The family Syrphidae is diverse and is subdivided into four subfamilies that vary in lifestyle, including between stages of development; therefore, there are differences in the number of species and specimens present in certain locations. In the study area, there was a greater number of species collected and a higher relative frequency for the subfamily Syrphinae, followed by the subfamilies Eristalinae and Microdontinae. These results corroborate the finding of other studies that found greater richness and abundance of species of the subfamily Syrphinae (Jorge et al. 2007; Souza et al. 2014; Djellab et al. 2019). The high richness and abundance of Syrphinae collected may be related to the wide distribution and activity of species that are abundant throughout the year. Additionally, the diversity in larval habits (phytophagous and predatory stages) favors the permanence of both larvae and adults of the species in these areas and their consequent capture. For instance, in the subfamily Eristalinae, representatives spend much of their time performing foraging activities and are therefore more easily collected (Morales and Köhler 2008).
The low number of species belonging to the subfamily Microdontinae that were captured was expected both for collection with entomological nets and for passive traps. Adults in this subfamily usually do not feed on floral resources (Reemer 2014) and, therefore, are not commonly found foraging on flowers and are rarely captured with entomological nets. According to Cheng and Thompson (2008), adults are commonly found in the vicinity of ant nests and are not distant from them, since the larvae have a myrmecophilic habit (Reemer 2013). For this reason, Microdontinae are usually rare in collections made with Malaise traps; in this study, the passive traps were not established near ant nests.
The most abundant and richest genus was Toxomerus Macquart, belonging to the subfamily Syrphinae, with 15 species and 260 specimens. The number of species sampled represents 38.46% of the species recorded for Brazil (39 species) (Morales and Marinoni 2022). Mengual et al (2012) state that Toxomerus species are probably the most abundant among the syrphids of the Neotropical region; however, few studies have collected biological information about this group. Another factor that may be linked to species abundance and richness is the great feeding plasticity of Toxomerus species. Among the sampled species, most were predators in the larval stage, except for Toxomerus politus Macquart, 1842, Toxomerus dispar (Fabricius, 1794), and Toxomerus floralis (Fabricius, 1798), which are polliniferous (Rojo et al. 2003; Nunes-Silva et al. 2010; Jordaens et al. 2015). In addition to these three species, it is believed that Toxomerus lacrymosus (Bigot, 1884), the most abundant species in this study, is also a predator. Although there is a lack of records on the feeding habits of the larvae of this species, Mengual et al. (2012) suggest that T. lacrymosus is phytophagous because it is phylogenetically related to other phytophagous species (i.e., it belongs to the politus group). It is worth noting that these authors pointed out the possibility of a misunderstanding of a record of T. lacrymosus predating Aleyrodidade cited by Oliveira et al. (2003) because, in a later study by the first author (Oliveira and Santos 2005), the image of a species of Ocyptamus s.l. is shown referring to T. lacrymosus.
The adults of Toxomerus laenas (Walker, 1852) and Salpingogaster nigra Schiner, 1868 are floral visitors, and the larvae are predators (Rojo et al. 2003). In turn, Ornidia obesa (Fabricius, 1775), the fourth most abundant species, is saprophagous as a larva, feeding on different types of decomposing organic material (Martins et al. 2010). It is relevant to note that S. nigra was more abundant in the rainy season, which was related to the habit of its larvae, which are predators of spittlebugs, whose nymphs occur only during the rainy season (Veríssimo et al. 2018).
For capturing adults, the Malaise trap was more efficient, considering the total number of specimens collected. Some species of Syrphidae have a short flight season, so passive traps assist in achieving greater capture of specimens (Terry and Nelson 2017). Marinoni and Bonatto (2002) also argue that the best place to install passive traps is in open fields, which also corroborates the fact that a higher number of specimens were collected in the Cerrado phytophysiognomy. Some syrphid species tend to remain close to the soil layer, while others exploit flowers in the treetops, flying away from the passive traps; therefore, a combination of different capture techniques may be the ideal approach to detect the largest possible number of species present in an ecosystem (Sommaggio 1999). This was also observed in the present study, since unique species were captured with an entomological net. However, it should be noted that the sampling effort differed between the collection techniques: the greater effort was made with the passive trap (i.e., seven consecutive days per site), while the collection made with the entomological net consisted of only one day in the study period. Therefore, despite there being more than one active collector, the sampling period covered by the active searching is not equal to that covered by passive trapping; thus, it is advisable to make more than one collection in the season of interest with an entomological net.
The richness and abundance of Syrphidae were compared among the phytophysiognomies, and the results showed that the Cerrado had greater abundance than the gallery forest, regardless of the collection technique, corroborating the results of a study by Augusto (2019), in which a higher abundance was found in the Cerrado (195) than in the gallery forest (62). This is probably because syrphids are generally associated with open and light areas, as these conditions favor the flight range of the insects in addition to enabling easy access to resources (Owen 1991; Jorge et al. 2007). The diversity index was significantly higher for the gallery forest than for the Cerrado. This is associated with the greater diversity in the forest, where 144 specimens from 96 different species were found, while in the Cerrado, of the 298 specimens collected, 146 are from only one species (T. lacrymosus).
The number of Syrphidae specimens is related to the amount of resources available in the environment, while the richness is linked to the diversity of those resources (Meyer et al. 2009). Gallery forests have evergreen vegetation, with humid and forested environments and water bodies. Additionally, gallery forests have a more representative species richness than other phytophysiognomies, with high environmental heterogeneity due to the presence of microenvironments (Ribeiro et al. 2001). These characteristics have a positive relationship with the diversity of Syrphidae species, which are positively influenced by the diversity of plants, in addition to providing several microhabitats favorable to the development of immature syrphid species (Rotheray et al. 2001). Studies conducted by Gaytán et al. (2020) and Budhathoki et al. (2021) observed a greater richness of syrphid species in a forest area than in pasture habitats. It is noteworthy that the area of the forest where the collections were performed with a net and the collection of insects by passive trapping was performed close to a transition area. This area is an intermediate space between open fields and closed forests, which favors the capture of grassland and forest species.
Some interpolation and extrapolation curves did not tend to stabilize, indicating the need for more sampling; in tropical environments, these curves generally do not stabilize due to the high richness in this region. Nevertheless, even if a small number of Malaise traps are used, the findings from long-term analyses can still be very useful because these traps can sample both rare species and the most abundant species in the study area (Fraser et al. 2008). However, it is critical to note that even if richness curves created according to the number of individuals do not tend to stabilize, they also do not tend to meet, which validates the results that the phytophysiognomies are actually different.
Lastly, during the collection period, there was a statistically significant negative correlation between the temperature and the composition of the Syrphidae community for the abundance of collected specimens. Other studies have also found a significant relationship with this factor. Groot and Kogoj (2015) found that temperature negatively affected the abundance of Cheilosia fasciata at temperatures between 7.3 and 12.2°C. This agrees with the study by Bashir et al. (2015), in which it was observed that the diversity of pollinator flies was lowest at 10°C. Ball and Morris (2015) reported that syrphids are active throughout the year but with greater prominence in the spring. This distribution is related to the periods in which there is greater availability of floral resources, which are necessary for adult nutrition (Djellab et al. 2019; Budhathoki et al. 2021).
In this study, the mean temperature ranged between 21 and 32°C and had a negative influence on the abundance and composition of the Syrphidae community. It is noteworthy that there are several possible factors that affect the abundance, including local and site-dependent biological factors; for example, very high temperatures can influence the floral structure, leaving plants more wilted, which indirectly affects visitation by flies. Terry and Nelson (2017) noted a decrease in the abundance of syrphids as the warmer months began, while Morales and Köhler (2008), analyzing the abundance of Syrphidae, identified temperatures between 28 and 32°C as preferential for syrphids. It is known that altitudinal temperature gradients have considerable influence on insect populations (Hodkinson 2005); therefore, the area studied by Morales and Köhler (2008), which has a lower altitude and consequently higher temperatures, presents different environmental characteristics than those studied here. However, the abundance patterns of Syrphidae seem to be more affected by high temperatures when historical temperatures are examined (Terry and Nelson 2017).
Importantly, studies cited in the discussion of abiotic factors were conducted in areas with different physiognomies, supporting the hypotheses and a discussion of the results found herein. Further studies on the influence of these factors on species abundance, especially in Cerrado areas, are needed to develop appropriate hypotheses that address the reason for these correlations. Further studies are also needed to determine the change that accompanies plant resources during times with decreased flower abundance to determine whether the effects of heat on flower abundance are physiological or indirect through its effect on vegetation