Diversity and Similarity Gomphocerinae (Orthoptera: Acrididae) Communities in the Brazilian Amazon

Inventories of Amazon invertebrates are relatively incipient and fragmented. The state of Amapá is one of the Amazonian states with a large knowledge gap regarding invertebrate biodiversity. Also, there is no record in the literature of systematic studies that focus mainly on the Acridofauna. Therefore the goal of this study was to understand the diversity and abundance of grasshoppers (Gomphocerinae) of the Environmental Protection Area of the Curiaú river, Macapá AP. Twelve samples were collected from October 2011 to September 2012 using the active search technique with sweep nets. A total of 508 Gomphocerinae individuals were sampled and classified into five genera and twelve species. The floristic composition of sites A1 and A3, and sites A5 and A6, are considered more similar since the locusts are closely related to the vegetation. This is the first study assessing the diversity of orthopteroides in the state of Amapá. However, this pioneering study is expected to be the basis of future studies.


Introduction
The main representatives of the order Orthoptera are commonly known as grasshoppers, crickets and bush-crickets. This order has approximately 25,000 species described worldwide (Naskrecki & Otte 1998) and is recognized primarily for its ecological and economic importance. The Orthoptera are ecologically important for being defoliators (primary consumers), contributing directly to the introduction of organic matter into the soil, and for stapling food for many vertebrates (frogs, lizards, birds, monkeys, etc.) and invertebrates (especially other insects), for this reason, they are considered essential elements in the food chain. The economic importance of Orthoptera is due to the fact that some species have a gregarious habit, forming "clouds" of locusts that can devastate crops in a short period of time (Song et al., 2018 andCigliano et al., 2018).
The Gomphocerinae are distributed along the Neotropics and can be found from the southern United States to Argentina (Otte 1981 andCigliano et al., 2018). Among this subfamily, the genus Rhammatocerus Saussure 1861 has been the least studied, with a wide geographical distribution in Brazil, and has a known record of nine species (Assis-Pujol 1997). The species Rhammatocerus schistocercoides (Rehn 1906), in particular, is a leading locust in the attack of plantations in the state of Mato Grosso (Batistella et al., 1997).
Invertebrate inventories in Amazon are relatively incipient and fragmented. The demand for biodiversity information, monitoring and quantification of the effect of impacts on biodiversity is huge and diverse. Such demands may range from economic and ecological zoning to conservation planning, bioprospecting, the development of alternative scenarios, natural resource management and monitoring of natural impacts (Matiotti da Costa et al., 2015).
This study tried to answer some important questions regarding a taxon of ecological and economic importance, which had not previously been systematically studied in the state of Amapá. Therefore, the goal of this study was to evaluate the diversity and abundance of grasshoppers (Gomphocerinae) of the EPA of the Curiaú River, (Macapá, State of Amapá), identifying the composition of species sampled in the study area and diversity estimates. The state of Amapá is one of Brazilian Amazon states with a large knowledge gap regarding invertebrate biodiversity. No systematic studies that focus primarily on the Acridofauna of the state have been reported in the literature, especially in the Environmental Protection Area of the Curiaú River (Área de Proteção Ambiental do Rio Curiaú -EPA of the Curiaú River).
The data were collected in 12 sampling campaigns using the active search technique for 60 minutes, with sweep nets, according to Borror & DeLong (1988) and Triplehorn & Johnson (2011) this is the most efficient type of collection for orthopteroids. Thus, a total of 144 samples were collected for each sampling site in the EPA of the Curiaú River, between October 2011 and September 2012. All material collected was identified to species and morphospecies levels using identification keys (Otte 1979a(Otte , 1979b(Otte , 1981Otte & Jago 1979;Donato & Cigliano 2000;Donato 2003;Assis -Pujol 1997;and Carbonell 1995) and through comparison of previously identified specimens from the IEPA (Institute of Scientific and Technological Research of the State of Amapá). Part of the identified specimens was labeled and stored in the scientific collections of IEPA and the teaching collection of the Federal University of Amapá (Universidade Federal do Amapá -UNIFAP). All data was stored in a metadata file.
Monthly averages of air temperature (C°), relative humidity (%), and accumulated precipitation for EPA of the Curiaú River were obtained by consulting the database of the National Institute for Space Research (Instituto Nacional de Pesquisas Espaciais -INPE), and National Environmental Data System (Sistema Nacional de Dados Ambientais -SINDA) available at http://sinda.crn2.inpe.br.
The statistical analyses were performed using the statistical package BiodiversityR of R software, version 2.15.2 (R Development Core Team 2011). Richness, abundance and diversity analyses were conducted following previously established parameters (Kindt & Koe 2005). Species richness was calculated to assess the effectiveness of Gomphocerinae sampling at the two environments, and to perform the analyses regarding species richness (Colwell & Coddington 1994;Chao & Lee 1992).
The indexes used were Shannon-Wiener (H') and Uniformity (U) (Krebs 1999). The Bray-Curtis distance was calculated to compare the similarities between each sampling site. The Kolmogorov-Smirnov test was used to assess if the samples follow a normal distribution (p>0.100). The Jaccard similarity index was used to evaluate the degree of similarity in species composition among areas.

Results
A total of 508 Gomphocerinae specimens were sampled and classified into five genera and twelve species. The Six species were recorded at site A1, seven at A2, and eight at A3 (all located within the pasture environment). A total of eight species were recorded at site A4, and five at A5 and A6 (both sites located within Amazon savanna). The results showed that site A2 had the highest grasshopper abundance (209; 44%), followed by site A3 with 77 specimens (15%). At site A1, 59 locusts were collected, corresponding to 11% of the total number of individuals. For sites located within the Amazonian Savannah, 72 specimens (13%) were recorded at A4, 52 (9%) at A6, and finally, 39 specimens at A5 (8%) ( Table 1).
The species Amblytropidia sp., O. concinnula, O. punctata and R. brasiliensis were recorded at all sampling sites, whereas C. pulcher, R. pictus, R. pseudocyanipes, R. schistocercoides and R. suffusus were recorded in only one. The other species were reported in more than one sampling site. In table I, the abundance of each species within the different sampling sites is shown.
The genus Amblytropidia was most abundant in site A2, the genus Rhammatocerus occurred in all sites and the species R. brasilienses was the only one that occurred in all sites. R. brasilienses was also had greater representation in site A2, and the Research, Society and Development, v. 10, n. 8, e54710817763, 2021 (CC BY 4. Mean monthly air temperature, relative humidity and accumulated precipitation values were obtained (Table 2).
Temperature and Relative Humidity values showed no significant change along the sampling period. However, accumulated precipitation was lower from August to December than from March to June. The highest precipitation values were recorded from March to June, a characteristic period called more and less rainy in the Amazon region. The species C. pulcher (n = 4) was only recorded in April 2012 at site A4. The species S. nigriceps was only recorded in May and June 2012 at site A3, and R. pictus (n = 1) in December 2011 at site A1. The species R. pseudocyanipes (n = 1) was found only at site A3 in April 2012. The other species were found in more than a month, throughout the sampling periods (Table   3). A total of 12 species were found in the sites studied. The highest species richness was observed in pasture environments.
The species accumulation curve is shown in Figure 1. For the Pasture environment an estimate of 16.7 species was given by the First order Jaccknife, and 36.5 by the first order Chao. For the Savana environment, the Jackknife gave an estimate of 12.3 species and Chao of 9 species. However, even showing that the pasture environment had higher species richness, it is noteworthy that the species accumulation curve did not reach its asymptote.  (Table 4). Table 4. Indicates the Shannon`s and Simpson's diversity index, highlighting the Shannon`s index, which gives greater weight to species richness than the Simpson index.
Tests carried out using the Bray-Curtis distance can confirm the existence of differences in the biological distance among sampling sites. Tests confirmed that the composition/abundance of species found in the pasture environment (sites A1, A2 and A3) differed from the savannah environments (sites A3, A4 and A5).
The Bray-Curtis distance test indicated that site A2 (site with the highest abundance) was closer to sites A1 and A3 (all within the Pasture environment). In turn, site A4 (site with the second-largest species richness) was closer to sites A5 and A6 (all comprising the savannah environment), as shown in Table 5.  Source: Authors.

Shannon Simpson
Sites A1 and A6 had the largest biological distance between each other (D= 0.566974), considering all sites. In contrast, A5 and A6 had the most similar community structure (lowest Bray-Curtis distance). However, sites A2 and A4 were similar, once A4 was closer to sites of the Pasture environment than to sites of the savannah environment, as shown in Figure 2.

Discussion
This study shows that the species richness values recorded at the sampling sites across the physiognomies were significantly different among the sampling sites. However, the pasture environment is more species rich than the Amazon  (Table 1 and Table 4). Thus, there was a similarity to studies carried out by Sánchez and Wysieck (1993) and Cigliano, Wysieckiv and Lange (2000) who found similar species richness values in pastures from Argentina.
However, in other studies, the species richness recorded seemed to be slightly higher than in this study. Kemp (1992) analyzed intraspecific associations of grasshoppers in a pasture environment of the Gallatin Valley, Montana, in the United States (from 1988 to 1990), and observed low species density in the region, and recorded high species richness for all sampling sites (Jaccknife and Chao index ranging from 10 to 17). Similar high species richness values were also found in a previous study carried out in Nebraska and northeastern Colorado (Joern & Pruess 1986;Capinera & Thomson, 1987).
It is noteworthy that grasshopper density was not reported in this study or in the studies cited, where species were recorded for over a year. However, there were 12 species at Curiaú. Thus, it is important to observe that sites A1, A2 and A3, consist of pasture environments and had the highest species richness (Table 1). This high species richness may be explained by the constant environmental changes made for grazing, cattle pasture and buffalo breeding on the Tábua Branca farm. In this farm there is a possible renewal of grass composition, favoring the high rate of species establishment in the three sites.
It is important to mention that among the pasture environment sites, A3 (site with the greatest species richness) is used less frequently than the other sites so that more time would be available for species establishment, favoring population growth.
Site A2, the furthest point used for farming purposes, had a high species richness value (S= 7) even when being used. The high species richness value observed for site A2 indicated that, despite the constant environmental change, the Gomphocerinae species present are able to become established in this environment.
Site A4 exhibited the highest species richness for the Amazon savannah environment. This pattern may have been detected due to the proximity to highway EAP-070. There may be changes affecting the richness and composition of locusts from this area, leading to this high species richness index. Sites A5 and A6, had the lowest species richness in this study and can be characterized by being located in remote areas and suffering no environmental changes. Thus, there were no changes in the floristic composition of the environment, which directly influences the food supply of these insects and consequently changes the composition and richness of species present in these two sites.
Site A2 exhibited the largest relative abundance of species in the area studied, with 209 specimens sampled. This large abundance may also be related to the floristic composition in the area. It is acknowledged that grass composition in the environment influences species composition, richness and abundance, in other words, the locusts are directly influenced by the floristic composition of the environment. Grasses are the main food of locusts. Thus, grass composition is a decisive factor in locust population establishment (Descamps, 1978 andCarvalho et al., 2012).
According to Barrientos (1995), farmers planted wheat (during winter) and soybeans (during summer) in the northwest and southeast regions of Rio Grande do Sul, from 1970 to 1985. The authors also reported the continuous use of chemicals for pest control. In 1986, with the introduction of cattle in the same areas and the use of "Pensacola" (Paspalum notatum Fluegge var. Saurae Parodi) grass, there was a great proliferation of grasshopper populations. This high proliferation, in the absence of natural enemies, caused a population outbreak.
In Brazil, there are records of several species of grasshoppers that are directly influenced by climatic factors such as temperature and accumulated precipitation (Embrapa, 2007). Environmental changes, such as the recent introduction of agriculture in regions, or ecological imbalance generated by chemical use also influence this community. These factors are linked to the appearance or disappearance of grasshopper populations.
The floristic composition was not assessed in this study, but it is known for being important for grasshopper occurrence.
The genus Orphulella was present in all samples and recorded throughout the study period. The species O. concinnula was the most abundant in the study, with 145 specimens, followed by O. punctata with 109 specimens. According to Matiotti da Costa et al., (2015), Serrao et al., (1997), Song (2010) and Rowell (2013), this genus has a wide geographical distribution, occurring from Canada to Argentina. Orphulella is one of the Gomphocerinae genera which is difficult to identify once its species are widespread and have phase polymorphism. The species O. punctata is known to be a fairly common Neotropical species, found in pastures and marginal forests from Central Mexico to Argentina (Matiotti da Costa et al., 2015, Otte 1978and Song et al., 2018. For Brazil, according to Silva et al. (1968), this genus can become pests of pastures and of Nicotiana tabacum. In São Carlos, Serrão et al. (1997) reported that this genus is found throughout the year without causing damage, showing that this species may show equilibrium patterns possibly related to the occurrence of predators and parasitoids. Tests were also carried out to identify the natural predators of this insect group, and predation mainly by spiders and lacewings was observed (Serrão et al., 1997).
A total of six species of the Rhammatocerus genus was recorded in this study. It is a so far rarely studied genus.
However, it has a wide geographic distribution and a known record of 9 species in Brazil (Assis-Pujol 1997). The species R.
brasiliensis was the third most abundant, with 100 individuals sampled. This species had a higher incidence from April to September 2012, suggesting that climatic conditions directly influence this species record. This occurrence pattern is explained, mainly by an increase in Accumulated precipitation in the same period, favoring population growth.
Importantly, the species R. schistocercoides, one of the major causes of damage to crops, was recorded among the species sampled. There are several studies reporting the damage this species causes to plantations. Lecoq (1991) reported that this species is causing damage to soybean, sugarcane and corn plantations in the states of Rondônia and Mato Grosso. Despite it being recorded only recorded at site A2, with the constant changes occurring in Curiaú until now, it may have not been possible for this species to experience a population outbreak able to generate economic damage. However, it is important to know that this species was recorded in the region studied. Another species recorded in this study, that so far shows no threat to crops, is the species R. pictus. The species R. pictus and R. pseudocyanipes had lower abundance and were only recorded once in December 2011 and in April 2012, indicating the existence of specificity of these species to climatic conditions that favor its appearance.
The species Compsacris pulcher, the only of the Compsacris genus recorded, occurred in only one site and during one month. This pattern suggests this species has specific requirements regarding the sampled environments, mainly for being sampled in environments near zones of transition between sites of open and closed vegetation structures (edge environments).
Species occurrence may also have been influenced by the climate.
Grasshoppers may exhibit specificity to certain vegetation types. Descamps (1978) and Carvalho et al. (2012) show states that many Acridoidea species may show specificity to exclusively open, closed, or partially sunny environments due to their feeding requirements. This statement is based primarily on the fact that there will be plant species best adapted to environments with varying degrees of light input. Consequently, the grasshopper fauna will respond to the vegetation types. This observation confirms the record of C. pulcher, once it was found in forest edge vegetation (access roads to the edges of tall vegetation, or adjacent areas of low and open vegetation). Vegetation with a plant community adapted to the direct solar incidence of some periods of the day is typical of such environments. This statement may indicate that this grasshopper species (C. pulcher) occurs specifically in the vegetation of partially sunny environments.
Correlations of the temporal variation data of each species indicated a population growth (higher abundance) during the period with the most precipitation for most of the species sampled. Also, there were no other records of this higher abundance, proving the specificity of locusts to the environmental and climate types, to which it is best adapted. The species S. nigriceps occurred in four of the six sampling sites, only in May and June 2012, showing the specificity of this species to climatic factors.
When evaluating species diversity according to the two indexes used in this study (Shannon and Simpson), site A2 was observed to have the highest diversity. This diversity can be explained by the high abundance recorded at this site mainly due to the environmental characteristics. Site A2 also had low species richness when compared with site A4, second in the diversity rank.
The floristic composition of sites A1 and A3, and A5 and A6 are more similar according to the analysis of similarity among sampling sites and considering that locusts are closely related to the vegetation. Also, de results of the dendrogram indicated similarity of 50% among these sites regarding Gomphocerinae species composition. This can be explained by the degree of human activity in each site (Carvalho et al., 2012). On the other hand, site A2 was similar only to site A4. Site A4 differed from sites A1 and A3. However, it exhibited biological proximity to sites A5 and A6, indicating a similarity in species composition among sites that belong to the same environment.

Conclusion
Currently, the knowledge regarding Orthopteroides in Brazil is still very limited, and most of the information is concentrated in the Northeast, Midwest, Southeast and Southern regions, where there is a larger problem with species that cause economic damage. Most studies carried out among the Brazillian regions are focused mainly near the states of Mato Grosso and Rondônia, due to the recently inserted soybean cultures. However, for the Northern region, little is known regarding the composition of this group. The information available became limited to the cities of Belém and Manaus. Besides, this information ends up being limited only to these two points. Other research centers have some difficulty.
This study was carried out to better understand the existing locust populations in the state of Amapá, and their patterns of diversity and similarity between different environments in the Brazilian Amazon. Therefore, it showed that the composition and distribution of these locusts need to be better evaluated. It is recommended that the sampling effort can be increased, as well as the number of study areas, with these changes, we believe that the composition and indices of the diversity of Gomphocerinae can be better understood.