Plot-level rapid screening for photosynthetic parameters using proximal hyperspectral imaging

Authors: MEACHAM-HENSOLD, KATHERINE - University Of Illinois; FU, PENG - University Of Illinois; WU, JIN - University Of Hong Kong; SERBIN, SHAWN - Brookhaven National Laboratory; MONTES, CHRISTOPHER - University Of Illinois; Ainsworth, Elizabeth - Lisa; GUAN, KAIYU - University Of Illinois; DRACUP, EVAN - Oak Ridge National Laboratory; PEDERSON, TAYLOR - University Of Illinois; DRIEVER, STEVEN - Wageningen University; Bernacchi, Carl

Submitted to: Journal of Experimental Botany
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/20/2020
Publication Date: 2/24/2020
Citation: Meacham-Hensold, K., Fu, P., Wu, J., Serbin, S., Montes, C., Ainsworth, E.A., Guan, K., Dracup, E., Pederson, T., Driever, S., Bernacchi, C.J. 2020. Plot-level rapid screening for photosynthetic parameters using proximal hyperspectral imaging. Journal of Experimental Botany. 71(7):2312-2328. https://doi.org/10.1093/jxb/eraa068.
DOI: https://doi.org/10.1093/jxb/eraa068

Interpretive Summary: The process plants use to grow, photosynthesis, is currently measured using leaf-level time laborious and/or destructive methods, which slows research and breeding efforts to identify crops with higher photosynthetic capacities. We present a tool that predicts photosynthetic capacity of an entire plot in less than one minute, by measuring light reflected back from the crop canopy with hyperspectral cameras, which ‘see’ far more colors than a traditional camera. Using field grown tobacco with genetically altered photosynthetic pathways over two growing seasons (2017 and 2018), we built an automated analysis pipeline to isolate reflected light from sunlit leaves in each plot. We use tobacco because it is a model crop species, allowing for easy experimentation. We used a “machine learning” regression technique (PLSR) to predict photosynthetic capacity and other important plant traits like carbon, nitrogen and chlorophyll content, from reflectance of the sunlit leaves. The analysis pipeline and methods can be used in any cropping system with modified species specific PLSR analysis to offer a high throughput field phenotyping screening for germplasm with improved photosynthetic performance.

Technical Abstract: Photosynthesis is currently measured using leaf-level time laborious and/or destructive methods, which slows research and breeding efforts to identify crop germplasm with higher photosynthetic capacities. We present a plot level screening tool for photosynthetic parameters and pigment contents that utilizes hyperspectral reflectance from sunlit leaf pixels collected from a plot in less than one minute. Using field grown Nicotiana tabacum with genetically altered photosynthetic pathways over two growing seasons (2017 and 2018), we built predictive models for a wide range of photosynthetic parameters and pigment traits. Using partial least squares regression (PLSR) analysis of plot-level sunlit vegetative reflectance pixels from a single VNIR (400-900nm) hyperspectral camera, we predict maximum carboxylation rate of Rubisco (Vc,max , R2 =0.79) maximum electron transport rate in given conditions (J1800 , R2 = 0.59), maximal light saturated photosynthesis (Pmax , R2 = 0.54), chlorophyll content (Chl, R2 = 0.87), the ratio of chlorophyll a to b (Chl a:b, R2 = 0.63), carbon content (C, R2 = 0.47) and nitrogen content (N, R2 = 0.49). Model predictions did not improve when using two cameras spanning 400-1800nm suggesting a robust, widely applicable and more ‘cost-effective’ pipeline requiring only a single VNIR camera. The analysis pipeline and methods can be used in any cropping system with modified species specific PLSR analysis to offer a high throughput field phenotyping screening for germplasm with improved photosynthetic performance.