What is digital agriculture?
Yeh: Digital agriculture focuses on interdisciplinary insights from statistics, computer science, crop and soil science, animal science, and economics. It draws from the technological aspects of computer science, artificial intelligence, and machine learning to help make farming more economically and environmentally sustainable over time.
How did you get involved with Cornell’s Nutrient Management Spear Program?
Walden: I'm from Long Island, so I didn’t grow up around agriculture, but I’ve always held an interest in gardening and sustainability. I discovered NMSP my sophomore year through an email listserv and applied for an open research position. I first worked with another honors student, Ben Lehman, and when COVID hit and fieldwork opportunities were limited, I switched over to working with Jason Cho, a graduate student working with farmer yield data, researching year-to-year variability in corn yields. After Jason graduated in 2021, I continued working with Data Analyst, Manuel Marcaida, on the evaluation of yield stability zones using various soil sampling approaches.
Yeh: It happened by chance! I grew up in the suburbs outside of NYC and had never considered doing anything remotely related to agriculture. But one day, I found a research assistant position being advertised via the Statistics department and I reached out. I began working on a project to clean up yield monitor data for corn silage and grain with Manuel Marcaida, and then joined the team of postdoctoral researcher Sunoj Shajahan, using machine learning to estimate yield from imagery. I became really interested in the impact that digital agriculture could have on society.
What project did you work on?
Walden: I just wrapped up my senior thesis project where I explored drivers of yield in corn fields. I looked at soil and landscape data to try to identify the most important factors in determining yield level. The goal was to figure out the main drivers of yield so that farmers can get insights as to how to manage their fields more efficiently, and therefore become more profitable and sustainable. In the future, I think it would be interesting to look at more factors that have the potential to affect yield, such as weather, microbial biomass, and nitrogen.
Yeh: Last spring, I started data cleaning for Manuel Marcaida, then in the summer I moved on to a new project working with drone images and corn yield prediction. The goal was to be able to make yield predictions, so that nitrogen fertilization management decisions can be improved. Farmers want to be strategic about when and where they apply nitrogen fertilizer, especially because over-application is expensive and can have tremendous environmental impacts. That's where digital agriculture comes in, knowing how and where to apply fertilizer using monitor systems and remote sensing.