Data from: Flight reconstruction of two European enantiornithines (Aves, Pygostylia) and the achievement of bounding flight in Early Cretaceous birds

Birds today follow different aerial strategies to deal with the high costs of flapping flight. Intermittent flight, through either flap-gliding or bounding, is commonly used among small birds as a strategy to optimize aerial efficiency. The broad morphological disparity of the different lineages of Mesozoic birds suggests that a range of aerial strategies could have evolved early in avian evolution. Based on biomechanics and aerodynamic theory, this study reconstructs the flying properties of two small enantiornithines from the Lower Cretaceous fossil site of Las Hoyas (Spain), Concornis lacustris and Eoalulavis hoyasi. Our results show that the short length of their wings in relation to their body masses were suitable for flying through strict flapping and intermittent bounds, but not through facultative glides. Aerodynamic models indicate that the power margins of these birds were sufficient to sustain bounding flight. Our results thus suggest that C. lacustris and E. hoyasi would have increased aerial efficiency through bounding flight, just as many small passerines and woodpeckers do today. Intermittent bounding appears to have evolved early in the evolutionary history of birds, at least 126 million years ago.Appendix S1. Concornis lacustris and Eoalulavis hoyasi measurements.Measurements of the fossil specimens from Las Hoyas included in the present study. Following Serrano et al. (2017).Appendix S1.docAppendix S2. Modern birds' dataset.Dataset of modern Neoaves used in the present study. Main flight modes were assigned following Viscor and Fuster (1987), Del Hoyo et al. (1992; vol 1-16) and Bruderer et al. (2010). Specimen code: Only number – Burke Museum of the University of Washington (Seattle, USA); UBCBM – Beaty Museum of the University of Bristish Columbia (Vancouver, Canada); BR – obtained from the dataset of Bruderer et al. (2010); VF - obtained from the dataset of Viscor and Fuster (1987); PN – obtained from the preset database of software Flight 1.24.Appendix S2.csvAppendix S3. Summary of Canonical Variates Analysis.Summary of Canonical Variates Analysis: 1. Eigenvalues and % variance explained by the canonical functions; 2. Lambda Wilk's test for mean groups' differences; 3. Coefficients of the canonical functions; 4. Reclassification of modern birds' dataset after leave-one-out validation. Group codes: continuous flapping (1); bounding (2); facultative gliding (3).Appendix S3.docAppendix S4. Difference in power margin between two modern birds with similar body mass.Flight reconstruction of two modern flying birds with similar BM to show difference in the power margin. Power curves indicating the relationship between Pmec and Vt were modeled for a continuous flapping strategy (i.e., q=1) with CDb calculated according the two approaches used in this study: A, CDb = 0.1; and B, CDb = 0.01 SL / Sb (see Methods). Pav is closely similar as both birds weighted almost the same, and thus the mean value of Pav was drawn (pointed horizontal line). The bar-tailed godwit Limosa lapponica (BM = 367g; CDb = 0.136) shows sufficient power margin to sustain flapping flight in the full-range of Vt, using the aerobic activity from the flight muscles. In contrast, the grey partridge Perdix perdix (BM = 372g; CDb = 0.125) cannot sustain a prolonged flapping flight because the Pav does not attain Pmec. Instead, partridges and other Galliformes fly using short bursts supported by anaerobic activity of the fiber muscles (see Bishop and Butler 2015).Appendix S4.do