Validation methods of LAI products on the basis of scaling effect  

作　　者：LIAO Yanran GAI Yingying YAO Yanjuan FAN Wenjie XU Xiru YAN Binyan 

机构地区：[1]Institute of Remote Sensing and GIS , Peking University, Beijing 100871, China [2]Beijing Key Laboratory of Spatial Information Integration & Its Applications, Beijing 100871, China [3]Institute of Oceanographic Instrumentation, Shandong Academy of Sciences, Qingdao 266001, China [4]Satellite Environment Center, Ministry of Environmental Protection, Beijing 100094, China [5]Jackson School of Geosciences, the University of Texas at Austin, Austin 78712, USA

出　　处：《遥感学报》2015年第1期134-142,共9页NATIONAL REMOTE SENSING BULLETIN

基　　金：National High Technology Research and Development Program of China(863 Program)(No.2009AA122103,2012AA12A304);National Natural Science Foundation of China(No.91025006,91325105,41271346);National Basic Research Program of China(973 Program)(No.2013CB733402)

摘　　要：Validation is one of the most important processes used to evaluate whether remotely sensed products can accurately reflect land surface configuration. Leaf Area Index( LAI) is a key parameter that represents vegetation canopy structures and growth conditions. Accurate evaluation of LAI products is the basis for applying them to land surface models. In this study,validation methods of coarse resolution MODIS and GLASS LAI products for heterogeneous pixels are established on the basis of the scaling effect and the scaling transformation. Considering spatial heterogeneity and growth difference,we transformed LAI from field measurements into a 1 km resolution scale with the aid of middle resolution images. We used average LAI and apparent LAI separately to validate the algorithms and products of MODIS and GLASS LAI. Two study areas,Hebi City and the Yingke Oasis,were selected for validation. Both MODIS and GLASS LAI products underestimate the true LAI in crop area. However,this result cannot be completely attributed to their algorithms. Instead,the primary reason is the heterogeneity and nonuniformity of the coarse pixels.Underestimation is evident in the Yingke Oasis,where heterogeneity is significant. Given that GLASS LAI product is the fusion of multiple LAI products,the mean value of this product is closer to the real situation,but the dynamic range is narrower than that of MODIS LAI product.Validation is one of the most important processes used to evaluate whether remotely sensed products can accurately reflect land surface configuration. Leaf Area Index（ LAI） is a key parameter that represents vegetation canopy structures and growth conditions. Accurate evaluation of LAI products is the basis for applying them to land surface models. In this study,validation methods of coarse resolution MODIS and GLASS LAI products for heterogeneous pixels are established on the basis of the scaling effect and the scaling transformation. Considering spatial heterogeneity and growth difference,we transformed LAI from field measurements into a 1 km resolution scale with the aid of middle resolution images. We used average LAI and apparent LAI separately to validate the algorithms and products of MODIS and GLASS LAI. Two study areas,Hebi City and the Yingke Oasis,were selected for validation. Both MODIS and GLASS LAI products underestimate the true LAI in crop area. However,this result cannot be completely attributed to their algorithms. Instead,the primary reason is the heterogeneity and nonuniformity of the coarse pixels.Underestimation is evident in the Yingke Oasis,where heterogeneity is significant. Given that GLASS LAI product is the fusion of multiple LAI products,the mean value of this product is closer to the real situation,but the dynamic range is narrower than that of MODIS LAI product.

关 键 词：VALIDATION leaf area index scaling effect scaling transformation 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]