Vertical integration of leaf area index in a Japanese deciduous broad-leaved forest

Abstract

Leaf area index (LAI) is an important quantity in the study of forest ecosystems, but field measurements of LAI often contain errors because of the vertical complexity of the forest canopy. In this study, we established a practical method for field measurement of LAI in the canopy of a deciduous broadleaved forest by accounting for its vertical complexity. First, we produced a semi-empirical model for the vertical integration of leaf dry mass per unit leaf area. We also quantified the litterfall for each tree species. These data enabled us to estimate the LAI of each species in autumn. By periodic in situ monitoring of some fixed sample shoots throughout the growing season, we were able to estimate the seasonality of leaf area (as a proportion of the annual maximum value at each point in time) of each species. By using this seasonality to extrapolate LAI values as a proportion of the LAI data in the leaf-fall season, we were able to estimate LAI throughout the year. We applied this method in a cool-temperate deciduous forest in central Japan (Takayama) in 2006 and validated conventional methods of LAI measurement: the plant canopy analyzer (LAI-2000) and the Tracing Radiation and Architecture of Canopies (TRAC) approach. LAI estimated by TRAC was in good agreement with our results, but LAI estimated using the LAI-2000 was only half the value estimated using our method. The use of basal area data as a proxy for species-specific leaf areas may save labor and time. Our method will be useful for studying the dynamics and interactions of multiple species because it can estimate LAI and its seasonal changes for each species.