Abstract

In this study, we aim to characterize synchronized global crop failures using remote sensing-based products, analyze their predictability and relationships with agroclimatic conditions using machine learning, and identify trends of the most influential agroclimatic indices revealed by machine learning over global croplands. We found that global synchronous crop failures showed strong interannual variability during 1982 to 2016. The most extreme global synchronous crop failure events occurred over 40% of global croplands in the years 2002 (rice and wheat) and 2012 (maize and soy), which had drier and warmer conditions compared to the normal years. Crop failure events can be accurately predicted using machine learning with agroclimatic indices. Of the four crops for both temperate and tropic regions, soy crop failure is most accurately predicted, with an Area Under the Curve (AUC) score of 0.8991 for the temperate region and 0.7892 for the tropics. The AUC score of maize failure in the temperate region is 0.8760, followed by wheat failure (0.8627) and rice failure (0.8025). In the tropics, the remaining crops performed similarly, with AUC scores of 0.7298 (maize), 0.7313 (rice), and 0.7337 (wheat). The machine learning model revealed that growing degree days, last spring frost, first fall frost, growing season precipitation, and optimal field conditions (represented by soil moisture) are the most influential agroclimatic indices, showing various nonlinear relationships with crop failure probabilities. The most influential agroclimatic indices present significant trends on more than 25% of global croplands, showing increasing growing degree days, earlier last spring frost, later first fall frost, while growing season precipitation and optimal field conditions are increasing. Our findings may inform food security predictions, selections of weather index for crop insurance, and climate adaptations.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.