Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Rangeland Resources & Systems Research » Research » Publications at this Location » Publication #359466
Title: Hyperspectral analysis of leaf pigments and nutritional elements in tallgrass prairie vegetation

Author
item LING, BOHUA - Kansas State University
item GOODIN, DOUGLAS - Kansas State University
item Raynor, Edward
item JOEM, ANTHONY - Kansas State University

Submitted to: Frontiers in Plant Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/28/2019
Publication Date: 3/20/2019
Citation: Ling, B., Goodin, D., Raynor, E.J., Joem, A. 2019. Hyperspectral analysis of leaf pigments and nutritional elements in tallgrass prairie vegetation. Frontiers in Plant Science. 10(142):1-13. https://doi.org/10.3389/fpls.2019.00142.
DOI: https://doi.org/10.3389/fpls.2019.00142

Interpretive Summary: Understanding the accuracy of hyperspectral measurement of plant attributes is critical to improving efficiency of data collection for grassland science at large spatial scales. Further, how fine-scale leaf-level characteristics such as nutritional content can be extrapolated to coarser spatial scales are unexplored aspects of grassland science that may inform grassland managers; most work on this subject has occurred in forest systems. This study examines foliar chlorophylls, carotenoids and nutritional elements across multiple tallgrass prairie functional groups at the leaf level using hyperspectral analysis. A method of spectral standardization was developed using a form of the normalized difference, which proved feasible to reduce the interference from background effects in the leaf reflectance measurements. Chlorophylls and carotenoids were retrieved through inverting the commonly employed physical model PROSPECT 5 and compared to empirical measurements of plant quality. We discuss how results, which showed that the retrieval of leaf biochemistry through hyperspectral analysis can be accurate and robust across different tallgrass prairie functional groups, may be incorporated into studies of large herbivore ecology that aim to understand the separation of plant quality from quantity. We also discuss how such an understanding can improve our knowledge of the roloe of nutritional ecology in driving ecological interactions in mesic grasslands.

Technical Abstract: Understanding the spatial distribution of forage quality is important to address critical research questions in grassland science. Due to its efficiency and accuracy, there has been a widespread interest in mapping the canopy vegetation characteristics using remote sensing methods. In this study, foliar chlorophylls, carotenoids and nutritional elements across multiple tallgrass prairie functional groups were quantified at the leaf level using hyperspectral analysis in the region of 470 – 800 nm, which was expected to be a precursor to further remote sensing of canopy vegetation quality. A method of spectral standardization was developed using a form of the normalized difference, which proved feasible to reduce the interference from background effects in the leaf reflectance measurements. Chlorophylls and carotenoids were retrieved through inverting the physical model PROSPECT 5. The foliar nutritional elements were modeled empirically. Partial least squares regression was used to build the linkages between the high-dimensional spectral predictor variables and the foliar biochemical contents. Results showed that the retrieval of leaf biochemistry through hyperspectral analysis can be accurate and robust across different tallgrass prairie functional groups. In addition, correlations were found between the leaf pigments and nutritional elements. Results provided insight into the use of pigment-related vegetation indices as the proxy of plant nutrition quality.