Does fertilization with dehydrated sewage sludge affect Terminalia argentea (Combretaceae) and associated arthropods community in a degraded area?

Nutrients from dehydrated sewage sludge play an essential role in the development of many plants such as Terminalia argentea, in the recovery of degraded areas. The aims were to assess the abundance, diversity and species richness of phytophagous, pollinators and predators arthropods, as well as the percentage of defoliation of T. argentea trees, fertilized (or not) with dehydrated sewage sludge in a degraded area. The abundance, diversity and species richness of phytophagous Coleoptera and total predators (predator insects + protocooperating ants + spiders); abundance and species richness of Diptera, pollinator insects, spiders, and predators (predator insects + spiders) were higher on trees fertilized with dehydrated sewage sludge. The abundance of phytophagous Coleoptera declined with the presence of phytophagous Hemiptera and protocooperating ants; population of phytophagous Orthoptera declined in response to phytophagous Coleoptera and total predators; the numbers of the leafminer Lyriomyza sp. directly increased with the numbers of spiders. The ecological indices of phytophagous, pollinators, and predator arthopods increased on Terminalia argentea trees fertilized with dehydrated sewage sludge; such a better ecological indices in fertilized than in unfertilized trees, show it more suitable for the recovery of degraded areas. We discuss the competition between phytophagous insects groups as well as herbivory reduction by predators.


Discussion
The highest ecological indices (abundance, diversity and species richness) of phytophagous, pollinator and predators arthropods on T. argentea, fertilized with dehydrated sewage sludge (DSS), are related to a higher nitrogen levels 6 and consequently a better development of these plants (e.g. > leaves/tree = > ETIB) 5,15 . The apparent competition between Coleoptera and Hemiptera for space and food, and the negative effect between protocooperating ants and phytophagous Coleoptera, are in accordance to findings on Caryocar brasiliense Camb. (Malpighiales: Caryocaraceae) trees 19,21 .
The highest ecological indices of phytophagous Cerambycidae, Cerotoma sp., Cratosomus sp., Euxesta sp., Lamprosoma sp., Lepidoptera caterpillars, Parasyphraea sp., Psiloptera sp., T. collaris and Hemiptera, ; pollinator T. spinipes; predators Araneidae, Salticidae, Pentatomidae and Polybia sp.; protocooperating ants and ecological processes (herbivory) on T. argentea trees fertilized with DSS, may be due to the highest numbers of leaves of this plant (> ETIB). Leaves are food resource with a better quality for these phytophagous insects, which in turn may attract a higher number of predators. Such an observation confirms the first hypothesis (i.e. ETIB), that the diversity and abundance of phytophagous insects, pollinators and their predators are usually higher on larger trees with higher leaf mass 5,15-17 . Thus, trees such as T. argentea, may seem as islands (as proposed by ETIB), and those with lower leaf mass present a higher chance to get extinct the endangered species 5,17,22,23 . In addition, the number of free amino acids and proteins in leaves, pollen and/or nectar production and quality (more protein and amino acids) in flowers, are superior in plants with higher nitrogen fertilization levels, e.g. T. argentea trees fertilized with DSS, increasing the attractiveness to phytophagous and pollinator insects 6,24-26 . Dehydrated sewage sludge used as a biofertilizer improved the macrofauna recovery, including scarab beetles' larvae and adults in degraded soils of the Cerrado (Brazilian Savanna) biome 27 .
The abundance of phytophagous Hemiptera and pollinators (e.g. T. spinipes) reduced the number of phytophagous Coleoptera; whilst this insect order reduced the numbers of Orthoptera and T. spinipes, as well as those of phytophagous Hemiptera and phytophagous Orthoptera reduced Clytini. These correlations confirm the second hypothesis that there was competition between those insect groups for space and feeding. Moreover, protocooperating ants, associated with phytophagous Hemiptera, for instance, may have attacked beetles. However, further , sucking and galling insect species for space and feeding was observed on C. brasiliense trees 17,19 .
Trigona spinipes, by flying in flocks with aggressive behavior, chases other pollinators, such as Apis mellifera L. and Tetragonisca angustula Latreille (Hymenoptera: Apidae) 28 , and also likely other insects (e.g. beetles); beyond damages shoot and plant growth tissues to remove fibers for nests construction 5,29,30 . Food web studies are intricate due to interactions among host plants, phytophagous, predators and parasitoids insects, soil and climatic conditions 31 . Only a few studies have examined food webs in complex ecosystems, such as in the Cerrado 18,31,32 .
Spiders, the dominant predators group (excluding the protocooperating ants), correlated negatively with some phytophagous insects (e.g. Lyriomyza sp. and Orthoptera), confirming the third hypothesis on the negative correlation between phytophagous insects and predators. On the other hand, T. spinipes is perhaps the major prey to spiders on T. argentea trees. Spiders are important in the biological control of phytophagous (r = − 0.73; P = 0.00) and leafminer insects (r = − 0.62; P = 0.01) on C. brasiliense trees 19,21 . Spiders are important in pest control in agroforestry systems, especially in tropical regions 21,33-35 since a wide range of pest insects can get caught in their webs, resulting in deaths 36 . The importance of these arthropods for biological control was confirmed by population reduction of Epiphyas postvittana (Walker) (Lepidoptera: Tortricidae) on Malus domestica Bork (Rosaceae) and Phyllocnistis citrella Stainton (Lepidoptera: Gracillariidae) on Citrus sinensis (L.) Osbeck (Rutaceae) 37,38 . In addition to spiders, the protocooperating ants were very abundant on T. argentea trees fertilized with DSS, probably due to the highest numbers of phytophagous insects-protocooperation [39][40][41] . The increased abundance of protocooperating ants reduced the numbers of phytophagous Coleoptera and T. spinipes on T. argentea trees, as observed in C. brasiliense, where the highest number of these ants reduced defoliation by beetles 19,21 . In addition, ants are bioindicators in the recovery of degraded area because they respond quickly to environmental complexity and by interacting mutually with other insects [42][43][44][45] . The abundance of the predatory wasp Polybia sp. was higher on fertilized plants probably due to a higher numbers of caterpillars (Lepidoptera) and the leafminer Lyriomyza sp.. Predatory wasps (Vespidae) are important natural enemies in agricultural systems such as Brassica campestris L. and kale B. oleracea L. var. acephala DC (Brassicales: Brassicaceae); Arabian coffee Coffea arabica L. (Gentianales: Rubiaceae) and tomato Solanum lycopersicon L. (Solanales: Solanaceae), preying mainly on caterpillars and leafminers (Lepidoptera) [46][47][48][49] .
Scientific RepoRtS | (2020) 10:11811 | https://doi.org/10.1038/s41598-020-68747-z www.nature.com/scientificreports/ generalist in their feeding habits, and the greatest complexity of canopy architecture increases niches options for phytophagous insects and consequently for the natural enemy diversity 50 . For example, sewage sludge increases the richness of the ground beetle Carabidae (Coleoptera) in the area of Oxford, USA 51 . The largest T. argentea tree canopy size (> ETIB) fertilized with DSS may explain the largest abundance of phytophagous insects (> defoliation), pollinators and predators, showing that this plant is adequate to recovery degraded areas. There was competition between groups of phytophagous insects and predator arthropods in high populations and consequent herbivory reduction.

Material and methods
Study. The study was conducted in a degraded area at the "Instituto de Ciências Agrárias (ICA)" of the "Universidade Federal de Minas Gerais (UFMG)", Montes Claros, Minas Gerais, Brazil (S 16º51′38″ W 44º55′00″ 943 m) from March 2015 to February 2017 (24 months; arthropod collection period). The area presents soil loss and changes in soil chemistry and hydrology due to degradation 52,53 . Köppen's climate classification 54 defines this area as tropical dry climate; annual rainfall, 1,000-1,300 mm, with dry winter; annual mean temperature, ≥ 18 °C. The type of soil is litolic neosoil 55 and chemical and physical details were described 5 .
Study design. Seeds were collected from five-years old Terminalia argentea trees at ICA/UFMG campus before sowing. Terminalia argentea seedlings were produced in March 2014 by sowing one seed per plastic polybag (8 × 12 cm), and these kept in a nursery covered with black shed net. The mixed substrate contained 30% organic materials (i.e. two parts of debris gardening pruning < 5 cm in length, and one of brown bovine manure), 30% clay soil, 30% sand, and 10% of mineral fertilizer (i.e. 160 g reactive natural phosphate per seedling) 5 . The  www.nature.com/scientificreports/ soil pH in the pits (40 × 40 × 40 cm) was corrected with dolomitic limestone with anhydrous carbonate mineral composed of calcium magnesium carbonate (90% relative total neutralization power) (187 g per pit), increasing base saturation to 50% 56 . Natural phosphate (80 g per pit), fritted trace elements (FTE) (10 g/pit), and marble roch dust (1 kg per pit) were added when needed. Thirty-centimeters tall T. argentea seedlings were planted in pits in a two-meters spacing , in six parallel lines on flat terrain with two-meters spacing lines, with four trees per treatment (fertilized or not with dehydrated sewage sludge-DSS) per line. The seedlings in experimental area were supplied with water until the beginning of the rainy season. The seedlings with five-cm long branches were pruned with a sterilized razor, eliminating the additional shoots (i.e. others different from the leader shoot) and branches up to 1/3 of crown height. The experimental design was in random blocks with two levels of fertilization (i.e. a single dose of 20 L of DSS per pit or none fertilization) and 24 replications with one plant each 5 . DSS (with 5% mean moisture content) was obtained from a sewage treatment plant (STP) in Juramento, Minas Gerais, Brazil. The STP is operated by the Minas Gerais Sanitation Company -"Companhia de Saneamento de Minas Gerais S.A. (COPASA)". The STP is highly efficient, removing more than 90% of the organic material from the domestic waste water. The sewage sludge is dumped off into coarse sand tanks, staying there for three months to reduce the amount of thermotolerant coliforms (and other pathogenic microrganisms) and reach the ideal levels for agricultural use that is < 10 3 of the most likely number per g of total solids (as recommended by the National Council for the Environment-"Conselho Nacional do Meio Ambiente-CONAMA"). The chemical and biological characteristics of the DSS were described 5,6 . Arthropods. Insects and spiders were visually counted, every two weeks, on the adaxial and abaxial surfaces of the leaves between 7:00 and 11:00 AM at the apical, middle and basal canopy in the northerly, southerly, easterly and westerly directions, in 12 leaves per plant (i.e. 27,648 leaves from 48 T. argentea trees) during 24 months. Only insects and spiders collected for identification were removed from trees during the assessment. At least three specimens per insect or spider species were collected using aspirator, stored in glass flasks with 70% ethanol or mounted, separated into morphospecies, and sent for identification. Insect defoliation was assessed visually as the leaf area loss on a 0-100% scales with 5% increments for removed leaf area 57,58 . Ecological indices. To avoid pseudoreplication, mean numbers of data per tree were ever used. Ecological indices (abundance, diversity, and species richness) were calculated for each species per tree in the treatments (fertilization or not with DSS) using the software BioDiversity Professional, Version 2 59 . The arthropod diversity was calculated using the Hill's formula 60,61 and the species richness with the Simpson indices 62,63 . The predator (i.e. insects and spiders) and prey ratio on T. argentea was calculated per tree. Predators were classified as spiders (most important group), predators (predators + spiders) and total predators (spiders + predators + protocooperating ants).
Statistical analyses. Data on defoliation percentage, abundance, diversity, and species richness of phytophagous insects, pollinators and predators were submitted to non-parametric statistical hypothesis, the Wilcoxon signed-rank test (P < 0.05) 64 , using the statistical program "Sistema para Análises Estatísticas e Genéticas" (SAEG), version 9.1 65 . Simple regression analyzes and parameters (P < 0.05) were performed with SAEG to test the interactions between groups of phytophagous, pollinators and predators, and foliar mass (see 41 ).
The Spearman correlation matrix, among the most significant characteristics, was calculated. The matrices were submitted to correlation networks 66 . Edge thickness was controlled by application of a cut value of 0.28 (from which the Spearman correlation becomes significant, meaning that only edges with |r ij |≥ 0.28 are highlighted). These analyses were performed in R version 3.4.1 67 . The correlation network procedure was performed using the package qgraph 66 .

Ethics.
No specific permits are required to Terminalia argentea tree in Brazil. The laboratory and field studies did not involve endangered or protected species.

Data availability
All data generated or analyzed during this study are included in this manuscript. License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creat iveco mmons .org/licen ses/by/4.0/.