From the publication just above (Introduction):
Wings designed to maximize the power-harvesting factor have a high aspect ratio (Echeverri et al., 2020a; Bauer et al., 2018; Fasel et al., 2017; Trevisi et al., 2021, among others) and use unconventional airfoils, which are designed to maximize the metric Cl³ / Cd² (Bauer et al., 2018; De Fezza and Barber, 2022; Porta Ko et al., 2023; Rangriz and Kheiri, 2025, among others).
Just after:
Recently, Trevisi (2024) introduced a new design methodology for windplanes, performing the aerodynamic design per a given wingspan. This is achieved by maximizing a new power coefficient², defined as the ratio between the generated power and the kinetic energy flux (or wind power density) passing through a disk with a radius equal to the wingspan (Trevisi et al., 2023a). This reference area does not correspond to an actual flow cross-section but is just used as a reference area to normalize the generated power. Wings designed to maximize this power coefficient have low aspect ratios (approximately between 4 and 7) and use conventional efficient airfoils (i.e., airfoils with high Cl / Cd).
See also a topic with a quote from @floba 's publication from Drag power kite with very high lift coefficient :
@fTrevisi and @floba, and others, could develop their respective points of view. Among other things: high (Cl / Cd)², or high Cl (Cl/Cd)², both being in Loyd’s formula?
(Other concerned topic: High lift coefficient and biplane kite)