Through this program, we will build a farmer-centric program of providing Midwest farmers insights, tools, and guidance on farm-level management practices using NASA satellite data and domain sciences.

By the end of the 4th-year of the project, this project aims to enroll 100 farmers across the state of Illinois in our program: 30 farmers from Northern Illinois, 40 farmers from Central Illinois, and 30 farmers from Southern Illinois.  

We aim to develop this project as a example to be scaled up to the whole US Midwest and beyond of effectively using NASA satellite data for improving farming practices and US agricultural prosperity.

Objective 1: Forming a research collaborative relationship with selective farmers in Illinois through working with Illinois Corn Growers Association (ICGA).

For Year 1 of the project, we plan to recruit 20 farmers in central Illinois (~100-200 fields). We will work with ICGA to select farmers; with the selected farmers, we will develop data-sharing relationships and also provide our research output back to them.

Objective 2: Developing field-level crop stress and nitrogen measurements based on NASA satellite data.

We will scale up the STAIR fusion algorithm (fusing MODIS, Landsat, and Sentinel-2 data) to develop 10-meter, daily, cloud-/gap-free surface reflectance fusion images for all the farmers’ fields from 2016 to present. We will develop algorithms to link STAIR satellite fusion data, along with other NASA satellite data (e.g. thermal data from ECOSTRESS, SIF data from TROPOMI) with farmers’ data (e.g. management data, yield map) to measure field-scale crop nitrogen stress and water stress.

Objective 3: Integrating satellite-based measurements and process-based modeling to infer nitrogen fertilizer application at field scales.

We will use “ecosys” model to model the agroecosystems for all the individual fields (~100-200 fields), driven by data of weather, soil properties, and farm management (e.g. planting date, fertilizer applications). We will use the satellite-based measurements for aboveground crop features (e.g. LAI, aboveground N, phenology) to constrain model performance such that we can better infer belowground processes, including belowground root-zone soil moisture, nitrate contents. This critical information will be used to assess nitrogen fertilizer application rates for various application timing alternatives, thus provide farmers guidance on better manage their field-level nitrogen fertilizer applications.