Abstract: It was studied that leaf area index (LAIs) were respectively estimated with sampling with a cubic frame (direct method) and using a plant canopy analyzer (indirect method) in more than 40 adult apple orchards (including unthinned, thinned and partially-cut-branch orchards) selected randomly in Shandong
Province. The results showed that LAIs estimated with direct and indirect methods in all orchards tested were 3.0 ± 0.2 and 4.3 ± 0.2，respectively. Compared to LAI with the direct method (LAIdi), average under estimation of LAIs with indirect method (LAIin) was 30.2%, in which the underestimation of LAIs were 32.0%, 35.5% and 25.0%，respectively in unthinned, thinned and partially-cut-branch orchards. In addition, the larger LAI was, the more serious underestimation was in various types of orchards. LAIs estimated with direct and indirect methods were strongly and linearly interrelated to each other (r = 0.91, 0.89 and 0.92, respectively, in three types of orchards, P＜0.00001). When omitting the reading of the external ring of the plant canopy analyzer and being recalculated, the linear correlation between LAIdi and LAIin was more significant (r = 0.93, 0.93 and 0.94, respectively, in three types of orchards, P＜0.00001), and the difference became less (less than 13%). When LAI of apple orchards is estimated by a plant canopy analyzer, further
correction of the data can improve its performance and produce reliable estimation through being modified by direct reference measurement. At this study, the modified coefficients of plant canopy analyzer were 1.3004, 1.2007 and 1.1762, respectively, in three types of orchards, that is to say, real LAIs in three types of orchards were about 1.3, 1.2 and 1.2 times respectively, more than estimated values (modified) of plant canopy analyzer.

Key words: apple, orchard, leaf area index, plant canopy analyzer, sampling with a cubic frame