Forest vertical and horizontal temperature similarity drives arthropod communities in a managed temperate forest

Manipulating the canopy structure is the core tool of silviculture operation, and with that, changing the light availability alters temperature dynamics from the forest floor to the canopy. This should affect communities of ectothermic organisms such as insects, but we lack information on insect distributions in the complex 3D space of forests. Therefore, we set up temperature loggers and insect traps vertically (flight-interception traps) and horizontally (pitfall traps) in forests with experimental thinning and gap felling 8 years after the intervention. By metabarcoding, we identified [~]10,600 Operational Taxonomic Units (OTUs) from 44 orders including [~]2450 arthropods assigned to species in our 426 samples. Arthropod community similarity matrices were quantified along the Hill numbers accounting for rare to dominant species and under consideration of incomplete samples. Arthropod communities were shaped by stratification (height above ground 0 m, 2 m, 10 m, 15 m), and by temperature similarity. Average nighttime temperature was the most important temperature variable for overall arthropod community similarity metrics. Restricted to flight interception traps, flying insect communities responded to daily temperature maximum and nighttime average temperature. Restricted to pitfall traps, on the other hand, arthropod communities were shaped by the overall temperature metric only when focusing on rare species. Additionally, all communities were strongly affected by season. Our results implies that management interventions establish different temperature heterogeneity within forest patches, which ultimately could drive species community similarity when including all arthropods in the area between forest floor and canopy.