The Leaf Angle Distribution （LAD） and Leaf Area Index （LAI） are important descriptors of the vegetation canopy. But direct field measurement of these variables is very labor-intensive and timeconsuming because of the complexity and variation in canopy architecture within and between species

especially for conifers. Upward-looking fisheye photos have been used to provide LAI estimates

but such estimates are subject to errors caused by unknown LAD and clumping of leaves and branches. In this paper

we present our new method on LAD and LAI estimation. The main approach is on how to introduce our intercrown gap probability model into LAI and LAD inversion algorithm

to express the effect of viewing direction on clumping index in normal gap probability model. Two validation experiments have been done. In one of them the estimated LAD （10-degree grouping） fits the directly measured LAD with a correlation coefficient 0.97. Another result is similar. Although the primary results are encouraging

more validation

especially under natural forest

are still needed to incorporate the method into our tree tomography procedure.