Arctiini Leach, [1815] (Lepidoptera, Erebidae, Arctiinae) of the Amazônia National Park, Pará, Brazil

This study evaluated the Arctiini fauna of the Amazônia National Park (Pará, Brazil) collected in four expeditions between July 11 to 15 and October 02 to 06, 2013 (dry season)


Introduction
There are many articles that highlight the Amazonian biodiversity and its mega-diversity in species, landscapes, and ecosystems, as well as its fundamental importance in global climate regulation. In fact, the Amazon is the largest rainforest in the world, covering an area of approximately 6,000,000 km 2 and stands out for having the highest diversity of animals and plants on the planet (BOUBLI; HRBEK, 2012).
The Protected Areas (PAs) have been key to the preservation in situ of nature. PAs have slowed down deforestation, which is more intense along the southern and eastern borders of the Amazon region and have mitigated land-use pressures such as illegal deforestation and the irregular occupation of land (BERNARD et al., 2014).
According to Carvalho et al. (2016), the municipality of Itaituba, where most of the Amazônia National Park (AMZNP) is located, is one of four municipalities in the state of Pará with the highest deforestation rates.
Lepidopterans are one of the main insect orders, and are particularly useful in studies of biodiversity owing to their species richness and economic importance. Due to the phytophagous eating habits of their larvae, they can be used as bioindicators of vegetation quality in a specific area (KITCHING et al., 2000;TESTON et al., 2006). Adults are easily collected with light traps and there are many different models (TESTON et al., 2006;CAMARGO, 2001).
Nocturnal Arctiini are one of the most used Lepidoptera groups for ecosystem monitoring, and many studies have documented its importance (DELFINA; TESTON, 2013;CORREA, 2015). Furthermore, some inventories evaluating the Arctiini fauna in protected areas (PAs) have already been made in the main biomes in Brazil (e. g. Amazon, Cerrado and Atlantic Forest) (FERRO; DINIZ, 2007;CORREA, 2015;MORENO;FERRO, 2016).
In the Neotropical region, there are 4,761 species of Arctiini (HEPPNER, 1991), andBrown Jr. andFreitas (1999) estimate the occurrence of at least two thousand species in Brazil. The state of Pará alone has 536 recorded species . Recently the first study on the Arctiini fauna was carried out in a protected area in the Amazon rainforest in Brazil (TESTON;CORREA, 2015).
With the aim of contributing to a greater understanding of Arctiini diversity in the Brazilian Amazon, primarily in PAs, a survey of the species of this lepidopteran subfamily was undertaken in the Amazônia National Park (AMZNP), Pará, Brazil. Until the present study, there were no records of the occurrence of Arctiini species in the AMZNP. Recently, political pressure has been put on the AMZNP and other federal conservation units to reduce their coverage, their degree of protection, and even their conservation status (BRASIL, 2012;BERNARD et al., 2014).

Material and Methods
The Amazonia National Park (AMZNP), a conservation unit maintained by the Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio), was created by the Decree of February 19, 1974. With an area of 1,070,736 ha, it is nested in the municipalities of Aveiro and Itaituba, in the state of Pará, and Maués in the state of Amazonas (Figure 1). It is limited to the north by the Andirá-Marau Indigenous Land, to the south by the Amanã National Forest, to the east by the Maués State Forest and the Pau-Rosa National Forest, and to the west by the Tapajós River. The park presents a mosaic of 14 different vegetation types, with the dense lowland rainforest occupying the largest portion, interspersed with a small portion of open rainforest (IBDF, 1978;KASECKER;SILVA, 2011;BRASIL, 2012).
In this region the rainy season extends from December to May, the dry from June to November (MORAES et al., 2005).
Collections were made in four expeditions, with four nights in length each, two during the rainy season and two during the dry season, totaling 16 samples. During the dry season, samples were collected July 11 to 15 and October 02 to 06, 2013, and during the rainy season, March 01 to 05 and June 25 to 29, 2014 ( Figure 1; Table 1). To that end, L-shaped light traps (2.0 m in width and 2.0 m length) were made with two white cloths. The moths were captured with killing jars containing Ethyl acetate. Two mixed lamps (250 watts) powered by a portable electric generator (TESTON; CORREA, 2015), one on each cloth, were used to attract the moths. These lights were left on all night, from 6pm to 6am of the following day, in a total sampling effort of 12h in every sampling site (SS).
The captured Arctiini were separated, counted, and recorded in specific worksheets for every location in the  Table 1 and satellite image by Google Earth Pro.
(voucher) were prepared following the traditional method for Lepidoptera, and subsequently incorporated into the laboratory collection (LELN).
In order to compare among rainy and dry seasons, a list of the occurrence and abundance (N) of Arctiini was organized in alphabetical order by subtribe, genus, and species. Richness (S), diversity (H') and evenness (E') Shannon indexes, Berger-Parker dominance (BP) (MAGURRAN, 2011), and estimate of species richness by the non-parametric methods Chao1, Chao2, ACE, ICE, Jackknife1, Jackknife2 and Bootstrap, using the program EstimateS (version 9.1.0) (COLWELL, 2013), are presented for rainy and dry seasons and total.

Results
In total, 804 specimens (N) were captured, distributed in 190 species (S) and representing all seven Arctiini subtribes, with N = 520 and S = 161 for dry season; and N = 284 and S = 102 for rainy season (Tables 2 and 3).
The Shannon diversity index (Table 3)  The richness estimators (Table 4) point to an increase in the number of Arctiini species in the AMZNP. In this study, between 84.8% (Bootstrap) to 65.7% (Jackknife2) of the total expected species were captured; and between 83% (Bootstrap) to 59% (ICE) for dry season and, between 80.3% (Bootstrap) to 44% (ICE) for rainy season. The accumulation curve of species (Figure 2) has not stabilized.

Discussion
The number of sampled species corresponds to 35.4% of the total known for the state, according to . A total of 22 species are new records for the state of Pará (denoted by an asterisk in Table  2), bringing the number of recorded species to 558. Of these, four are new records for the Brazilian Amazon: Glaucostola maroniensis Joicey & Talbot, 1918, Lophocampa albescens (Rothschild, 1909, Pseudalus salmonaceus (Rothschild, 1909) and Trichromia lophosticta (Schaus, 1911).
Only seven species represented one third of the total abundance (30.5%). On the other hand, 64 species (33.6%) were represented only by one captured specimen. In previous research, including those conducted in the state of Pará, many Arctiini species were represented by only a few individuals (HAWES et al., 2009;DELFINA, 2010;TESTON et al., 2012;TESTON, 2013;CORREA, 2015;VALENTE et al., 2018). The same applies to the dry and rainy seasons. In the dry season, eight species represented 31.5% of the abundance and 68 species (42%) with one captured specimen and, in the rainy season, four species represented 28.9% of the abundance and 55 species (54%) with one captured specimen (TESTON; DELFINA, 2010;TESTON et al., 2012;TESTON, 2013;VALENTE et al., 2018).
The diversity, equitability and dominance indices follow the trend already observed in other studies in Pará, being more relevant in the dry season (TESTON; DELFINA, 2010;TESTON et al., 2012;TESTON, 2013).
When comparing the richness with other inventory in protected areas (PAs) of other Brazilian biomes (e. g. Cerrado and Atlantic Forest), we note the relevance of the richness found in this study, which was 61% higher than that found in the Emas National Park (S = 117) in the cerrado (MORENO; FERRO, 2016); and only 19% fewer species than in the Boraceia Biological Station (S = 235) in the Atlantic Forest, with greater sampling effort (410 nights) (FERRO; DINIZ, 2007).
The accumulation curve of species did not show an asymptote, indicating the need to enlarge the sampling effort. In the same manner, the estimators show that the observed richness varied from 65.7% (Jackknife2) to 84.8% (Bootstrap) in relation to the estimated richness, indicating that the inventory must be broadened, consistent with recent Arctiini study carried out in the Serra do Pardo National Park, an important Protected Area of Pará (TESTON; CORREA, 2015), which points to an increase in the observed richness as a function of increased sampling efforts.
Arctiini's continuous inventory in the Amazonia National Park will contribute to a better understanding of lepidoptera biodiversity in the Amazon biome, and will serve as basic knowledge for conservation strategies for the group in this protected area. Teston and Ferro (2019) suggest intensive collection efforts for tiger moths in this biome.