2.1 | Field and experiment work
The study site was located at Hutu Ranch, Qingcheng Mountain, Dujiangyan City, Sichuan Province, China (N 30.868449, E 103.495669; Alt. 1100 m). The study area was composed of subtropical evergreen broad-leaf forest with a subtropical monsoon climate that is identified as one of the world’s top 36 biodiversity hotspots (Habel et al., 2019; Myers et al., 2000)
From early November 2022 to mid-March 2023, we used mist net surveys and bird banding to capture birds. In total, 15 mist nets were installed. Two types of mist nets were used: 15 m long * 4 m high and 8 m long * 3 m high. The hole size of both mist nets was 1.5 * 1.5 cm. Mist surveys were conducted from 8:00 am to 6:00 pm every day, and the nets were monitored at least once per hour to avoid injury and death of captured birds. Captured birds were banded with a colored plastic ring with a unique number on the right foot and then released. The body mass of each bird was measured using a digital scale (± 0.01 g). Eight morphological traits were measured using a digital caliper (± 0.01 mm), including body length, bill length, bill crack length, tail length, flat wing length, second secondary feather length (S2) and third primary feather length (P3). The visual fat score was assigned a number based on the fat stored in the furcular and abdominal regions, referencing Kaiser’s songbird fat score system (Kaiser, 1993). Sex was initially determined by external characteristics if possible (for example, the blue plumage of adult males). For robins with olive-brown plumage, blood samples were collected from their brachial vein and stored in 100% ethanol for analysis in the lab, as their sex could not be identified based on morphology alone. DNA in blood samples was extracted using an Animal Genomic DNA kit (manufactured by Mei5 Biotech, Co. Ltd, Beijing, China), and PCR amplification was performed using primer pairs sex1’/sex2 (Wang et al., 2010) and ATP5A1 (Bantock et al., 2008), respectively. Sex was considered successfully determined only if two primers showed identical results.
In this study, the beginning of winter was defined from the day that the daily mean temperature consistently measured at below 10 °C for at least one week. The fine-scale climate data (temperature and humidity) was collected by a COS-03-X recorder (manufactured by Shandong Renke Control Technology Co.,Ltd). To investigate the insect food resource change in winter, the pitfall trap method was used to locally quantify the ground arthropod biomass. We set up five 10 m x 10 m plots near the mist nets, and buried 15 plastic cups in the ground of each plot. Each plastic cup was half-filled with glycerol-ethanol. We collected the trapped arthropods weekly and weighed the corresponding biomass of each plot (±0.001 g) after washing and drying the arthropods in the lab.