Assessment of leaf chlorophyll content models for winter wheat using Landsat-8 multispectral remote sensing data

Authors: ZHOU, XIANFENG - Hangzhou Dianzi University; ZHANG, JINGCHENG - Hangzhou Dianzi University; CHEN, DONGMEI - Hangzhou Dianzi University; Huang, Yanbo; KONG, WEIPING - Chinese Academy Of Sciences; YUAN, LIN - Zhejiang University; YE, HUICHUN - Chinese Academy Of Sciences; HUANG, WENJIANG - Chinese Academy Of Sciences

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/3/2020
Publication Date: 8/11/2020
Citation: Zhou, X., Zhang, J., Chen, D., Huang, Y., Kong, W., Yuan, L., Ye, H., Huang, W. 2020. Assessment of leaf chlorophyll content models for winter wheat using Landsat-8 multispectral remote sensing data. Remote Sensing. 12(2574):1-18. https://doi.org/http://dx.doi.org/10.3390/rs12162574.
DOI: https://doi.org/10.3390/rs12162574

Interpretive Summary: Crop leaf chlorophyll content is a critical index to characterize crop growth condition. Scientists of Hangzhou Dianzi University, USDA-ARS Crop Production Systems Research Unit at Stoneville, MS, and Chinese Academy of Sciences have developed methods of estimating crop leaf chlorophyll content from NASA's Landsat-8 multispectral imagery over the cropping areas in the suburb of Beijing, China. The results indicated that the Landsat-8 imagery are suitable for crop leaf chlorophyll estimation and this study provides the basis for crop leaf chlorophyll estimation from other similar multispectral imagery datasets.

Technical Abstract: Leaf chlorophyll content (LCC) is a critical index to characterize crop growth condition, photosynthetic capacity, and physiological status. Its dynamic change characteristics are of great significance for monitoring crop growth conditions and understanding the process of material and energy exchange between crops and the environment. Extensive researches have been focused on LCC retrieval with hyperspectral data onboard various sensor platforms. Nevertheless, limited attention has been paid to LCC inversion from multispectral data, such as the data from Landsat-8, and the potentials and capabilities of the data on crop LCC estimation has not been fully explored. The present study made use of Landsat-8 Operational Land Imager (OLI) imagery and the corresponding field experimental data to evaluate its capabilities and potentials for LCC modeling using four different retrieval methods, including vegetation indices (VIs), machine learning regression algorithms (MLRAs), lookup-table (LUT) based inversion, and hybrid regression approaches. Results showed that modified triangular vegetation index (MTVI2) exhibited the best estimate accuracy for LCC retrieval with a root mean square error (RMSE) of 5.99 µg/cm2 and a relative RMSE (RRMSE) of 10.49%. Several other vegetation indices that established from Red and Near-InfraRed (NIR) bands also exhibited good accuracy. Models established from gaussian process regression (GPR) achieved the highest accuracy for LCC retrieval (RMSE = 5.50 µg/cm2, RRMSE = 9.62%) compared with other MLRAs. Moreover, Red and NIR bands outweighed other bands for GPR modelling. LUT-based inversion methods with “K(x) = -log (x) + x” cost function that belongs to “minimum contrast estimates” family showed the best estimating results (RMSE = 8.08 µg/cm2, RRMSE = 14.14%), and the adding of multiple solution regularization strategies effectively improved the inversion accuracy. For hybrid regression methods, The using of active learning (AL) techniques together with GPR for LCC modelling significantly increased the estimating accuracy, and the combination of entropy query by bagging (EQB) AL and GPR performed with the best accuracy for LCC estimation (RMSE = 12.43 µg/cm2, RRMSE = 21.77%). Overall, our study suggest that Landsat-8 OLI data are suitable for crop LCC retrieval and could provide basis for LCC estimation with similar multispectral datasets.