| GLIBERTY-DSAIL耦合模型反演南方混交林植被LAI |
| |
| 引用本文： | 郭云开,刘建琴,郭燕青,曹骁,谢琼.GLIBERTY-DSAIL耦合模型反演南方混交林植被LAI[J].测绘通报,2020,0(11):39. |
| |
| 作者姓名： | 郭云开 刘建琴 郭燕青 曹骁 谢琼 |
| |
| 作者单位： | 1. 长沙理工大学交通运输工程学院, 湖南 长沙 410076;2. 长沙理工大学测绘遥感应用技术 研究所, 湖南 长沙 410076;3. 湖南工程职业技术学院测绘地理学院, 湖南 长沙 410151;4. 长沙理工大学公路地质灾变预警空间信息技术湖南省工程实验室, 湖南 长沙 410114 |
| |
| 基金项目： | 国家自然科学基金（41471421；41671498）；长沙理工大学公路地质灾变预警空间信息技术湖南省工程实验室开放基金（kfj190603） |
| |
| 摘 要： | 针对南方丘陵地区针叶-阔叶混交林植被叶面积指数（leaf area index，LAI）反演精度低且研究较少的问题，本文提出了一种GLIBERTY-DSAIL耦合模型组合多元线性回归反演LAI的方法。本研究以GLIBERTY-DSAIL模型模拟光谱和植被实测高光谱为数据源，通过相关性分析，选取与LAI相关性高的植被指数作为反演因子，构建多元线性回归模型定量反演植被LAI并进行精度评定。结果表明：与LAI显著相关的RVI、DVI、GNDVI、MSAVI这4种植被指数作为反演因子，结合本文提出的组合模型反演LAI，模型预测决定系数R2为0.708 6，均方根误差RMSE为0.302 1，精度整体较高。该组合方法可较好地用于反演针叶-阔叶混交林植被LAI，为南方地区混交林LAI的研究提供新思路。
|
| 关 键 词： | GLIBERTY-DSAIL模型 多元线性回归 混交林 叶面积指数 定量反演 |
| 收稿时间： | 2020-06-04 |
| 修稿时间： | 2020-07-07 |
GLIBERTY-DSAIL coupled model inversion of vegetation LAI in southern mixed forest |
| |
| GUO Yunkai,LIU Jianqin,GUO Yanqing,CAO Xiao,XIE Qiong.GLIBERTY-DSAIL coupled model inversion of vegetation LAI in southern mixed forest[J].Bulletin of Surveying and Mapping,2020,0(11):39. |
| |
| Authors: | GUO Yunkai LIU Jianqin GUO Yanqing CAO Xiao XIE Qiong |
| |
| Abstract: | In order to solve the problem of low leaf area index (LAI) inversion accuracy and little research in the coniferous-broadleaved mixed forest in the southern hilly areas, this paper proposes a GLIBERTY-DSAIL coupling model combined with multiple linear regression inversion LAI method. In this study, the GLIBERTY-DSAIL model selects simulated spectrum and vegetation measured hyperspectral as data sources. Through the correlation analysis, the vegetation index with high correlation about LAI is selected as the inversion factor, a multiple linear regression model is constructed to quantitatively invert vegetation LAI. The paper evaluates the accuracy. The results show that: RVI, DVI, GNDVI, and MSAVI vegetation indexes that are significantly related to LAI are used as inversion factors, combined with the model proposed in this paper to invert LAI, the model prediction coefficient R2 is 0.708 6, and the root mean square error RMSE is 0.302 1. The accuracy is higher overall. This combined method can be used to invert vegetation LAI of coniferous-broadleaved mixed forests, and provide new ideas for the study of mixed forest LAI in southern areas. |
| |
| Keywords: | GLIBERTY-DSAIL model multiple linear regression mixed forest LAI inversion |
|
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
|
