Narrow but robust advantages in two-big-leaf light use efficiency models over big-leaf light use efficiency models at ecosystem level

This study aims to (1) investigate whether two-big-leaf light use efficiency (LUE) models (TL) outperform big-leaf LUE (BL) by incorporating different gross primary productivity (GPP) responses in sunlit and shaded leaves; (2) explore the robustness of using leaf area index (LAI), clumping (Ω) spherical angle distribution partition canopies into leaves across canopy architectures; (3) identify optimal response forms models. To exclude influences drivers GPP other than radiation, we collected various formulations functions temperature, vapor pressure deficit, CO2, soil water supply, intensity cloudiness construct 5600 BLs 1120 TLs. These were evaluated at 196 globally-distributed eddy covariance sites from FLUXNET observational network Nash-Sutcliffe model (NSE), root mean squared error Bayesian information criterion. Across all sites, best (BL*; NSE=0.82) was statistically equal TL (TL*; NSE=0.84). However, daily dynamics under hot dry conditions described TL* 17% highlighting local importance separating leaves. approaches represent effective LAI, approach relies on normalized difference vegetation with a or flexible rather satellite LAI Ω. We also observed similar performance between non-rectangular hyperbola reciprocal Models degrade when maximum is not differentiated leaves, but saturation levels are same. Despite functional differences, five TLs agree larger contribution total GPP, resulting higher LUE. Overall, these results suggest marginal robust selection compared BL.