Abstract
Leaf area index (LAI) is an important parameter used to characterize the canopy structure of forest ecosystems. Disturbances such as thinning change the canopy structure and influence various ecosystem functions. A larch plantation in Hokkaido, Japan was thinned during the winter of 2003-2004, decreasing its basal area density by 23% from 29.7 to 22.8 m2 ha-1. To evaluate the LAI change caused by the thinning, we measured LAI in 2003 and 2004 using a direct method with litter collection (LC) and two indirect methods using a plant canopy analyzer (PCA) and a ground-based laser scanner (LS), although a September 2004 typhoon disaster halted LC measurements. In 2003, the annual maximum LAI measured via the LC (LAILC) was 5.6 m2 m-2, of which the LAILC of broadleaf trees accounted for about half. Using rings 1-5 of five sky sectors, the annual maximum LAI indicated by the PCA (LAIPCA) was 3.4 m2 m-2, demonstrating that the PCA underestimated LAI. However, using rings 1-4 and 1-3, LAIPCA increased to 4.4 and 4.8 m2 m-2, respectively. This result shows that eliminating the outer rings of the PCA improves LAI underestimation because some scattering light reflected by leaves engenders underestimation of LAIPCA when the zenith angle is large. The maximum LAI by the LS (LAILS) was equal to LAILC. However, the LS greatly overestimated LAI in the leaf-fall and leafless seasons. The laser beam is thought to have increased in diameter with distance and consequently underestimated the gap fraction.
Maximum LAIPCA and LAILS after thinning were each half of that before thinning. The rates of decrease were greater than the basal area density, which is linearly related with LAI, because canopy inhomogeneity is expected to cause LAI underestimation after thinning. LAI probably decreased to about 4.6 m2 m-2, assuming that the ratio of LAI to basal area remained constant despite thinning.
