Arthropod communities in the tropics are increasingly impacted by rapid changes in land use. Because species showing distinct seasonal patterns of activity are thought to be at higher risk of climate-related extirpation, global warming is generally considered a lower threat to arthropod biodiversity in the tropics than in temperate regions. To examine changes associated with land use and weather variables in tropical arthropod communities, we deployed Malaise traps at three major anthropogenic forests (secondary reserve forest, oil palm forest, and urban ornamental forest (UOF)) in Peninsular Malaysia and collected arthropods continuously for 12 months. We used metabarcoding protocols to characterize the diversity within weekly samples. We found that changes in the composition of arthropod communities were significantly associated with maximum temperature in all the three forests, but shifts were reversed in the UOF compared with the other forests. This suggests arthropods in forests in Peninsular Malaysia face a double threat: community shifts and biodiversity loss due to exploitation and disturbance of forests which consequently put species at further risk related to global warming. We highlight the positive feedback mechanism of land use and temperature, which pose threats to the arthropod communities and further implicates ecosystem functioning and human well-being. Consequently, conservation and mitigation plans are urgently needed.

The natural areas of the Coastal Plain of Rio Grande do Sul (CPRS) have suffered fragmentation due to anthropic action. The faunal surveys offer a low-cost method to quickly evaluate environmental alterations, and Syrphidae flies are often used as models in this kind of study. We aimed to ascertain the diversity of Syrphidae in the South region of Brazil by estimating its species’ richness, and to use this data to identify new areas for conservation. In this survey Malaise traps were installed for 8 days in the CPRS, which was divided into five regions. Each region was subdivided into seven collecting areas and each of those areas received four traps, totaling 140 traps. A total of 456 Syrphidae individuals from 18 genera and 49 species were collected. In Region 1, there were nine exclusive species; in Region 2, there were three; in Region 3, there were 13, ten of which came from Estação Ecológica do Taim (ESEC Taim). In the Individual-based rarefaction analysis, Region 1 possessed the largest number of expected species out of the regions in the CPRS; we found 97% of these species. This insect collection effort, as one of the first in the CPRS, has broadened the known geographic distributions of 11 species of Syrphidae, and also indicated areas to be conserved. Additionally, it gave support for expanding ESEC Taim and creating new areas of conservation in Region 1, in Arroio Pelotas and Arroio Corrientes.

Metabarcoding, the coupling of DNA-based species identification and high-throughput sequencing, offers enormous promise for arthropod biodiversity studies but factors such as cost, speed and ease-of-use of bioinformatic pipelines, crucial for making the leapt from demonstration studies to a real-world application, have not yet been adequately addressed. Here, four published and one newly designed primer sets were tested across a diverse set of 80 arthropod species, representing 11 orders, to establish optimal protocols for Illumina-based metabarcoding of tropical Malaise trap samples. Two primer sets which showed the highest amplification success with individual specimen polymerase chain reaction (PCR, 98%) were used for bulk PCR and Illumina MiSeq sequencing. The sequencing outputs were subjected to both manual and simple metagenomics quality control and filtering pipelines. We obtained acceptable detection rates after bulk PCR and high-throughput sequencing (80–90% of input species) but analyses were complicated by putative heteroplasmic sequences and contamination. The manual pipeline produced similar or better outputs to the simple metagenomics pipeline (1.4 compared with 0.5 expected:unexpected Operational Taxonomic Units). Our study suggests that metabarcoding is slowly becoming as cheap, fast and easy as conventional DNA barcoding, and that Malaise trap metabarcoding may soon fulfill its potential, providing a thermometer for biodiversity.

The horseflies in a patch of gallery-forest in a savannah of French Guiana were collected by 1°) for 24 hr with a Malaise trap, and 2° for 2 hr at dawn and at dusk, with a Malaise trap baited with carbon dioxide. The crepuscular horseflies, Cryptotylus unicolor, Chlorotabanus mexicanus and Chlorotabanus inanis, were more abundant in the baited trap (average: 55% of the flies collected) than in the unbaited trap (3%). Seasonal changes in relative abundance of the crepuscular species were evident in the baited trap only. The baited trap caught more flies and more species (averages: 315 flies and 7.5 species per 2hr catch) than the unbaited trap (averages 64 flies and 5.1 species per 24 hr catch). However, the overall number of species collected in each trap during the study did not differ greatly (16 species for 126 catches of 24 hr for the unbaited trap vs. 14 species for 29 catches of 2 hr for the baited trap).