Effects of land use and microclimate changes on ant foraging in two Neotropical biomes

<b>Methodology</b>We collected data on ant foraging activity in the state of Minas Gerais, Brazil. Our sampling was conducted in Atlantic rainforest and Cerrado biomes. These regions are characterized by a wet summer (October to March) and a dry winter (April to September). We sampled between January and March of 2022 and 2023, during the summer, when the climate is hotter and more humid. In the summer, Cerrado has an average precipitation of 212.87 mm and average temperature of 22.16 °C, while Atlantic rainforest has an average precipitation of 134.54 mm and an average temperature of 20.41 °C (INMET, 2024). In total, we sampled 32 sites: six woodland savannah (cerradão), six tropical forest, and 20 coffee plantations (coffee plantations in Cerrado n = 8 and coffee plantations in Atlantic rainforest n = 12).In each site, we installed five pitfall traps in a transect and five sampling points spaced 50 meters apart to assess ant species richness and composition. The traps consisted of plastic pots with a diameter of 12 cm and a depth of 11 cm, installed at ground level. We added an aqueous solution composed of 200 ml of water mixed with salt and detergent to capture and preserve the collected ants (Jiménez‐Carmona et al., 2020). Each trap was left sampling for a period of 48 hours. After this period, the captured ants were stored in 70% alcohol.In the same transect, we assessed four liquid nutrients: amino acids, carbohydrate, lipid, and sodium. Moreover, we used water as a control in this experiment. In each site, between 7 a.m. and 8 a.m., we created a transect with 25 sampling points spaced 10 meters apart. In each transect, we sampled ants using the nutrients in Fisher 50 mL tubes. Each tube contained 10 ml of an aqueous solution of each type of nutrient. The nutrients included 20% amino acids (AA, made with Whey Protein Isolate and neutral containing L-glutamine and other amino acids such as leucine, isoleucine, and valine), 20% carbohydrate (CHO, made with sucrose), lipids (100% extra virgin olive oil, without water), 1% sodium solution (NaCl), and tap water (Lasmar et al., 2023). The tubes were placed horizontally on the ground. The tubes were closed at the end of sampling, and the ants inside them were collected and stored in 70% alcohol for preservation.After the nutrient experiment, in the same transect, we assessed foraging for larvae as a proxy measure for insect predation by ants. We sampled between 8 a.m. and 12 p.m., however, the transect had only 5 sampling points spaced 50 m apart. At each sampling point, we installed two beetle larvae (<i>Tenebrio molitor </i>Linnaeus, 1758). The larvae were attached with Velcro to an ethylene-vinyl acetate (EVA) surface to prevent them from escaping. At each sampling point, we observed foraging activity in four rounds of five minutes each. First, we attached the larvae to the ground at the first sampling point and immediately observed larvae use for five minutes. Then, we attached larvae at the second sampling point and observed predation activity for five minutes. We repeated this process until completing the transect and then started again for the second, third, and fourth rounds. This totaled 20 minutes per sampling point and 100 minutes per transect. We considered the foraging activity by ants when they attacked the larvae, either by stinging (using the sting) or biting (using the mandibles) (Wilker et al., 2023). The ants that engaged in predation were collected and stored in 70% alcohol for later identification.We measured temperature and humidity alongside the larvae experiment, at the same five sampling points over four rounds, between 8 a.m. and 12 p.m., using a digital thermo-hygrometer, we measured air temperature and humidity at 5 cm from the ground at the sampling point. The temperature was measured in degrees Celsius (°C), and the humidity was measured as a percentage (%). For the humidity data, we converted the values to proportion data, ranging between 0 and 1.<b>References</b>INMET. (2024). Instituto Nacional de Meteorologia. Banco de Dados Meteorológicos para Ensino e Pesquisa - BDMEP. Brasília, DF, Brazil. Available in: http://www.inmet.gov.br/. (Accessed 30 May 2024).Jiménez‐Carmona, F., Carpintero, S., &amp; Reyes‐López, J. L. (2020). Ant sampling: the importance of pitfall trap depth as a bias factor. <i>Entomologia Experimentalis et Applicata, 168(9),</i> 703–709. https://doi.org/10.1111/eea.12965Lasmar, C. J., Bishop, T. R., Parr, C. L., Queiroz, A. C., Wilker, I., Feitosa, R. M., Schmidt, F. A., &amp; Ribas, C. R. (2023). Testing the context dependence of ant nutrient preference across habitat strata and trophic levels in Neotropical biomes. <i>Ecology, 104(4),</i> e3975. https://doi.org/10.1002/ecy.3975Wilker, I., Lasmar, C. J., Schmidt, F. A., da Costa, M. M. S., Almeida, D. D. F., Dutra, D. B. D. S., Alves, A. L., Silva, L. L., &amp; Ribas, C. R. (2023). Land‐use change in the Amazon decreases ant diversity but increases ant‐mediated predation. <i>Insect Conservation and Diversity, 16(3),</i> 379–392. https://doi.org/10.1111/icad.12632