The Composition and Seasonal Variation of Epigeic Arthropods in Different Types of Agricultural Crops and Their Ecotones

Abstract

Changes in the structures of epigeic arthropods reflect changes in the ecological status of their habitats. The aim of this research was to assess the influence of ecotones and environmental variables (pH soil, soil moisture, potassium, phosphorus, nitrogen) on the abundance of epigeic arthropods. Between 2018 and 2020, an investigation of different types of crops (Brassica napus L., Pisum sativum L., Triticum aestivum L., T. spelta L., Zea mays L., grass mixture) recorded 31,315 individuals belonging to 14 taxonomic groups and their ecotones, using pitfall traps. The abundance of epigeic arthropods was affected by moisture, pH of soil, and levels of phosphorus, potassium and nitrogen. We observed a greater average number of individuals in the ecotones than in the fields in the months of September and October and in Triticum aestivum crops. During other months and for the remaining crops, we did not notice this difference. This contributed new information regarding the ecotone rule. Our results suggest that agricultural intensification affects epigeic arthropods in the crop Triticum aestivum, which is important for the production of biomass and also affects crop yields.

1. Introduction

Agricultural landscapes vary with the cropping system, topography and intensity of management. Agriculture is considered one of the key drivers of biodiversity loss [1,2]. Agricultural interventions affect the level of agroecosystem biodiversity and mainly contribute to a reduction in abundance of edaphic groups [3]. The use of hedges and field margins (ecotones) in the landscape creates a specific habitat for insects, birds, plants, and other animals [4,5]. Ecotones, serving as biodiversity refuges, can help lessen the effects of agricultural practices. They benefit biodiversity by harbouring diverse plant communities that support and act as shelters or overwintering sites for invertebrate communities [6]. This means that flora has natural value, because it can support agricultural production by attracting pollinating insects or other beneficial animals that can regulate pest populations [7,8]. Field margins also act as ecological corridors, assuring connectivity between non-crop areas and fields and preventing separation from important landscape patches [9].

Soil communities play an important role in the decomposition of organic matter in the biogeochemical cycles of the biogenic elements carbon, nitrogen, sulphur and phosphorus. Therefore, they also are important for the sustainability of the soil ecosystem via their contribution to the transformation and degradation of waste and toxic substances. The response of soil organisms is a key part of the sustainability of the soil ecosystem [10,11], being considered a bioindicator of the environment [12,13,14,15]. Epigeic arthropods constitute an important part of the biodiversity in agricultural areas, fulfilling a wide variety of ecological roles that influence the function of the ecosystem. [16]. Predator and parasitoid species are viewed as beneficial for agriculture, and introducing them as biological control agents is common [17]. Phytophagous species have economic implications, as they may act as pests and become the target of insecticides. In general, arthropod communities are fundamental to agricultural ecosystems, with seasonal variation being a common trait, especially in areas with strong climatic seasonality [18].

One criterion for evaluating the sustainability of ecosystems is assessment of the fluctuations in soil quality, an important natural resource; few phenomena on our planet match the complexities of the processes within it. Agricultural land is a type of modified ecosystem with a significant human influence, wherein the course of natural processes has been disturbed. The soil reflects ecosystem metabolism, as within soils the bio-geochemical processes of all different ecosystem components are combined [19]. Soil quality can be evaluated using chemical, physical and biological indicators. To determine soil quality, appraisals of its organic matter [20], bulk density [21] and aggregate stability [22] are required. Biological monitoring is also required to correctly assess soil degradation and correlated risks [23].

The purpose of this study was to analyse the effect of an ecotone on the conditions of Brassica napus L., Pisum sativum L., Triticum aestivum L., T. spelta L., Zea mays L. and grass mixture with respect to the abundance and composition of epigeic arthropods. It was also intended to discover the influence of environmental variables (soil humidity, soil pH, and levels of potassium, phosphorus and nitrogen) on epigeic arthropods.

2. Materials and Methods

Research took place during the years 2018–2020, during which we collected epigeic arthropods from six types of agricultural crops. These types of agricultural crops were examined throughout each year, as the position of crops in the fields changed every year. In winter-planted crops (Brassica napus, Pisum sativum, Triticum aestivum, T. spelta), epigeic arthropods were collected from November to July. In the spring-planted crop (Zea mays), epigeic arthropods were trapped from April to October. In the grass mixture, epigeic arthropods were collected all year round (Figure 1). We used five pitfall traps (750 mL) for each site, which were placed in a line at distances of 10 m. The placement of the pitfall traps varied; the first and second pitfall traps were located in the ecotones (meadow type), and the other traps were inside the fields (Figure 2). Ecotone-placed traps were always at the edge of the fields, next to the white line; the average width of the ecotones was 15 m. Formaldehyde solution (4%) for material fixation during regular collection at two-week intervals was used. Pitfall traps were always in the fields and were collected at two-week intervals. The nomenclature of epigeic arthropods was established according to the work of [24,25]. The study area was located in the Podunajská pahorkatina—Danubian upland geomorphological unit (in southwestern Slovakia) in the cadastral territory of Nitra. The altitude of the monitored area was approximately 130 m above sea level with a brown soil type. The study area is considered a warm arid climate with mild winters. The mean temperature ranges during each month were as follows: January −5–5 °C, February −3–6 °C, March 0–12 °C, April 10–20 °C, May 15–22 °C, June 18–27 °C, July 22–29 °C, August 20–29 °C, September 15–23 °C, October 8–15 °C, November −3–7 °C, December −5–5 °C. The average precipitation for each month was as follows: January 30 mm, February 26 mm, March 35 mm, April 12 mm, May 65 mm, June 77 mm, July 41 mm, August 57 mm, September 64 mm, October 54 mm, November 40 mm and December 55 mm.

The insecticide FORCE (Syngenta, Basel, Switzerland), a granular insecticide intended for soil application to control soil pests, was applied to the crops. Insects were killed through respiratory and tactile poison ingestion. The preparation had a fast effect and a strong residual (repellent) action against a wide range of soil pests from the orders of Coleoptera, Aranea, and Hymenoptera. The applied dose was administered uniformly at a concentration of 12–15 kg per ha each year for the duration of the research. Solinure FX fertilizer (Medilco Hellas S.A., Athens, Greece) containing chlorides and urea, was applied to the crops and was intended for field fertility. Due to its acidifying effect, it contributed to lowering the soil pH.

Conventional tillage was based on annual tillage plowing, incorporating crop residues and weeds into the soil. The soil was plowed three times and turned. Pre-sowing preparation and sowing were combined. Machines with driven working tools were used in conjunction with a seed drill. When sowing, it was possible to use seed coulters with an obtuse angle of penetration into the soil.

At each pitfall trap location we removed stones and fallen leaves from crops and sampled the soil to a depth of 15 cm for analysis. Five samples (one from each of five sites) were taken from each field every two weeks over the three years of the study period. Subsequently, environmental variables (N, P, K, pH, moisture) were analysed using a soil moisture meter (Rapitest 3 1835, Luster Leaf, Illinois, IL, USA) and a pH meter (Dexxer PH-03, Luboň, Poland). We thoroughly wetted the broken-up soil with water (ideally distilled or de-ionised water) to a muddy consistency. We wiped the meter probe clean with a tissue or paper towel and inserted it into the soil up to the probe base (7–10 cm). We waited one minute and wrote down the value. We converted the measured values into units of mg. The average values of environmental variables at traps for all crops during the years 2018–2020 are shown in

Table 1. Average values of environmental variables during 2018–2020 for the studied crops.

Crops/Environmental Variables	Pitfall Traps
	1	2	3	4	5
Brassica napus
potassium	14.09	16.36	16	16.88	14.38
phosphorus	1.13	1.31	1.28	1.35	1.15
nitrogen	14.09	16.36	16	16.88	14.38
pH	7.01	7.10	7.11	6.98	6.98
moisture	2	2.09	1.90	2.13	2.13
Grass mixture
potassium	8	13.50	13.50	14	12.50
phosphorus	0.64	1.08	1.08	1.12	1
nitrogen	8	13.50	13.50	14	12.50
pH	6.99	6.93	6.82	6.90	6.92
moisture	1.30	1.70	1.70	1.60	1.60
Pisum sativum
potassium	11.54	14.17	12.08	13.33	15.42
phosphorus	0.92	1.13	0.97	1.07	1.23
nitrogen	11.54	14.17	12.08	13.33	15.42
pH	7.06	6.81	6.88	6.90	6.85
moisture	1.54	1.58	1.75	1.50	1.58
Triticum aestivum
potassium	16.67	17.50	16.67	19.17	15
phosphorus	1.33	1.40	1.33	1.53	1.20
nitrogen	16.67	17.50	16.67	19.17	15
pH	7.02	6.95	7	7	6.98
moisture	1.50	1.67	1.83	1.83	1.67
Triticum spelta
potassium	14	14	10	14	10
phosphorus	1.12	1.12	0.80	1.12	0.80
nitrogen	14	14	10	14	10
pH	6.96	7	7	7	7.13
moisture	1.40	1.60	1.40	1.80	1.50
Zea mays
potassium	10	15.45	12.73	12.73	13.18
phosphorus	0.80	1.24	1.02	1.02	1.05
nitrogen	10	15.45	12.73	12.73	13.18
pH	7	7	6.86	6.89	7.02
moisture	1.45	1.91	1.55	1.55	1.73

.

2.1. Database Quality

The data obtained were saved in the Microsoft SQL Server 2017 database program (Express Edition) (Microsoft Corporation, Washington, DC, USA), consisting of frequency tables for collections and measured environmental variables (pH, soil moisture, potassium, phosphorus, nitrogen). The database also contained code tables for study sites and their variables (crops, habitat, locality name, cadastral area, altitude, coordinates of localities). Matrices for statistical calculations using [26] were programmed.

2.2. Statistical Analyses

Multivariate analysis (redundancy analysis, RDA) was employed to determine the dependencies between objects (abundance of epigeic arthropods, soil characteristics, crops). The number of individuals of each crop in pitfall traps 1–5 was used as a matrix. We tested the statistical significance of soil pH, soil moisture, and levels of potassium, phosphorus and nitrogen with the Monte Carlo permutation test (iterations = 499) in the Canoco5 program [27].

To determine the normality of the data distribution, we used the Shapiro–Wilk W-test for analysis in the statistical program [28], As the data distribution did not achieve normality (p-value = 0.00), we used the nonparametric Kruskal–Wallis test (ANOVA) to test the differences in abundance between pitfall traps.

3. Results

Over the three-year research period, we found a total of 31,315 individuals belonging to 14 taxonomic groups in the studied area. Taxa of Coleoptera (45.65%), Collembola (17.41%) and Julida (15.59%) had an eudominant representation of individuals. Araneae (6.65%) demonstrated the dominant group of individuals in crops. Other groups of epigeic arthropods had subdominant to subrecendent representation (

Table 2. Distribution of the arthropods in the agricultural crops during the years 2018–2020.

Arthropods/Crops	Pitfall Traps					∑ Ind.
	1	2	3	4	5
Grass mixture/∑ individuals	3494	3940	3526	3622	3800	18,382
Araneae	355	356	253	382	353	1699
Opilionidea	3	9	7	24	14	57
Collembola	1011	1161	838	835	1603	5448
Isopoda	60	39	56	32	22	209
Julida	773	1062	1103	1013	490	4441
Lithobiomorpha	15	27	15	17	13	87
Coleoptera	751	833	806	908	752	4050
Dermaptera	0	0	0	2	0	2
Diptera	144	160	172	110	156	742
Hemiptera	36	34	37	44	61	212
Hymenoptera	124	67	57	97	51	396
Orthoptera	98	112	70	76	58	414
Haplotaxida	25	29	18	10	9	91
Pisum sativum/∑ individuals	281	287	229	211	148	1156
Araneae	23	42	20	13	20	118
Isopoda	9	9	9	16	0	43
Julida	31	54	17	17	12	131
Coleoptera	175	145	156	150	96	722
Dermaptera	12	15	13	1	14	55
Diptera	2	0	0	0	0	2
Hymenoptera	10	22	1	7	6	46
Orthoptera	19	0	13	7	0	39
Zea mays/∑ individuals	1426	2556	1398	2565	1017	8962
Araneae	15	24	38	10	19	106
Opilionidea	33	24	24	21	20	122
Collembola	0	0	3	0	0	3
Isopoda	123	144	131	121	120	639
Julida	21	25	14	0	0	60
Lithobiomorpha	21	29	17	16	20	103
Coleoptera	1126	2178	1109	2329	777	7519
Dermaptera	0	1	0	0	0	1
Diptera	54	65	26	14	21	180
Hymenoptera	16	6	16	23	14	75
Orthoptera	11	53	13	31	21	129
Haplotaxida	6	7	7	0	5	25
Triticum aestivum/∑ individuals	315	256	192	226	145	1134
Araneae	2	1	2	11	0	16
Isopoda	0	0	0	2	0	2
Julida	34	52	34	59	59	238
Coleoptera	225	156	101	124	56	662
Dermaptera	10	4	4	12	9	39
Diptera	20	19	14	2	9	64
Hymenoptera	17	0	4	4	0	25
Orthoptera	4	23	25	12	12	76
Haplotaxida	3	1	8	0	0	12
Brassica napus/∑ individuals	210	379	151	125	129	994
Araneae	16	22	26	14	17	95
Isopoda	1	2	1	1	0	5
Julida	0	4	2	0	0	6
Lithobiomorpha	2	0	0	2	0	4
Coleoptera	171	313	106	80	100	770
Diptera	1	12	1	1	4	19
Hemiptera	1	2	0	0	0	3
Hymenoptera	6	2	4	23	1	36
Orthoptera	12	21	11	4	7	55
Haplotaxida	0	1	0	0	0	1
Triticum spelta/∑ individuals	113	162	286	117	9	687
Araneae	7	10	14	14	4	49
Isopoda	10	16	12	6	0	44
Julida	2	0	1	0	2	5
Coleoptera	85	132	256	97	3	573
Hymenoptera	4	3	0	0	0	7
Orthoptera	5	1	3	0	0	9
∑ individuals from all crops	5839	7580	5782	6866	5248	31,315

).

Multivariate analysis of arthropod abundance over the study period was determined using redundancy analysis (RDA, SD (length of gradient) = 1.10 on the first ordination axis). We observed relationships between the abundance of epigeic arthropods and environmental variables (pH of the soil, soil moisture, potassium, phosphorus, nitrogen). The values of the explained variability in taxonomic data were 54.6% on the first ordination axis and 66.5% on the second ordination axis. The cumulative variability of the set of species explained by environment variables was on the first ordination axis 73.8% and on the second axis 89.9%. Using the Monte Carlo permutation test, we identified statistically significant effects of soil moisture (p = 0.0267, F(1.3376) = 2.0521, df = 3), pH of soil (p = 0.0188, F(1.9583) = 2.0297, df = 3), phosphorus (p = 0.0458, F(1.1012) = 3.0885, df = 3), potassium (p = 0.0278, F(1.2125) = 1.9810, df = 3) and nitrogen (p = 0.0485, F(1.5986) = 2.2546, df = 3) on the structure of arthropod distribution. The selected variables were not mutually correlated with the maximum value of the inflation factor = 4.9440. The ordination graph (triplot) had arthropod predominance ordered around pitfall traps one and two, located in the ecotone. The number of taxa of arthropods decreased in the pitfall traps located inside the fields (3–5). Environmental variable pH correlated with the prevalence of Diptera, Hymenoptera, Haplotaxida, Araneae, Lithobiomorpha, Isopoda, Orthoptera at pitfall trap 1. Environmental variables moisture (%), potassium (mg), phosphorus (mg), nitrogen (mg) correlated with the prevalence of Coleoptera, Julida, and Opilionida at pitfall traps 2, 3 and 4 (Figure 3).

In the second stage of multivariate analysis using redundancy analysis (RDA, SD (length of gradient) = 1.20 on the first ordination axis), we observed relationships between epigeic arthropods groups and crops (Brassica napus, Pisum sativum, Triticum aestivum, T. spelta, Zea mays, grass mixture). The values of the explained variability in taxonomic data were 55.4% on the first ordination axis and 84.3% on the second ordination axis. The cumulative variability of the set of species explained by environmental variables (crops) was on the first ordination axis 60.1% and on the second axis 92.7%. The selected variables were not mutually correlated with the maximum value of the inflation factor = 1.3164. The ordination graph (triplot) had arthropod predominance ordered around pitfall traps one and two, located in the ecotone, and pitfall trap three, located in a field on the border with the ecotone. This trend was reflected in all studied crops. Pitfall traps four and five were located deeper into the field, which was associated with a reduced number of taxa caught in these pitfall traps (Figure 4).

We confirmed statistically significant differences in the number of individuals among pitfall traps 1–5 in Triticum aestivum (p = 0.0462, F(1.5716) = 4.99, df = 4), Triticum spelta (p = 0.049, F(3.0531) = 4.41, df = 4), and Zea mays (p = 0.0075, F(1.0148) = 4.329, df = 4). No significant difference was found in the number of individuals among pitfall traps 1–5 in grass mixture (p = 0.9331, F(4.393) = 0.8391, df = 4), Pisum sativum (p = 0.8227, F(4.103) = 1.5223, df = 4), or Brassica napus (p = 0.9113, F(4.84) = 0.9898, df = 4) (Figure 5). The results showed significantly more individuals in pitfall traps 1, 2 (located in the ecotones) than in pitfall traps 3, 4, 5 (located in the fields) in the crop Triticum aestivum. In the grass mixture, Pisum sativum, Triticum spelta and Zea mays we did not find a decline in the number of individuals in traps located inside the fields.

We confirmed a statistically significant difference in the number of individuals among pitfall traps 1–5 in May (p = 0.05, F(1.3636) = 4.73, df = 4), July (p = 0.0449, F(3.7223) = 4.134, df = 4), August (p = 0.0082, F(1.449) = 4.120, df = 4), September (p = 0.05, F(2.992) = 4.148, df = 4), and October (p = 0.0372, F(4.254) = 4.82, df = 4). A statistically insignificant difference was observed in the number of individuals among pitfall traps 1–5 in March (p = 0.6082, F (4.72) = 2.7056, df = 4), April (p = 0.7222, F(4.128) = 2.0735, df = 4), June (p = 0.5951, F(4.212) = 2.781, df = 4), and November (p = 0.6743, F(4.80) = 2.3353, df = 4) (Figure 6). The results showed a greater average number of individuals in pitfall traps 1, 2 (located in the ecotones) than in pitfall traps 3, 4, 5 (located in the fields) during September and October. In March, April, May, July and August there was no such difference observed.

4. Discussion

In intensively managed agricultural landscapes, ecotones survive as remnants, of which field margins are usually the most common structures. Boundaries (ecotones) have an important agricultural function, and their role in environmental and conservation functions is acquiring greater significance. Agricultural operations render them susceptible to disturbance, particularly from pesticide drift and eutrophication from fertiliser misplacement [29]. Epigeic arthropods living in agricultural landscapes have a greater tolerance than the epigeic arthropods of natural habitats [30]. They also achieve high local density due to the influence of agriculture, and field margins support the most abundant and diverse communities of soil arthropods [31]. We recorded an arthropod community dominated by Coleoptera, Collembola, Julida and Araneae. The great abundance of these groups influenced the maintenance of the natural balance and the substance cycle of the biogenic elements in ecosystems such as carbon, nitrogen, sulphur and phosphorus. The dominance of Hymenoptera (Formicidae) and Coleoptera has been described as a general trait of ground-dwelling assemblages [32]. Their activities accelerate the decomposition of plant residues, aerate the soil and improve soil structure and quality [33]. The presence of other epigeic arthropods is heterogeneous and may depend on the management regime and the surrounding vegetation [34].

The scale, structure, use and management of land can affect the composition of fields. These features may be sources of weeds, pests and disease; ecotones may be of great importance in providing habitats for species beneficial for agriculture [35]. Using a multivariate model, we demonstrated the positive influence of environmental variables within ecotones on the abundance of soil arthropods. Thus, our results agree with the results of [36], who observed a decline in arthropods with increasing land use. In conditions of complex land use, ecotones are numerous, whilst fewer ecotones may be available in less complex systems. As a result, coexistence through resource allocation in simplified systems is likely to be limited, resulting in a reduction in species richness. More complex habitat and structure may allow greater access to food resources [37]. The explanation for greater species richness in less disturbed habitats is that in frequently or intensely disturbed environments, community composition cannot progress beyond early pioneer stages. The results are environments that favour early successional species and disadvantage later successional species [38]. Whilst more data on the impact of introduced margin strips on different taxa are required, the beneficial effects of botanical diversity, particularly wild flowers, have an effect on invertebrate diversity [39,40,41].

Traps located near the field margins captured more arthropods than those located at medium and large distances from the ecotone. Ecotones bordering crop fields provide stable shelter, food sources, and microclimates to a wide range of arthropods [8]. In our study traps, the number of individual arthropods captured in Triticum aestivum crops was distinct from the number of individual arthropods captured in grass mixture, Pisum sativum, Triticum spelta, and Zea mays regarding differences between pitfall traps 1, 2 (located in the ecotones) and 3, 4, 5 (located in the fields), which may have been affected by the width of the ecotone. Common crop management practices may serve to increase the frequency and severity of disturbance regimes [42]; for example, tillage disrupts a number of beneficial invertebrate groups [43,44,45].

The prominence of arthropod abundance from month to month is related to fluctuations in climatic factors (such as temperature, precipitation and day length) [46]. These factors, in combination with distinct life histories, are thought to greatly influence arthropod assemblages [47]. The number of individual arthropods captured in September and October was distinct from the number of individuals of arthropods captured in March, April, May, July and August regarding differences between pitfall traps 1, 2 (located in the ecotones) and 3, 4, 5 (located in the fields), which may have been affected by the weather and climate. The trend of decreasing numbers of individual arthropods towards the centre of fields was not observed in March, April, May, July and August, when the ecotone rule did not manifest. Differences in the number of arthropods were affected by the different weather conditions during the seasons and months [48]. The monthly trends regarding abundance of arthropod groups, with individual taxa exhibiting distinct seasonal patterns, reflected the high variability among the life-cycles of the captured groups [49].

5. Conclusions

Our results provided new knowledge regarding the preference of epigeic arthropods in conditions of Brassica napus, Pisum sativum, Triticum aestivum, T. spelta, Zea mays, and grass mixture, and of the influence of ecotones on their occurrence. We confirmed a decrease in the average number of individuals in pitfall traps located further from the ecotone during the months September and October and in the crop Triticum aestivum. Interestingly, this trend did not apply to the other months or for the remaining crops, which did not confirm the ecotone rule (that the highest diversity is found in semi-natural areas). Based on the above facts, our results have provided new information regarding this ecotone rule. The dispersion of epigeic arthropods was also positively influenced by soil moisture, pH of soil, and levels of phosphorus, potassium and nitrogen. Epigeic arthropods are important drivers of ecosystem functions such as nutrient cycling, pest control and maintenance of soil structure. Ecotones are important for increasing the number of epigeic arthropods, thereby increasing biomass, which affects crop yields. Therefore, it is important to promote strategies for addressing the conservation of arthropods in agricultural landscapes.