Estimation of Leaf Area Index using Photosynthetically Active Radiation Measurements from Flux Tower Networks Across North America

Abstract

Leaf Area Index (LAI) is a critical biophysical variable that influences the exchange of carbon, water, and energy between vegetation and the atmosphere. However, continuous and automated LAI measurements remain limited. To address this gap, a Photosynthetically Active Radiation (PAR)-based approach derived by Lang (1987) and later refined by Gonsamo et al. (2018) was employed to estimate daily LAI from half-hourly PAR data collected at flux towers across Canada and USA. These data, spanning four to 23 years, cover various vegetation types, including Deciduous Broadleaf Forests (DBF), Evergreen Needleleaf Forests (ENF), and Mixed Forest (MF) sites. This method leverages the attenuation of PAR between sensors positioned above and below the canopy, which mimics the probability of sunlight penetrating the canopy, enabling direct estimation of LAI using ceptometry-based techniques. Seasonal trends derived from the resulting time-series graphs closely tracked expected phenological phases, showing consistent patterns across winter, spring, summer, and autumn. Validation of the estimated LAI against ground-based measurements from seven sites yielded strong positive correlations, with an R² of 0.77 and a slope of 0.62. Comparisons with MODIS LAI, aligned within ± two weeks of ground measurements, also revealed robust correlations. However, MODIS tended to under/overestimate LAI in contrast to the ground and Lang-Gonsamo estimates. Data inconsistencies, particularly gaps in the time series, posed challenges and highlighted the need for gap-filling techniques in future work to ensure completeness and identification of detailed trends. Furthermore, accounting for factors like scattering will be crucial to refining the accuracy of LAI estimates. This study emphasizes the potential of continuous PAR-based LAI monitoring to enhance the understanding of phenological transitions and environmental drivers and improve satellite product validation.