摘要: |
以水稻叶片光谱和光合入射有效辐射(PARin)为数据源,分别使用8种叶绿素相关植被指数和其与PARin的乘积对珞优9348和丰两优4号品种水稻叶片在3种氮肥浓度下的净光合速率进行反演,建立不同氮肥水平下,品种无关的水稻叶片净光合速率估算模型,并验证精度。结果表明:1)仅基于叶绿素相关植被指数反演叶片净光合速率是不可行的,模型R2均<0.48;2)考虑区分品种与氮浓度,使用叶绿素相关植被指数与PARin的乘积来反演叶片净光合速率可以显著提高模型精度,植被指数CI(Chlorophyll Index,CI)反演效果最好,R2均>0.75,RMSE不超过1.85 μmol/(m2·s);3)区分高、低氮水平进行反演,基于CI的模型R2分别为0.80和0.89,RMSE分别为1.25和1.37 μmol/(m2·s),可满足反演需求。因此,基于叶绿素植被指数CI的模型可适用于不同品种的水稻叶片净光合速率统一反演。 |
关键词: 高光谱反射率 叶绿素相关植被指数 入射光合有效辐射 净光合速率 |
DOI:10.11841/j.issn.1007-4333.2020.01.07 |
投稿时间:2019-01-16 |
基金项目:中央高校基本科研基金(2042017kf0236);国家高技术研究发展计划(863 计划)项目(2013AA102401) |
|
Remote estimation of rice leaf net photosynthetic rate based on hyperspectral reflectance |
LIU Cong,PENG Yi,FANG Shenghui
|
(School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China) |
Abstract: |
Based on the rice leaves hyperspectral reflectance and incident photosynthetically active radiation(PARin)data, 8 vegetation indices and their products with PARin were used to construct the variety-irrelevant leaf net photosynthetic rate retrieval model, respectively. Two rice varieties, Luoyou 9348 and Fengliangyou 4, were taken as study materials and experiments were conducted under 3 different nitrogen content conditions. The results showed that: 1)Leaf net photosynthetic rate estimation based on chlorophyll-related vegetation indices only was infeasible with R2 below 0. 48; 2)Considering both variety and nitrogen, Chlorophyll Index(CI)multiplied by PARin performed best among the 8 indices with R2 above 0. 75, RMSE no more than 1. 85 μmol/(m2·s)in training and validation set; 3)Considering nitrogen content, the CI models also performed well with R2 of 0. 80 and 0. 89, RMSE of 1. 25 and 1. 37 μmol/(m2·s)in low and high nitrogen content condition in training set, respectively. In conclusion, the model met remote-based retrieval accuracy requirements and could be used for unified retrieval of leaf net photosynthetic rate of rice. |
Key words: hyperspectral reflectance chlorophyll-related vegetation indices incident photosynthetically active radiation net photosynthetic rate |