Litter vs. Lens: Evaluating LAI from Litter Traps and Hemispherical Photos Across View Zenith Angles and Leaf Fall Phases

Abstract. Leaf area index (LAI) is a key parameter for modeling ecosystem productivity, climate interactions, and hydrological processes, as well as monitoring vegetation health. Reliable assessment of LAI and its temporal dynamics is therefore essential for understanding forest functioning and seasonal phenology. Among available ground-based approaches, digital hemispherical photography (DHP) offers a practical, reproducible, and spatially flexible means to estimate LAI. Yet, it remains uncertain how robustly DHP can capture temporal LAI dynamics compared to direct reference methods. Here, we evaluate DHP-derived LAI time series with litter trap (LT)-derived LAI in a temperate deciduous broad-leaved forest. First, by comparing DHP-derived LAI estimates with LT-derived LAI across varying view zenith angles (VZA) ranging from 10 to 90°, we investigate how well both methods align. To evaluate whether factors other than the spatial footprint contributed to differences between DHP- and LT-derived LAI, we analyzed how VZA influenced canopy structure metrics (Gap Fraction, Clumping Index, G(θ)). Although these parameters changed with VZA and affected LAI estimation errors in a random forest model, the sampling radius derived from the VZA had the greatest effect on the errors. Using 15 sample locations, we found the highest average correlation (R2 = 0.88) at a VZA of 20°, suggesting that the litter traps capture a relatively narrow spatial footprint, consistent with the model analysis highlighting the dominant role of sampling radius. To overcome uncertainties for individual litter traps attributed to varying site conditions, we applied a generalized linear mixed model. This location-specific calibration using the litter traps showed an significant increase of the correspondence to R2 = 0.97. Our results confirm that, when appropriately calibrated and acquired at an optimal VZA (20°), DHP provides reliable, high-resolution monitoring of seasonal LAI dynamics in deciduous forests, offering a valuable complement to remote-sensing products for ecosystem research and management.