CAJ | 학술논문

以祁连山区排露沟流域为研究区,利用高分辨率的遥感数据获取多种植被指数,并与观测的叶面积指数进行回归分析,发现叶面积指数LAI与归一化植被指数NDVI的相天性最好(R~2=0.3766),且以LAI与NDVI的关系建立的模犁精度最高(RMSE=0.381).通过t检验,证明NDVI模型明显优于其他植被指数模型,因此把它选为最佳模型,模拟整个研究区青海云杉林叶面积指数的空间分布,为小流域分布式生态水文模型提供重要的参数.
이기련산구배로구류역위연구구,이용고분변솔적요감수거획취다충식피지수,병여관측적협면적지수진행회귀분석,발현협면적지수LAI여귀일화식피지수NDVI적상천성최호(R~2=0.3766),차이LAI여NDVI적관계건립적모리정도최고(RMSE=0.381).통과t검험,증명NDVI모형명현우우기타식피지수모형,인차파타선위최가모형,모의정개연구구청해운삼림협면적지수적공간분포,위소류역분포식생태수문모형제공중요적삼수.
Leaf area index values from Picea crassliolia plots were acquired in Pailugou basin of Qilian Mountain. Using a QuickBird multispectral image, the mean values for the vegetation indexes(NDVI, RVI, ARVI, MSAVI, EVI, MCAVI) were also calculated for each plot. Regression analyses of LAI with all vegetation indices revealed the most significant positive relationships(R~2=0.766) between LAI and NDVI. The root mean square errors and the t tests revealed that NDVI model was the most accurate one (RMSE=0.381) and statistically better than the other models. It is thus suggested that NDVI model can be employed as the most valuable tool for monitoring LAI in Picea crassliolia forests and provide an important parameter for the distributed eco-hydrological models on small catchments.