Count LDA LR NB CART KNN SVM RFC

Sáenz Oviedo, Mayra A., Kuhn, William R., Rondon Sepulveda, Martin A., Abbott, John, Ware, Jessica L. & Sanchez-Herrera, Melissa, 2022, Are wing contours good classifiers for automatic identification in Odonata? A view from the Targeted Odonata Wing Digitization (TOWD) project, International Journal of Odonatology 25, pp. 96-106 : 103-105

Genus Count LDA LR NB CART KNN SVM RFC Acanthagrion 16 x x x x x x x Aeshna 131 x x x x x x x Amphiagrion 1 x

Amphipteryx 1 x

Anax 39 x x x x x x x Anisagrion 2 x x x x x x x Aphylla 33 x x x x x x x Archilestes 20 x x x x x x x Argia 85 x x x x x x x Arigomphus 81 x x x x x x x Basiaeschna 27 x x x x x x x Boyeria 11 x x x x x x x Brachymesia 22 x x x x x x x Brechmorhoga 4 x x x x x x x Calopteryx 149 x x x x x x x Cannaphila 4 x x

Castoraeschna 1 x

Celithemis 149 x x x x x x x Cordulegaster 90 x x x x x x x Cordulia 8 x x x x x x x Coryphaeschna 25 x x x x x x x Crocothemis 2 x x

Cyanallagma 3 x x x x x x x Diastatops 13 x x x x x x x Didymops 6 x x x x x x x Dorocordulia 17 x x x x x x x Drepanoneura 1 x x x x x x x Dromogomphus 45 x x x x x x x Dythemis 14 x x x x x x x Enallagma 264 x x x x x x x Epiaeschna 22 x x x x x x x Epipleoneura 2 x

Epitheca 130 x x x x x x x Erpetogomphus 57 x x x x x x x Erythemis 174 x x x x x x x Erythrodiplax 107 x x x x x x x Euthore 1 x x x x x x x Fluminagrion 1 x x x x x x x Gomphaeschna 17 x x x x x x x Gomphurus 160 x x x x x x x Gynacantha 29 x x x x x x x Hagenius 17 x x x x x x x Helocordulia 13 x x x x x x x Hesperagrion 15 x x x x x x x Hetaerina 75 x x x x x x x Heteragrion 2

Hylogomphus 62 x x x x x x x Idiataphe 1 x x x x x x x Iridictyon 3 x x x x x x x Ischnura 83 x x x x x x x Ladona 55 x x x x x x x Lanthus 21 x x x x x x x Leptobasis 2 x

Lestes 187 x x

are an array of flight behaviors (fliers, gliders and perchers; Corbett & May, 2008) and these flight styles can be reflected in the similarities found in wing contours within both our observed confusion groups. For example, in migratory species of libellulids’ hindwings can show convergence towards a wing planform that favors the gliding flight as an energy saving strategy (SuarezTovar & Sarmiento; 2016). For zygopterans, their flight is more passive, and their ability to disperse might be associated with slow flight or overflight ( Bomphrey et al., 2016), which would explain any similarities in wing contours for coenagrionids and lestids. Comparisons of the damping ratios and natural frequencies of two dragonfly and two damselfly species, shows that for the anisopterans damping properties between fore- and hindwings were significantly different, while in zygopterans there were no or very weak differences in the damping ratios between both wings, suggesting that the structural design and wing shape can influence the aerodynamics of their flight behaviors (Lietz et al; 2021). In addition, functional morphology traits of the wings, like types of joints of the wing veins, spines and presence of resilin, a protein that gives certain flexibility to the wings of insects can be evaluated in this groups, like previously done by Appel and Gorb (2014) to understand the wing contour similarities in these taxa.

Overall, our results suggest that the wing contours by themselves can discriminate with a moderate accuracy and precision, in comparison with other wing attributes obtained using high resolution images. In addition, we tested multiple classifying algorithms for the contours, where LDA had the best performance.