Investigation of the influence of leaf thickness on canopy reflectance and physiological traits in upland and Pima cotton populations

Authors: PAULI, DUKE - Cornell University; White, Jeffrey; ANDRADE-SNACHEZ, PEDRO - University Of Arizona; Conley, Matthew; HUEN, JOHN - University Of Arizona; Thorp, Kelly; French, Andrew; Hunsaker, Douglas - Doug; CARMO-SILVA, ELIZABETE - Lancaster University; WANG, GUANGYAO - Bridgestone Americas Tire Operations; GORE, MICHAEL - Cornell University

Submitted to: Frontiers in Crops Science and Horticulture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/28/2017
Publication Date: 8/17/2017
Publication URL: https://handle.nal.usda.gov/10113/5801856
Citation: Pauli, D., White, J.W., Andrade-Snachez, P., Conley, M.M., Huen, J., Thorp, K.R., French, A.N., Hunsaker, D.J., Carmo-Silva, E.A., Wang, G., Gore, M.A. 2017. Investigation of the influence of leaf thickness on canopy reflectance and physiological traits in upland and Pima cotton populations. Frontiers in Crops Science and Horticulture. 8:1405. doi: 10.3389/fpls.2017.01405.

Interpretive Summary: Research to support plant breeding increasingly seeks ways to accelerate field evaluations of genetically diverse plant materials. This need is driven by the spectacular advances in our ability to characterize genetic variation, which has led to measurement of traits, or “phenotyping”, limiting progress in breeding. Many systems for field-based, high throughput phenotyping quantify and characterize the radiation (from sunlight or artificial illumination) that is reflected from the crop canopy to infer crop structure, function and health. However, it remains difficult to characterize the biophysical attributes of crops from reflectance data alone, and furthermore, to interpret how such properties impact deeper, more mechanistic interpretations. Because of these uncertainties, we assessed relations between the actual thickness of leaves and several canopy-associated traits, including normalized difference vegetation index (NDVI), which is among the mostly widely used parameters to characterize crop growth from reflectance data. Other measured traits included carbon isotope discrimination (CID) and chlorophyll content. In addition, we phenotyped key cotton fiber traits to evaluate the impact of canopy properties on traits that determine end-product quality and economic value of the crop. Two distinct cotton populations were evaluated, an upland (Gossypium hirsutum L.) recombinant inbred line (RIL) population and a Pima (G. barbadense L.) population composed of released cultivars from the last several decades. Both populations were grown under contrasting irrigation regimes, water-limited and well-watered conditions, for three years in a hot, arid environment in Arizona. Leaf thickness showed large genotypic differences in both populations, with high heritabilities, suggesting that thickness is under strong genetic control. Leaf thickness and NDVI were strongly correlated (maximum r = - 0.73) for the Pima population but not for the upland population. Additionally, leaf thickness relations with other physiological traits varied across years and irrigation regimes. Overall, our results support considering how simple crop morphological traits like leaf thickness affect canopy reflectance. This perspective should improve the efficiency of field-based phenotyping for crop improvement, leading to more rapid development of improved varieties, and ultimately benefiting both producers and consumers.

Technical Abstract: Field-based, high-throughput phenotyping (FB-HTP) methods are becoming more prevalent in plant genetics and breeding because they enable the evaluation of large numbers of genotypes under actual field conditions. Many systems for FB-HTP quantify and characterize the reflected radiation from the crop canopy to derive phenotypes as well as infer plant function and health status. However, despite decades of research on canopy reflectance, it remains difficult to uniquely characterize the biophysical and physiological attributes of germplasm from reflectance data alone, and furthermore, to interpret how these intrinsic properties may impact downstream interpretation of canopy reflectance data. Because of these uncertainties, we assessed relations between leaf thickness and several canopy-associated traits, including normalized difference vegetation index (NDVI), which was collected via active reflectance sensors carried on a mobile FB-HTP system, carbon isotope discrimination (CID), and chlorophyll content. In addition to these physiological traits, we phenotyped key cotton fiber traits to evaluate the impact of canopy properties on traits that determine end-product quality and economic value of the crop. To examine the associations among traits, we used two distinct cotton populations, an upland (Gossypium hirsutum L.) recombinant inbred line (RIL) population and a Pima (G. barbadense L.) population composed of released cultivars from the last several decades. Both populations were grown under contrasting irrigation regimes, water-limited (WL) and well-watered (WW) conditions, across three years in a hot, arid environment. We observed significant variation in leaf thickness among genotypes in both populations and high estimates of broad-sense heritability (averages of 0.74 and 0.79 for upland and Pima populations, respectively), suggesting that the majority of phenotypic variance for thickness was due to genetic effects. Leaf thickness and NDVI were strongly correlated (maximum r = - 0.73) for the Pima population but not for the upland population. Additionally, leaf thickness exhibited varying relationships with other physiological traits across years and within irrigation regimes in both populations. Overall, our results support the consideration of how basic crop morphological properties, such as leaf thickness, impact canopy reflectance and thus influence observed variation in traits like NDVI as quantified by current and developing FB-HTP methodologies.