Understanding the impacts of climate change on agroecosystem and integrating satellite data for precision agriculture.

Feeding the growing global population is challenging, especially when there are increasing competitions for land and water to maintain other essential ecosystem services. Our research thus integrates ecology, computational modeling, remote sensing, and machine learning approaches to advance the science that guides sustainable agricultural management, and to develop tools that can help farmers and regulators apply this science more effectively.

With the rapid progress in earth observatory power and a range of modern tools, we’re particularly interested the following:

  1. Remote sensing and deep learning for agriculture management for mapping agriculture elements and patterns.
  2. Scalable quantification technology for high-resolution greenhouse gas (GHGs) fluxes.
  3. Process-based hybrid modeling and knowledge-guided machine learning for agroecosystem prediction.
  4. Understanding the impacts of climate change on agroecosystem.
  5. Controlled environment and urban agriculture.
A Nature Communications paper led by Licheng Liu: KGML can improve carbon cycle quantification in agroecosystems
A cashew mapping research led by Leikun Yin: Mapping smallholder tree crops in West Africa with spatiotemporal transferable domain adversarial deep learning
A digital-twin of maize led by Qi Yang and Junxiong Zhou
A Nature Climate Change paper led by Zhenong Jin: Critical benefits of snowpack for winter wheat are diminishing.