loading page

Knowledge-based Artificial Intelligence for Agroecosystem Carbon Budget and Crop Yield Estimation
  • +10
  • Licheng LIU,
  • Wang Zhou,
  • Kaiyu Guan,
  • Bin Peng,
  • Chongya Jiang,
  • Jinyun Tang,
  • Sheng Wang,
  • Robert Grant,
  • Symon Mezbahuddin,
  • Xiaowei Jia,
  • Shaoming Xu,
  • Vipin Kumar,
  • Zhenong Jin
Licheng LIU
University of Minnesota Twin Cities

Corresponding Author:[email protected]

Author Profile
Wang Zhou
University of Illinois at Urbana Champaign
Author Profile
Kaiyu Guan
University of Illinois at Urbana Champaign
Author Profile
Bin Peng
University of Illinois at Urbana Champaign
Author Profile
Chongya Jiang
University of Illinois at Urbana Champaign
Author Profile
Jinyun Tang
Lawrence Berkeley Natl Lab
Author Profile
Sheng Wang
University of Illinois at Urbana Champaign
Robert Grant
University of Alberta
Symon Mezbahuddin
University of Alberta
Xiaowei Jia
University of Pittsburgh
Author Profile
Shaoming Xu
University of Minnesota Twin Cities
Author Profile
Vipin Kumar
University of Minnesota Twin Cities
Author Profile
Zhenong Jin
University of Minnesota-Twin Cities
Author Profile

Abstract

Improving the estimation of CO2 exchange between the atmosphere and terrestrial ecosystems is critical to reducing the large uncertainty in the global carbon budget. Large amounts of the atmospheric CO2 assimilated by plants return to the atmosphere by ecosystem respiration (Reco), including plant autotrophic respiration (Ra) and soil microbial heterotrophic respiration (Rh). However, Ra and Rh are challenging to be estimated at large regional scales because of the limited understanding of the complex interactions among physical, chemical, and biological processes and the resulting high spatio-temporal dynamics. Traditional approaches for estimating Reco including process-based (PB) models are limited by human knowledge resulting in limited accuracy and efficiency. Accumulation of the in situ observation of net ecosystem exchange (NEE), weather, and soil, and satellite data of GPP, LAI and soil moisture make it possible for applying data driven machine learning (ML) approaches. But the ML model approach has disadvantages of omission of domain knowledge and lack of interpretability. Here we propose a novel knowledge guided machine learning (KGML) method for predicting daily Ra and Rh in the US crop fields. With Gated Recurrent Unit (GRU) as the basis, we develop the KGML models constructing the hierarchical structure of ML with a mass balance constraint. The KGML models were pre-trained using synthetic data generated by an advanced agroecosystem model, ecosys, and re-trained with real-world FLUXNET observation data. We extrapolate the best KGML model to crop fields over the US with the help of satellite data, reanalysis climate forcings, and soil database to reveal the spatio-temporal variations and key controlling factors. We believe this study advances the interpretable machine learning concept for carbon cycle estimation and will shed light on many other process-based biogeochemistry research.