160 1090 182 751 87 13 141 79 6 53 17 38 26 14 8 9 3 0 0 3677 1495 240 0 2 1737 2436 421 325 5 113 5 32 7 8 10 17 2 3 1 8 1 2 0 1 3397 Total

10300 1615 333 128 12376 5820 2984 2583 795 642 227 209 92 82 71 58 41 32 21 19 11 5 3 1 13696

layer) (Fig. 3c+d). On 12 August, the density of F. fimetaria had decreased considerably and there were no significant treatment effects. Instead, the density of Tullbergia spp. has continued the increasing trend in the fodder radish plots (Fig. 3e+f). The mites belonged mainly to the two groups Prostigmatic and Mesostigmatic mites with the former being clearly the most abundant one (Table 2). On the first sampling date there was a significant treatment effect on the densities of Prostigmatic mites with hairy vetch and fallow plots having lower densities than the two other treatments (P < 0.05). However, in the 5–10 cm soil layer the density in the winter rye plots was not significantly different from the other plots (Fig 4a +b). The significant

562

J?rgen Aagaard Axelsen and Kristian Thorup Kristensen

treatment effects on the Prostigmatic mites almost vanished on the following sampling dates, the only exception being a significantly higher density in the hairy vetch plots than in both the fodder radish and the fallow plots in the lower soil layer on the 4 June (P < 0.05) (Fig. 4c). When the density of Mesostigmatic mites is concerned, there was a significantly lower density in the hairy vetch and fallow plots than in the fodder radish plots in the 5–10 cm horizon on the first sampling date (P < 0.05). This difference was not found on the second sampling date, and on the third sampling date, there was significantly higher numbers in the hairy vetch plots than in the fodder radish and fallow plots (P < 0.05). There were no significant effects of the treatments on the Mesostigmatic mites in the upper 5 cm horizon at any of the sampling dates.

Fig. 3. The mean densities of the four most abundant Collembola species in the four treatments in the two sampling horizons 0–5 cm and 5–10 cm. a + b: 16 May, c + d: 4 June and e + f: 12 August. The vertical bars show the standard error on the mean. Note different Y-axes. Results that are significantly different, are marked with different letters

Green manuring and Collembola and mites

563

Fig. 4. The mean densities of the two most abundant mite groups in the four treatments in the two sampling horizons 0–5 cm (a, c, e) and 5–10 cm (b, d, f). a + b: 16 May, c + d: 4 June and e + f: 12 August. The vertical bars show the standard error on the mean. Note different Y-axes. Results that are significantly different are marked with different letters

564

J?rgen Aagaard Axelsen and Kristian Thorup Kristensen

Discussion
The densities of both Collembola and mites were surprisingly high compared to the results from other investigations of the microarthropod fauna in agricultural soils. Thus, Filser (1995) found no more than 10,000 Collembola m-2 in green manured hops, Andersen et al., (in prep) found up to 20,000 Collembola m-2 from organically managed fields and Krogh (1994) found an average density of 16,900 Collembola m-2 and 15,700 mites m-2 in an organic cropping system. However, Pimentel and Warneke (1989) mention a few higher records of microarthropod densities from systems with extremely high inputs of organic matter. In comparing the figures from this investigation with data from the literature it must be taken into consideration that only the upper 10 cm of the soil was sampled, and the green manure was incorporated into the upper 15 cm. The occurrence of higher densities in the green manured plots compared to the fallow plots may have two reasons. Firstly, input of organic matter is known to enhance the population development of these animals (Pimentel & Warneke 1989). Furthermore, Filser (1995) observed between two and three times as many Collembola in green manured fields compared to fields fertilised with mineral fertiliser on sandy soil, but the densities were much lower than the densities in the present study. Secondly, the weed control in the fallow plots during the autumn which is likely to have reduced the density of animals, since soil tillage is known to harm the animals (Wardle 1995). It is common practice to carry out mechanical weed control on fallow fields in organic farming, while the soil with catch crops is, of course, left in peace. Therefore, the observed differences are likely to be representative for the situation in real life organic farming. The very high densities of animals in summer are most likely due to both the input of organic matter from the green manures grown in the previous autumn and the effect of a dense cover of the undersown clover grass during the sampling period. A cover of living mulches is known to affect the soil fauna positively (Wardle 1995). This is consistent with the observed development in the populations; after the green manures, the population density was high already in the spring, but in the control plots the population was lower in the spring but increased as the plant cover developed during the growing season. The difference in densities of mites and Collembola between the three types of green manure may be due to the fact that fodder radish dies during winter. This is supported by Scholte & Lootsma (1998) who investigated the number of Collembola and nematodes in fields where the soil had been amended by three different catch crops and found clear differences in Collembola densities between the catch crop types. The white mustard (Sinapis alba) plots had the highest density of Collembola. Thus, it is tempting to conclude that the quality of green manured soils as a habitat for the mesofauna seems to depend on whether the green manure dies during winter or not. However, both white mustard and fodder radish are crucifers and may contain rather high levels of mustard oils, glucosinolates and sulphur. These factors may also play a role. The difference in Collembola density between the fodder radish plot and the hairy vetch and winter rye plots was due to differences in one species only, namely F. fimetaria. Thus, it is tempting to assume that fodder radish supports the growth of fungi species, which are very suitable as food for F. fimetaria, and that this resource is depleted during the summer causing the subsequent dramatic decrease in the F. fimetaria density. This idea is strongly supported by (Lootsma & Scholte 1997) who added

Green manuring and Collembola and mites

565

dried fodder rape material and F. fimetaria to the soil in bioassay experiments with suppression of Rhizoctonia stem canker on potatoes. In this experiment they found a clear stimulatory effect of fodder rape on the F. fimetaria population. Furthermore, (Lootsma & Scholte 1997, 1998) found a suppression of Rhizoctonia stem canker by F. fimetaria. Thus, it appears that the growth of fodder rape as catch crop/green manure could have a positive effect on the suppression of plant pathogenic fungi, at least on Rhizoctonia stem canker. This is in accordance with several other studies, where crucifer crops have been found to reduce the incidence of soil borne diseases more efficiently than other plant species (Chan & Close 1987; Muehlchen et al. 1990; B?dker & Thorup-Kristensen 1999). The food preference of F. fimetaria has been investigated by Hall and Hedlund (1999) who found the following ranking of the investigated food items (fungi): Cladosporium cladosporioides over Alternaria alternata, Mucor hiemalis, Trichoderma viridae and Penicilinium sp., and A. alternata over Fusarium oxysporum and T. viridae. The remains of fodder radish may favour one of the higher ranked food items. It is difficult to explain the different significant effects of green manure on the mites because there are almost no clear trends. However, the significantly higher density of Mesostigmatic mites in the green manured plots 4 June coincides with the same pattern for Collembola. This makes sense, because most Mesostigmatic mites are predators, which may have developed their population size as a response to the high abundance of available prey, e.g. Collembola. The results show that autumn grown green manures can have a significant effect on the population density of Collembola and mites in the soil. Thus, when organic farms include autumn grown green manure or nitrogen catch crops in their crop rotation to improve the N supply for the crops, they are also likely to increase the population density of these groups of soil animals. As these catch crops also strongly reduced the risk of N leaching losses, increased the N uptake and yield of the subsequent barley crop (Thorup-Kristensen 1994; Thorup-Kristensen & Nielsen 1998), it shows that soil cover in the autumn may have very significantly positive effects both for agriculture and environment. The strongest effects in this experiment, both on reduction in N leaching, yield of the barley crop and population densities of the soil animals were obtained with the fodder radish crop as also found in previous experiments (Thorup-Kristensen 1994). Thus, the results indicate that it is not only important to grow an autumn cover, but it is also very important which species is grown.

Acknowledgements
This work was financed by the Danish National Environmental Research Programme through the Danish Research Centre of Organic Farming.

References
B?dker, L., Thorup-Kristensen, K. (1999) Effect of green manure crops on root rot and arbuscular mycorrhizal fungi in pea roots. In: (eds) Olesen, J.E., Eltun, R., Gooding, M.J., Jen-

566

J?rgen Aagaard Axelsen and Kristian Thorup Kristensen

sen, E.S., K?pke, U. (eds) Designing and testing crop rotations for organic farming. F?JO Report no. 5. Bolton, H., Elliot, L. F., Papendick, R. I., Bezdicek, D. F. (1985) Soil microbial biomass and selected soil enzyme activities: Effect of fertilization and cropping practices. Soil Biology and Biochemistry 17, 297–302. Chan, M. K. Y., Close, R. C. (1987) Aphanomyces root rot of peas 3. Control by the use of cruciferous amendments. New Zealad Journal of Agricultural Research 30, 225–233. Filser, J. (1995) The effect of green manure on the distribution of collembola in a permanent row crop. Biology and Fertility of Soils 19, 303–308. Fjellberg, A. (1980) Identification Keys to Norwegian Collembola. Norsk Entomolgisk Forening, Norway. Fraser, D. G., Doran, J. W., Sahs, W. W., Lesoing, G. W. (1988) Soil microbial populations and activities under conventional and organic management. Journal of Environmental Quality 17, 585–590. Hall, M., Hedlund, K. (1999) The predatory mite Hypoaspis aculeifer is attracted to food of its fungivorous prey. Pedobiologia 43, 11–17. Heinonen-Tanski, H. (1990) Conventional and organic cropping systems at Suitia III: Microbial activity in soils. Journal of Agricultural Science in Finland 62, 321–330. Krogh, P. H. (1994) Microarthropods as bioindicators. A study of disturbed populations. In: Terrestrial Ecology. Silkeborg, Natural Environmental Research Institute, pp. 96. Lootsma, M., Scholte, K. (1997) Effect of soil moisture content on the suppression of Rhizoctonia stem canker on potato by the nematode Aphelenchus avanae and the springtail Folsomia fimetaria. Plant Pathology 46, 209–215. Lootsma, M., Scholte, K. (1988) Effect of soil pH and amendments with dried fodder rape on mycophagous soil animals and Rhizoctonia stem canker of potato. Pedobiologia 42, 215–222. Muehlchen, A. M., Rand, R. E., Parke, J. L. (1990) Evaluation of crucifer green manures for controlling Aphanomyces root rot of peas. Plant Disease 74, 651–654. Pfiffner, L., Mader, P. (1997) Effects of Biodynamic, Organic and Conventional Production Systems on Erthworm Populations. Biological Agriculture & Horticulture 15, 3–10. Pimentel, D., Warneke, A. (1989) Ecological effects of manure, sewage sludge and other organic wastes on arthropod populations. Agricultural Zoology Reviews 3, 1–30. Scholte, K., Lootsma, M. (1998) Effect of farmyard manue and green manure crops on populations of mycophagous soil fauna and Rhizoctonia stem canker of potato. Pedobiologia 42, 223–231. Thorup-Kristensen, K. (1994) Effect of nitrogen catch crop species on the nitrogen nutrition of succeeding crops. Fertilizer Research 37, 227–234. Thorup-Kristensen, K., Bertelsen, M. (1996) Green manure crops in organic vegetable production. In: Kristensen, N. H., H?eg-Jensen, H. New Research in Organic Agriculture. Proceedings from the 11th International Scientific IFOAM Conference, Copenhagen, pp. 75–79. Thorup-Kristensen, K., Nielsen, N. E. (1998): Modelling and measuring the effect of nitrogen catch crops on nitrogen supply for succeeding crops. Plant and Soil 203, 79–89. Wardle, D. A. (1995) Impacts of disturbance on detritus food webs in gro-ecosystems of contrasting tillage and weed management practices. Advances in Ecological Research 26, 105–184.


相关文章:
更多相关标签: