Pauli Lab

The University of Arizona (UA) Field Scanner is the worlds most advanced and complete phenotyping solution offering unparalleled insights into the plant lifecycle. The UA Field Scanner is a fully automated system that is capable of measuring crop growth and development efficiently, accurately, and quantitatively across the growing season. This longitudinal, continuous collection of high spatiotemporal resolution data enables the capture of deep phenotypes that can elucidate the mechanics and dynamics of plant development and physiology in relation to environmental conditions. To deliver these data, a suite of high-quality cameras and sensors are housed in a weather-proof cabinet that is moved over the field area by the supporting 30-ton robotic gantry. This provides accurate and consistent placement of the sensors relative to the crop canopy delivering robust and highly detailed data of unprecedented amount; two to six terabytes of data are capable of being generated per day. By coupling these cutting-edge sensors with newly developed, high-throughput data processing pipelines, traits such as ground cover, canopy height, plant geometry, growth and biomass, counting features, growth stages, vegetation indices, canopy composition, photosystem performance, and many more can be quantifiably described and recorded. Additionally, data fusion techniques are being used to integrate data collected from multiple sensors with different spatial and spectral resolutions to produced fused data sets that contain more information compared to the individual sensors. Through the integration and synthesis of these data streams, novel biological insight is being generated to help solve basic science questions as well as delivered applied solutions for the breeding and development of new crop cultivars and meeting the challenges facing a growing global population.

Phenotyping is one of the greatest limiting factors upon the rate of crop improvement. To ensure that agricultural production can increase in the face of a changing climate, it is imperative that researchers identify and implement more effective means of phenotyping plants. Emerging technologies, including remote sensing via unmanned aerial systems, or UAS, are attractive alternatives that could greatly increase the efficiency of crop improvement programs. Nonetheless, efforts must be made to determine how to implement these technologies and use the data that they generate. To that end, various high-throughput phenotyping projects are currently underway within the program. Rotary-wing UAS equipped with payloads including red, green, and blue (RGB) and thermographic, or infrared, sensors are currently in use, and these devices are being flown or will be flown over economically important crops including cotton (Gossypium hirsutum L.), sorghum (Sorghum bicolor L. Moench), and lettuce (Lactuca Sativa L.). We are using photogrammetry software and algorithms to process these data and extract phenotypes of interest. RGB data is being used to elucidate phenotypic characteristics such as canopy coverage, plant height, estimated yield, growth rate, and more. Thermographic data is being used to determine canopy temperature which is indicative of drought stress levels but also of pathogens such as fusarium wilt in lettuce. Through harnessing UAS imagery with quantitative genetics, we are working towards elucidating the genetic mechanisms that give rise to phenotypic plasticity.

Irrigation for agriculture accounts for 70% of human freshwater use. Without capacity to increase available water, this constraint poses the greatest challenge facing future food and natural resource security. Increases in plant productivity conferred by more effective use of water by plants is thereby needed. However, lack of understanding of the biophysical processes and genetic architecture of traits controlling whole-plant water movement (termed hydraulics) and their relationships and potential trade-offs to yield and quality limits development of more water efficient varieties that have the potential to reach market. Our research in the area of ecophysiological modeling addresses this need by linking genetics, phenomics and physiology via biophysical process-based models. We are leveraging expertise in field-based phenomics and quantitative genetics, genotype x environment modeling (Dr. Wang; Purdue Univ.), stress physiology (Dr.Ewers; Univ. of Wyoming), and biophysical modeling (Dr. Mackay; Univ. at Buffalo), targeting cotton (Gossypium hirsutum L.), a key species for plant-based fiber production. Our teams research aims to elucidate the genetic and physiological mechanisms that underlie their drought response by evaluating diverse germplasm under large-scale field experiments using high-throughput phenomics to develop and test biophysical plant models that simulate across different genotypes.