Type of organic fertilizer rather than organic amendment per se increases abundance of soil biota

Introduction

Soil biodiversity is highlighted in the European soil thematic strategy as a key component of soil quality (European Commission, 2002), because soil organisms provide key ecosystem functions, such as decomposition, nutrient and carbon (C) cycling and water regulation (Barrios, 2007; Kulmatiski et al., 2014). Worldwide, the quality of agricultural soils is threatened by the loss of soil organic matter (SOM) and by soil compaction (Montanarella, 2015). A practice commonly used that offsets these threats and enhances soil functions is the application of carbon-containing amendments, i.e., organic fertilizers. A first step towards understanding how fertilization regulates soil quality is to examine its effects on the soil biota (Bender, Wagg & Van der Heijden, 2016). Organic fertilizers increase SOM and have positive immediate within season effects on the abundance of several soil organisms at different trophic levels, e.g., microorganisms (Birkhofer et al., 2008; Paterson et al., 2011), soil fauna such as nematodes (Ferris & Matute, 2003; Paterson et al., 2011), mites (Axelsen & Kristensen, 2000; Cao et al., 2011), collembolans (Axelsen & Kristensen, 2000; Haubert et al., 2009) and earthworms (Lofs-Holmin, 1983), as well as aboveground predators (Hines, Megonigal & Denno, 2006; Riggi & Bommarco, 2019). On the other hand, mineral nitrogen (N) fertilization has been reported to have minimal immediate impact on soil microbes (Geisseler & Scow, 2014) and inconsistent effects on soil fauna e.g., microarthropods (Siepel & Van de Bund, 1988; Curry, 1994; Postma-Blaauw et al., 2010). However, while the general within season response to organic fertilization is enhanced diversity and abundance of soil biota when compared with no or mineral fertilization, it is less well known how the overall soil communities react to long-term organic fertilization.

In contrast to the documented positive immediate effects of organic amendments on soil organisms, their long-term effects are inconsistent. For example, the biomass of microbial primary decomposers (Wander et al., 1995; Marschner, Kandeler & Marschner, 2003; Bünemann, Schwenke & Van Zwieten, 2006) and the abundances of secondary decomposers (Andrén & Lagerlöf, 1983; Moore, 1994; Alvarez, Frampton & Goulson, 2001; Kautz, López-Fando & Ellmer, 2006; Ngosong et al., 2009) can be both positively and negatively affected by long-term organic fertilization. These incongruences could be due to differences in the quality of the applied organic matter. In particular, labile carbon sources with a low C/N ratio constitute a high-quality C source for the lower trophic levels in the soil food web (Bardgett et al., 2005; Haynes, 2005; Cotrufo et al., 2013; de Vries & Caruso, 2016). Organic fertilizers that increase labile C inputs are therefore used more efficiently by microbes than recalcitrant fertilizers, and in the long term, high-quality labile litter results in greater SOM formation and accumulation (Cotrufo et al., 2013). Nevertheless, to our knowledge comparative investigations of organic fertilizers with respect to their qualitative effects of available C (i.e., equal amounts of C) on soil organisms are scarce (Leroy, 2008), and even lacking for long-term conditions.

To assess whether organic fertilizers are better at enhancing soil biota than mineral fertilizers in the long-term, we used a 17-year-old fertilization experiment in which all management practices other than type of fertilizer were similar. The amount of added C was comparable across all organic fertilizers, but the quality (i.e., C/N) of the organic amendment varied depending on type of organic matter added (manure, hay, compost or sewage sludge). To primarily assess the long-term effects of the organic amendments, soil samples were collected 1.5 years after the last application of organic amendments. We expected (1) mineral fertilizers to have an indirect positive effect on the abundance of soil biota as a consequence of increased crop growth from fertilization (Kätterer, Börjesson & Kirchmann, 2014), and (2) organic fertilizers to have an even greater positive effect on the abundance of soil biota as they provide a direct source of C in addition to the indirect C input via crop growth (Bünemann, Schwenke & Van Zwieten, 2006). We also anticipated (3) the organic fertilizers to differ in their effect due to differing N concentrations, i.e., C/N ratios. Microorganisms, in particular bacteria, likely respond positively to the amendments with lower C/N ratio (Leite et al., 2017), i.e., sewage sludge (C/N = 8.8) and compost (C/N = 9.4). On the other hand, earthworms should respond positively to the amendments with higher C/N (Aira, Monroy & Domínguez, 2006), i.e., hay (C/N = 20.5) and manure (C/N = 16.8), whereas nematodes, springtails and mites were expected to show varying responses due to differing food sources within each of these faunal groups.

Materials & Methods

Results

Microorganisms

There was an effect of fertilization on abundance of both fungi and bacteria (F = 4.06, p = 0.010 and F = 3.60, p = 0.016, respectively). Compared with the unfertilized soil, fungi were similarly stimulated by both organic (hay and manure) and mineral fertilizers (ammonium sulphate) (Fig. 1A). However, fungal abundance in unfertilized plots was similar to that in calcium nitrate, compost and sewage sludge plots. On the other hand, bacteria were generally enhanced by organic fertilizers compared with mineral fertilizers (F = 4.64, p = 0.044), and especially by the addition of hay (Fig. 1B). However, bacterial abundance in unfertilized plots was similar to that in compost and sewage sludge plots. Overall, there was no effect of treatment on the fungi:bacteria ratio, which ranged between 0.04 and 0.14.

Nematodes

The total abundance of nematodes was affected by fertilization (F = 2.81, p = 0.041). Some mineral (ammonium sulphate) and some organic (sewage sludge, manure) fertilizers increased nematode abundance compared with unfertilized soil (Fig. 2A). Taken separately, none of the nematode feeding groups were affected by the fertilizer treatments (Table S2).

Correlations with soil factors

Generally, the measured soil factors only explained a minor portion of the variation in the overall abundances (20.9%) and biomasses (24.8%) of the soil organism groups (Fig. 3). Regarding abundances, bacteria, mesostigmatic mites and earthworms correlated positively with the amount of carbon and nitrogen in the soil, while Entomobryomorpha and oribatid mites correlated positively with soil C/N (Fig. 3A). The abundances of the Podumorpha and Entomobryomorpha correlated negatively with soil pH (Fig. 3A). None of the biomasses was clearly correlated to the measured soil factors (Fig. 3B). There were also few correlations with the C/N of the organic amendments; the fungal abundance correlated positively while total springtail abundance and the abundance of Podumorpha correlated negatively (Table 4).

Table 4:

Spearman rank correlation coefficients between soil biota abundance/biomass and the C/N ratio of the organic amendments.

Significant correlations (p < 0.05) are indicated in bold.

Soil biota group	C/N amendment
Microorganisms
Fungi (copies/g DW soil)	0.59
Bacteria (copies/g DW soil)	0.45
Nematodes
Total abundance (no./g DW soil)	−0.11
Plant feeders	−0.42
Fungal feeders	−0.17
Bacterial feeders	−0.16
Omnivores	−0.42
Predators	−0.09
Springtails
Total abundance (ind./m2)	−0.55
Podumorpha	−0.61
Entomobryomorpha	−0.09
Symphypleona	−0.32
CWB1 (µg/ind.)	−0.28
Mites
Total abundance (ind./m2)	0.39
Oribatida	−0.04
Astigmata	0.46
Prostigmata	0.18
Mesostigmata	0.26
CWB1 (µg/ind)	0.06
Earthworms
Abundance (ind./m2)	0.02
Biomass (g DW/m2)	0.00

DOI: 10.7717/peerj.11204/table-4

Notes:

1CWB, community weighted mean biomass.

Discussion

As hypothesized, long-term addition of both mineral and organic fertilizers enhanced the abundances of most groups of soil organisms compared with unfertilized soil, but contrary to our second hypothesis, the mineral fertilizers had an equally stimulating effect on soil biota as organic fertilizers. The lack of major differences in the abundances of various soil community members between organic and mineral fertilizer treatments could be due to positive priming in the plots treated with mineral fertilizers. Mineral fertilization is known to increase the turnover of soil organic matter shortly after application, thus providing easily available nutrients that can improve conditions for a variety of soil biota (Kuzyakov, Friedel & Stahr, 2000). In the long term, mineral fertilizers generally increase the microbial biomass in cropping systems (Geisseler & Scow, 2014), and this effect can propagate in the soil food web and increase soil fauna and erase differences between fertilization treatments. In addition, the increased crop production (i.e., plant dry weight) in the plots with mineral fertilization compared with the other treatments (Kätterer, Börjesson & Kirchmann, 2014) also adds labile C to the soil via root exudates and decaying crop roots, which would positively influence soil biota. The fact that the mineral fertilizer treatments were also supplied with P and K, while the organic amendments were not, could potentially create nutrient imbalances in the soil, although previous studies have shown that systems with organic fertilizers have higher levels of these nutrients than conventional systems (e.g., Clark et al., 1998). Unfortunately, we do not have data on soil nutrient levels, other than nitrogen, and future studies evaluating effects of organic amendments should address effects of possible nutrient imbalances.

In opposition to our third hypothesis, the C/N of the organic amendment did not influence the soil biota, as the more N-rich organic fertilizers had effects similar to those with higher C/N. Nevertheless, fertilization affected the organisms in different ways depending on the type of fertilizer and taxa considered. Both fungi and bacteria were affected by the fertilizers although the ratio between them was not, which is in line with previous findings (Bittman, Forge & Kowalenko, 2005; Kätterer, Börjesson & Kirchmann, 2014). Bacteria and fungi are both taxonomically and functionally extremely diverse groups of organisms. Therefore, there were probably context-dependent responses by the different bacterial and fungal species and functional guilds to the different fertilization treatments that were not captured by only looking at the total community size. This could explain why reported responses in terms of microbial abundances and biomass to increasing soil C after addition of organic matter have been inconsistent (i.e., Witter, Mårtensson & Garcia, 1993; Houot & Chaussod, 1995; Leita et al., 1999; Enwall et al., 2007; Hallin et al., 2009; Cederlund et al., 2014; Kätterer, Börjesson & Kirchmann, 2014). For example, arbuscular mycorrhizal fungal abundance has been shown to be promoted by manure addition (Liu et al., 2020; Ozlu et al., 2019). In addition, microbial performance depends on multiple soil properties, e.g., pH and bulk density, that are also affected by organic amendments (Bünemann, Schwenke & Van Zwieten, 2006). Organic matter can also be more or less readily accessible to microorganisms as it is degraded. Thus, the increased abundance of bacteria in the hay treatment two years after amendment may be due to the recalcitrant nature of hay; this suggestion is supported by long-term results showing that bacterial abundances and biomass only increase in straw or peat amended soils when supplemented with mineral N compared with mineral fertilized soils (Wessén et al., 2010; Cederlund et al., 2014). Overall, our results suggest that in agricultural systems with annual crop rotations and annual tillage, biennial amendments with organic matter promote fungal and bacterial abundances only marginally in the long-term compared with annual addition of mineral fertilizers.

Although there was an overall effect of the fertilizer treatments on the total abundance of nematodes, surprisingly this was not apparent on the nematode feeding group level. In particular, bacterivorous nematodes are known to respond positively to fertilization, especially to organic amendments (e.g., Forge, Bittman & Kowalenko, 2005a; Leroy et al., 2007; Villenave et al., 2010; D’Hose et al., 2018; Grabau, Vetsch & Chen, 2018). Nevertheless, an increase in bacterivore abundance is greatest in the season when the fertilizer is applied (Grabau, Vetsch & Chen, 2018) and in our study, the last application may have been too long ago to see an effect. The lack of responses in bacterial- and fungal-feeding nematodes could also be a reflection of the marginal effects of the treatments on their bacterial and fungal food sources, respectively. For root-feeding nematodes, positive (Weiss & Larink, 1991; Forge, Bittman & Kowalenko, 2005b), neutral (Bulluck, Barker & Ristaino, 2002; D’Hose et al., 2018) and negative effects (Akhtar, 1999; Leroy et al., 2007) of organic fertilizers have been found. However, often only a limited number of fertilizers have been compared in the same study, and in many cases only one type of organic fertilizer was investigated. Altogether, our comprehensive comparison of several organic and inorganic fertilizers indicates that long-term fertilization is a minor factor affecting the occurrence of root-feeding nematodes, which was also suggested by Schmidt et al. (2017). Both long-term mineral (ammonium sulphate) and organic (manure and sewage sludge) fertilizers appeared to promote nematodes in agricultural soils compared with no fertilization. Due to the differences between these fertilizers and the fact that no effects were found at the feeding group level, the biotic and abiotic ecological processes underpinning the observed pattern probably differ between the amendment types.

The mesofauna showed contrasting responses to fertilization depending on the taxa, which, in the case of springtails, was related to differences in soil pH among the treatments. Previous studies have shown increasing numbers of springtails with several different organic amendments (Wang et al., 2015; Platen & Glemnitz, 2016; Leroy et al., 2007), but not with sewage sludge (Andrés et al., 2011). Surprisingly, we found the highest springtail abundance in sewage sludge-treated plots, particularly within the sub-family Podumorpha. Possibly, it was the combination of increasing organic matter and low pH caused by the addition of sewage sludge that benefited the Podumorpha. Mites did not benefit from fertilization per se as the total abundance did not differ between unfertilized and fertilized plots, but their abundance was generally enhanced by organic fertilizers compared with mineral fertilizers. The response of mites to fertilizers may be complex because of their diversity of feeding habits and habitat preferences within and among the different sub-orders (Schneider et al., 2004). Nevertheless, manure additions supported relatively high abundances of all mite sub-orders, which agrees with previous results (Minor & Norton, 2004; Cao et al., 2011; Gruss, Twardowski & Hurej, 2018). Our results suggest that the mesofauna is only marginally promoted by long-term fertilization, whether with mineral or organic fertilizers. Whether this depends on small direct effects of the fertilizer treatments or indirect effects on their food sources or other soil fauna is not possible to determine from the present study.

Earthworms feed directly on organic matter and are therefore expected to respond positively to the addition of organic amendments, as well as the quality of the amendments. In accordance with our results, addition of manure has been shown to increase abundance and biomass of earthworms (Edwards & Lofty, 1982; Whalen, Parmelee & Edwards, 1998; Sizmur et al., 2017; D’Hose et al., 2018). Adding hay as an organic amendment has been studied far less, but earthworms do feed on both hay and straw (Al-Maliki & Scullion, 2013). However, field observations have indicated that the size of the straw pieces is important and that straw milled into smaller pieces (<3 mm) is more easily ingested (Sizmur et al., 2017). Studies with rice straw have also shown that the moisture level in the soil is important for its decomposition by earthworms (Gorbunova et al., 2020). The hay used in the present study was chopped into 20 cm long pieces, but other factors related to the conditions in the soil probably also contributed to the response of the earthworms to the addition of hay. Earthworms readily feed on sewage sludge and can even be used for vermicomposting it (Sinha et al., 2009). Therefore, it is not surprising that in the present study the abundance of earthworms increased in the plots with added sewage sludge. The same pattern has also been found in previous studies evaluating addition of sewage sludge in the field (Hamilton & Dindal, 1989; Stöven & Schnug, 2009).

Although presented and mostly discussed separately here, soil organisms interact with each other and influence each other’s populations. As the organism groups were not sampled at the same time during the season, we did not investigate co-occurrences and correlations between them. Even when sampled at the same time, however, it is difficult to draw conclusions about the mechanisms behind co-occurrence patterns (Horner-Devine et al., 2007), especially when analyzing the soil biota at low taxonomic resolution. Nevertheless, we believe that our results based on abundances of all major soil organism groups are adequate for the general questions we are addressing related to the long-term effects of fertilizers on soil biota. To also address questions related to soil health, a higher taxonomic resolution would be needed to decipher the responses to management and performance of the detrimental species compared with the beneficial species within the individual soil organism groups. More long-term studies covering a wider range of C/N ratios than in the present study are also needed, as well as evaluation of long-term annual organic amendments.

Conclusions

Long-term addition of organic and mineral fertilizers had beneficial effects on the abundances of most soil organisms compared with unfertilized soil, but the responses differed between the soil biota groups. That the effect of organic amendments was similar to mineral fertilizers for some of the groups indicates that biennial organic amendments are insufficient by themselves to boost soil biota and annual additions or combining additions with other practices affecting soil quality, such as no or reduced tillage and other crop rotations, are required to deliver this effect. There had probably been positive transient effects of the amendments on some of the more rapidly-responding groups, like bacteria and bacterial-feeding nematodes. The only organism group that showed a clear positive response to the organic amendments was the earthworms, a group that directly feeds on the added material. Overall, the specific type of fertilizer was more important in determining the numerical and compositional responses of soil biota than whether it was mineral or organic in nature. Our results indicate that both chemical properties, such as pH and C and N levels in the soil, as well as biotic interactions between the soil biota groups are responsible for the observed patterns, and these factors are in turn affected by the fertilizer treatments.

We highlight the fact that accurate assessments of the influence of fertilizers on soil biota communities will be achieved only if a wide range of soil organisms and soil chemistry are investigated concurrently.

Supplemental Information