Spatial Modeling of Sentinel Plants: UAV-Based Phenotyping in Cotton and Tomato

Nitrogen runoff from fertilized cropland fuels toxic algae blooms which contaminate surrounding water supplies. CEPD:RUBY tomato plants are a newly genetically engineered line of sentinel plants, also known as biosensors or reporter plants, which utilize the activation of the C-Terminally Encoded Peptide Downstream (CEPD) promoter under nitrogen stress. By inserting absorption pigments and fluorescent reporters after the CEPD promoter, CEPD:RUBY tomato releases a red pigment to its leaves in low nitrogen conditions, effectively communicating nitrogen status in the field. Coupled with remote sensing imagery from drone flight, sentinel plants enable field nitrogen mapping, which can reduce overfertilization and limit nitrogen runoff. We tested CEPD:RUBY for the minimum planting density that would provide useful nitrogen information for the entire field. Normalized Difference Vegetation Index (NDVI) from a 6 x 33 cotton dataset served as a proxy for the sentinel signal. Using the autoregressive parameters estimated from the fitted model, we constructed a spatial covariance matrix to simulate a 9 × 33 field. Sentinel densities between 4-30% were evaluated, and 900 iterations of random sampling were performed at each density. Finally, ordinary kriging was used to interpolate nitrogen values across the full field. Accounting for the possibility of a 15% error in sentinel reporting, the experiment found that 15% sentinel density is needed in order to achieve a 95% kriging success rate and R^2 > 0.5. Combined with future yield trials on sentinel plants, these results provide a benchmark density for the deployment of sentinel plants, aiming to reduce excessive nitrogen application and promote sustainability in agriculture.

I am grateful to the NSF-funded CROPPS program and BTI for the opportunity to work in Professor Kelly Robbins’ lab this summer. I would especially like to thank my amazing mentor Dinesh Ghimire. Dinesh gave me great advice when it came to abstract writing, poster designing, and so much more. Thanks to his mentorship, my research project felt like smooth sailing from the very beginning. In addition to having a positive mentor-mentee relationship, it was great to have a friend working with me in the lab.

My research project was coded entirely in the R programming language, which I had never used before this summer. Now that the program has come to an end, I am comfortable in my ability to use R for research. The first few weeks of the program taught me how to quickly pick up new skills, an ability that I am sure will come in handy in my future career. In addition to programming, I learned about other technologies used in computational plant breeding research. I watched Dinesh fly a drone equipped with state-of-the-art cameras for high-throughput phenotyping across a large field. I also witnessed firsthand how my project contributed to the forefront of plant research. One of my most memorable experiences this summer was planting the CEPD:RUBY tomatoes that I modelled in my project. Through fieldwork, I gained insight on the many components of conducting agricultural experiments and was reminded of the tangible, hands-on nature of research.

In the upcoming academic year, I will be a rising third-year computer engineering student at the University of California, Los Angeles (Go Bruins!). Engaging in plant research this summer revealed the beauty of an interdisciplinary education and the excitement of collaborating with researchers from other fields. Although I came in with a limited background in plant biology, it was fulfilling to apply my programming abilities to plant science because I could see the impact of my work beyond just the computer screen. After graduating with my bachelor’s, I intend to pursue a doctorate degree in electrical and computer engineering. Although I plan on staying in my current field of study, I would be excited to participate in a project that involves another field of science, such as plant biology, because the BTI/CROPPS REU program has shown me how enjoyable such an experience can be.