**Comparison of the E**ff**ect of Perennial Energy Crops and Arable Crops on Earthworm Populations**

#### **Beata Feledyn-Szewczyk \*, Paweł Radzikowski, Jarosław Stalenga and Mariusz Matyka**

Department of Systems and Economics of Crop Production, Institute of Soil Science and Plant Cultivation—State Research Institute, Czartoryskich 8 Str., 24-100 Puławy, Poland; pradzikowski@iung.pulawy.pl (P.R.); stalenga@iung.pulawy.pl (J.S.); mmatyka@iung.pulawy.pl (M.M.)

**\*** Correspondence: bszewczyk@iung.pulawy.pl; Tel.: +48-081-478-6803

Received: 1 October 2019; Accepted: 22 October 2019; Published: 24 October 2019

**Abstract:** The purpose of the study was to compare earthworm communities under winter wheat in different crop production systems on arable land—organic (ORG), integrated (INT), conventional (CON), monoculture (MON)—and under perennial crops cultivated for energy purposes—willow (WIL), Virginia mallow (VIR), and miscanthus (MIS). Earthworm abundance, biomass, and species composition were assessed each spring and autumn in the years 2014–2016 using the method of soil blocks. The mean species number of earthworms was ordered in the following way: ORG > VIR > WIL > CON > INT > MIS > MON. Mean abundance of earthworms decreased in the following order: ORG > WIL > CON > VIR > INT > MIS > MON. There were significantly more species under winter wheat cultivated organically than under the integrated system (*p* = 0.045), miscanthus (*p* = 0.039), and wheat monoculture (*p* = 0.002). Earthworm abundance was significantly higher in the organic system compared to wheat monoculture (*p* = 0.001) and to miscanthus (*p* = 0.008). Among the tested energy crops, Virginia mallow created the best habitat for species richness and biomass due to the high amount of crop residues suitable for earthworms and was similar to the organic system. Differences in the composition of earthworm species in the soil under the compared agricultural systems were proven. Energy crops, except miscanthus, have been found to increase earthworm diversity, as they are good crops for landscape diversification.

**Keywords:** soil biota; invertebrates; farming systems; bioenergy; biodiversity; wheat; ecosystem

#### **1. Introduction**

Earthworms constitute the largest component of animal biomass in the soil, and they are termed "soil engineers" [1] or even "ecosystem engineers" [2]. The diversity and abundance of earthworms are an important criterion of soil fertility [3]. Earthworms play a major role in the processes of decomposition of plant and animal organic residues, soil humus formation, creation of a crumbly soil structure, water regulation, pathogen and pest control, degradation of pollutants, as well as nitrogen binding [2,4]. Earthworms provide many ecosystem services, such as maintenance of proper soil structure, soil humus formation, nutrient cycling, erosion control, and biological crop protection [5].

Activity of earthworms and formation of burrows improve the aeration and water infiltration and therefore reduce surface runoff [6]. There are many mineral and organic ingredients in earthworm excrements that are beneficial for the growth and development of plants. These fractions are well mixed in worm casts, and the nutrients are present in a readily available form [1,5]. Soil passing through the digestive system of earthworms is enriched with beneficial microorganisms binding free nitrogen and activating phosphorus, which become available to plants. Both excreta and secretions (metabolic water and mucus) of earthworms contain plant growth stimulators (auxin, gibberellin, and cytokine), affecting the quality and quantity of the crop yield [7]. By pulling fallen leaves into the soil, foliar pathogens and pests are biologically degraded. Earthworms distribute insect-killing

nematodes (*Steinernema* sp.) and fungi (*Beauveria bassiana*) in the soil, thus contributing to better natural regulation of soil-borne pests. Earthworms ingest organic residues of different C:N ratios, convert them to a lower C:N ratio, and finally contribute to carbon sequestration [5]. A rich earthworm fauna is a key in maintaining and safeguarding soil health and in fostering many essential ecosystem functions of soil.

Agricultural practices, such as tillage, crop rotation, cultivation of catch crops (defined here as additional fast-growing crops grown between successive plantings of main crops), the use of mineral fertilizers, and pesticides all have significant impacts on wild flora and fauna, including soil organisms [8–10]. The sensitivity of earthworms to unfavourable changes in agrocenosis makes them a good indicator of the ecological footprint of agricultural systems [11]. According to Pfiffner [5], the following measures are prerequisites for the flourishing of earthworms in agricultural soils: provision of sufficient food, abstinence from the use of pesticides harmful to earthworms, application of soil-conservation methods such as reduced tillage and no-till, avoidance of soil compaction and promotion of well-structured and aerated soils, appropriate fertilization, and balanced humus management within the crop rotation. Different farming systems can support or reduce biodiversity and soil settlement by earthworms. High-input conventional agriculture reduces the biodiversity of earthworms, whereas conservation and organic agriculture benefit this group of organisms [12–14].

Cultivation of perennial crops for energy purposes is still a new agricultural system so little scientific evidence appears to be available on the effects of these types of crops on the environment, including on earthworm populations [15,16]. The positive effect of energy crops on biodiversity is connected with the lower agrochemical input as compared to the intensive production used in annual crops [15,17–19]. Due to the unknown impact of many plant species used for energy purposes on the environment and biodiversity, there should be wide-ranging and long-term ecological monitoring conducted for these crops [19,20]. According to Verdade et al. [21], biodiversity monitoring programs are needed to help the decision-making process concerning the conflict between the expansion of energy crops and the conservation of biodiversity. These programs should take into account comparisons with neighbouring agricultural crops [22,23]. Such a comparison has been done in this current study.

Our hypothesis was that the cultivation of certain species of perennial energy crops stimulate earthworm diversity and abundance more than farming systems on arable land.

On this basis, the aim of the present research was to compare the impact of various agricultural systems on arable land (organic, integrated, conventional, and winter wheat monoculture) and the impact of cultivation of perennial energy crops (miscanthus, Virginia mallow, and willow) on earthworm diversity, abundance, and biomass under the conditions pertaining in Eastern Poland (Central Europe).

#### **2. Materials and Methods**

#### *2.1. Site Description and Experimental Design*

The assessment of earthworms was carried out as part of a long-term field experiment (1994–until now) with four crop production systems: organic (ORG), integrated (INT), conventional (CON), and monoculture-conventional (MON) (Table 1). Experimental plantations of perennial energy crops (2008–until now)—miscanthus (*Miscanthus* sp.) (MIS), Virginia mallow (*Sida hermaphrodita*) (VIR), and willow (*Salix viminalis*) (WIL)—have been established 70–100 m away from the crop production systems on the arable land described above. The experimental sites are located at the Agricultural Research Station of the Institute of Soil Science and Plant Cultivation—State Research Institute (IUNG-PIB) in Osiny (Poland, Lublin voivodeship, N: 51◦28 , E: 22◦30 ) on *Haplic Luvisol* soil [24] with a texture of loamy sand. The chemical properties of the soil in the different farming systems are presented in Table 2. Average annual total precipitation of the experimental site was 586 mm with a mean air temperature of 7.5 ◦C (data for the years 1950–2013).


**Table 1.** Crop management in winter wheat in different farming systems and three perennial energy crops (2014–2016).

**Table 2.** Soil chemical properties of the 0–30 cm layer of *Luvisol*.


<sup>1</sup> ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow.

The tested farming systems on arable land were characterised by different crop rotations and agricultural management (Table 1). In the organic system, no synthetic pesticides and natural phosphorus (P) and potassium (K) fertilizers such as crude potassium salt or kainite as well as compost, applied once in a crop rotation, under potato (30 t ha−1) were applied. High-input conventional systems included two variants: (1) 3-field crop rotation: winter oilseed rape, winter wheat, and spring wheat and (2) winter wheat monoculture. In both objects, crops were cultivated intensively, i.e., with high rates of synthetic mineral fertilizers and pesticides (Table 1). In the integrated system balanced mineral and organic fertilization (about 20–30% lower than in conventional system), adaptation to the crop requirements and soil fertility were used. Soil tillage was similar in all crop production systems; in general, it was a traditional plough system. Before sowing of winter crops, a four-furrow reversible plough was used at depths of 15–20 cm. Before sowing of spring crops, winter ploughing was carried out at a depth of minimum 20 cm. After late crops such as maize or potato or after cultivation of a catch crop, winter ploughing was done usually in the second half of November. For other crops harvested in July/August, winter ploughing was performed at the end of September. Before sowing, a compact

seedbed cultivator was usually used. Number and time of ploughing treatments in particular systems depended on the crop rotation structure.

Crop protection consisted of nonchemical (mechanical) measures, and only a limited number of herbicides and other plant protection products were applied based on harmfulness thresholds. Most herbicides used were applied on crops, and only a few based on glyphosate were sprayed on stubble and incorporated into soil.

The analyses were carried out in the soil under winter wheat cv. Jantarka, which grew in all of these crop production systems. The area of each field was 1 ha so it was possible to manage them using real agricultural practices. In energy crops, no chemical crop protection measures have been performed during the study period. The area of the energy crop fields was from 200 m2 (miscanthus and Virginia mallow) to 500 m<sup>2</sup> (willow).

#### *2.2. Earthworm Collection*

Earthworms were collected in the years 2014–2016, according to the method of soil blocks [25]. Samples were taken twice in each season: in spring (April) and in autumn (October), when soil temperature and moisture are usually suitable for earthworm activity. Soil blocks 25 cm × 25 cm × 25 cm (0.0625 m2 of arable soil layer) were dug out of each field in 5 replications. The blocks were separated by at least 5 m from each other and 5 m away from the field margin to avoid the edge effect. Soil blocks were placed on a sheet on which the earthworms were caught and kept in collecting containers. The earthworms were transported in cool boxes to the laboratory, where they were washed, weighed, and preserved in 4% formalin for further investigations. In later terms, earthworms were classified into species according to an identification guide [26]. The species was recognized on the basis of morphological features. The structure of the mouth, distribution of bristles, number of segments, structure and location of the reproductive organs, and the location of secretory holes were thoroughly analyzed. Due to longer storage in the formalin solution, the colour of individuals was not taken into account. Both mature and young forms were considered, but only adult individuals were defined to the species [5]. Earthworm species richness, their abundance, and biomass (fresh weight) were taken into account as biodiversity indicators. Earthworm density and biomass were calculated per 1 m2. For each investigated agricultural system, the results from all terms of collection were presented (3 years × 2 terms × 5 replications = 30 samples for each studied field).

#### *2.3. Statistical Analyses*

Because this trial was without replication, we chose to consider each sample point as a replicate, though this approach could have induced some bias because of pseudoreplications. We are, however, quite confident that, for several reasons, our sampling design allowed us to use this approach: (i) sample points within a field were far enough away from each other (20 m) to ensure replicate independence, and (ii) the possibility that the sample fields were affected by confounding factors due to limited randomization cannot be excluded but was limited as the trial was evenly affected by the same management before the trial setup, topographic and pedologic gradients were controlled, and a preliminary assessment of the trial spatial heterogeneity was found to be very low within the block soil heterogeneity. This methodological approach has been used earlier by Henneron et al. [14]. There are examples of field trials that are not replicated, for example, one of the oldest long-term experiments established in 1843 Rothamsted (UK) [27], but, with the help of specific statistical methods, permits reliable comparisons. Lack of replications is due to different reasons, but usually, it is caused by physical constraints such as land availability or plot size. Sometimes financial or social limitations are also important [28].

In order to check the normality of the distributions, the Shapiro–Wilk test was used. The obtained data sets were characterized by a non-normal distribution; therefore, the nonparametric Kruskal–Wallis rank test was used to assess the significance of differences in the examined features at the significance level *p* ≤ 0.05. Statistical analyses were performed using Statistica 10 software (Stat. Soft. Inc., Tulsa, OK, USA).

In order to group the earthworm communities, a cumulative hierarchical classification was done using MVSP 3.1. software, Kovach Computing Services, Anglesey, Wales [29]. The quantitative Sorensen's similarity index (Percent Similarity) was used to classify similarities between earthworms under different farming systems and types of energy crops [30].

In order to classify the samples based on earthworm species composition and species based on their participation in the samples, ordination techniques were used [31]. As first, Detrended Correspondence Analysis (DCA) was applied, which is recommended for the preliminary ordering of data [31,32]. The length of variance gradient calculated in this analysis characterizes the data structure and constitutes the criterion for selecting further ordination methods to assess the significance of the tested environmental or agrotechnical factors. Due to the fact that the length of the first axis gradient in DCA analysis was lower than 2 standard deviations, which showed that the distribution of species was not compatible with the Gaussian curve, linear method Principal Component Analysis (PCA) was used to perform direct ordination [33]. The results of this ordination were presented graphically on diagram (PCA diplot). The analyses were performed in the Canoco 4.5 program [32].

#### **3. Results**

#### *3.1. Species Richness and Abundance*

A total of 11 species of earthworms were recorded in the compared agricultural systems (Table 3). The largest number of species occurred in the willow (10 species), and the lowest number was in the high-input crop production systems: wheat monoculture and the conventional system (7 species). In the organic system, there were many juvenile unspecified individuals (Lumbricidae sp.). It should be noted that *Allobophora chlorotica* was found only in crop production systems on arable land and that *Proctodrilus antipai* was only found under perennial energy plants.



<sup>1</sup> ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow. <sup>2</sup> Different letters indicate significant differences between treatments according to the Kruskal–Wallis test at *p* ≤ 0.05 (*n* = 30).

In crop production systems on arable land, the largest number of earthworm individuals (88.6 indv. m<sup>−</sup>2) was recorded in the soil under winter wheat cultivated in organic system (Table 3). Over twice less individuals (35 indv. m<sup>−</sup>2) were found in monoculture of winter wheat. Earthworms density decreased in the order of ORG > CON > INT > MON. Among compared energy crops, willow had the largest earthworm abundance (74 indv. m<sup>−</sup>2, only 16% less than in the organic system) while the smallest was found in miscanthus field (43 indv. m<sup>−</sup>2, 23% more than in wheat monoculture).

Total earthworm abundance was significantly higher in the organic system as compared to winter wheat monoculture (*p* = 0.001; *z* = 3.85) and to miscanthus (*p* = 0.008; *z* = 3.65) (Table 3). There were no significant differences between the organic system and willow in earthworm density.

Earthworm species number per sample was higher under winter wheat cultivated organically than under monoculture (*p* = 0.002; *z* = 2.68), than the integrated system (*p* = 0.045; *z* = 3.64), and than under miscanthus (*p* = 0.039; *z* = 3.24) (Figure 1). Earthworm diversity in the organic system and in Virginia mallow did not differ significantly. Among the tested energy crops, differences in the species richness were found between miscanthus and Virginia mallow (*p* = 0.032; *z* = 2.55).

**Figure 1.** Average earthworm species number per sample (0.0625 m2) in winter wheat cultivated in four crop production systems on arable land and perennial energy crops (average for 2014–2016, mean ± SE, n = 30). ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow. <sup>1</sup> Different letters indicate significant differences between treatments according to the Kruskal–Wallis test at *p* ≤ 0.05.

The biomass of earthworms was 3 times larger in the organic system in comparison with the wheat monoculture, miscanthus, and willow systems (Figure 2). There were no significant differences between the organic, conventional, and integrated systems, which could be the result of using compost and catch crops twice in the integrated system and of using straw ploughing in the conventional system (Table 1).

**Figure 2.** Earthworm biomass (fresh weight g m<sup>−</sup>2) in winter wheat cultivated in four crop production systems on arable land and perennial energy crops (average for 2014–2016, mean ± SE, n = 30). ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow. <sup>1</sup> Different letters indicate significant differences between treatments within arable systems and energy crops according to the Kruskal–Wallis test at *p* ≤ 0.05.

#### *3.2. The Relationships between Agricultural Systems and Species Composition*

In the organic system, the indeterminate individuals (Lumbricidae sp.), mainly juvenile (27%) and *O. lacteum* (26%), constituted the most numerous groups (Figure 3). The highest density of *L. terrestris* was found in the organic system and Virginia mallow. The earthworm communities in the integrated system and the wheat monoculture system were dominated by *A. caliginosa* (39% and 60% share, respectively). The highest share of *A. rosea* were observed in earthworm communities under miscanthus and willow. In the soil under willow and Virginia mallow, earthworms unspecified for species, mainly epigeic juveniles of low biomass, dominated. In the miscanthus cultivation, *A. caliginosa* and *A. rosea* were the most numerous.

In order to confirm the relationship between land use and the abundance of earthworm species, the ordination method PCA (Principal Component Analysis) was used (Figure 4).

Along the gradients of the axes of particular crop production systems, species most closely associated with a given type of farming and energy crop were grouped together (Figure 4). *O. lacteum* was most closely connected with crops cultivated organically. *L. terrestris*, *A. longa, L. rubellus*, and unidentified Lumbricidae sp. were located close to the gradients of the axes of the organic system and Virginia mallow, which indicates that they were characteristic for both crop production systems. *P. antipai* and *A. rosea* were strongly positively correlated with willow. Species on the left-hand side of the diagram in Figure 4—*A. caliginosa, O. cyaneum, A. chlorotica*, and *A. georgii*—were associated with more intensive farming systems (CON, MON, and INT) and miscanthus.

Gradients of the systems and the species distribution in Figure 4 showed similarity of the earthworm communities for the organic system and Virginia mallow (right, upper site of the diagram) and dissimilarities from other, more intensive crop production systems (INT and MON) as well as miscanthus (left, down site of the diagram). Along the gradient of axis I, the highest positive correlation between the tested systems and the location of species occurred for willo and negative occurred for wheat monoculture. The most positively correlated with the gradient of axis 2 was the organic system while that negatively correlated was willow (Table A1, Figure 4).

**Figure 3.** Relative abundance of earthworms species in communities in different agricultural production systems. ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow.

**Figure 4.** Ordination diagram of agricultural systems and earthworm species abundance in relation to the first and second axes of PCA (PCA biplot). ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow.

The hierarchical classification confirmed the similarity of the earthworm communities in the organic and Virginia mallow systems and their dissimilarity from integrated system, wheat monoculture, and miscanthus (Figure 5). In these analyses, the dissimilarity of earthworm communities under miscanthus from the other two energy crops and their similarity to wheat monoculture and the integrated system have been shown.

**Figure 5.** The results of the hierarchical cumulative classification using the unweighted pair group method with arithmetic mean (UPGMA) of samples representing earthworm abundance in different types of land use using the quantitative Sorensen's coefficient of similarity. ORG—organic; INT—integrated; CON—conventional; MON—monoculture; MIS—miscanthus; VIR—Virginia mallow; WIL—willow.

#### **4. Discussion**

#### *4.1. Species Richness and Abundance of Earthworms*

In tested farming systems, on arable land, and among energy crops, 11 species of earthworms were recorded. Neirynck et al. [34] and Pfiffner [5] have shown that, generally, in cropland, only from 1 to 11 species are commonly found from among the about 40 earthworm species occurring in Central Europe. Most of them are species with a wide range and large adaptation abilities. The most common species in Poland are *Dendrobaena octaedra* (Sav.), *Lumbricus terrestris* L., *L. rubellus* Hoffm., *Aporrectodea caliginosa* (Sav.), *Aporrectodea rosea* (Sav.), and *Eiseniella tetraedra* (Sav.) [26].

The relative abundance of earthworms depends upon soil type, topography, and vegetation, and it is influenced by land use [1,34]. According to Lavelle [35], in terrestrial ecosystems, density of earthworms may reach 10<sup>6</sup> ha <sup>−</sup><sup>1</sup> and their biomass may reach 2 t ha <sup>−</sup>1. They are present everywhere except in arid and frozen regions. In Central Europe, about 120–140 earthworms per 1 m<sup>2</sup> of arable soil are considered satisfactory in terms of soil fertility [5]. In our own research, a smaller number of earthworms were recorded, i.e., between 88 indv. m−<sup>2</sup> in the organic system up to 35 indv. m−<sup>2</sup> in monoculture. The lower earthworm abundance could be the result of being sandier, having higher acidity, and having low humus content of the investigated soils compared to the other European countries, which are not favourable to earthworms [5]. Very low earthworm abundance on sandy soils was also observed on conventional arable dairy farm situated in Peer in the sands of the Campine region (Belgium) [36].

Almost all of the earthworm species found are typical of the region and land use [26]. *A. caliginosa* was most abundant in arable systems. It is a species found in a wide range of environments. It is highly tolerant of habitat quality, including acidification and low levels of organic matter. The species tolerates intensive tillage to some extent, making it the most common species on arable land. Other highly represented species, *L. terrestris*, *O. lacteum, O. cyaneum, A. longa, A. rosea*, and *L. rubellus*, have slightly higher habitat requirements and are typical rather for meadows and pastures habitats than for arable land [35]. The species with the highest tolerance for low pH is *A. chlorotica*, which was found to have one individual in a cereal monoculture. Among the rare species, *A. georgii* was found. This species was previously found in the neighbouring region of the Mazovian voivodeship [26]. A rare, South European species of the earthworm, *P. antipai*, is usually found on heavy soils in the Vistula river valley, whereas in the studies, it was found in poor light soil. Heavy hydrogenic soils also lie in the

farm serving the experiments, which suggests that both species could be transferred with the soil on agricultural equipment.

#### *4.2. The E*ff*ect of Crop Rotation and Catch Crops*

The number of earthworms depends on the availability and quality of organic matter in the soil [6,26]. On the arable land, manure and compost as well as the protein-rich residues of legumes (*Fabaceae*) are particularly preferred by earthworms. A diversified crop rotation with long-lasting and deep-rooted catch crops rich in clover or green manure crops as well as diversified crop residues are essential to maintain or increase earthworm populations [5]. This was confirmed by our own results showing that earthworm species richness and density were significantly higher in the organic system than in other systems (by 126% higher mean number of species and by 156% more individuals than in wheat monoculture). The higher species number and abundance of earthworms in the organic system could be the effect of the diversified crop rotation with clover and grasses mixture as well as of the compost and catch crop used, which provided food for earthworms. In a study by Schmidt et al. [37], a wheat-clover cropping system supported earthworm communities that were twice as large and had between one and five times more earthworm species than that found under conventional wheat mono-cropping. The above authors recorded between one and five more earthworm species in the wheat-clover system than in the pure wheat system.

Increased earthworm density was also observed as a result of mulching of crop residues on the field surface, especially during winter [5,38]. Post-harvest residues are an important source of organic matter, and they also affect the microclimatic properties of the soil. In mulched soil, there are more earthworms feeding close to the soil surface. In addition, such soil is more resistant to freezing than the soil without plant cover, which has an impact on the mortality of earthworms during winter, late autumn, and spring [12]. In our study, larger number and biomass of earthworms in the conventional system than in the integrated system could have been caused by a large biomass of post-harvest residues from the incorporation of wheat and rape straw.

The content and type of humus in soil are some of the most important factors determining the species composition of earthworms. *L. terrestris* has a tendency to occur in soil rich in organic carbon [39]. In our own research, *L. terrestris* was the most frequent in wheat in the organic system and Virginia mallow, where the content of organic matter was the highest. According to Schmidt et al. [37], wheat-clover cropping especially favoured species belonging to the epigeic and epigeic/anecic ecological groups such as *L. rubellus* and juvenile *Lumbricus* and was also observed in our own study, whereas *A. caliginiosa* may occur even in soils poor in carbon [39]. In the presented study, *A. caliginosa* was the most numerous in the intensive systems on arable land: integrated, wheat monoculture, conventional, as well as willow. A higher share of species typical for meadows and pastures such as *L. terrestris, O. lacteum*, *O. cyaneum*, and *A. longa* in the organic system is the result of 1.5 years of clover and grass cultivation preceding winter wheat in rotation. During this period, species building permanent vertical tunnels are not disturbed by soil tillage and receive some fresh plant residues on the soil surface, which promote development of such earthworm communities [35].

#### *4.3. The Influence of Di*ff*erent Farming Systems*

The effect of farming systems on earthworm species richness is a result of different agricultural practices, such as of crop rotation, crop protection, and fertilization.

Application of pesticides in conventional crop production systems can decrease density and biomass of earthworm population [12]. Pesticides may disrupt enzymatic processes, increase individual mortality, decrease fecundity and growth, or even change individual behavior such as feeding rate [5]. Anectic earthworms such as *L. terrestris* are most susceptible to surface application of pesticides, which, in our research, corresponds with a higher density of this species in the organic system in comparison with other agricultural systems where pesticides were used. Since *L. terrestris* forms permanent burrows, it does not come into contact with subsurface soil in its burrows. However,

this species collects plant residues and pulls it into its tunnels, which is why it is directly exposed to the use of pesticides. On the contrary, endogenic species such as *A. caliginosa*, which continuously extend their burrows as they feed in the subsurface soil, are the most susceptible when toxic pesticides are incorporated into the soil [5]. According to Irmler's research [12], the change from conventional to organic management has positively influenced species that form deep tunnels such as *L. terrestris*. The conversion from conventional to organic did not significantly affect the species penetrating shallow, horizontal tunnels, such as *A. caliginiosa* and *A. rosea*.

Herbicides probably do not damage earthworms directly, but they can reduce earthworm populations by decreasing availability of organic matter coming from weeds on the soil surface [40]. According to some authors, soil fauna is more threatened by the use of insecticides than herbicides. Active substances, such as carbofuran, forat, and terbufos, used to control soil-dwelling pests are also extremely toxic to earthworms [41].

Some synthetic mineral fertilizers, especially ammonium sulfate, can be harmful to earthworm populations, probably due to an acidifying affect [5]. On the contrary, the use of manure increases both the number and biomass of earthworms in arable land [42]. This was confirmed by our own study where the earthworm density was significantly the highest in the organic system. Similarly, Pfiffner's and Mader's [43] studies showed that, in conventional fields, where mineral fertilization and integrated plant protection were applied, the number and biomass of earthworms was about 40% lower compared to the fields where mineral-organic fertilization and integrated plant protection was applied. However, in the organic system, the number of earthworms was additionally 80% higher and the biomass was 40% higher in comparison to the object with mineral-organic fertilization and integrated plant protection. Moreover, the quality and quantity of manure could be important factors affecting the earthworm population due to salinity stress [44]. In our study, earthworm abundance under the organic system was higher than under the integrated, conventional, and wheat monoculture systems by 56%, 36%, and 156%, respectively. Similarly, Pfiffner and Luka [45] found about a 50% higher abundance of earthworms in an organic system as compared to integrated ones. Comparisons between organic and conventional systems have shown from an 80% [43] to a 400% higher earthworm density in the organic system [14]. In our research, the biomass of earthworms was 3 times larger in soil under the organic system than in the wheat monoculture, miscanthus, and willow. Similarly, Henneron et al. [14] found a 3 times larger biomass of earthworms in an organic system in comparison with high-input conventional system. In the study by Pfiffner and Mäder [43], the difference in earthworm biomass between organic and conventional systems with mineral fertilization and Integrated Pest Management (IPM) was about 75% and only 35% when the conventional system with mixed mineral and organic fertilization and IPM was compared. Stopping the use of synthetic fertilizers and pesticides are important factors that stimulate the population size and condition of earthworms in the organic system [5,14,26]. In more intensive systems, simplified crop rotation and high input of synthetic fertilizers and pesticides negatively affected the earthworm populations [12].

#### *4.4. The Influence of Perennial Energy Crops*

The influence of perennial energy crops on fauna diversity depends on cultivated species and agricultural practices [23,46]. Our study showed that mean earthworm species number, density, and biomass in energy crops were intermediate between wheat in organic system and high-input, intensive wheat monoculture and were dependent on the type of energy crop. Among tested energy crops, Virginia mallow created the best habitat in terms of species richness and biomass due to high amount of crop residues suitable for earthworms. The small amount of plant residues in miscanthus resulted in low earthworm indicators and high similarity to intensive crop production systems. In Felten and Emmerling's [16] research, the number of earthworm species under miscanthus was placed between intensively cultivated crops (rapeseed, cereals, and maize) and grassland/fallow. In comparison to annual cropping systems, miscanthus plantation enhanced higher densities and diversity of soil invertebrates but not of ground-dwelling invertebrates [47]. In miscanthus stand,

earthworm diversity and abundance were improved in arable soils although biomass may be reduced through poorer food quality [48]. Miscanthus leaf litter does not provide a particularly useful food resource due to its low-nitrogen, high-carbon nature [49], and earthworms feeding on this kind of low-nitrogen material have been found in other studies to lose overall mass [50]. In contrast, though, the extensive litter cover at ground level under miscanthus compared to the bare soil under annual cereals was suggested to be a potentially significant advantage for earthworms in soil surface moisture retention and protection from predation [48].

Another study confirmed that miscanthus created a poorer habitat for fauna than did willow [51]. In our own research, earthworm density was the highest in the soil under willow. Studies conducted in Great Britain and Sweden confirmed the positive impact of willow on the diversity of invertebrates in comparison with crops cultivated in intensive conventional systems on arable land [19,52].

According to Hedde et al. [53], the change in land use from typical annual crops to perennial energy plants resulted in an increase in the density of invertebrates in the soil, which may be caused by a smaller amount of synthetic fertilizers and chemical plant protection chemicals used, with no significant changes in richness and species composition of tested invertebrates. In our study, earthworm indicators were dependent on the type of energy crop which correspond with our working hypothesis. In the cultivation of energy crops, more earthworms depended on the presence of mulch. *L. rubellus* as well as species that burrow deep tunnels, *L. terrestris, A. longa,* and *O. lacteum*, were present, which was also a consequence of the lack of tillage. The composition of earthworm species found in energy plants resembled that found in orchards [35].

Further research should be warranted to design and assess innovative cropping systems including the range of candidate bioenergy crops, possibly grown in alternative lands, and also in the face of future climate changes [54].

#### **5. Conclusions**

It can be concluded that the organic crop production system with a diversified crop rotation including grass-clover mixtures, catch crops, and manure application in the conditions of Eastern Poland (Central Europe) supported the largest diversity and abundance of earthworms in comparison to the high-input cereal monoculture and integrated systems. The organic system favoured the population of *L. terrestris* and juvenile Lumbricidae. The intensification of agricultural production by simplified crop rotation and the input of synthetic fertilizers and chemical plant protection products caused a decrease in the number of species and abundance of earthworms. *A. caliginiosa* dominated in the earthworm community in the monoculture of wheat. On plantations of energy crops, the earthworm population indices were located between the organic system and the high-input, intensive wheat monoculture. The effect of energy crop cultivation on earthworm abundance and ecosystem services which are provided depends on the respective crop species. Among the compared energy crops, Virginia mallow created the best habitat for earthworms. In miscanthus, earthworm community was the poorest and the most similar to wheat monoculture. The composition of earthworm species found in energy plants resembled that found in orchards. Proper management of energy crops can support biodiversity and ecosystem services supplied by earthworms, such as humus production..

**Author Contributions:** Conceptualization, M.M., J.S., and B.F.-S.; methodology, B.F.-S., P.R., and J.S.; validation, M.M. and J.S.; investigation, B.F.-S. and P.R.; data curation, B.F.-S. and P.R.; writing—original draft preparation, B.F.-S. and P.R.; writing—review and editing, M.M. and J.S.; visualization, B.F.-S. and P.R.; supervision, M.M. and J.S.; project administration, J.S. and M.M.; funding acquisition, J.S. and M.M.

**Funding:** The study was carried out under the statutory project of IUNG-PIB no 1.13. (2014–2017). This work has been cofinanced by the National (Polish) Centre for Research and Development (NCBiR) entitled "Environment, agriculture and forestry" project: BIOproducts from lignocellulosic biomass derived from MArginal land to fill the Gap in Current national bioeconomy, No. BIOSTRATEG3/344253/2/NCBR/2017.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **Appendix A**


**Table A1.** Inter set correlations of agricultural systems variables with Principal Component Analysis (PCA) axes in study of earthworms communities.

<sup>1</sup> ORG—organic; INT—integrated; CON—conventional; MON—monoculture, MIS—miscanthus, VIR—virginia mallow, WIL—willow.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Communication* **Do Long-Term Continuous Cropping and Pesticides A**ff**ect Earthworm Communities?**

#### **Kinga Treder \*, Magdalena Jastrz ˛ebska, Marta Katarzyna Kostrzewska and Przemysław Makowski**

Received: 4 March 2020; Accepted: 16 April 2020; Published: 20 April 2020

Department of Agroecosystems, Faculty of Environmental Management and Agriculture, University of Warmia and Mazury in Olsztyn, Plac Łódzki 3, 10-718 Olsztyn, Poland; magdalena.jastrzebska@uwm.edu.pl (M.J.); marta.kostrzewska@uwm.edu.pl (M.K.K.); przemyslaw.makowski@uwm.edu.pl (P.M.) **\*** Correspondence: kinga.treder@uwm.edu.pl

**Abstract:** Earthworm species composition, the density of individuals, and their biomass were investigated in spring barley and faba bean fields in a long-term (52-year) experiment conducted at the Production and Experimental Station in Bałcyny, in north-eastern Poland (53◦40 N; 19◦50 E). Additionally, post-harvest residues biomass, soil organic matter (SOM), and soil pH were recorded. The above traits were investigated using two experimental factors: I. cropping system—continuous cropping (CC) vs. crop rotation (CR) and II. pesticide plant protection: herbicide + fungicide (HF+) vs. no plant protection (HF−). A total of three species of *Lumbricidae* were found: *Aporrectodea caliginosa* (Sav.) in both crops, *Aporrectodea rosea* (Sav.) in spring barley, and *Lumbricus terrestris* (L.) in faba bean. The density and biomass of earthworms were unaffected by experimental treatments in spring barley fields, whereas in faba bean CC increased and HF+ decreased earthworm density and biomass in comparison with CR and HF− respectively. Total post-harvest residues in faba bean fields were higher under CC in relation to CR and under HF+ compared with HF− treatment in both crops. Compared to CR, CC increased soil pH in spring barley fields and decreased in faba bean fields. Experimental factors did not affect SOM. Earthworm density and biomass were positively correlated with SOM content.

**Keywords:** soil organic matter; soil pH; post-harvest residues; crop rotation; *Hordeum vulgare* L.; *Vicia faba* L. ssp. *minor*

#### **1. Introduction**

Earthworms are strategic invertebrates in agroecosystems. The drillosphere, composed of horizontal and vertical burrows and casts created by earthworms, significantly affects soil structure and enhances gas exchange, water infiltration, and root penetration across the soil profile [1–4]. Earthworms improve the content of soil organic matter, contribute to humus formation processes and form a mull soil by burrowing large quantities of surface organic matter to belowground and relocating soil from depths to the top by casting. These invertebrates have an impact on the structure, concentration, and activity of soil microbial communities involved in organic matter decomposition and mineralization [5,6]. Earthworms casts are characterized by higher pH, C, Ca2<sup>+</sup>, Mg2<sup>+</sup>, and K<sup>+</sup> contents than surrounding soil aggregates and incorporate nutrients available for plants [7,8]. The N mineral availability increases with earthworm abundance [9,10]. Earthworms, through their interaction with microorganisms, are essential factors influencing soil organic carbon and its dynamics [10,11]. In the presence of earthworms, greater production of plant growth regulators was observed [12,13]. The non-negligible role of earthworms on improving plant tolerance to parasitic nematodes and a reduction in the severity of take-all disease was also reported [14–16]. The above-mentioned benefits

of earthworm activity contribute to plant growth and production as seen by aboveground biomass and crop yield increases [17–19].

Density, diversity, structure, and activity of earthworm populations in agroecosystems are dependent on agricultural management [20]. Intensification of agricultural practices based on multiple tillage treatments, simple crop sequence, minor organic fertilization, and chemical methods of plant protection have a negative effect on earthworm populations. Long-term, intensive, and deep tillage can decline earthworm (mainly anecic) abundance [21–24] whereas shallow plowing with residue mixing and conservation cultivation techniques can increase their number [25–27]. Organic fertilizers such as manure, crop residues, and mulch are the source of food supply for soil biota and have a positive impact on their populations [28,29]. The conclusion from the literature on the effect of pesticides on earthworm populations is still ambiguous. The effect of pesticides on soil organisms is closely related to their active substances and doses. Some of them, especially fungicides and insecticides, are toxic or lethal to earthworms and cause a decrease in cocoon production and density of juveniles, a delay in growth and a mortality increase [3,30–32]. Herbicides were also found to have an adverse impact on earthworms by causing histological changes in their body tissues and increasing mortality [33–37]. However, earthworms can develop adaptation mechanisms against the toxic effect of pesticides [38,39].

The importance of earthworms in agroecosystems is well-recognized, but the long-term effects of continuous cropping and pesticide use on earthworm populations are much less documented. The objective of this study was to examine the impact of the above-mentioned factors on species composition, density, and biomass of earthworms, post-harvest residue biomass, soil organic matter, and soil pH. The alternative hypothesis assumed that long-term continuous cropping and chemical plant protection have an impact on earthworm communities was tested against the null hypothesis that the above factors do not affect the analyzed parameters.

#### **2. Materials and Methods**

#### *2.1. Experimental Design and Crop Management*

The field experiment was initiated in autumn 1967 in the Production and Experimental Station in Bałcyny in north-eastern Poland (53◦40 N; 19◦50 E). During the first five years, nine crops were sown in a continuous cropping system (growing of the same crop on the same field each year). In 1972, two crop rotations were included to analyze continuous cropping impact. Crop rotation varied throughout the experiment. Currently, twelve crops in continuous cropping and in two crop rotations (growing different crops one after the other on the same field) are being sown. The crop rotations are A. sugar beet, maize, spring barley, peas, winter rape, and winter wheat; B. potato, oats, fiber flax, winter rye, faba bean, and winter triticale.

Fertilization in the first sixteen years was the only mineral. Since 1983 farmyard manure was included in doses: 30 t ha−<sup>1</sup> on potato/sugar beet field and 15 t ha−<sup>1</sup> every three years in a continuous cropping system. Mineral fertilizers are applied in terms and doses respective to each crop's needs.

In one part of every crop field, no pesticides have been ever applied, which provides a unique chance to study no plant protection effect after 52-years of a continuous cropping and crop rotation system. To have a comparison for these results on the other parts of each crop field, herbicides (since 1972 till now) and fungicides (since 1983 till now) have been included. Throughout the experiment, the use of pesticides has been updated according to The Institute of Plant Protection National Research Institute recommendations.

The results presented in this paper were based on two crops differing in their biology and agricultural importance: spring barley (cultivar Radek) and faba bean (cultivar Amigo). Spring barley is grass with a short root system and short vegetation period whereas faba bean has a deep, well-developed root system with nitrogen-fixing nodules. Faba bean, by nitrogen fixation and high mass of residues, increases soil biological activity, organic matter content, porosity and soil moisture, which has an impact on earthworm communities [40].

Basic agricultural data for spring barley and faba bean in 2019 are presented in Table 1.


**Table 1.** Basic agricultural data for spring barley and faba bean in 2019.

1—before sowing, 2—at stem elongation stage; \* trade name; active ingredient; rate; crop growth stage.

Two experimental factors were investigated, each with two levels: I. cropping system: continuous cropping (CC) vs. crop rotation (CR), II. plant protection: herbicide + fungicide (HF+) vs. no plant protection (HF−). In both spring barley and faba bean, particular experimental treatments (CC-HF+, CC-HF–, CR-HF+, CR-HF−) were performed in 3 replications (i.e., plots). From each plot 3 samples were taken. That brought the total to 9 samples for each treatment. Each plot size was 27 m2.

#### *2.2. Soil Characteristics*

The experiment was established on Luvisol medium soil, derived from light loam lying on loamy sand. At the beginning of the lasting rotation (2016) soil contained an average (mg kg<sup>−</sup>1) of available forms of phosphorous—289.3, potassium—258.5, magnesium—55.0, total nitrogen—800 with 1.1% Corg, and pH—5.7.

#### *2.3. Meteorological Data*

The climate in this region is temperate humid, with annual total precipitation around 587.5 mm and a mean annual air temperature of 7.9 ◦C (data for the years 1981–2015). Weather conditions of the July–September 2019 period are presented in Figure 1. High air temperature in combination with small rainfall before sampling in August could have reduced earthworm activity and biomass. In September, the rainfall and temperature were more favorable for earthworms [41].

**Figure 1.** Daily temperature and rainfall before sampling.

#### *2.4. Sampling*

#### 2.4.1. Post-Harvest Residue Sampling and Preparation

Just after harvest, the post-harvest residues (crop and weed bottom stalks with roots) were removed from a surface area of 0.40 m2 and a depth of 0.30 m in three replications from each plot. Residues were transported to the laboratory in plastic bags. Crop and weed residues were separated, washed under tap water and air-dried for several days. Crop residues were split into shoots and roots. The dry mass of post-harvest residues was weighed.

#### 2.4.2. Soil Sampling and Preparation

Three soil samples from each plot were selected on the days of earthworm collection. Soil samples were taken in three replications from each plot from a 0–30 cm depth with a hand-held twisting probe (Egner's soil sampler) and returned in plastic bags to the laboratory. Stones and plant residues were removed from the soil. Soil samples were air-dried in plastic trays and then passed through a 2 mm sieve to prepare homogenous samples.

#### Soil Organic Matter Analysis

The soil organic matter (SOM) was determined using the loss-on-ignition method [42]. Soil samples (each of mass 10 g) were oven-dried in crucibles at 105 ◦C for 2 h to remove hygroscopic water and then cooled and weighed. Afterward, soil samples were heated at 360 ◦C for 2 h and cooled and weighed. SOM (%) was calculated as the mass lost during combustion:

$$\text{LSOM} = \text{[(soil weight 105 °C - soil weight 360 °C)/soil weight 105 °C]} \times 100 \tag{1}$$

#### Soil pH Determination

Soil pH was determined using the potentiometric method after water extraction (pH (H2O)). In glass beakers, 10 g of the air-dried and sieved soil samples were mixed with 25 mL of distilled water and shaken. After two hours of equilibration, electrical conductance was measured in soil-water suspension for 10 min using a Hanna HI 221 (Hanna Instruments) pH meter.

#### 2.4.3. Earthworm Sampling and Identification

On 12 plots of each crop, three soil samples (25 cm × 25 cm × 40 cm depth) were sampled after each crop harvest: spring barley—1.08.2019 and faba bean—26.09.2019. Earthworms were selected from soil blocks by the hand-sorting method. In the laboratory, earthworms were rinsed, dried, weighed (data refers to live biomass), narcotized in 35% ethyl alcohol and preserved in 4% formalin and 75% ethyl alcohol. Clitelated individuals were classified into species level, and juveniles were classified into species or genera level by external morphology using keys [41].

#### *2.5. Statistical Analysis*

The data were analyzed statistically by a two-factorial analysis of variance ANOVA in the STATISTICA (data analysis software system), version 12, StatSoft. Homogeneous groups were estimated by Duncan's test at a *p* < 0.05. The Shapiro–Wilk W-test was used for testing the normality of variable distribution and Levene's test for homogeneity of variance. The correlation coefficients were calculated to measure the strength and direction of the relationship between the variables. The coefficients were determined based on the data from all treatments, separately for spring barley and faba bean.

#### **3. Results and Discussion**

#### *3.1. Post-Harvest Residues*

In spring barley fields, the cropping system did not affect the dry mass of post-harvest residue (Table 2). However, CC related to CR increased faba bean mass of shoots, weeds and total residues. Positive influence of CC on faba bean shoots residues biomass was an unexpected result because growing faba bean in the same field year after year leads to the accumulation of autotoxic compounds (mainly phenolic acids) in the soil that inhibits plants growth [43]. In previous studies based on this experiment, Rychcik and Zawi´slak [44] reported lower faba density under CC than CR treatment. Though, the lower density of plants may result in their higher biomass. This effect was reported by Kotecki in the experiment with faba bean, were biomass of one plant increased with density decrease [45]. Post-harvest biomass of weeds is the aftereffect of weeds density. The current consensus is that crop rotation, in contrast to continuous cropping, decreases weeds density, which was confirmed by Rychcik in faba bean fields [46].

HF+ compared to HF− treatment resulted in higher production of residues in both crops: in spring barley—shoots, in faba bean—shoots and roots. Furthermore, the biomass of weeds residues was significantly higher in HF− cereal and legume fields in relation to HF+. The presented results are obvious, considering the effects of herbicide use, which was in line with previous studies [46].

In spring barley fields, CC-HF+ increased the total residue biomass in comparison to CC-HF−, while there was no difference between CR-HF+ and CR-HF−. CR-HF− resulted in less spring barley shoots mass production related to CC-HF+ but it was higher than under CC-HF−. The biomass of weed residues under CC-HF− and CR-HF− was higher than in CR-HF+. The achieved results are in agreement with a previous study concerning weed infestation of spring barley in this experiment [47].

Similar results were noted in faba bean fields. The total residue biomass in CC-HF− and CR-HF+ was significantly lower than in CC-HF+ and higher in relation to CR-HF− interaction.

Moreover, HF+ in relation to HF− increased the biomass of faba bean shoots residues in CR, but not as strong as with CC. In relation to other treatments, CC-HF−increased the mass of weed residues. The biomass of faba bean roots was unaffected by treatment interactions.


**Table 2.** Dry mass of post-harvest residues (t ha<sup>−</sup>1).

\* CR—crop rotation, CC—continuous cropping, HF−—no plant protection, HF+—herbicide + fungicide; \*\* values with different letters vary significantly (Duncan's test, *p* < 0.05), x ± sem—mean ± standard error of mean.

#### *3.2. Soil Organic Matter Content*

Experimental factors and their interactions did not affect the SOM in spring barley and faba bean fields (Table 3). These results are in line with the previous study based on this experiment where the Corg content in spring barley was comparable to faba bean fields in both cropping systems [48]. The same amounts of manure during every six-year-lasting rotation were applied in all treatments, so this may be the major cause of undifferentiated SOM levels. In contrast to the presented results, some authors [49–51] assert that crop rotation, especially with legumes, gives preferential conditions for soil C increase.



\* CR—crop rotation, CC—continuous cropping, HF−—no plant protection, HF+—herbicide + fungicide; \*\* values do not differ significantly (Duncan's test, *p* < 0.05), x ± sem—mean ± standard error of mean.

#### *3.3. Soil pH*

In comparison with CR, CC increased soil pH in spring barley (Table 4). Hickman [52] reported lower pH in maize continuous cropping than in maize—wheat and maize—soybean rotations and soybean continuous cropping. The author suggested that these results may be explained by the long-term use of anhydrous ammonia in maize crops fields. HF+ lowered soil pH in relation to HF−. Spring barley under HF+ achieved higher yields than HF− (data not published). With higher yields, larger amounts of macroelements were removed from the soil. In spring barley fields, the interaction between cropping system and plant protection was proved. Under CR-HF+ treatment lower soil pH than under CR-HF− was noted, while there was no difference in soil pH under CC-HF− and CC-HF+. In faba bean fields, the values of soil pH were lower in CC than in CR. Dinitrogen-fixing legumes, including faba bean, are considered to generate soil acidification by releasing H+ to rhizosphere [53–55]. Lee [56] reported that continuous legume cultivation increased soil acidity. Comparably, Williams [57] noted soil pH decrease in a long-term experiment with clover pastures.



\* CR—crop rotation, CC—continuous cropping, HF−—no plant protection, HF+—herbicide + fungicide; \*\* values with different letters vary significantly (Duncan's test, *p* < 0.05), x ± sem—mean ± standard error of mean.

HF+ caused pH decrease only in spring barley fields. Spring barley under HF+ produced higher yields than under HF− (own data not published). Thus with higher yields, larger amounts of macroelements were removed from the soil.

In faba bean fields, no interaction of cropping system and plant protection on soil pH was revealed.

#### *3.4. Earthworm Species Richness and Structure*

In the experimental fields, only three species of *Lumbricidae* were found: *Aporrectodea caliginosa* (Sav.), *Aporrectodea rose*a (Sav.), and *Lumbricus terrestris* (L.) (Figure 2). The same species composition was reported by Valchovski [58] in cultivated and non-cultivated Vertic Luvisol. The species richness of earthworms in agroecosystems is usually lower than in natural ecosystems and depends mainly on soil type, soil humidity, organic matter content, cultivation treatments, and crop type [59–61]. In both crops, the most numerous species was *Aporrectodea caliginosa* (Sav.), which occurs commonly in arable lands in all temperate zones [62–64]. This endogenic earthworm can adapt to unfavorable environmental conditions like low soil moisture, low organic matter content, or tillage practices [22,65–67]. Although *Aporrectodea rosea* (Sav.) is very common in agroecosystems in different crops as well as in pastures [20,62,68], in the current study it was recorded only in spring barley fields under CR-HF+ and CC-HF− treatments. It is worth noting that *Lumbricus terrestris* (L.) was found only in CR-HF+ and CC-HF+ faba bean fields. Edwards [62] suggests that one of the limiting factors for *Lumbricus terrestris* (L.) abundance is soil organic matter. Faba bean plants have a well-developed, extensive fibrous root system, which gives preferential treatment to anecic earthworms.

0% 20% 40% 60% 80% 100% CR-HF- CR-HF+ CC-HF- CC-HF+ *Aporrectodea caliginosa Aporrectodea rosea Aporrectodea juveniles*

In both crops, the majority of the earthworms found were juveniles representing the genera *Aporrectodea*.

(**b**)

**Figure 2.** Relative abundance (%) of earthworm species (based on individuals m<sup>−</sup>2) on (**a**) spring barley and (**b**) faba bean (B) fields; CR—crop rotation, CC—continuous cropping, HF−—no plant protection, HF+—herbicide + fungicide.

#### *3.5. Earthworm Density and Biomass*

The density of individuals and biomass of collected earthworms was lower than reported in other works [64,69,70], but comparable with results achieved in other researchers conducted in The Experimental Station in Bałcyny [71–74]. Experimental treatments did not affect the density or biomass of earthworms in spring barley fields (Table 5). Spring crops do not create favorable environmental conditions for earthworm abundance because of low organic matter input and agrotechnical works on fields in early spring when the activity of earthworm communities increases.

In faba bean fields, both experimental treatments affected earthworm abundance. In comparison with CR, CC increased the density of individuals and biomass of earthworms. The reason for this could be the higher mass of post-harvest residues. In a study by Edwards [62], the abundance of earthworms was two times higher in fields with continuous wheat (for 136 years) than with wheat—root crops. The negative influence of HF+ on the earthworm population in relation to HF− on faba bean fields was apparent in the decrease in individual density and their biomass. Many authors reported a reduction in growth, biomass loss, decreased cocoon production or higher mortality after pesticide application [75–79]. Biomass reduction may be an effect of reduced food intake as a strategy to avoid contamination. Nonetheless, the response to agrochemicals differs depending on earthworm species, the concentration of toxic substances, environmental conditions (soil type, temperature, humidity, organic matter content, etc.) and the duration of exposure [77,80,81]. CC-HF− treatment had a positive influence on earthworm density and biomass in relation to other experimental treatments.


**Table 5.** Density (individuals m<sup>−</sup>2) and biomass (g m−2) of earthworms in spring barley and faba bean fields.

\* CR—crop rotation, CC—continuous cropping, HF−—no plant protection, HF+—herbicide + fungicide; \*\* values with different letters vary significantly (Duncan's test, *p* < 0.05), no letters—no significant differences, x ± sem—mean ± standard error of mean.

#### *3.6. Relationship Between Earthworm Abundance and Post-Harvest Residues, SOM, and Soil pH*

In spring barley fields, no relationship between earthworm abundance and post-harvest residues, SOM or soil pH was observed (Table 6). A strong positive association of SOM and earthworm abundance was noted in faba bean fields This suggests that organic matter left in the field by a legume, characterized by high protein content may be beneficial for earthworm populations. The above is in the line with Kladivko [82], where a higher density of earthworms in continuous soybean than in continuous corn was reported. The association between soil organic matter content and the presence of the earthworms was also observed in other findings [83,84]. Soil pH and dry mass of post-harvest residues were not significantly correlated with earthworm density or biomass.


**Table 6.** The correlation coefficient (r) between density and biomass of earthworms with SOM, pH, and post-harvest residues of spring barley and faba bean fields.

\* coefficient significant at *p* < 0.05.

#### **4. Conclusions**

In the current study post-harvest residues were unaffected by cropping system, but HF+ compared to HF− increased spring barely shoots residues and total post-harvest biomass and decreased weed post-harvest biomass. Experimental factors did not differentiate SOM in spring barley fields. CC, in relation to CR, increased soil pH whereas HF+ decreased it in comparison with HF−. In spring barley fields, two earthworm species were found: *Aporrectodea caliginosa* (Sav.) and *Aporrectodea rosea* (Sav.). CC and HF+ did not affect earthworm density or biomass in relation to CR and HF−. No correlation between earthworm abundance and post-harvest residues, SOM, or soil pH was noted.

In faba bean fields above-ground post-harvest residue biomass was increased by CC in relation to CR. In turn, HF+ compared with HF− increased total and faba bean shoots and roots post-harvest biomass but decreased weeds post-harvest biomass. In faba bean fields, SOM stayed at the same level regardless of experimental factors. Lower pH values were noted under CC than CR treatment, whereas HF+ did not affect it. *Aporrectodea caliginosa* (Sav.) and *Lumbricus terrestris* (L.) were recorded in faba bean fields. CC increased earthworm density and biomass in comparison with CR whereas HF+ decreased these features in relation to HF−. Positive correlations between earthworm density and biomass and SOM content were noted in faba bean fields.

**Author Contributions:** Conceptualization, K.T., M.J., and M.K.K.; methodology, K.T., M.J., and M.K.K.; validation, M.K.K. and P.M.; investigation, K.T. and P.M.; formal analysis, K.T., M.J., and P.M.; data curation, K.T. and P.M.; visualization, K.T., M.J., and M.K.K.; writing—original draft preparation, K.T.; writing—review and editing, M.J. and M.K.K.; funding acquisition, K.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was financially supported by the Minister of Science and Higher Education within the framework of the program entitled "Regional Initiative of Excellence" for the years 2019–2022, Project No. 010/RID/2018/19, amount of funding 12.000.000 PLN.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Communication* **Phosphorus Fertilizers From Sewage Sludge Ash and Animal Blood Have No E**ff**ect on Earthworms**

#### **Magdalena Jastrz ˛ebska \*, Marta K. Kostrzewska and Kinga Treder**

Department of Agroecosystems, Faculty of Environmental Management and Agriculture, University of Warmia and Mazury in Olsztyn, Plac Łódzki 3, 10-718 Olsztyn, Poland; marta.kostrzewska@uwm.edu.pl (M.K.K.); kinga.treder@uwm.edu.pl (K.T.)

**\*** Correspondence: jama@uwm.edu.pl; Tel.: +48-89-523-4829

Received: 9 March 2020; Accepted: 5 April 2020; Published: 7 April 2020

**Abstract:** Soil invertebrates are crucial for agroecosystem functioning yet sensitive to agricultural practices, including fertilization. Considering the postulates of circular phosphorus economy, the use of fertilizers from secondary raw materials is likely to return and increase and may even become obligatory. The effects of recycled fertilizers on soil fauna communities, however, remain poorly understood. In this paper, the effect of phosphorus fertilizer (RecF) and biofertilizer (RecB) from sewage sludge ash and dried animal (porcine) blood on earthworm's occurrence in soil is discussed. RecB is RecF activated by phosphorus-solubilizing bacteria, *Bacillus megaterium.* Waste-based fertilizers were assessed in field experiments against commercial superphosphate and no P fertilization. Three levels of P doses were established (17.6, 26.4, and 35.2 kg P ha<sup>−</sup>1). Earthworms were collected after the test crop harvest (spring or winter wheat). In the experiments two earthworm species, *Aporrectodea caliginosa* and *Aporrectodea rosea*, were identified. A large proportion of juvenile individuals were recorded in 2017. The recycled fertilizers used in the experiments used in recommended doses, similarly to superphosphate, did not alter the density, biomass, species composition, and structure of earthworms. Further long-term field research is recommended.

**Keywords:** *Lumbricidae*; *Aporrectodea caliginosa*; *Aporrectodea rosea*; phosphorus fertilizers; phosphorus-solubilizing microorganisms; renewable resources; heavy metals; Luvisols; wheat

#### **1. Introduction**

Earthworms (*Lumbricidae*) are listed among the most important soil-dwelling invertebrates [1]. They constitute a major component of soil fauna communities in most ecosystems [2]. The role of earthworms in soil fertility has been known for over a century [2]. So far, a great number of studies have been undertaken which highlight direct and indirect effects of their activity on biotic and abiotic soil properties, and, consequently, plant productivity. Due to their services, earthworms are referred to as ecosystem engineers [3,4] and indicators of biological soil health [5,6].

The occurrence, distribution, and abundance of earthworms can be affected by a range of environmental factors, including climate, soil conditions, food sources, metal concentration, and predator pressure [5]. In addition, in agroecosystems, agricultural practices such as irrigation, tillage, lime application, fertilizer and pesticide use, drainage, crop rotation, and cover crops influence earthworm abundance and activity [7] because they change one or more of the factors listed above [5,8].

Despite potential soil pollution [9], increased use of inorganic fertilizers to enhance crop yields is a common practice in modern agriculture. Both beneficial and harmful effects of inorganic fertilizers on earthworm populations have been observed [10]. The positive effect is believed to be an indirect consequence of increased crop biomass production and the resulting increase in organic residues [11]. On the other hand, the toxic effects of inorganic fertilizers on earthworms, especially upon direct contact, have been reported [12,13].

Modern European agriculture faces a shortage of primary phosphorus (P) sources. Phosphate rock was included in the EU list of critical resources in 2014 [14]. A circular P economy, including recycling, seems to be a necessity in this part of the world. Inorganic and organic waste are often a source of nutrients in fertilizers [15,16]. As has been proved in numerous scientific centers, phosphate rocks can be replaced with P-rich secondary raw materials [17–19]. Municipal and industrial byproducts such as sewage sludge ash (SSA), animal bones, and blood may constitute the basis for alternative fertilizers [19]. An innovative approach, initiated to activate P from raw material, is the inclusion of phosphorus-solubilizing microbes (PSM) into waste-based preparations [20]. The use of recycled fertilizers is expected not only to provide satisfactory yields in terms of quantity [21,22] and quality, but also not to cause negative changes in the soil environment. Concerning the latter, it should be taken into account that the introduction of nutrient carrier and PSM to the soil could alter soil properties both directly (nutrient content and availability, pH, possible presence of toxic elements) and indirectly (e.g., through microbial activity modification or plant growth stimulation) [23]. Changes in habitat conditions could affect earthworm populations. It is also crucial to be aware that the consequences of recycled fertilizer use, while being invisible in the short term, may lead to significant environmental changes in the long term [24,25].

The aim of this research has been to determine the impact of the fertilizers produced from SSA and animal blood on earthworm occurrence in the soil. The recycled fertilizer (RecF) and biofertilizer (RecB), i.e., RecF activated by *Bacillus megaterium* bacteria (PSM) were assessed against superphosphate, a commercial phosphorus fertilizer. It was hypothesized that the impact of the recycled fertilizers on soil earthworms would be similar or more favorable/less harmful than that of the traditional P fertilizer.

#### **2. Materials and Methods**

#### *2.1. Fertilizers*

In field experiments, the recycled P fertilizer (RecF) and biofertilizer (RecB) were compared to a commercial fertilizer superphosphate (SP). These preparations were manufactured from sewage sludge ash (ash from the incineration of sewage sludge biomass from wastewater treatment; SSA) and dried animal (porcine) blood. During RecB production, raw material (SSA + blood) was biologically activated by phosphorus-solubilizing bacteria, *Bacillus megaterium*. Both products were in the form of granules.

RecF and RecB were produced at the Institute of New Chemical Syntheses in Puławy (Poland), according to a concept developed at the Wrocław University of Science and Technology (Wrocław, Poland). The SSA originated from the Municipal Wastewater Treatment Plant 'Łyna' in Olsztyn (Poland), and dried blood was obtained from the meat processing industry. The bacteria strains were obtained from the Polish Collection of Microorganisms at the Institute of Immunology and Experimental Therapy of the Polish Academy of Sciences in Wrocław (Poland). The elemental composition of the recycled fertilizers is presented in Table 1. The production process was described by Rolewicz et al. [26].


**Table 1.** Elemental composition of the recycled fertilizers.

According to the Department of Advanced Material Technologies of the Wrocław University of Science and Technology (Wrocław, Poland).

Superphosphate FosdarTM 40 (Gda ´nsk Phosphorus Fertilizer Plant 'Fosfory' Sp. z o.o., Gda ´nsk, Poland) was purchased on the market. This P fertilizer contains 17.6% P, 7.15% Ca, 2.00% S, and microelements (B, Co, Cu, Fe, Mn, Mo, and Zn), according to the commercial information provided on the label.

#### *2.2. Soil and Meteorological Conditions*

Three field experiments with spring (2016, 2017) or winter (2017; sown in autumn 2016) common wheat (*Triticum aestivum* ssp. *vulgare* MacKey) were conducted. In each experiment, the soil on which wheat was grown met the requirements of the species (Table 2) and was within the range of soils preferred by earthworms [27].


**Table 2.** Soil characteristics before the start of the experiments.

<sup>1</sup> According to World reference base for soil resources 2014 [28].

Meteorological conditions in the period of one month before earthworm sampling are presented in Table 3. In both growing seasons, fairly heavy rainfall and moderate temperatures in July and early August could have stimulated earthworm activity at the time of earthworm sampling [29,30].


**Table 3.** Atmospheric precipitation and air temperature during the study period according to the Meteorological Station in Bałcyny, Poland.

#### *2.3. Experimental Design and Agronomic Management*

In the field experiments, RecF and RecB were assessed against SP and no phosphorus (No P) treatments. In addition, three different P levels were established: (1) 17.6, (2) 26.4, and (3) 35.2 kg P ha–1; therefore, finally, ten treatments of P fertilization were compared (Table 4).


**Table 4.** Fertilization treatments compared in the experiments.

Phosphorus fertilizers were applied before the sowing of wheat. They were manually scattered on the soil surface and then mixed with the soil by harrowing. Other basic agrotechnical data for the experiments are presented in Table 5.


**Table 5.** Basic agricultural data for the experiments.

<sup>1</sup> potassium chloride, <sup>2</sup> ammonium sulphate, – not applied.

Experiments were established in a randomized block design. In each experiment, particular experimental treatments were performed in four replications (plots) (Figure S1). The area of a single experimental plot was 20 m2.

#### *2.4. Earthworm Sampling and Identification*

Earthworms were harvested mechanically 2–3 days after the wheat harvest. Soil columns with a surface area of 0.0625 m<sup>2</sup> (0.25 m <sup>×</sup> 0.25 m) and a depth of 0.4 m were dug out of each plot, then crushed and passed through a sieve, and individuals of *Lumbricidae* were collected. Afterwards, the earthworms were transported to the laboratory, where they were washed, counted, and weighed. Anaesthetized in a 30% ethanol (Czempur, Piekary Sl ˛ ´ askie, Poland) solution, earthworms were preserved in a 4% formalin (Czempur, Piekary Sl ˛ ´ askie, Poland) and 75% ethanol solution for the subsequent analysis of the species composition. The earthworms were sorted into adults and juvenile forms. The adult individuals were further classified into species using an identification key to soil-dwelling oligochaetes [31]. The species composition, number and biomass of earthworms in the 0–0.4 m soil layer were expressed per 1 m2 of plot area.

#### *2.5. Statistical Analysis*

The normality of variable distribution was checked using the Shapiro–Wilk W-test, and the homogeneity of variance was checked using Levene's test. Since the assumptions of the analysis of variance were not met, the results were processed by the alternative nonparametric Kruskal–Wallis test. The calculations were performed using Statistica 12.0 software [32].

#### **3. Results and Discussion**

Both in 2016 and 2017, thermal and rain conditions in July and early August promoted earthworm presence in the 0–0.4 m soil layer. Having found convenient habitat moisture at this level of the soil profile, the individuals of *Lumbricidae* did not enter into diapause or migrate deeper into the soil seeking better conditions [30]. The density of earthworms found in the studied soil columns ranged from 6 to 44 individuals and the biomass from 1.1 to 21.5 g per m<sup>2</sup> (Table 6). These values are similar to those presented by Tiwari [33] from a sandy loam Oxisol in India, but smaller than the values reported by other authors from different arable soils in Poland [4] and Slovakia [34]. The abovementioned differences may have been caused by different timing of sampling, which did not correspond to the periods of the highest earthworm activity (spring and autumn) indicated in the literature [4,34]. In 2016, the average earthworm biomass was relatively higher than in 2017 due to a greater share of adult individuals in the community.



No significant differences between treatments according to the Kruskal–Wallis test at *p* ≤ 0.05.

In all experiments, only two earthworm species were identified, i.e., *Aporrectodea caliginosa* and *Aporrectodea rosea* (Figure 1), which is hardly surprising. These species are among the most common in Poland [31] and Europe [35], and they were the only ones recorded by Kanianska et al. [34] in some study sites in Slovakia. In 2016, mainly adult earthworms were noted, and on average, *A. caliginosa* and *A. rosea* occurred in similar proportions (42% and 39%, respectively). In 2017, among the earthworm individuals found after spring wheat harvest, juvenile forms dominated, often constituting 100% of the community. Adults were found sporadically. A large proportion of juvenile forms (mostly over 50%) were also recorded in the soil after the winter wheat harvest. In this experiment, *A. rosea* was predominant. A high number of juvenile individuals is often thought to be an indicator of suitable conditions for earthworm development [29,36]. A dominance of juvenile forms over adult earthworms has also been noticed by other authors [4,34].

**Figure 1.** Species composition and structure of earthworms (based on the density of individuals). No significant differences between treatments according to the Kruskal–Wallis test at *p* ≤ 0.05.

In none of the conducted experiments did the earthworm density and biomass depend on the type of P fertilizers used or their doses (Table 6). Moreover, earthworm abundance (density and biomass) under no P treatment did not differ from that under fertilizers. In addition, no evident link between the species composition and structure of earthworms and the applied P fertilization was observed (Figure 1).

To compare, in the study by Tiwari [33] conducted in an Oxisol (India), the single superphosphate applied at P dose of 25 kg ha–1 did not change the earthworm density and biomass in comparison to control treatment (no fertilizer). An increase in the number and biomass of earthworms with the addition of superphosphate to pastures in Australia and New Zealand was reported [37]; however, the authors argued that P fertilizer led to an increase in plant production in these ecosystems and, hence, available food. In contrast, in other studies [34,38,39], a negative relationship between earthworm biomass and P content in soil was found. Some authors proved that inorganic fertilizers, including superphosphate, can be toxic to earthworms upon direct contact [12,13].

In the current study, the SSA is the main raw material for the fertilizers produced, and one that may raise concerns about the heavy metal presence [18]. The issue of toxic element occurrence is key since Khan et al. [40], based on a pot experiment, claimed that the high content of heavy metals in the tested fiber and chemical industry sludge ashes was the reason for the decrease in the number of adults, juveniles, cocoons, and fresh weight of the earthworm *Pheretima posthuma* found four months after the waste application. Using animal blood as a fertilizer for organic farming [41,42] and a fertilizer binder [43] was recommended. The content of potentially toxic elements in fertilizers tested in the current study was low (Table 1), and the fertilizer doses used were not excessive. According to other research, metals such as copper (Cu), zinc (Zn), and iron (Fe), which are contained in RecF and RecB fertilizers, may also be toxic to earthworms [13,44,45], although they play the role of microelements for plants. Neuhauser et al. [44] proved that Cu and Zn were more toxic to *Eisenia fetida* than cadmium (Cd) and lead (Pb). Toxicity of aluminum (Al) to earthworms was reported as well [46]. Additional reflections (and caution) should also be prompted by studies on long-term use of sewage sludge documenting the negative impact of metal accumulation in the soil on soil microorganisms [24,25,47].

To date, only a few studies have examined the effect of SSA-based fertilizers on earthworms. Rastetter et al. [48] ecotoxicologically analyzed three crystallization products and five ash products of recovered phosphate-containing materials, obtained from treated sewage sludge, sludge liquors or sludge ashes from municipal wastewater treatment plants in Europe. The phosphate recyclates were compared with a conventional phosphate fertilizer (triple superphosphate). The avoidance test with the earthworm *Eisenia fetida* was used to determine the effects of chemicals on behavior of earthworms. The authors concluded that relevant agronomical application amounts of all phosphate recyclates and triple superphosphate might not have an acute toxic effect on the soil invertebrates. In contrast to endogeic species found in the current study, *E. fetida* is epigeic, and some research has suggested that the sensitivity of ecologically different earthworm species to chemicals/pollutants may vary [49,50]. The earlier field studies by Jastrz ˛ebska et al. [23,51–53] showed that suspension and granular fertilizers from SSA and/or animal bones with a low content of toxic elements and applied in recommended doses did not alter the abundance (density and biomass), species composition, and structure of soil earthworms. In the cited studies, only endogeic species were found, both in fertilized and nonfertilized soil. The current study is in line with the above results. It is also worth highlighting that the peculiar impact of PSM included in biofertilizer on earthworms was not noticed. The same results were obtained by Jastrz ˛ebska et al. [53] when fertilizer and biofertilizer from SSA and animal bones were compared. It can thus be concluded that PSM introduced into the soil in the amounts required for biofertilizers do not significantly alter the earthworm habitat conditions.

In the presented experiments, chemical plant protection was used. This may create the assumption that pesticides affected earthworms and masked the effects of fertilizers. However, in the earlier study with SSA-based suspension fertilizer, Jastrz ˛ebska et al. [23] did not observe the effect of pesticides (applied at recommended doses) on earthworms, nor the interaction between phosphorus fertilizations and plant protection (no plant protection vs. chemical plant protection). Considering the abovementioned results, we believe that this phenomenon did not occur in the presented study either.

#### **4. Conclusions**

Recycled fertilizers produced from secondary raw materials, such as sewage sludge ash with a low content of toxic elements and dried animal blood, applied in reasonable doses, similarly to superphosphate, did not pose a threat to earthworms. The impact on these organisms is not a limitation to their use. However, taking into account the potential toxicity of waste, relevant studies preceding the recommendation of each new recyclate-based product and long-term field ones are postulated.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/4/525/s1, Figure S1: Scheme of experimental design.

**Author Contributions:** Conceptualization, M.J.; methodology, M.J., M.K.K., and K.T.; formal analysis, M.J.; investigation, K.T., M.J., and M.K.K.; resources, M.J. and K.T.; data curation, M.J.; writing—original draft preparation, M.J.; writing—review and editing, M.K.K.; visualization, M.J.; funding acquisition, M.J., M.K.K., and K.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Centre for Research and Development, Poland, grant number PBS 2/A1/11/2013.

**Acknowledgments:** The Institute of New Chemical Synthesis in Puławy is highly acknowledged for providing fertilizers for field experiments and Agnieszka Saeid from the Wrocław University of Science and Technology for the information on the fertilizers' chemical composition.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
