ObjectiveExploring the spatial heterogeneity of leaf area index (LAI) and controlling factors is to provide a solid foundation for obtaining the distribution of LAI in local or region scales.

MethodBased on a 30 ha (500m×600m) mixed broadleaved-Korean pine (Pinus koraiensis) forest at Fenglin National Nature Reserve in Xiaoxing'an Mountains of northeastern China, which was divided into 750 subplots by 20m×20m, all woody species with a diameter at breast height≥1cm were identified and surveyed. We first obtained LAI values of six major species (e.g., P. koraiensis, Abies nephrolepis, Tilia amurensis, Betula costata, Ulmus laciniata and Acer mono) based on empirical models between LAI and basal area for each subplot. We then analyzed the spatial heterogeneity of LAI for major species and the forest stand by geo-statistic methods (semivariogram and Kriging interpolation), and the correlations between LAI and topographic variables (e.g., elevation, slope degree, slope aspect and curvature) were also explored.

ResultThe variation coefficient of LAI for six major species was all larger than 10%, and can be considered relatively moderate or strong, and degrees of variation were U. laciniata > B. costata > T. amurensis > P. koraiensis > A. mono > A. nephrolepis. The ratio value of nugget (C₀) by sill (C₀+C) was 0.50 for P. koraiensis, showing moderate spatial autocorrelation; in contrast, the ratio values for other five species were all lower than 0.25, showing strong spatial autocorrelation. The range (A) values ranged from 24m (T. amurensis) to 126m (P. koraiensis) of LAI for six major species.The clearer anisotropic structures were exhibited by LAI values for P. koraiensis, U. laciniata and A. mono, in contrast to other species; additionally, the spatial heterogeneity of LAI in east-west (0°) direction was clearly larger than that in south-north (90°) within 240m for P. koraiensis, and showed opposite results when larger 240m. The LAI was significantly positively correlated with all elevation, slope degree, slope aspect and curvature for P. koraiensis (P < 0.01), and different correlations between four topographic variables and LAI values were showed for other five species.

ConclusionSpatial heterogeneity of LAI is correlated with not only study scale but also direction. The influence of topographic variables on LAI distribution varies with species, but the four topographic variables all significantly affected LAI distributions for a forest stand.