Epigeal invertebrate fauna in monoculture and integrated systems in the cerrado

Epigeal fauna are considered key biological indicators of soil quality, which can be used as parameters for determining sustainable agricultural production systems. This study aimed to evaluate the epigeal fauna under different monoculture and integrated production systems in a Yellow Latosol in the Cerrado biome. The epigeal fauna was sampled using pitfall traps in areas of no-tillage, pasture, exclusive eucalyptus cultivation, integrated livestock-forest system and in a native cerrado. The following variables were determined: number of individuals trap day, total richness and the indices of Shannon and Pielou. The number of individuals trap day and richness were higher in native cerrado, while the no-tillage area presented the lowest values. For the Shannon and Pielou indices, native cerrado and integrated livestock-forest system presented the highest values while exclusive eucalyptus cultivation and no-tillage presented the lowest values. The pasture, exclusive eucalyptus cultivation and integrated livestock-forest system favored the ecological indexes of the epigeal fauna, while no-tillage was the system that least contributed to the establishment of these organisms. In areas explored with pasture, exclusive eucalyptus cultivation and integrated livestock-forest system, the community and the diversity of groups of fauna tend to increase. The association with different groups of fauna is more expressive in environments of native cerrado and exclusive eucalyptus cultivation.


Introduction
In the Cerrado biome located in Piauí, Northeast Brazil, there has been intensive soil use with continuous applications of correctives (e.g. lime, gypsum), fertilizers and agrochemicals. Such practices cause severe changes in soil properties (Braz et al., 2013), compromising the sustainability of the ecosystem and deteriorating the biological component of the soil, such as the composition of the epigeal invertebrate fauna (Lima et al., 2019;Bartz et al., 2014).
The different groups of the epigeal soil fauna are of fundamental importance for the maintenance of soil quality because they participate in important ecological processes, performing various environmental functions, such as the litter fragmentation, organic matter decomposition, biorecycling and redistribution of nutrients between the different soil layers and gallery opening (Choudhary et al., 2018;Bartz et al., 2014).
In addition to its ecological function, the invertebrates are considered important bioindicators of soil quality due to their high sensitivity to impacts caused by different management and land use systems. Thus, the epigeal fauna can be used as parameters for determining sustainable agricultural production systems (Santos et al., 2016). The indications of soil management systems that promote greater biological conservation can be done by assessing the disturbance levels of invertebrate fauna, observing a decline the abundance and diversity of species, in addition to the change in the fauna composition from o non disturbed, as well as the presence of certain specific groups (Souza et al., 2016;Silva et al., 2012).
The importance of these invertebrates has been emphasized for maintaining soil quality and the environment in the Cerrado biome (Lima et al., 2019;Santos et al., 2016;Portilho et al., 2011;Silva et al., 2011). However, there is still a lack of information on their behavior in anthropized areas and especially with eucalyptus cultivation (Souza et al., 2016;Cortez et al., 2015;Vargas et al., 2013). Moreover, more studies are needed regarding areas involving production systems with the use of eucalyptus, both in monoculture and in integrated systems, aiming at understanding the effects of different management systems and land uses on epigeal fauna.
This study aimed to evaluate the epigeal fauna community under different monoculture and integrated production systems in a Yellow Latosol in the Cerrado biome located in Piauí, Brazil.

Methodology
The study was conducted at Fazenda Chapada Grande, located in Regeneração, Piauí state, Brazil (06°21'03" S, 42°28'79" W, 374 m of altitude). The climate of the region is classified as Aw, according to the Köppen classification, composed of hot and humid tropical weather, with rainy summers and dry winters. The annual average temperature is 26.4 °C and the annual average rainfall is 1,371 mm. The chemical attributes of the soil, which is classified as Dystrophic Yellow Latosol (Jacomine et al., 1986), are shown in Table 1.  Four different soil management systems were selected for the study: no-tillage (NT), pasture (PAS), exclusive eucalyptus cultivation (EEC), integrated livestock-forest system (LFI) and a native Cerrado (NC) area, which was used as reference.
In the NT area, deforestation occurred in 2010, and the initial tillage consisted plowing and root incorporation of 4 t ha -1 of dolomitic limestone and fertilization with 250 kg ha -1 of NPK. The area was cultivated with rice in 2011 and with soybean in 2012 under a conventional system, which weas later converted into no-tillage and crop rotation (soybean and corn) systems.
During sampling, the area in this system was cultivated with soybeans.
The Pasture (PAS) was implemented in 2008 by using plowing and liming applications. The pasture was formed with Urochloa brizantha grass, which was subjected to rotational grazing by Nellore cattle with a stocking rate of 1 UA ha -1 year -1 for 15 days in each paddock.
For the exclusive eucalyptus cultivation (EEC) system, planting was carried out in 2007 using MA-2000 clones, with spacing of 2 m between plants in rows and 3 m between rows after soil preparation. In this system, 2 t ha -1 of dolomitic limestone, were applied. In this system, rotational grazing with Nellore cattle with a stocking rate of 1 UA ha -1 year -1 for 15 days in each paddock was adopted. The Cerrado area is composed of herbaceous, shrub, arboreal vegetation and vines typical of this biome, especially trees with tortuous stems.
Sampling of epigeal fauna was carried out in February 2017, using pitfall traps, according to the methodology described by Moldenke (1994), adapted by Aquino et al. (2006). The traps constituted of plastic containers 10 cm high and diameter of 10 cm, which were buried in the ground with the edge at the level of the soil surface. Seven traps were installed per soil management system. In each trap, 200 mL of a preservative solution based on 4% formaldehyde were added to prevent deterioration of the invertebrates. After seven days, the traps were removed from the soil and the collected individuals were identified at the level of order or class (Dindal, 1990;Gallo, 1988). where: H is the Shannon-Weaner index; S is the richness.
The organisms were divided into functional groups based on characteristics of habitat use and the main form of food resources use, being classified into: Decomposers, Engineers and Predators, according to the classification of (Merlim et al., 2005) and (Brown et al., 2009).
The number of individuals trap -1 day -1 and the richness of the groups were subjected to the Shapiro-Wilk and Barllet's tests to verify the normality and homogeneity of the data. As they did not meet the assumptions of the analysis of variance, the data were subjected to non-parametric analysis, using the Friedman test at 5% significance level. Principal component multivariate analysis (PCA) was carried out between the taxonomic groups and the soil management systems, using the statistical software R (R Development Core Team, 2016).

Diversity e population of fauna
The number of individuals trap -1 day -1 , which represents the abundance of epigeal fauna organisms, differed significantly (p≤0.05) between the systems, with the highest value found in the NC area, while the lowest value was found in NT (Table 2). The highest values of individuals per trap observed in NC can be attributed to the presence of litter from a diversity of plant species, which can be used as a source of food for edaphic organisms . According to Hedde et al. (2015), the more diverse the vegetation cover, the greater the heterogeneity of the litter, which may favor the establishment of a larger population fauna organisms. In addition, in areas of native vegetation there is a greater number of ecological niches and less competition between species. In the PAS, what may have contributed to the high values, may be the presence of the brachiaria that produces a large amount of organic waste to the soil due to the high biomass production of the aerial part and its extensive root system and in constant renovation (Nunes et al., 2012). These plant residues serve as a food source often creating a microclimate favorable to the soil invertebrates, resembling a natural forest environment. The lower abundance of epigeal fauna in the NT, system might be due to the applications of herbicides and insecticides, as reported by Mentone et al. (2011). In addition, the short time of adoption of this system and the low amount of plant residues left in the soil is still not configured as an environment favorable to the emergence of habitats for the colonization of the fauna.
The richness differed between treatments, with the highest values being observed in NC, followed by the PAS. The lowest value was found in NT. These results were similar to those found by Silva et al. (2011), Regarding the relative distribution of epigeal fauna, the groups Coleoptera and Formicidae were present in greater proportions in all the management systems studied. In the EEC, in addition to the predominance of the Coleoptera and Formicidae groups, there was also a great predominance of the Collembola group (Table 3).   (Farias et al., 2015) occurring in different ecological niches and that present quite diversified eating habits. Due to this characteristic, it is one of the most abundant orders in Brazilian soils. The results observed in the present study are in agreement with other studies for Cerrado conditions, which showed great abundance of this group, regardless of the management methods and the time of assessment (Portilho et al., 2011;Vargas et al., 2013). Although some families of this order are considered pests in agriculture both at larval and adult stages, they can be beneficial for soil fertility and physics (Portilho et al., 2011).

Groups
The high frequency of the Formicidae group might be associated with the fact that these organisms are very abundant and dominant in terrestrial ecosystems, indicating that this group has the capacity to adapt to different changes in the environment (Mentone et al., 2011). Other authors have also reported a high frequency of the Formicidae group in studies of epigeal fauna in different soil management systems (Pasqualin et al., 2012;Vargas et al., 2013). This group is of relevant importance for maintaining soil quality because it performs several processes. In addition, the complexity of galleries and chambers of the anthills increases the porosity and the drainage capacity of the soil, decreasing soil density (Mentone et al., 2011).
The predominance of the Collembola group in EEC is due to the presence of a dense litter layer composed of decomposing organic waste. These organisms feed mainly on decomposing organic materials and microorganisms, such as fungi and bacteria (Verma et al., 2014). Martins et al. (2017) also verified high abundance of Collembola in monoculture of eucalyptus.

Classification of epigeal fauna organisms
Regarding the functional groups of the soil epigeal fauna, the decomposers were the ones with the highest density. The main representatives were the Acari, Collembola and Diplopoda groups (Table 4).  Merlim et al. (2005) and Brown et al. (2009). Averages were calculated using the Friedman test and compared on the same row. ns: not significant * significant (p≤0.05). NC: native cerrado; NT: no-tillage; PAS: pasture; EEC: exclusive eucalyptus cultivation; LFI: integrated livestock-forest integration. Source: Authors.
The highest occurrences were recorded in EEC for the Acari group, differing (p ≤ 0.05) from the other studied management systems. Similar behavior was also observed for Collembola in this same management system, which differed significantly (p ≤ 0.05) from the others. The Acari edaphic group feeds on decomposing organic materials, which might explain its greater predominance in EEC. Mite species are largely considered parasites of plants and animals, however, some species show great relevance in the decomposition of organic matter, and presence of this group in large quantities can be considered as an indicator of soil biological condition .
The ecosystem engineers were represented by the groups Formicidae, Isoptera and Oligochaeta. The first one differed (p≤0.05) between the systems, presenting greater abundance in NC and PAS. The order Isoptera had the highest occurrence in NC. The Oligochaeta group was not captured in all systems, and did not differ (p≤0.05) between the systems. Engineer organisms play an essential role in maintaining soil quality due to their effects on their aggregation, with the creation of biogenic structures, which can be galleries, nests, chambers and fecal acorns. It can significantly affect the physical properties of the soil and the availability of resources for other organisms (Anderson, 2009).
As for Predators, the presence of the groups Araneae, Dermaptera and Diptera, Diplopoda and Scorpionida was also observed, with Araneae showing greater predominance among the systems with greater density in NC. The greater amount of litter in this system promoted better conditions for the greater abundance of natural enemies (predators). The Araneae group plays a fundamental role in regulating populations of edaphic fauna, being considered an important tool in pest control. However, its density is strongly influenced by agricultural practices. According to Baretta et al. (2014), the areas with less anthropic action, such as natural vegetation, favor the appearance of spiders. In these areas, there is greater availability of food and conditions for web formation. On the other hand, the use of insecticides and chemicals in agricultural areas causes a reduction in the availability of prey (food) and the residual and repellent effect of some chemical substances that can provide a significant decrease in the spider community in agro-ecosystems.

Principal component analysis
Two principal components were decomposed from PCA, which explained 72.09% of the total variance of the data, which allowed for a two-dimensional ordering of the management systems and taxonomic groups of the epigeal fauna. The first component (PC1) presented 41.96% and component 2 (PC2) accounted for 30.13% of the data variability ( Figure 1). to the great occurrence of the Acari group, which was much higher than that observed in the other systems and still because it has a low occurrence of the other groups in this system. The eucalyptus forest, despite being an environment with only one species of arboreal size, presents several shrub and herbaceous species in its understory, leaving the soil practically covered and providing a diversity of plant residues. In addition, the high density of some groups in EEC, such as Collembola and Acari, could have played a major role regarding this result.
Although the methodology used in the present study is not specific for the collection of organisms that live inside soil, the Oligochaeta group was collected at PAS. The capture of this group in this management system occurred as a result of the presence of feces from the animals that grazed in the area months before the installation of the traps. During the decomposition stage, these wastes favored even more the incidence of these organisms in this management system. According to Martínez and Sánchez (2002), animal waste increases the quality of organic matter, promoting greater carbon mineralization, favoring the development and establishment of these invertebrate fauna groups. The low relationship of NT with fauna groups might be a consequence of management practices carried out in this system in previous years. In addition, the low amount of residues for soil cover and the short adoption time for this management system were not sufficient to improve soil epigeal invertebrate fauna indicators.

Conclusion
In areas explored with pasture systems, exclusive eucalyptus cultivation and integrated livestock-forest, the community and the diversity of groups of epigeal fauna tend to increase.
The association with different groups of epigeal fauna is more expressive in environments of native cerrado and exclusive eucalyptus cultivation. The exclusive eucalyptus cultivation favors the increase of the community and the diversity of groups of epigeal fauna, as well as the association of this environment with different groups of this fauna.
For future research, it is recommended to use the pitfall and monolith methodologies together, in order to obtain greater robustness to the results. In addition, perform a multivariate analysis of main components involving soil fauna data and the physical, chemical and biological attributes of the soil.