归一化植被指数的应用

中国农业科学 2008,41(10):3350-3356 Scientia Agricultura Sinica doi: 10.3864/j.issn.0578-1752.2008.10.057

估算水稻叶面积指数的调节型归一化植被指数

刘占宇

1,2

，黄敬峰

1,3

，王福民

1,2

，王 渊

1,2

（1浙江大学农业遥感与信息技术应用研究所，杭州 310029；2浙江大学环境与资源学院环境修复与生态健康教育部重点实验室，杭州 310029；

3

浙江省农业遥感与信息技术重点研究实验室，杭州 310029）

摘要：【目的】叶面积指数（LAI）是描述植物冠层结构、群落生长分析和陆地生态系统研究的重要参数，提高叶面积指数的估算精度是遥感工作者的重要研究方向之一。【方法】通过不同氮素营养水平的水稻小区实验，利用2004年中稻高光谱反射率数据，模拟中等分辨率成像光谱仪（MODIS）前四个通道，提出包含蓝、绿、红和近红外四个谱段的调节型归一化植被指数ANDVI（adjusted-normalized difference vegetation index）。对ANDVI和归一化植被指数NDVI（normalized difference vegetation index）、增强型植被指数EVI（enhanced vegetation index）、绿波比值指数GRI（green ratio index）、红边比值指数RRI（red-edge ratio index）等5个光谱植被指数与水稻LAI的相关关系进行了分析。利用2004年晚稻试验数据，对与LAI相关关系较好的ANDVI进行验证。【结果】ANDVI指数模型预测效果最好，均方根误差为1.771，估算精度达到63.1%。【结论】说明ANDVI具有进行在大面积范围内监测水稻LAI的能力。

关键词：水稻；叶面积指数；植被指数；中等分辩率成像光谱仪；高光谱反射率

Adjusted-Normalized Difference Vegetation Index for

Estimating Leaf Area Index of Rice

LIU Zhan-yu1,2, HUANG Jing-feng1,3, WANG Fu-min1,2, WANG Yuan1,2

(1Institute of Agricultural Remote Sensing & Information System Application, Zhejiang University, Hangzhou 310029; 2Ministry of Education Key Laboratory of Environmental Remediation and Ecological Health, Zhejiang University, Hangzhou 310029; 3Key

Laboratory of Agricultural Remote Sensing & Information System in Zhejiang Province, Hangzhou 310029)

Abstract:【Objective】Leaf area index (LAI) is an important parameter for describing vegetation canopy structure and community development condition in the terrestrial ecosystem. To improve the estimation accuracy of leaf area index is the main research direction for the remote sensed scholars.【Method】 In this study the mid and late maturing rice at the different nitrogen fertilizer level was planted in 2004 yr, and the hyperspectral canopy reflectance of mid rice was obtained. The front four bands of Moderate Resolution Imaging Spectrometer (MODIS) (i.e. red,near-infrared,blue and green regions) were simulated to develop, and the adjusted-normalized difference vegetation index (ANDVI). A comparison of ANDVI with other VIs, such as Normalized Difference Vegetation Index (NDVI), enhanced Vegetation Index (EVI), Green Ratio Index (GRI) and Red-edge Ratio Index (RRI) for estimation of LAI was made. The efficiency of ANDVI was validated with the hyperspectral canopy reflectance data of late maturing rice in 2004. 【Result】Results show that the estimation precision of exponential equation of ANDVI was 63.1%, the root mean square error was 1.771, 【Conclusion】and ANDVI has the potential of monitoring LAI in an extensive area.

Key words: Rice; Leaf area index (LAI); Vegetation index; MODIS; Hypsespectral canopy reflectance

0 引言

【研究意义】叶面积指数（LAI）是一个重要的

植物学参数和评价指标，是陆地生态系统的一个十分重要的结构参数，它控制着植被的许多生物物理过程，如光合、呼吸、蒸腾、碳循环和降水截获等，在农学、

收稿日期：2006-08-21；接受日期：2007-05-22

基金项目：国家自然科学基金（40571115）和国家高技术发研究展计划项目（2007AA10Z205） 作者简介：刘占宇（1979-），男，河南南阳人，博士，研究方向为高光谱遥感。Tel：0571-86971830；E-mail：zdrsbond@zju.edu.cn