Abstract: [Objective] The objective of this study was mainly to quantify the bias in measuring the seasonal variation of optical LAI due to different error sources and to explore which one is the most important for affecting the accuracy of LAI measurements by different optical methods in planation forests in the Heilongjiang Liangshui National Nature Reserve, in order to provide the theoretical basis for accurately estimating LAI by optical methods. [Methods] This study was conducted in a Korean pine (Pinus koraiensis) plantation and a Dahurian larch (Larix gmelinii) plantation. The seasonal changes of LAI were estimated using direct and optical methods including digital hemispherical photography (DHP) and LAI-2000 plant canopy analyzer. Further, the measurements bias of seasonal changes were analyzed due to error sources (i.e., woody materials, clumping effects beyond shoot level, clumping effects within shoot or automatic exposure setting) with optical methods. [Result] As for Korean pine plantation, the LAI values measured with DHP and LAI-2000 were all lower than that measured with direct LAI measurement method in different periods by 55%~68% and 19%~35%, respectively. As for Dahurian larch plantation, the LAI values measured with DHP and LAI-2000 were also lower than that measured with direct LAI measurement method in most periods, but the LAI from DHP and LAI-2000 were larger than that with direct LAI measurement in late leaf fall period. The bias due to different error sources with the optical methods showed clear seasonal dynamics. For Korean pine plantation, the total bias due to all error sources during all periods approximately explained averagely 84% and 79% of the differences between DHP or LAI-2000 and direct LAImeasurement, respectively. For Dahurian larch plantation, in the case of correcting the error caused by xylem by Photoshop (PS) method, the total bias due to all error sources during all periods approximately explained averagely 80% and 65% of the differences between DHP or LAI-2000 and direct LAI measurement, respectively; In contrast, when correcting the error caused by xylem using a background method, the corresponding explanation ratios were all less than 5%. [Conclusion] For the two coniferous forests, DHP is more suitable for estimating the seasonal changes of LAI than LAI-2000 after correlation, and PS method is more useful in correcting the error caused by xylem in estimating seasonal changes of LAI using the two optical methods than the background method.

Key words: leaf area index (LAI), digital hemispherical photography (DHP), LAI-2000 plant canopy analyzer, woody-to-total area ratio, clumping effects, automatic exposure